Fuel 118 (2014) 329–334

Contents lists available at ScienceDirect

Fuel

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

Combustion of vegetable oils under optimized conditions of atomizationand granulometry in a modified fuel oil burner

0016-2361/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.fuel.2013.11.009

⇑ Corresponding author. Tel.: +33 467 615 762; fax: +33 467 616 515.E-mail address: gilles.vaitilingom@cirad.fr (G. Vaitilingom).

Tizane Daho a, Gilles Vaitilingom b,⇑, Oumar Sanogo c, Salifou K. Ouiminga a, Augustin S. Zongo a,Bruno Piriou d, Jean Koulidiati a

a LPCE, département de physique, Université de Ouagadougou, 03 BP 7021 Ouagadougou, Burkina Fasob CIRAD, unité de recherche Biomasse-Energie, TA B-42/16, 73 rue JF Breton, 34398 Montpellier, Francec IRSAT, Centre National de la Recherche Scientifique et Technique, 03 BP 7047 Ouagadougou, Burkina Fasod INRA, UMR 1208 IATE, Supagro Montpellier-Université Montpellier II-CIRAD-INRA, 34060 Montpellier, France

h i g h l i g h t s

� Combustion of heating fuel oil and cottonseed oil in a modified burner is compared.� Atomization and granulometry optimization are necessary for vegetable oil combustion.� Riello 40N10 can achieve spray conditions and particle size recommended for burners.� Cottonseed oil must be sprayed at 28 bars and preheated up to 125 �C.� Non-condensable gases and most organic compounds emissions are close for both fuels.

a r t i c l e i n f o

Article history:Received 12 June 2011Received in revised form 4 November 2013Accepted 5 November 2013Available online 16 November 2013

Keywords:Vegetable oilAtomizationGranulometryBurnerCombustion

a b s t r a c t

The use of vegetable oils in burners represents an attractive alternative to the use of heating fuel oil (HFO)in heat production for domestic heating, small industrial units, drying of various products etc. In thiswork, a characterization of the combustion of cottonseed oil in a modified burner (type Riello 40N10)was performed to assess its ability to achieve proper combustion of vegetable oils in optimized condi-tions of atomization and granulometry. The quality of the combustion has been evaluated by the analysisof combustion products (CO, O2, CO2, NO, NO2, SO2) and organic compounds including PAHs. Results showthat the modifications made on the burner type 40N10 can achieve suitable spray conditions and giveparticle size within the recommended values for burners. In the case of Riello 40N10 burner, a fuel pres-sure of 28 bars is adequate and the minimum temperature required for oil preheating is 125 �C. Whenthese conditions are achieved, cottonseed oil combustion leads to the emission of non-condensable gasesand the organic compounds species as well as their concentration close to those of HFO.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The growing energy needs in developing and emergingcountries, even in rural areas, give more and more importance tobiofuels in general and to vegetable oils in particular. This trendis supported by the declining petroleum resources and the role thatbiofuels could play in reducing emissions of greenhouse gases. Atthe local scale, biofuels called ‘‘first generation Biofuels’’ mayappear as a real factor of development. As such, the use of purevegetable oils as alternative fuels to diesel oil or heating fuel oil(HFO) for local (domestic heating), as well as industrial use orapplications such as farming, electrification and drying of productscan prove to be interesting if used with some precautions.

Many studies on the characteristics of vegetable oils or theirderivatives and their use in diesel engines have been achieved overthe last four decades. These studies showed a behaviour close tothat of diesel oil [1–12].

However, diesel fuel, HFO and each vegetable oil or animal fathave their own characteristics and specific behaviour that distin-guish it from another. This is related to their specific physicaland chemical nature. In particular, the fatty acid composition, highviscosity and low volatility are key factors of differences in thebehaviour of vegetable oils [2]. This can lead to ignition problemsas well as coking of the colder parts of the combustion chamberdue to thermal decomposition and polymerization under certainconditions of temperature [13–16].

