Effect of the Degree of Unsaturation of Biodiesel ... - ACS Publications

Dec 6, 2010 - To gain further insight into the effect of the biodiesel chemical structure, specifically its degree of unsaturation, on engine performa...
0 downloads 3 Views 819KB Size
Energy Fuels 2011, 25, 77–85 Published on Web 12/06/2010

: DOI:10.1021/ef101096x

Effect of the Degree of Unsaturation of Biodiesel Fuels on Engine Performance, Combustion Characteristics, and Emissions Pedro Benjumea,*,† John R. Agudelo,‡ and Andres F. Agudelo‡ †

Faculty of Mines, Universidad Nacional de Colombia, Medellı´n AA 1027, Colombia, and ‡Faculty of Engineering, Universidad de Antioquia, Medellı´n AA 1226, Colombia Received August 16, 2010. Revised Manuscript Received November 18, 2010

Biodiesel fuels derived from different feedstocks may have significantly different fatty acid profiles and physicochemical properties. To gain further insight into the effect of the biodiesel chemical structure, specifically its degree of unsaturation, on engine performance, combustion characteristics, and emissions, an experimental investigation was conducted on a high-speed direct-injection automotive diesel engine fueled with three mixtures of fatty acid methyl esters. The fuel matrix was designed such that the effect of the degree of unsaturation of the tested biodiesel fuels was isolated. This allowed for the maximization of the effect of the cetane number, while the other properties, such as the chain length, oxygen content, density, viscosity, and volatility, varied within a small range. Results indicated that the degree of unsaturation of biodiesel fuels did not significantly affect engine performance and the start of injection, but it had a noticeable influence on combustion characteristics and emissions, via its effect on the cetane number. A higher degree of unsaturation of biodiesel fuels led to a longer ignition delay and, consequently, a more retarded start of combustion. Regardless of the engine-operating mode, an almost fixed start of injection was attained, while the premixed portion of combustion, peak heat release rate, maximum pressure gradient, peak in-cylinder bulk-gas-averaged temperature, total hydrocarbon (THC) emissions, smoke opacity, and nitrogen oxides (NOx) emissions increased with the degree of unsaturation.

disparity reported in measured exhaust NOx emissions may be partly a consequence of the variability in the experiments.2-4 Experimental tests may differ in engine technology, engine speed and load, ambient conditions, principle and resolution of the measurement apparatuses and sensors, etc. A simple systematic issue, often not sufficiently weighted, is the nature of the tested fuels: the specific biodiesel fuel used and the diesel fuel used either as a comparison base or for blend preparation. The term biodiesel fuels is intentionally used in this paper to highlight that biodiesel is not a unique compound. From a chemical standpoint, a biodiesel fuel is a mixture of monoalkyl esters of fatty acids. An essential feature of a biodiesel fuel is that its fatty acid profile corresponds to that of its parent oil or fat. Thus, biodiesel fuels derived from different feedstocks may have significantly different compositions and properties. The structural features that influence the physicochemical properties of a fatty acid alkyl ester molecule are the chain length and branching of the alkyl group corresponding to the monohydric alcohol used in the transesterification or esterification reactions, the length, branching, and number of double bonds of the fatty acid chain, the position and geometric configuration (cis or trans) of the double bonds, and the presence of additional functional groups in the fatty acid chain. Most commercial biodiesel fuels are mainly composed of natural medium- to long-chain (C16-C18) fatty acid methyl

1. Introduction Biodiesel fuels have inherent characteristics that allow for their use in modern diesel engines without introducing significant modifications to their design and calibration. This fact has been a key driving force for the deployment of biodiesel use in the transportation sector. However, differences in the chemical nature between petroleum diesel fuels and biodiesel fuels lead to differences in their physicochemical properties, which ultimately affect fuel quality, engine performance, efficiency, combustion characteristics, and emissions. Numerous experimental and theoretical research studies have been carried out to compare engine performance, combustion characteristics, and emissions when pure or blended biodiesel fuels are used instead of conventional petroleum diesel fuels. Comprehensive reviews analyzing the results of most representative studies have been presented by Graboski and McCormick in 19981 and Lapuerta et al. in 2008.2 With regard to engine emissions, there is a wide consensus that biodiesel fueling tends to decrease significantly the emissions of particulate matter (PM), total hydrocarbons (THCs), and carbon monoxide (CO). In the case of nitrogen oxides (NOx), although the highest consensus lies in a slight increase of NOx emissions with biodiesel fueling, some contradictory trends have been reported. It has been argued that much of the *To whom correspondence should be addressed. E-mail: pbenjume@ unal.edu.co. (1) Graboski, M. S.; McCormick, R. L. Combustion of fat and vegetable oil derived fuels in diesel engines. Prog. Energy Combust. Sci. 1998, 24, 125–164. (2) Lapuerta, M.; Armas, O.; Rodrı´ guez-Fernandez, J. Effect of biodiesel fuels on diesel engine emissions. Prog. Energy Combust. Sci. 2008, 34, 198–223. r 2010 American Chemical Society

(3) Sun, J; Caton, J. A.; Jacobs, T. J. Oxides of nitrogen emissions from biodiesel-fuelled diesel engines. Prog. Energy Combust. Sci. 2010, 36, 677–695. (4) Assessment and Standards Division, Office of Transportation and Air Quality, United States Environmental Protection Agency (U.S. EPA). A comprehensive analysis of biodiesel impacts on exhaust emissions. Draft Technical Report EPA420-P-02-001; U.S. EPA: Washington, D.C., 2002.

