Redalyc.MODELING and OPTIMIZATION of an OTTO

Home , Otto cycle

Revista Mexicana de Ingeniería Química ISSN: 1665-2738 [email protected] Universidad Autónoma Metropolitana Unidad Iztapalapa México

Gómez, A.; Zacarías, A.; Venegas, M.; Vargas, R.O.; Carvajal, I.; Aguilar, J.R. MODELING AND OPTIMIZATION OF AN OTTO CYCLE USING THE ETHANOL- GASOLINE BLEND Revista Mexicana de Ingeniería Química, vol. 16, núm. 3, 2017, pp. 1065-1075 Universidad Autónoma Metropolitana Unidad Iztapalapa Distrito Federal, México

Available in: http://www.redalyc.org/articulo.oa?id=62053304030

How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Vol. 16, No. 3 (2017) 1065-1075 Revista Mexicana de Ingeniería Química

CONTENIDO MODELING AND OPTIMIZATION OF AN OTTO CYCLE USING THE Volumen 8, númeroETHANOL-GASOLINE 3, 2009 / Volume 8, number BLEND 3, 2009

MODELADO Y OPTIMIZACION´ DE UN CICLO OTTO UTILIZANDO LA MEZCLA ETANOL-GASOLINA 213 Derivation and application of the Stefan-Maxwell equations A. Gomez´ 1, A. Zacar´ıas1*, M. Venegas2, R.O. Vargas1, I. Carvajal3, J.R. Aguilar1 1 Instituto PolitécnicoNacional, (Desarrollo ESIME y aplicación Azcapotzalco. de las ecuaciones Avenida de de Stefan-Maxwell) las Granjas 682, Santa Catarina, 02250 Ciudad de México, Mexico. Stephen Whitaker 2Departamento de Ingenier´ıaTérmicay de Fluidos, Universidad Carlos III de Madrid, Avda. Universidad 30, 28911, Leganés, Madrid, Spain. 3 BiotecnologíaInstituto PolitécnicoNacional, / Biotechnology ESIME, UPALM, Ciudad de México07738, México Received: April 9, 2017; Accepted: July 29, 2017 245 Modelado de la biodegradación en biorreactores de lodos de hidrocarburos totales del petróleo Abstract intemperizados en suelos y sedimentos The use of bio-fuels, like bio-ethanol mixed with gasoline, is growing interest in the automotive industry, to reduce the fossil fuels dependence. However, (Biodegradation the air/fuel ratiomodeling for any of sludge blend isbioreactors not found of in total the petroleum literature. hydrocarbons It is found only weathering for specific in soil values such as: E10, E50 and E85 forand example. sediments) For this reason, in the present work a mathematical model is presented to determine the stoichiometric air/fuel ratio of the ethanol-gasoline blend in the range between 0% to 100% of ethanol molar percentage in the S.A. Medina-Moreno, S. Huerta-Ochoa, C.A. Lucho-Constantino, L. Aguilera-Vázquez, A. Jiménez- blend. The model is based on combustion chemical analysis for any composition ethanol-gasoline. The optimum compression González y M. Gutiérrez-Rojas ratio for maximum network is also obtained, considering isentropic processes during compression and expansion. The analysis of the Otto cycle with the259 air Crecimiento,/fuel model sobrevivencia and optimum y compressionadaptación de ratioBifidobacterium is developed. infantis Results a condiciones show that ácidas the stoichiometric air/fuel ratio of the blend ethanol-gasoline is not linear. The maximum difference between air/fuel ratios predicted and experimentally (Growth, survival and adaptation of Bifidobacterium infantis to acidic conditions) reported is 7% in the whole range of the blend. The analysis of the Otto cycle, using the equation derived shows that the power L. Mayorga-Reyes, P. Bustamante-Camilo, A. Gutiérrez-Nava, E. Barranco-Florido y A. Azaola- and the torque decrease when the ethanol mole fraction grows. The equations obtained in this work can be used to predict the performance of internal combustionEspinosa engines using the ethanol-gasoline blend in the continuous range of ethanol molar percentage between 0% and 100%.265 Statistical approach to optimization of ethanol fermentation by Saccharomyces cerevisiae in the Keywords: modeling, Otto cycle, ethanol-gasoline blend. presence of Valfor® zeolite NaA Resumen La relacion´ aire/combustible (Optimización para cualquier estadística mezcla de no la estfermentacióna´ reportada etanólica en la literatura, de Saccharomyces solo´ para valorescerevisiae espec en ´ıficos.presencia En de este trabajo, se presenta un modelo matemzeolitaatico´ Valfor® para zeolite determinar NaA) la relacion´ estequiometrica´ aire/combustible de la mezcla etanol-gasolina en el intervalo de 0% a 100%G. del Inei-Shizukawa, porcentaje molar H. A. deVelasco-Bedrán, etanol en la mezcla. G. F. Gutiérrez-López El modelo se basa and enH. Hernández-Sánchez el analisis´ qu´ımico de la combustion´ para cualquier composici on´ etanol-gasolina. Tambien´ se obtiene la relacion´ de compresion´ optima´ para el trabajo neto maximo,´ considerando procesos isentropicos´ durante la compresion´ y la expansion.´ Se realiza el analisis´ del ciclo Otto con el modelo Ingeniería de procesos / Process engineering aire/combustible y la relacion´ de compresion´ optima.´ Los resultados muestran que la relacion´ estequiometrica´ aire/combustible de la mezcla etanol-gasolina271 Localización no es lineal. de una La planta diferencia industrial: maxima´ Revisión reportada crítica entre y adecuación las relaciones de los aire criterios/combustible empleados que en se predicen y experimentales es de 7%esta en decisión todo el intervalo de la mezcla. El analisis´ del ciclo Otto, utilizando la ecuacion´ derivada muestra que la potencia y el torque (Plant decrecen site selection: cuando Critical la fracci reviewon´ mol and deadequation etanol aumenta. criteria used Este in trabajothis decision) se puede utilizar para predecir el rendimiento de los motoresJ.R. de Medina, combusti R.L.on´ Romero interna y utilizando G.A. Pérez la mezcla etanol-gasolina en el intervalo continuo entre 0% y 100%. Palabras clave: modelado, ciclo Otto, mezcla etanol-gasolina.