These differences in characteristics require specific operatingconditions of use for their proper combustion in diesel engines orburners. Indeed, the use of pure vegetable oils in standard

330 T. Daho et al. / Fuel 118 (2014) 329–334

domestic burners leads to cooking of cold parts of the burner(especially the air deflector) and draining of unburned fuel [17].So, most of the studies on the use of vegetable oils on burners havefocused on the use of preheated oil or pure biodiesel or their blendswith heating oil on standard domestic burners [18–26]. This use instandard domestic burners requires several adjustments [26]. Asshown by [22], blending of vegetable oil with a minimum of 30%HFO, under specific conditions, is necessary for a proper combus-tion in a standard domestic burner.

To overcome these constraints and limitations to the use of veg-etable oils in standard domestic burners, a modified burner forvegetable oils has been designed (derived from Riello 40 N seriesburner). This burner was set up specifically to allow the use of var-ious vegetable oils.

The first objective of this work was to assess the ability of thisburner to achieve the required spray conditions of atomizationand granulometry with a refined cottonseed oil (RCO). Then, underthese suitable conditions of atomization and granulometry, thesecond objective was to verify the quality of the combustion, atthe same thermal power output, of this refined cottonseed oilcompared to heating fuel oil by analysis of non-condensablecombustion products (CO, O2, CO2, NO, NO2, SO2), and of organiccompounds including PAHs.

2. Materials and Method

Tests were carried out in the CIRAD Biomass Energy laboratoryin Montpellier (France), with the collaboration of the ‘‘Laboratoirede Physique et de Chimie de l’Environnement’’ (Burkina Faso) andthe ‘‘Institut de Recherche en Sciences Appliquées et Technologies’’(Burkina Faso).

2.1. Equipment

Fig. 1 shows the schematic diagram of the experimental setup.It includes a modified burner, a combustion chamber and two de-vices for analyzing combustion products. The burner used is a Riel-lo burner (type 40N10) [27] with a preheating and a recirculationsystem of the fuel similar to that used for heavy fuel. Its main tech-nical characteristics are indicated in Table 1. A 1 kW electrical hea-ter and a recirculation system of the fuel (Fig. 2) are locateddownstream of the pump. The fuel can be heated up to tempera-tures from 50 �C to 200 �C.

When the burner is switched on, the temperature of the heaterincreases gradually up to the temperature fixed by the setpoint.

Fig. 1. Schematic diagram of experimental setup.

Once this temperature is reached, the pump allows the fuel flowingand recirculating through the preheating circuit (path 1, 4 and 5 inFig. 2). This leads to an increase of the temperature of the fuel inthe nozzle at a temperature very close to the setpoint. In these con-ditions, an electrovalve ‘‘VR’’ allows the spraying of the fuelthrough the nozzle and its return through path 3. The combustionoccurs in the combustion chamber which is a horizontal cylindricalsteel enclosure. Its dimensions are: length 120 cm and diameter60 cm. A cone with 8 openings of 10 cm diameter closes the cham-ber; it has a diameter restriction from 60 to 20 cm over a depth of22 cm. This chamber, designed for various uses, especially dryingof agricultural products, has the advantage of relatively high wallstemperature (temperatures above 500 �C). This is favourable toevaporation and complete combustion of unburnt vegetable oilsdroplets escaped from the spray.

The analyzing system of the combustion products comprisestwo devices: The first system is a combustion analyzer, the Kane-May Quintox KM9106, equipped with electrochemical cells. Thisallowed for real time measurement of the O2, NO, NOx, CO andSO2 level in dry flue gases. The CO2 content was calculated bythe analyzer depending on the characteristics of the fuel and thepercentage of residual oxygen measured in the emissions. Theraw gas stream passed through a blow-pipe and a duct heated toa temperature of 115 �C to avoid condensation and dissolution ofcertain gases (SO2, NOx). Gases were fed through a combustiongas conditioner (KaneMay KM9008), thus allowing for suddencooling of wet exhaust gases. Gas composition data from the ana-lyzer were then transferred to a computer using the KaneMay FIRE-WORKS software. The measuring ranges and the uncertainties ofthe cells are provided in Table 2.