77

pubs.acs.org/EF

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

esters, for which the main structural difference is consequently the number of double bonds. The degree of unsaturation of an alkyl ester molecule is an indicator of the number of double bonds present in its fatty acid chain, with a higher number of double bonds representing a higher degree of unsaturation. A strong correlation between the degree of unsaturation and the properties of alkyl esters has been reported.5,6 While the heating value, melting point, cetane number, viscosity, and oxidation stability decrease, density, bulk modulus, fuel lubricity, and iodine value increase with the degree of unsaturation. It has been argued that it is possible to design or engineer fatty acid profiles to optimize biodiesel fuel properties.6,7 As the interest for widening the possible raw materials for biodiesel production has increased worldwide and the research community has realized the significance of the effect of the molecular structure on biodiesel fuel properties, the number of research studies trying to explain the consequent effect of the structural features of the fatty acid alkyl esters, mainly chain length and degree of unsaturation, on engine performance, combustion characteristics, and emissions has grown. Results are not conclusive yet and, in some cases, are contradictory. McCormick et al.8 tested several pure fatty acid methyl esters and 21 biodiesel fuels derived from different raw materials in a diesel engine following the U.S. heavy-duty federal test procedure, to investigate the effect of the length and number of double bonds of the fatty acid chain on NOx and PM emissions. A direct correlation between the number of double bonds, quantified as the iodine value, and NOx emissions was found, while PM emissions remained practically unchanged. On the other hand, for fully saturated fatty acid chains (C12:0, C16:0, and C18:0), NOx emissions increased with a decreasing chain length, while PM emissions did not show significant variations. Tat et al.9 compared the NOx emissions from a John Deere 4045T direct-injection diesel engine fueled with regular soybean oil biodiesel (high content of polyunsaturated fatty acid methyl esters) and a high-oleic soybean biodiesel (high content of monounsaturated fatty acid methyl esters), produced from a genetically modified oil. Results showed a significant reduction in NOx emissions using the high-oleic soybean biodiesel. With regard to unburned hydrocarbon and smoke emissions, no significant differences were found. Lapuerta et al.10 carried out an experimental study to determine the effect of the degree of unsaturation of biodiesel fuels on NOx and PM emissions from a direct-injection diesel engine typical of those used in European cars. Four biodiesel fuels produced from different feedstocks and having iodine values ranging from 90 to 125 were tested. Results showed that

the degree of unsaturation of biodiesel fuels had significant effects on the NOx and PM emissions. As the biodiesel fuel became more unsaturated, NOx emissions increased and PM emissions decreased. Puhan et al.11 tested three types of biodiesel fuels with different molecular weights and degrees of unsaturation, to study the effect of the molecular structure on the combustion characteristics and emissions of a single-cylinder directinjection diesel engine. Biodiesel fuels were prepared from linseed, jatropha, and coconut oils. More unsaturated biodiesel fuels exhibited higher advancements in fuel injection timing, ignition delay (ID), heat release rate (HRR), and maximum gas pressure and exhaust gas temperature. Emissions of NOx, smoke, THCs, and CO also increased with the degree of unsaturation. Results also showed that the use of linseed oil biodiesel as a sole fuel in the diesel engine tested should not be recommended because its fueling led to exaggerated high NOx emissions and low thermal efficiency. Sch€ onborn et al.12 conducted a series of experiments on a single-cylinder research engine, investigating the influence of the molecular structure on the combustion behavior of fatty acid alkyl ester molecules under diesel engine conditions. The tested fuels comprised eight samples of pure individual fatty acid alkyl ester molecules of different structure, as well as four biodiesel fuels produced by the transesterification of rapeseed oil, palm oil, jatropha oil, and tallow. All fuel samples were subjected to three series of experiments characterized by keeping the injection timing, ignition timing, or ID constant. It was observed that the molecular structure of the fuel significantly influenced the formation of NOx and PM and their respective concentrations in the exhaust gases. The influence on the formation of NOx in particular appeared to be exerted first through the effect that the molecular structure had on the ID and second through the flame temperature at which the various molecules were burned. The emission of PM, on the other hand, showed a correlation with the number of double bonds in the fuel molecules for the case of larger, accumulation-mode particles and with the boiling point of the fuel samples for the case of the smaller, nucleation-mode particles. With regard to the effect of pure alkyl esters on autoignition and combustion processes, Zhang et al.13 conducted an experimental study on the premixed ignition behavior of four C9 fatty acid esters in a motored cooperative fuel research (CFR) engine. To gain insight into the low-temperature oxidation process of biodiesel-relevant compounds, the engine exhaust was sampled and analyzed at various compression ratios. Combustion analysis showed that the tested fuels evidently exhibited different ignition behaviors. The magnitude of low-temperature heat release (LTHR) followed the order: ethyl nonanoate > methyl nonanoate > methyl 2-nonenoate > methyl 3-nonenoate. The lower oxidation reactivity for the olefinic methyl esters in the low-temperature regime was explained by the reduced amount of six- or sevenmembered transition-state rings formed during the oxidation of the unsaturated esters because of the presence of a double

(5) Knothe, G. Dependence of biodiesel fuel properties on the structure of fatty acid alkyl esters. Fuel Process. Technol. 2005, 86, 1059–1070. (6) Refaat, A. A. Correlation between the chemical structure of biodiesel and its physical properties. Int. J. Environ. Sci. Technol. 2009, 6, 677–694. (7) Knothe, G. “Designer” biodiesel: Optimizing fatty ester composition to improve fuel properties. Energy Fuels 2008, 22, 1358–1364. (8) McCormick, R. L.; Graboski, M. S.; Alleman, T. L.; Herring, A. M.; Tyson, K. S. Impact of biodiesel source material and chemical structure on emissions of criteria pollutants from a heavy-duty engine. Environ. Sci. Technol. 2001, 35, 1742–1747. (9) Tat, M. E.; Wang, P. S.; Van Gerpen, J. H.; Clemente, T. E. Exhaust emissions from an engine fueled with biodiesel from high-oleic soybeans. J. Am. Oil Chem. Soc. 2007, 84, 865–869. (10) Lapuerta, M.; Armas, O.; Rodrı´ guez-Fernandez, J. Effect of the degree of unsaturation of biodiesel fuels on NOx and particulate emissions. SAE Int. J. Fuels Lubr. 2009, 1, 1150–1158.