The second device includes a cooling bath for trapping organiccompounds in the combustion products and a micro gas chromato-graph (micro-GC) as shown in Fig. 3. The collected gas bubble in aseries of bottles (which are immersed in the cooling bath) contain-ing isopropanol, and then move into the micro GC for the analysisof non condensable gases (O2, CO2, CO). The detection and thequantification of these gases are carried out by mass spectrometry.The organic compounds dissolved in isopropanol were analyzedwith an Agilent 6890 chromatograph and a mass spectrometer Agi-lent 5975. The characteristics of the column used are: type: DB-1701, length: 60 m and inner diameter: 250 lm. A mixture of1 ml of the sample and 1 ml of phenanthrene (internal standard,concentration: 25 mg/l) is injected into the chromatograph. Andthen, the identification and quantification of organic compoundsfollow.

2.2. Fuels used

The heating fuel oil (HFO) was taken as reference. To avoid apossible influence of minor compounds contained in crude vegeta-ble oils on the quality of the combustion, a commercial refined cot-tonseed oil (RCO) from Burkina Faso was used.

These fuels were used at different fuel pressures (16–28 bars)and following different primary air/secondary air ratios inpreliminary tests to determine at first the optimum operatingpoint based on lowest emissions in flue gases (an operating pointis defined by setting primary air, secondary air and fuel pressure).The optimum operating point corresponds to an equivalence ratioof 0.55 and a pressure of 28 bars. At this operating point, theconditions of atomization and the granulometry were evaluatedby varying the temperature of the vegetable oil from 50 to150 �C. So, only cottonseed oil temperature was varied when eval-uating the atomization conditions and fuels droplets size; fuelmass flow and fuel–air equivalence ratio were kept constant.HFO did not require any preheating. However, given the minimumtemperature of 50 �C, necessary for the operation of the burner, the

Table 1Technical characteristics of the Riello N10 burner and performance during tests.

Parameter Specification

Manufacturer Riello (Italy)Type 40N10 (equipped for heavy fuel use)Power output 34–102 kWAcceptable fuel

viscosity25–50 at 50 �C (mm2/s)

Fuel outlet pressure 16–28 barsNozzle Type HAGO, 1 GPH–angle 45� (B) CF JV

Orifice diameter: 0.15 mmPower output Consumption during tests

66 kW at 28 barpressure

6.85 kg/h refined cottonseed oil; 5.8 kg/h heatingfuel oil

Pre-heating of fuel 125 �C refined cottonseed oil; 50 �C heating fuel oil

T. Daho et al. / Fuel 118 (2014) 329–334 331

HFO was used at a temperature of 50 �C. Then, the temperatures of125 �C and 50 �C were respectively used for cottonseed oil andheating fuel oil for emissions comparison.

Finally, the combustion of cottonseed oil and HFO was per-formed and an analysis of combustion products was carried out.The results presented were obtained after several sets of measure-ments, at least five (5), such that the relative differences are lessthan 10%.

3. Results and discussion

3.1. Atomization conditions and granulometry of the fuel spray

To verify that the modifications applied on the Riello burnertype 40N10 provide good conditions for proper atomization andsuitable granulometry, the approach adopted was based on:

Fig. 2. Schematic diagr

Table 2Technical characteristics of the Quintox KM9106 gas analyzer.

Parameter Measuring range

Temperature of smokes 0–1100 �COxygen (O2) 0–25%Carbon oxide (CO) 0–10,000 ppm

Nitric oxide (NO) 0–5000 ppm

Nitrogen dioxide (NO2) 0–1000 ppm

Sulfur dioxide (SO2) 0–5000 ppm

Carbon dioxide 0–Fuel value

– the mechanism of disintegration of a liquid jet and classifi-cation of regimes of spray and the criteria for transitionbetween these regimes established by Reitz and Bracco[28] and Hiroyasu and Arai [29]

– determining the Sauter mean diameter (SMD) using Elkotbcorrelation [30] given by Eq. (1)

The Reynolds number (Rej) and Ohnesorge number (Ohj) of thejet, defined by Eqs. (2) and (3), allow in a first approximation tocharacterize the regime of disintegration of the fuel jet and charac-terize the transitions between the different regimes.

�D32 ¼ 6156 � m0:385fl � ðqfl � rflÞ0:737 � DP�0:54 � qa ð1Þ

Rej ¼qfl � Vj � dtr

lflð2Þ

Ohj ¼lflffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

qfl � rfl � dtr

q ð3Þ

qa denotes air density; vj, DP, dtr, respectively denote the jet veloc-ity, the difference between the fuel pressure in the injection lineand the pressure in the combustion chamber, the diameter of thenozzle orifice; qfl, lfl, mfl, rfl are respectively fuel density, dynamicviscosity, kinematic viscosity and surface tension (RCO oil or HFO).

The primary atomization depends largely on the physical andchemical characteristics of the fuel (viscosity, surface tension anddensity of the fuel). This is the same for the Sauter mean diameter.Thus, the physical and chemical characteristics of the refined cot-tonseed oil and those of HFO have been determined and the resultsare reported in Table 3.

am of fuel circuit.

Precision Resolution

1.0 ±0.3% of reading 0.1 �C�0.1% + 2% 0.1%±20 ppm < (400 ppm) 1 ppm5% of reading (<2000 ppm)±10% of reading (>2000 ppm)±5% of reading (>100 ppm) 1 ppm±5 ppm (<100 ppm)±5% of reading (>100 ppm) 1 ppm±5 ppm (<100 ppm)±5% of reading (>100 ppm) 1 ppm±5 ppm (<100 ppm)±0.3% 0.1%

Fig. 3. Schematic diagram of analyzing combustion products by micro-GC.

Table 3Physical and chemical characteristics of RCO and HFO.

Parameter RCO HFO Method/Norm

C (%) 77.39 85.10 CombustionH (%) 11.90 14.90 CombustionO (%) 10.10 0 PyrolysisN (%) <0.30 <0.30 CombustionS (%) 0 <0.20 NFT 60 106Viscosity at 40 �C (mm2/s) 35.7 3.3 NF EN ISO 3104Water content (ppm) 656 96 NF EN ISO 12937Acid value (mg KOH/g) 0.17 – ASTM D 664Density at 15 �C (kg/m3) 929 854 NF EN ISO 12185Sediments (ppm) 23 – NF EN 12662Phosphorus (ppm) 0.2 – NFT 60 106Magnesium (ppm) 0.2 – NFT 60 106Calcium (ppm) 2.1 – NFT 60 106LHV (kJ/kg) 36000 42583 ASTM D 240Surface tension at 25 �C (m N/m) 32.4 28.6 NF EN 14 370Flash point (�C) 229.5 62.5 NF EN ISO 2719

332 T. Daho et al. / Fuel 118 (2014) 329–334

The Reynolds number (Rej) and Ohnesorge number (Ohj)presented in Table 4, show that atomization begins at the nozzleoutlet giving a ‘‘complete spray’’.

Conditions of atomization of cottonseed oil for different oil tem-peratures and for the operating point defined previously on theRiello 40N10 burner are shown in Fig. 4. This figure shows thatthe heating of cottonseed oil and the relatively high fuel pressurelead to a ‘‘complete spray’’ for most of the heating temperatures.Only the temperature of 50 �C leads to a regime close to thetransition zone between the regimes of ‘‘complete spray’’ and‘‘incomplete spray’’.

The increase in temperature of RCO leads to a reduction in itsviscosity, density and surface tension. Consequently, there is a

Table 4Reynolds number and Ohnesorge number for RCO in the Riello 40N10 burner.

Temperature (�C) Viscosity (mm2/s) Surface tensio

50 23.9 0.030985 10.1 0.029095 8.45 0.0284

105 7.2 0.0278115 6.3 0.0272125 5.75 0.0267150 4.8 0.0252

decrease of Ohnesorge number and an increase of Reynolds num-ber. This leads to better conditions of atomization of RCO jet veryclose to those of HFO maintained at a temperature of 50 �C.