(11) Puhan, S.; Saravanan, N.; Nagarajan, G.; Vedaraman, N. Effect of biodiesel unsaturated fatty acid on combustion characteristics of a DI compression ignition engine. Biomass Bioenergy 2010, 34, 1079–1088. (12) Sch€ onborn, A.; Ladonmmatos, N.; Williams, J.; Allan, R.; Rogerson, J. The influence of molecular structure of fatty acid monoalkyl esters on diesel combustion. Combust. Flame 2009, 156, 1396–1412. (13) Zhang, Y.; Yang, Y.; Boehman, A. Premixed ignition behavior of C9 fatty acid esters: A motored engine study. Combust. Flame 2009, 156, 1202–1213.

78

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

Table 1. Chemical Composition of Test Fuels (% Mass) type of methyl ester lauric miristic palmitic stearic palmitoleic oleic linoleic linolenic total saturated total unsaturated

12:0 14:0 16:0 18:0 16:1 18:1 18:2 18:3

Table 2. Test Engine Characteristics and Operating Conditions

BP100

BPL50

BL100

0.31 1.03 43.3 4.20 0.15 41.8 9.10 0.15 48.8 51.1

0.17 0.60 25.9 4.10 0.12 32.1 12.8 24.3 30.8 69.2

0.00 0.00 5.30 3.90 0.07 20.5 17.1 53.1 9.2 90.8

reference type

Isuzu 4JA1 turbocharged, direct injection, four in-line cylinders 2499 bowl type 93  92

swept volume (cm3) combustion chamber diameter (mm)  stroke (mm) intake valve timing exhaust valve timing fuel injection pump

bond in the aliphatic chain of the esters. The inhibition effect of the double bond on the low-temperature oxidation reactivity of fatty acid esters became more pronounced as the double bond moved toward the central position of the aliphatic chain. The aim of the present paper is to contribute to the understanding of the effect of the degree of unsaturation of biodiesel fuels on the performance, combustion characteristics, and emissions of automotive diesel engines. Therefore, an experimental work was conducted on a high-speed direct-injection diesel engine fueled with three mixtures of fatty acid methyl esters covering a wide range of concentrations of the most common molecules contained in commercial biodiesel fuels (methyl esters of palmitic acid C16:0, oleic acid C18:1, linoleic acid C18:2, and linolenic acid C18:3). Because the degree of unsaturation is a general concept, several alternatives to quantify it were taken into account.

injection pressure (MPa) injector nozzle hole (mm) compression ratio rated power (kW) maximum torque (N m)

open: 24.5 before top dead center (BTDC) close: 55.5 after bottom dead center (ABDC) open: 54.0 before bottom dead center (BBDC) close: 26.0 after top dead center (ATDC) distributor Bosch VE type with mechanical governor (dynamic fuel injection timing) 19.1 (first injection)/25.5 (second injection) 0.190, five holes 18.4 59 (80 hp) at 4100 rpm 170 at 2300 rpm

test conditionsa

M (N m)

n (rpm)

SOI (deg)

mf (g/cyc-cyl)

EGR (%)

M1 M2

95 26

2420 1770

23.5 25.0

0.01165 0.00509

0 0

a

M, torque; n, engine speed; SOI, start of injection; mf, injected mass per cycle and cylinder (average); and EGR, exhaust gas recycling rate.

indices based on the number of reactive positions: the allylic position equivalent (APE) index and the bisallylic position equivalent (BAPE) index. One APE is defined as 2 for a concentration of 1% methyl oleate (or other monounsaturate), methyl linoleate, or methyl linolenate, because these compounds contain two allylic positions. One bisallylic position equivalent (BAPE) is defined as 1 for a concentration of 1% linoleate. Thus, 1% methyl linolenate yields 2 bisallylic position equivalents. The APE and BAPE indices of the test fuels were estimated by multiplying the mass fraction of each component by its respective number of allylic or bisallylic positions, respectively. Cetane numbers of the test fuels were estimated as a function of the reported cetane numbers of the individual methyl esters by a weighted mass average and using the empirical correlation proposed by Bamgboye and Hansen.18 Additionally, the so-called biodiesel cetane index (BCI) proposed by Lapuerta et al. was calculated as a simple exponential function of the biodiesel density.19 A comparative analysis of the properties of the test fuels is presented in the Results and Discussion. Special attention is focused on the effect of the degree of unsaturation on the key properties that affect engine performance, combustion characteristics, and emissions. 2.2. Experimental Equipment. Tests were carried out in an instrumented automotive diesel engine, whose specifications are given in Table 2. The engine was coupled to a 230 kW eddy current dynamometer (Schenck W230). Air and fuel consumption were measured with a hot-wire sensor (Magnetrol TA2) and an electronic balanced mass flow sensor, respectively. With regard to exhaust emissions, NOx emissions were measured by a Horiba EXA 240CL chemiluminescense analyzer, smoke opacity was measured by an AVL DICOM 4000 optical opacimeter, and THCs were measured by a flame ionization detector ThermoFID model 19 ES. In-cylinder combustion diagnosis was carried out using a two species (air and combustion products), single-zone model based