Therefore, the use of viscous fuel, such as vegetable oils, in thistype of burner and for specific conditions provides good conditionsfor atomization.

Regarding the granulometry of the ‘‘spray’’ of RCO, it can beclose to the values recommended for burners if RCO is heated upto a temperature of 125 �C (Fig. 5). Indeed, spray granulometryfor burners must be in the range of 50–100 lm with a dispersionsuch as the droplets of larger diameter (200–300 lm) represent asmall fraction of the mass of the jet [31,32].

Thus, these results allow defining a minimum preheating tem-perature of RCO to obtain both a ‘‘atomized spray’’ and a granulom-etry suitable for this type of burner. As this minimum temperatureis defined, the combustion quality of RCO can be evaluated. Thecombustion tests are thus performed at the operating point consid-ered for the study of atomization conditions with heating of RCO atthe minimum temperature of 125 �C. All fuels were used at thesame atomization pressure of 28 bars.

3.2. Emissions

3.2.1. Non-condensable gasesThe results of analysis of non condensable gases are shown in

Table 5. These results were obtained with the QUINTOX and thenvalidated with the micro-GC.

For all measured gases during the combustion of RCO, the con-centrations are very close to those measured for the HFO. The re-sults are in agreement with those obtained in a previous studywith rapeseed oil [17].

The CO values obtained are of the same order of magnitude asthe uncertainties of the instrument (±20 ppm for CO) so that we

n (N/m) Reynolds number Ohnesorge number

1062 0.3432521 0.1482767 0.1253089 0.1073356 0.0943561 0.0874193 0.074

0.00

0.01

0.10

1.00

10.00

1 10 100 1000 10000 100000

Ohn

esor

ge n

umbe

r

Reynolds number

RaM-FWIFWI-SWISWI-atomizationRCO - 50 CRCO - 85 CRCO - 95 CRCO - 105 CRCO - 115 CRCO - 125 CRCO - 150 CHFO - 50 C

RaM

FWISWI

Atomisation

°°°

°°°°

°

Fig. 4. Atomization conditions on the modified burner 40N10 (RCO: refinedcottonseed oil; HFO: heating fuel oil; RaM: Rayleigh Mechanism zone; FWI: FirstWind Interaction zone; SWI: Second Wind Interaction zone).

0

50

100

150

200

250

50 85 95 105 115 125 150

Saut

er m

ean

diam

eter

(µm

)

Fuel temperature (°C)

RCO HFO

Fig. 5. Sauter mean diameter of HFO and RCO versus fuel temperature.

Table 5Emissions of non-condensable gases.

O2

(%)CO2

(%)CO(ppm)

NO(ppm)

NOx

(ppm)SO2

(ppm)

RCO 10.2 8.1 13 84 88 0HFO 8.5 9.3 0 75 78 45

T. Daho et al. / Fuel 118 (2014) 329–334 333

cannot conclude to a difference in CO emission between RCO orHFO. Indeed, the granulometry (at the heating temperature consid-ered), the atomization conditions of RCO and the temperature con-ditions in the combustion chamber are sufficient for completeevaporation of the spray and the complete oxidation of CO–CO2.In fact, evaporation of preheated vegetable oil droplets is better

Table 6Concentration of organic compounds: light hydrocarbons and PAHs species.

Compound

Pyridine, 2 -picolinep-xylene; o-xylene; 2,3-benzofuran; Indene; 2-metoxy-phenol. (gaiacol), o-cresol; 2,43-picoline; 4-picoline; p-cresol; 2-metoxy-4-methyl-phenol; 2-me-Naphtalene; Isoqu