2. Experimental Section 2.1. Test Fuel Characterization. The test fuels were palm oil biodiesel (BP100), linseed oil biodiesel (BL100), and their blend at 50% by volume (BPL50). The methyl ester content of the neat biodiesel fuels was above 97% by weight. The chemical composition of the test fuels, quantified by gas chromatography, is given in Table 1. Palm oil biodiesel is characterized by having a balance between saturated (43.3% C16:0) and monounsaturated (41.8% C18:1) methyl esters. On the other hand, linseed oil biodiesel is predominantly unsaturated (90.8%), having significant contents of diunsaturated (17.1% C18:2) and especially triunsaturated (53.1% C18:3) methyl esters. The physicochemical properties of the test fuels were determined according to recommended standard methods or by means of simple mixing rules or empirical correlations.14 The iodine value (IV) of pure methyl esters was calculated as a function of their molecular weight and number of double bonds.15 For the test fuels, the IV was estimated by multiplying the mass percentage of the unsaturated methyl esters by their respective IV. Therefore, the mixture IV is simply a measure of its average amount of unsaturation. In assessing their oxidative stability, it has been established that the rate of oxidation of fatty compounds depends upon the number and also the location of the double bonds along the fatty acid chain.16 The oxidation reactions are usually initiated at the ethylene groups (-CH2-), whose position is immediately adjacent to a double bond (allylic position) or two double bonds (bisallylic position).17 Knothe16 proposed two alternative unsaturation (14) Benjumea, P.; Agudelo, J. R.; Agudelo, A. F. Basic properties of palm oil biodiesel-diesel fuel blends. Fuel 2008, 87, 2069–2075. (15) American Oil Chemists’ Society (AOCS). AOCS recommended practice Cd 1c-85, calculated iodine value. Official Methods and Recommended Practices of the AOCS, 5th ed.; Firestone, D., Ed.; AOCS: Champaign, IL, 1998. (16) Knothe, G. Structure indices in FA chemistry. How relevant is the iodine value. J. Am. Oil Chem. Soc. 2002, 79, 847–854. (17) Bouaid, A.; Martinez, M.; Aracil, J. Long storage stability of biodiesel from vegetable and used frying oils. Fuel 2007, 86, 2596–2602.

(18) Bamgboye, A. I.; Hansen, A. C. Prediction of cetane number of biodiesel fuel from the fatty acid methyl ester (FAME) composition. Int. Agrophys. 2008, 22, 21–29. (19) Lapuerta, M.; Armas, O.; Hernandez, J. J. Correlation for the estimation of the density of fatty acid esters fuels and its implications. A proposed biodiesel cetane index. Chem. Phys. Lipids 2010, 163, 720–727.

79

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

Table 3. Chemical Characteristics and Parameters for Quantifying the Degree of Unsaturation compound

chemical formula

molecular weight

oxygen content (% mass)

IV (g of I2/100 g)

APE

BAPE

TPE

methyl oleate methyl linoleate methyl linolenate BP100 BPL50 BL100

C19H36O2 C19H34O2 C19H32O2 C18.04H34.92O2 C18.40H34.25O2 C18.88H33.49O2

296.50 294.49 292.47 283.96 287.58 292.53

10.79 10.87 10.94 11.27 11.13 10.93

85.6 172.4 260.4 52.0 112.7 185.4

200 200 200 102 138 182

0 100 200 9 61 123

200 300 400 111 199 305

Table 4. Physicochemical Properties of Test Fuels property

method

BP100

BPL50

BL100

density at 15 °C (kg/m ) kinematic viscosity at 40 °C (mm2/s) higher heating value (MJ/kg) lower heating value (MJ/kg) lower heating value (MJ/m3) T10, temperature at 10% recovered (°C) T50, temperature at 50% recovered (°C) T90, temperature at 90% recovered (°C) cloud point (°C) cold filter plugging point (°C) pour point (°C) Rancimat oxidation stability (h) stoichiometric air/fuel ratio cetane number (mass weighted average) cetane number (Bamgboye-Hansen correlation) BCI average reported cetane number

ASTM D1298 ASTM D445 ASTM D240

871.6 4.6 39.86 37.14 32444 322 328 336 18 15 12 12.83 12.57 63.7 63.9 71.4 62.7

885.2 4.4 39.75 37.13 32866 327 334 347

893.5 4.2 39.63 37.12 33169 327 338 361 9 -6 -9 0.1 12.43 36.5 36.5 41.8 30.0

3

ASTM D86 ASTM D2500 ASTM D6371 ASTM D97 EN 14112

on the approach proposed by Lapuerta et al.20 Atmospheric pressure was taken into account in the pressure signal processing. Heat transfer was calculated using the correlation by Woschni,21 adjusting its constants to the specific engine by means of energy balances.22 To record the instantaneous in-cylinder pressure, a Kistler 6056A piezoelectric pressure transducer installed in the glow plug and a Kistler 5011B charge amplifier were used. A total of 100 pressure curves were registered in each engine operation mode to guarantee confidence in the combustion diagnosis. That number of pressure curves led to a coefficient of variation (COV) of the indicated mean effective pressure under 3% for all tests performed. The instantaneous piston position was determined using an angular encoder with a resolution of 1024 pulses/ revolution (Heidennhain ROD 426) coupled to the crankshaft at the opposite extreme of the fly wheel. The angle of the start of injection (SOI) was measured with a clamp-on transducer connected to the AVL DICOM 4000 analyzer. High-speed data were acquired using Tone LabView-based software and National Instruments data acquisition system (model PCIMIO-16E-4 board). The final results of combustion diagnosis were obtained from software called Caribe, which was previously used by the authors for evaluating the effect of altitude and palm oil biodiesel fueling on the combustion characteristics of a high-speed direct-injection diesel engine.23 2.3. Test Conditions. As shown in Table 2, test conditions of 2420 rpm and 95 N m (M1) and 1770 rpm and 26 N m (M2) were selected as the engine-operating modes. The M1 mode yielded the minimum air/fuel ratio and maximum smoke opacity among the collection of steady operation modes that reproduce the

0.51 12.50 51.3 51.3 52.1

Federal Test Procedure (FTP-75) homologation cycle. The dynamic characteristics of the vehicle Chevrolet Luv 2.5 Turbo, equipped with the Isuzu 4JA1 diesel engine used in this study, and the FTP-75 speed profile were programmed in software developed by authors to translate this transient cycle into steady operation modes (torque - speed).24 All points were weighted according to its frequency in the cycle, and finally, M1 and M2 were selected as the most representative high- and low-load modes, respectively. Before starting experiments with a new fuel, lines were drained, fuel was then added, and the engine operated for at least 1 h to purge any fuel remaining from previous experiments. Runs with different fuels were conducted without any modification of the engine or fuel injection system. Each fuel was randomly tested 3 times on different days. Experiments were carried out at 1500 m above sea level (atmospheric pressure of 640 mmHg).

3. Results and Discussion 3.1. Fuel Property Comparison. Table 3 presents the chemical formula, molecular weight, oxygen content, and four parameters taken into account for quantifying the degree of unsaturation of the test fuels and pure unsaturated methyl esters. Table 4 lists the main physicochemical properties of the test fuels. As seen in Table 3, the average chain length of the biodiesel fuels varied within a narrow range (from 18.04 to 18.88), indicating that the main structural difference among them was their degree of unsaturation. For pure methyl esters, the oxygen content increases slightly with the degree of unsaturation because of the decrease in the molecular weight associated with the displacement of two hydrogen atoms by each double bond. In the case of the test fuels, the more unsaturated biodiesel fuel had the lowest oxygen content as a consequence of the higher sensitivity of the molecular weight to chain length.

(20) Lapuerta, M.; Armas, O.; Hernandez, J. J. Diagnosis of DI diesel combustion from in-cylinder pressure signal by estimation of mean thermodynamic properties of the gas. Appl. Therm. Eng. 1999, 19, 513–529. (21) Woschni, G. A universally applicable equation for the instantaneous heat transfer coefficient in the internal combustion engine. SAE Trans. 1967, 76, 3065–3083. (22) Hsu, B. D. Practical Diesel-Engine Combustion Analysis; Society of Automotive Engineers (SAE) International: Warrendale, PA, 2002. (23) Benjumea, P.; Agudelo, J. R.; Agudelo, A. F. Effect of altitude and palm oil biodiesel fuelling on the performance and combustion characteristics of a HSDI diesel engine. Fuel 2009, 88, 725–731.

(24) Agudelo, J.; Moreno, R.; Perez, J. Dynamic and energy behavior of a bus fuelled with natural gas. Rev. Fac. Ing., Univ. Antioquia 2009, 51, 79–87.

80

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

Consequently, with its high content of polyunsaturated methyl esters, linseed oil biodiesel (BL100) had an IV (185.4 g of I2/100 g) significantly higher than that of the palm oil biodiesel (52 g of I2/100 g) and the maximum limit specified in the European standard EN 14214 (120 g of I2/100 g). As expected, a similar trend was followed by the other indices of unsaturation. The oxidation stability of the test fuels, quantified by the Rancimat induction period in hours, is strongly influenced by their content of more unsaturated methyl esters, especially, methyl linolenate. Neat palm oil biodiesel (BP100) with the lowest content of methyl linolenate (0.15%, m/m; see Table 1) and, consequently, the lowest BAPE index had an induction period (12.83 h; see Table 4) significantly higher than that of the other test fuels and the minimum specifications stipulated in the European standard EN 14214 (6 h) and the American Society for Testing and Materials (ASTM) standard D6751 (3 h). This indicated a greater sensitivity of the BAPE index to methyl esters responsible for increased reactivity. However, this index does not take into account the allylic positions, which, although to a less extent, also contribute to reactivity. For this reason, a parameter called the total position equivalent (TPE) index, defined as the sum of the APE and BAPE indices, is proposed in this work for quantifying the degree of unsaturation of biodiesel fuels. Density, viscosity, heating value, volatility, and cetane number are the main physical properties of the fuel that affect the diesel combustion process. As inferred from Table 4, these properties are affected to different extents by the degree of unsaturation. The density of the test fuels increased with the degree of unsaturation. Assuming a linear behavior, if the degree of unsaturation increases by 100 TPE, the density increases in 1.3%. Values of this property for neat biodiesel fuels lay within the range limited in the European standard EN 14214 (860-900 kg/m3). Other important fuel properties that are more difficult to measure have been correlated with density. Tat et al.25,26 calculated the isentropic bulk modulus of pure methyl esters and several biodiesel fuels as a function of the density and speed of sound at elevated pressures. They found that the speed of sound and the isentropic bulk modulus of biodiesel fuels also tend to increase as the degree of unsaturation increases. The higher heating value (HHV) of the test fuels underwent a slight decrease with the degree of unsaturation (0.14%/100 TPE). Because the values of the HHV of the most common saturated and unsaturated long-chain methyl esters vary within a narrow range,5 the test fuels as well as commercial biodiesel fuels must have similar values of this property. The kinematic viscosity of the test fuels decreased with the degree of unsaturation (4.5%/100 TPE). The values of this property for the pure biodiesel fuels lay within the range limited in the ASTM standard D6751 (1.9-6 mm2/s) and also in the narrower range considered in the European standard EN 14214 (3.5-5 mm2/s).

Table 5. Average Engine Performance Parameters

M1 M2

fuel

BSFC (g/kWh)

BFCE

equivalence ratio

BP100 BPL50 BL100 BP100 BPL50 BL100

283.5 ( 0.95 280.7 ( 2.34 281.3 ( 2.07 457.5 ( 3.18 466.7 ( 2.10 453.7 ( 1.73

0.342 ( 0.001 0.345 ( 0.003 0.345 ( 0.003 0.212 ( 0.001 0.208 ( 0.001 0.213 ( 0.001