Fluorene; Alpha-naphtol; Beta-naphtol; Phenanthrene; Anthracene; Fluoranthene;Benzo(k)fluoranthene; Benzo(a)pyrene; Dibenzo(ah)anthracene; Indeno(123 cd)py

phenolNaphthalenem-cresol

when the oil temperature is relatively high. Indeed, the latent heatof vaporization of a liquid decreases rapidly with increasing tem-perature [33]. The preheating of cottonseed oil could lead to short-er evaporation time of oil droplets which is similar to what isobserved when a vegetable oil droplet is introduced into anincreasing temperature atmosphere [34]. This could be favourableto oil combustion and explain the relatively low levels of CO ob-served. In addition, the experimental combustion chamber hasno boiler to evacuate heat produced by the combustion. There isno convective transfer between the walls of the combustion cham-ber and water. Convective transfer occurs only between the wallsof the combustion chamber and ambient air for which the heattransfer coefficient is lower than that of water. This allows thechamber walls having a relatively high average temperature. Thismay results in a significant increase in radiation phenomena inthe combustion chamber. Such phenomena are favourable to thevaporization and combustion of vegetable oils. This positive effectof the increase in temperature of the combustion chamber on theability of vegetable oils to evaporate and burn properly has beenobserved in previous studies [34,35]. Results obtained with a con-ventional burner show that vegetable oils have a behaviour close tothat of HFO for adequate temperature conditions (high thermalpower) [22]. These observations (similar behaviour at high temper-atures) have been carried out during tests on engines or in tests ofevaporation of droplets of vegetable oils or their derivatives [2,34–36]. It may be noted the same result between RCO and HFO interms of NOx emissions. There is no fundamental difference be-tween the NOx emissions levels of the two fuels.

CO2 emissions for RCO are slightly lower than those of HFO. Thiscould be partly due to the H/C ratio higher for vegetable oils rela-tive to HFO. SO2 emissions are zero for RCO in contrast to HFO. In-deed, SO2 emissions are proportional to the sulfur content of thefuel (Table 3).

3.2.2. Organic compoundsThe results of chromatographic analysis of soluble organic com-

pounds contained in the products of combustion are presented inTable 6. Among these organic compounds, those that are the sub-ject of special attention are the Polycyclic Aromatic Hydrocarbons(PAHs) for their carcinogenicity. Although PAHs are subject toattention, mainly the Benzo (a) pyrene is the subject of a standardin some regions. It is usually used as PAHs indicator.

The results of PAHs obtained in this study are for overall, and forboth fuels, below the detection limit of the measuring device ex-cept for naphthalene (Table 6). It is likely that the values of PAHsare very low with pure vegetable oil, a similar observation has beendone by the study conducted by [21] who found values of 30 ng/N m3 for biodiesel with Riello burners. Regarding Naphthalene, itis classified possibly carcinogenic to humans, even if it is lighterand less toxic than the heavy PAHs [37]. These emissions of naph-thalene could be due to lean combustion (equivalence ratio of0.55). Indeed, emissions of this compound are performed substan-

Concentration(mg/m3 gas)

HFO RCO

<2.33 <2.33-dimethylphenol; Quinoline; 3,4 Dimethylphenol; <0.02 <0.02inoline; 1-Methylnaphthalene; Acenaphtylene; Acenaphtene;Pyrene; Benzo(a)anthracene; Benzo(b)fluoranthene;

rene; Benzo(ghi)perylene; Chrysene

<0.23 <0.23

0.23 0.220.50 0.710.08 <0.02

334 T. Daho et al. / Fuel 118 (2014) 329–334

tially under lean combustion conditions (for equivalence ratio be-low 0.6) [37,38]. Values are slightly higher for the RCO than forHFO. The unsaturation of vegetable oils such as RCO may promotethe formation of PAHs [39].

4. Conclusion

This study revealed the possibility of using pure cottonseed oilin an adapted burner which sustained appropriate adjustments.The Riello 40N10 can achieve proper atomization conditions anda suitable spray granulometry close to the values recommendedfor burners. In the case of this burner, the proper fuel pressure is28 bars and the minimum temperature for preheating cottonseedoil is 125 �C.

When the conditions for proper atomization are carried out andwhen the granulometry is suitable, RCO and HFO have very closecombustion characteristics. For the temperatures reached in thecombustion chamber used in this study, there is no difference be-tween the two fuels in term of non-condensable gases at the ex-haust as well as in term of organic compounds. The results ofanalysis of PAHs are for overall below the detection limit of themeasuring device except for naphthalene for both fuels. Morestudy must be carried out to accurately quantify PAHs and to verifythe influence of lean combustion conditions on PAHs emission.This highlight the need to use specific means for detecting PAHswhich thresholds were too low to be measured by the equipmentused in the current study.