0.42 ( 0.005 0.41 ( 0.004 0.41 ( 0.005 0.23 ( 0.007 0.24 ( 0.002 0.23 ( 0.001

The distillation profile of the test fuels showed some differences, which became more noticeable at the mediumtemperature (T50) and high-temperature (T90) points. The T50 and T90 points were higher for the more unsaturated fuels. However, that increase in distillation temperatures may be partly due to the slight differences in chain length and density among fuels. Cetane number data shown in Table 4 indicate that this key fuel property is one of the most sensitive to the degree of unsaturation. The variation of the estimated cetane number of the test fuels with their TPE index is practically linear, indicating that an increase in 100 TPE leads to a 22% decrease in the cetane number. The values of estimated cetane numbers were in agreement with average measured reported values for palm oil18 and linseed oil11 biodiesel fuels. The cetane number of linseed oil biodiesel is significantly lower than the minimum specifications stipulated in the European standard EN 14214 (51) and in the ASTM standard D6751 (47). The linearity between the degree of unsaturation and cetane number indicates that biodiesel fuels with a TPE index higher than 200 are not suitable as neat fuels for modern diesel engines. 3.2. Engine Performance. Table 5 compares the average values resulting from three measurements of brake-specific fuel consumption (BSFC), brake fuel conversion efficiency (BFCE), and equivalence ratio. Results indicate that engine performance was not significantly affected by the type of biodiesel fuel or its degree of unsaturation. This was expected because the heating value, stoichiometric air/fuel ratio, and oxygen content were very similar for the biodiesel fuels tested. Additionally, the BFCE in each operating mode was almost the same. This indicates that the engine converted the chemical energy of the test fuels into mechanical energy with a similar efficiency. Because the stoichiometric air/fuel ratio of the fuels and the engine BSFC were very similar, the equivalence ratio in the engine was practically the same at each operating mode, regardless of the fuel tested. The attainment of a nearly equivalent engine performance for all test fuels allows for a meaningful analysis of the effect of the fuel nature, specifically degree of unsaturation, on combustion characteristics and engine emissions. 3.3. Combustion Characteristics. Table 6 presents the main parameters that characterize the in-cylinder combustion process when the engine was fueled with the test fuels at both operating modes. Figures 1-4 show the HRR and the cumulative heat release (CHR) curves for the test fuels at the M1 and M2 operating modes. The angle of SOI was almost the same for all test fuels at each operating mode. The small differences found in this parameter were probably caused by the experimental error and accuracy of the measurement instruments rather than any effect of the biodiesel type or degree of unsaturation. This result indicates that the differences in density, viscosity, velocity of sound, and bulk modulus, which are the main fuel

(25) Tat, M. E; Van Gerpen, J. H.; Soylu, S.; Canakci, M.; Monyem, A.; Wormley, S. The speed of sound and isentropic bulk modulus of biodiesel at 21 °C from atmospheric pressure to 35 MPa. J. Am. Oil Chem. Soc. 2000, 77, 285–289. (26) Tat, M. E; Van Gerpen, J. H. Speed of sound and isentropic bulk modulus of alkyl monoesters at elevated temperatures and pressures. J. Am. Oil Chem. Soc. 2003, 80, 1249–1256.

81

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al. Table 6. Combustion Characteristics

M1 M2

fuel

SOI (deg)

ID (ms)

maximum dp/dR (kPa/deg)

ΔR (deg)

peak in-cylinder average temperature (°C)

BP100 BPL50 BL100 BP100 BPL50 BL100

23.5 ( 0.15 23.7 ( 0.10 23.6 ( 0.28 24.8 ( 0.08 25.0 ( 0.18 24.8 ( 0.22

0.90 ( 0.05 0.96 ( 0.01 0.98 ( 0.01 1.09 ( 0.02 1.21 ( 0.03 1.32 ( 0.06

463.54 ( 1.4 528.87 ( 1.2 621.66 ( 1.3 584.69 ( 1.8 678.89 ( 2.0 691.98 ( 2.4

29.3 ( 0.02 30.2 ( 0.08 32.0 ( 0.05 21.9 ( 0.04 22.0 ( 0.08 23.0 ( 0.50

1717 ( 4.2 1731 ( 5.2 1780 ( 3.1 1374 ( 8.2 1384 ( 5.0 1395 ( 9.7

Figure 1. HRR of test fuels at the high-load/high-speed mode.

Figure 3. HRR of test fuels at the low-load/low-speed mode.

Figure 2. CHR of test fuels at the high-load/high-speed mode.

Figure 4. CHR of test fuels at the low-load/low-speed mode.

properties that affect the SOI in engines equipped with conventional pump-line-nozzle fuel injection systems, were not high enough to cause any advance or retard at the SOI. Additionally, because the differences in the heating value among the test fuels were not significant, the injection pump did not have to compensate for the energy input by altering the SOI to increase the amount of fuel injected to the cylinders. Because the SOI was practically constant for all test fuels in each operating mode, the differences in the angle of start of combustion (SOC), which can be noticed in Figures 1 and 3, were mainly determined by the ID, which in turn was mainly affected by the cetane number of the fuel. As the cetane number of the biodiesel fuels increased, the ID was shortened and, thus, the SOC was advanced. ID, which was defined for measurement purposes as the time corresponding to the angular interval occurring between the SOI and 5% of the CHR, increased with the degree of unsaturation at both operating modes (see Table 6). An increase in the degree of unsaturation of 100 TPE led to increases in the ID by 4.5 and 10.9% at the M1 and M2 operating modes, respectively. This result is consistent with

the higher reactivity of the saturated alkyl chains of BP100, because these can undergo low-temperature branching, typical of long-chain hydrocarbons.27 These reactions are hindered in the unsaturated esters, where the double bonds add radicals instead, decreasing the possibility of isomerization of the peroxide radical that leads to low-temperature branching.28 Another reason could be the stability of the allylic-type radicals formed from the abstraction of hydrogen atoms.29 This result is also in agreement with the findings by Bennadji et al.,30 who experimentally investigated the autoignition properties of C4-C6 saturated and unsaturated alkyl esters in shock tubes. ID calculations and measurements elucidated some details of the reactivity for unsaturated and saturated species. They argued that the slower ignition of the unsaturated molecules was promoted by the stability provided by the double bond. (28) Vanhove, G.; Ribaucour, M.; Minetti, R. On the influence of the position of the double bond on the low-temperature chemistry of hexenes. Proc. Combust. Inst. 2005, 30, 1065–1072. (29) Vanhove, G.; Petit, G.; Minetti, R. Experimental study of the kinetic interactions in the low-temperature autoignition of hydrocarbon binary mixtures and a surrogate fuel. Combust. Flame 2006, 145, 521–532. (30) Bennadji, H.; Biet, J.; Coniglio-Jaubert, L.; Billaud, F.; Glaude, P. A.; Battin-Leclerc, F. Experimental autoignition of C4-C6 saturated and unsaturated methyl and ethyl esters. Proceedings of the 4th European Combustion Meeting (ECM); Vienne, Austria, April 14-17, 2009.

(27) Curran, H. J.; Gafuri, P.; Pitz, W. J.; Westbrook, C. K. A comprehensive modeling study on n-heptane oxidation. Combust. Flame 1998, 114, 149–177.

82

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

Figure 6. Smoke opacity as a function of the degree of unsaturation.

Figure 5. Specific THC emissions as a function of the degree of unsaturation.

the biodiesel fuel, quantified as the TPE index, increased. This result is in agreement with the findings by Graboski et al.,32 who obtained higher THC emissions when the degree of unsaturation of pure methyl esters was increased. They used a DDC series 60 11.1 L diesel engine. A possible explanation for this trend was given by Gai et al.,33 who showed, via kinetic modeling, that a more unsaturated ester [methyl (E)-2-butenoate, -MC-] formed higher levels of hydrocarbons when compared to a saturated ester (methyl butanoate, -MB-). The chemical reaction path showed that MC mostly reacts through hydrogen-atom abstraction and decomposition. There is a wide consensus that a higher oxygen content leads to a more complete and cleaner combustion. This explains to a great extent the general trend toward significantly lower PM and THC emissions with biodiesel fueling in comparison to non-oxygenated petroleum diesel fueling. In the case of the test fuels, the oxygen content was similar and its slight variation was due to the slight differences in chain length instead of the degree of unsaturation. The higher THC emissions produced by the more unsaturated test fuels could also be a consequence of their higher T90 point of the distillation curve (see Table 4). A high T90 point indicates that the fuel may comprise fractions or components that may not be completely vaporized and burnt, thereby increasing THC emissions. This behavior became more noticeable at the low-load/low-speed operating mode for which in-cylinder temperatures were lower. Another fact that could explain this is the lower cetane number of the more unsaturated test fuels, because excessively longer IDs could produce overmixing of the fuel spray with its surrounding air.31 This fact can also explain the higher THC emissions obtained for all test fuels at the M2 operating mode, where the equivalence ratio was lower than in the M1 mode (see Table 5). 3.4.2. Smoke Opacity. Figure 6 shows that, regardless of the operating mode, smoke opacity slightly increased with the degree of unsaturation of the biodiesel fuels. This result was unexpected because of the well-known decrease tendency in soot formation as premixed combustion phasing increases.31 As shown in Figures 1 and 3, the premixed combustion phasing increased with the degree of unsaturation of the biodiesel fuels. A possible explanation for this was exposed by Gai et al.33 and Sarathy et al.,34 who showed

As expected, the ID of each test fuel was lower at the M1 mode because of the higher in-cylinder pressures and temperatures corresponding to a higher load.31 The SOC and ID had a significant influence on global heat release patterns. As seen in Figures 1 and 3, the typical phenomenology of conventional diesel combustion was displayed by all test fuels. The common “two-stage” heat release with premixed burn followed by diffusion burn was especially noticed at the high-load/high-speed condition. With regard to the operating mode, as the degree of unsaturation of the test fuels increased, the premixed portion of combustion, peak HRR, maximum pressure gradient (dp/dR), and peak in-cylinder bulk-gas-averaged temperature also increased (see Table 6). Premixed combustion corresponds to the fuel that is mixed with air and prepared to burn during the ID period. Unsaturated biodiesel fuels with a longer ID allow for more air/fuel mixing to take place before autoignition. Because this flammable mixture usually burns very quickly, the HRR for premixed combustion is substantially higher than that of diffusion combustion. Accordingly, an increase in the premixed fraction results in more energy being released over a short time scale close to top dead center (TDC) with a nearly constant combustion chamber volume, resulting in higher pressure gradients and in-cylinder gas temperatures. With regard to the effect of the operating mode on combustion characteristics, it can be inferred from Table 6 and Figures 1 and 3 that, regardless of the biodiesel fuel, at the M2 mode, almost all of the fuel was burned under premixed conditions and the global combustion event was shorter than at the M1 mode. For measurement purposes, combustion duration (ΔR) was defined as the angular interval between 5 and 90% of the CHR curve. This behavior was a consequence of the longer ID and the less mass of a given fuel that had to be injected per cycle and cylinder at this operating mode. This combustion pattern led to higher maximum pressure gradients but lower peak in-cylinder bulk-gasaveraged temperatures (see Table 6). 3.4. Exhaust Emissions. 3.4.1. THC Emissions. As seen in Figure 5, the specific THC emissions were relatively low for all test fuels, especially at the high-load/high-speed condition. Regardless of the operating mode, a slight increase in THC emissions was noticed as the degree of unsaturation of

(33) Gaı¨ l, S.; Sarathy, S. M.; Thomson, M. J.; Dievart, P.; Dagaut, P. Experimental and chemical kinetic modeling study of small methyl esters oxidation: Methyl (E)-2-butenoate and methyl butanoate. Combust. Flame 2008, 155, 635–650. (34) Sarathy, S. M.; Gail, S.; Syed, S. A.; Thomson, M. J.; Dagaut, P. A comparison of saturated and unsaturated C4 fatty acid methyl esters in an opposed flow diffusion flame and a jet stirred reactor. Proc. Combust. Inst. 2007, 31, 1015–1022.

(31) Heywood, J. B. Internal Combustion Engine Fundamentals; McGraw-Hill: New York, 1988. (32) Graboski, M. S.; McCormick, R. L.; Alleman, T. L.; Herring, A. M. The effect of biodiesel composition on engine emissions from a DDC series 60 diesel engine. Final Report NREL/SR-510-31461; National Renewable Energy Laboratory (NREL): Golden, CO, 2003.

83

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

Figure 7. Specific NOx emissions as a function of the degree of unsaturation.