References

[1] Pryde EH. Vegetable oils as fuels alternatives. Symp overview JAOCS1984;61:1609–10.

[2] Vaitilingom G. Huiles végétales – biocombustibles diesel: influence de lanature des huiles et en particulier de leur composition en acides gras sur laqualité-carburant. Ph.D. Thesis, France: University of Orleans; 1992.

[3] Murayama T. Evaluating vegetable oils as a diesel fuel. Inform1994;5:1138–45.

[4] Altin R, Çetinkaya S, Yücesu HS. The potential of using vegetable oil fuels asfuel for diesel engines. Energy Convers Manage 2001;42(5):529–38.

[5] Kalam MA, Husnawan M, Masjuki HH. Exhaust emission and combustionevaluation of coconut oil-powered indirect injection diesel engine. RenewEnergy 2003;28:2405–15.

[6] Rakopoulos CD, Antonopoulos KA, Rakopoulos DC, Hountalas DT, GiakoumisEG. Comparative performance and emissions study of direct injection dieselengine using blends of diesel fuel with vegetable oils or biodiesels of variousorigins. Energy Convers Manage 2006;47:3272–87.

[7] Wang YD, Al-Shemmeri T, Eames P, Mcmullan J, Hewitt N, Huang Y, et al. Anexperimental investigation of the performance and gaseous exhaust emissionsof a diesel engine using blends of a vegetable oil. Appl Therm Eng 2006;26(14–15):1684–91.

[8] S�ehmus A, Hüsamettin B, Cengiz Ö. The comparison of engine performance andexhaust emission characteristics of sesame oil–diesel fuel mixture with dieselfuel in a direct injection diesel engine. Renew Energy 2008;33:1791–5.

[9] Agarwal AK, Rajamanoharan K. Experimental investigations of performanceand emissions of karanja oil and its blends in a single cylinder agriculturaldiesel engine. Appl Energy 2009;86:106–12.

[10] Rakopoulos DC, Rakopoulos CD, Giakoumis EG, Dimaratos AM, Founti MA.Comparative environmental behavior of bus engine operating on blends ofdiesel fuel with four straight vegetable oils of Greek origin: sunflower,cottonseed, corn and olive. Fuel 2011;90:3439–46.

[11] Leenus MJM, Varuvel EG, Singh DKJ, Nagalingam B. A comparative analysis ofdifferent methods to improve the performance of cotton seed oil fuelled dieselengine. Fuel 2012;102:372–8.

[12] Daho Tizane, Vaitilingom Gilles, Ouiminga SalifouK, Piriou Bruno, ZongoAugustinS, Ouoba Samuel, et al. Influence of engine load and fuel droplet size

on performance of a CI engine fueled with cottonseed oil and its blends withdiesel fuel. Appl Energy 2013;111:1046–53.

[13] Korus RA, Mousetis TL, Lloyd L. Polymerization of vegetable oils. In:Proceedings of the international conference on plant and vegetables oils asfuels, vols. 4–82. Fargo (USA): ASAE, Publication August 1982. p. 218–23.

[14] Srivastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sustain EnergyRev 2000;4:111–33.

[15] Kalam MA, Masjuki HH. Emissions and deposit characteristics of a small dieselengine when operated on preheated crude palm oil. Biomass Bioenergy2004;27:289–97.

[16] Ramadhas AS, Jayaraj S, Muraleedharan C. Characterization and effect of usingrubber seed oil as fuel in the compression ignition engines. Renew Energy2005;30:795–803.

[17] Vaitilingom G, Perilhon C, Liennard A, Gandon M. Development of rape seed oilburners for drying and heating. Ind Crops Prod 1998;7(2–3):273–9.

[18] Holt GA, Hooker JD. Gaseous emissions from burning diesel, crude and primebleachable summer yellow cottonseed oil in a burner for drying seed-cotton.Bioresource Technol 2004;92:261–7.

[19] San José Alonso J, López Sastre JA, Romero-Ávila C, López Romero EJ.Combustion of rapeseed oil and diesel oil mixtures for use in the productionof heat energy. Fuel Process Technol 2006;87:97–102.

[20] San José Alonso J, López Sastre JA, Romero-Ávila C, López Romero E. A note onthe combustion of blends of diesel and soya, sunflower and rapeseed vegetableoils in a light boiler. Biomass Bioenergy 2008;32:880–6.

[21] Macor A, Pavanello P. Performance and emissions of biodiesel in a boiler forresidential heating. Energy 2009;34:2025–32.

[22] Daho T, Vaitilingom G, Sanogo O. Optimization of the combustion of blends ofdomestic fuel oil and cottonseed oil in a non-modified domestic boiler. Fuel2009;88(7):1261–8.

[23] San José J, Al-Kassir A, López Sastre JAy, Gañán J. Analysis of biodieselcombustion in a boiler with a pressure operated mechanical pulverisationburner. Fuel Process Technol 2011;92:271–7.

[24] Bazooyar B, Ghorbani A, Shariati A. Combustion performance and emissions ofpetrodiesel and biodiesels based on various vegetable oils in a semi industrialboiler. Fuel 2011;90:3078–92.

[25] Ghorbani A, Bazooyar B, Shariati A, Jokar SM, Ajami H, Naderi A. Comparativestudy of combustion performance and emission of biodiesel blends and dieselin an experimental boiler. Appl Energy 2011;88:4725–32.

[26] San José Alonso J, Romero-Ávila C, San José Hernández LM, Al-Kassir A.Characterising biofuels and selecting the most appropriate burner for theircombustion. Fuel Process Technol 2012;103:39–44.

[27] http://www.rielloburners-france.com, or: http://new.rielloburners.com.[28] Reitz RD, Bracco FB. On the dependence of spray angle and other spray

parameters on nozzle design and operating conditions. SAE paper; 1979. p.790494.

[29] Hiroyasu H, Arai M. Structures of fuel sprays in diesel engines. SAE paper;1990. p. 900475.

[30] Elkotb MM. Fuel atomization for spray modelling. Prog Energy Combust Sci1982;8(1):61–91.

[31] Monnot G, Feugier A, Le Bouc F, Mauss F, Meyer C, Perthuis E, et al. Lacombustion des les fours et les chaudières. Paris: Ed. Technip; 1978.

[32] Borghi R, Destriau M. La combustion et les flammes. Paris: Ed. Technip; 1995.[33] Watson KM. Prediction of critical temperatures and heats of vaporization. Ind

Eng Chem 1931;23:360–4.[34] Daho T, Vaitilingom G, Sanogo O, Ouiminga SK, Segda BG, Valette J, et al. Study

of droplet vaporization of various vegetable oils and blends of domestic fueloil–cottonseed oil under different ambient temperature conditions. BiomassBioenergy 2012;46:653–63.

[35] Daho T, Vaitilingom G, Sanogo O, Ouiminga SK, Segda BG, Valette J, et al. Modelfor predicting evaporation characteristics of vegetable oils droplets based ontheir fatty acid composition. Int J Heat Mass Trans 2012;55:2864–71.

[36] Morin C, Chauveau C, Gökalp I. Droplet vaporisation characteristics ofvegetable oil derived biofuels at high temperatures. Exp Thermal Fluid Sci2000;21(1–3):41–50.

[37] Yilmaz N, Donaldson AB. Evidence of PAH production under lean combustionconditions. Fuel 2007;86(15):2377–82.

[38] Atal A, Levendis YA, Carlson J, Dunayevskiy Y, Vouros P. On the survivabilityand pyrosynthesis of PAH during combustion of pulverized coal and tirecrumb. Combust Flame 1997;110(4):462–78.

[39] Morin C. Vaporisation et oxydation à haute température et haute pression degouttes de combustibles liquides, application aux n-alcanes et estersméthyliques d’huiles végétales. Ph.D. Thesis, France: University of Orleans;1999.