Figure 8. In-cylinder bulk-gas-averaged temperatures at the high-load/ high-speed mode.

through experiments and kinetic modeling of an opposed flow diffusion flame and a jet stirred reactor that the following concentrations of soot precursors were much higher with unsaturated (MC) than with saturated (MB) pure methyl esters: pC3H4 (propyne), aC3H4 (allene), 1,3-butadiene, and benzene concentrations. The modeling analysis let them identify the double bond as the key reaction pathway that led to the unsaturated ester forming higher levels of soot precursors and unsaturated hydrocarbons, when compared to the saturated ester. This result was also corroborated by Garnet et al.,35 who observed that the acetylene yield in pyrolysis increased with the number of carbon-carbon double bonds. Their experimental study was performed with methyl octanoate and methyl octenoate. 3.4.3. NOx Emissions. Figure 7 shows that, regardless of the operating mode, specific NOx emissions increased with the degree of unsaturation of the biodiesel fuels. On the other hand, for each test fuel, NOx emissions increased with the engine load. At higher loads, characterized by a greater diffusion portion of combustion, the flame temperature effect could be dominant.36,37 This result is in agreement with those reported in the literature.4, 8,10,11,32,38 As discussed earlier, a lower degree of unsaturation led to a shorter ID (higher cetane number) and, consequently, more advanced SOC. Although, these conditions could favor the formation of NOx inside the combustion chamber because of a possible increase of the residence time of the combustion products at high temperatures, the higher cetane number of the less unsaturated biodiesel fuels also led to less premixed burning and, therefore, lower combustion rates, which finally resulted in lower in-cylinder bulk-gas-averaged temperatures during the entire combustion event, which could inhibit thermal NOx formation (see Table 6). Figure 8 shows the variation of the in-cylinder bulk-gas-averaged temperatures with the crank angle, for all test fuels at the high-load/high-speed mode. The trend was similar at the M2 operating mode.

Szybist et al.,39 testing B20 blends of soybean biodiesel with different concentrations of methyl oleate with an ultralow sulfur biodiesel fuel, found that the timings at which the peak in-cylinder bulk-gas-averaged temperatures and HRRs occurred could have a most dominant role in NOx formation than their magnitudes. In this work, those timings were very similar for all test fuels at each operating mode. According to the experimental results obtained in this work, the cetane number appears to be the key property that determined the net NOx emissions of the biodiesel fuels. However, it is necessary to acknowledge that the degree of unsaturation also affects the local combustion temperature (the higher the degree of unsaturation, the higher the adiabatic flame temperature) and might also affect the NO formation via prompt mechanisms, because this is highly influenced by intermediate combustion products, although it is widely accepted that it has secondary importance with respect to the thermal one. It has been suggested that the double bonds could contribute to the formation of higher levels of certain hydrocarbon radicals in the premixed fuelrich mixture, which could result in more prompt NO formation during subsequent combustion.8 As inferred from the previous discussion, NOx formation in diesel engines is a very complex phenomenon, because it is not quantitatively determined by a change in a single fuel property but rather the result of a number of coupled mechanisms, whose effects may tend to reinforce or cancel one another under different conditions, depending upon engine technology and specific combustion and fuel characteristics.40 4. Conclusions From the experimental work carried out to investigate the effect of the degree of unsaturation of biodiesel fuels on engine performance, combustion characteristics, and emissions, the following conclusions can be drawn: (1) While the degree of unsaturation of biodiesel fuels did not significantly affect engine performance and SOI, it had a noticeable influence on combustion characteristics and emissions. (2) A higher degree of unsaturation led to a longer ID and consequently more retarded SOC. Regardless of the engine-operating mode, the premixed portion of combustion, peak of HRR, maximum pressure gradient, and peak of in-cylinder bulk-gas-averaged

(35) Garner, S.; Sivaramakrishnan, R.; Brezinsky, K. The highpressure pyrolysis of saturated and unsaturated C7 hydrocarbons. Proc. Combust. Inst. 2009, 32, 461–467. (36) Ban-Weiss, G. A.; Chen, J. Y.; Buchholz, B. A.; Dibble, R. W. A numerical investigation into the anomalous slight NOx increase when burning biodiesel; A new (old) theory. Fuel Process. Technol. 2007, 88, 659–667. (37) Eckerle, W. A.; Lyford-Pike, E. J.; Stanton, D. W.; LaPointe, L. A.; Whitacre, S. D.; Wall, J. C. Effects of methyl ester biodiesel blends on NOx emissions. SAE Int. J. Fuels Lubr. 2009, 1, 102–118. (38) Wyatt, V. T.; Hess, M. A.; Dunn, R. O.; Foglia, T. A.; Hass, M. J.; Marmer, W. N. Fuel properties and nitrogen oxide emission levels of biodiesel produced from animal fats. J. Am. Oil Chem. Soc. 2005, 82, 585–591.

(39) Szybist, J. P.; Boehman, A. L.; Taylor, J. D.; McCormick, R. L. Evaluation of formulation strategies to eliminate the biodiesel NOx effect. Fuel Process. Technol. 2005, 86, 1109–1126. (40) Mueller, C. J.; Boehman, A. L.; Martin, G. C. An experimental investigation of the origin of increased NOx emissions when fuelling a heavy-duty compression-ignition engine with soy biodiesel. SAE Int. J. Fuels Lubr. 2009, 2, 789–816.

84

Energy Fuels 2011, 25, 77–85

: DOI:10.1021/ef101096x

Benjumea et al.

in-cylinder bulk-gas-averaged temperatures during the entire combustion event and, therefore, less thermal NOx formation. (5) The TPE index appears to be an adequate parameter to quantify the degree of unsaturation of fatty acid compounds, such as biodiesel fuels.

temperature increased with the degree of unsaturation. (3) THC emissions and smoke opacity slightly increased with the degree of unsaturation, regardless of the engine-operating mode. For all test fuels, ID became longer and THC emissions increased as the engine load decreased. (4) NOx emissions were mainly controlled by the cetane number of the fuel or the ID and, threfore, the relative amount of heat released during the premixed combustion phase. The higher cetane number of the less unsaturated biodiesel fuels led to less premixed burning and lower combustion rates, which finally resulted in lower

Acknowledgment. The authors thank the Colombian Ministry of Agriculture and Rural Development and the Environmental  Authority of Medellı´ n and its metropolitan area (Area Metropolitana del Valle de Aburr a) for the financial support of Project 003 2007D3608-67 (Biodiesel Project).

85