1 Introduction performed by Rocha-Mart´ınez et al. (2002), while the desires properties for a rocket fuel were studied by Miranda (2003). The synthesis and characterization of perovskites for fuel cells as an alternative energy Deployment of fossil fuels has increased the interest source was analyzed by Chavez-Guerrero´ et al. in using new alternatives energy sources. Otto and (2003), while Carbon nanotubes produced from Diesel engine models with cyclic variability were * Corresponding author. E-mail: [email protected] Tel. 57296000 ext. 64511

Publicado por la Academia Mexicana de Investigacion´ y Docencia en Ingenier´ıa Qu´ımica A.C. 1065 G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

hexane and ethanol has been used by Mendoza et al. gasoline blend; one of them is taking into account (2006). The use of bio-fuels, like biodiesel studied the volume percentage considering that for gasoline by Gonca and Dobrucali (2016) o bio-methanol and is 14.6 and for ethanol 9.0, according to Pulkrabek bio-ethanol mixed with gasoline, is getting growing (2003). On the other hand, as it can be observed in interest in the automotive industry, to reduce the Chen et al. (2012), the vaporization enthalpy respect fossil fuels dependence Murali et al. (2012). The to the fraction of ethanol in gasoline is linear when it use of ethanol blend gasoline is demonstrated also is based on the volume. On the contrary, its tendency by Yucesu¨ et al. (2006), who give results of the is quadratic if it is based on the molar mass. Further, ethanol use to reduce the gasoline consumption, the the air/fuel ratio for the ethanol-gasoline blend has emissions during the cold start up of a flex fuel been compiled in Kasseris (2011), based on different engine, the experimental determination of ethanol bibliographic data. There, the ratio is presented for physical properties, and the heat flux characteristics the mixtures E0, E10, E20, E50, E85 and E100 of spray wall impingement with ethanol, respectively. (the number represents the percentage of ethanol The economy and emissions of light vehicles, and the in gasoline). Further, the air/fuel ratio, AF, can be dynamic analysis of the supply chain feasibility for obtained by different ways as discussed briefly in the the ethanol-gasoline mixture in Mexico, are shown in following: Hernandez´ et al. (2014). The use of this fuels blendin motorcycles was reported by Yao et al. (2013), where 1. By means of chemical analysis, considering: the authors show the good results obtained. flame velocity, problem geometry, mass flow rate, time, species concentration, etc. The air/fuel ratio is of significant importance This analysis is complex, due to partial for the analysis and measurement of combustion, differential equations are coupled and they as shown by Clements and Smy (1976) for the require numerical solution. This process may measurement of the ionization density. Someone increase the cost of the research, mainly for papers have been found on this issue. Polymeropoulos the computing time that can be excessive. and Sernas (1977) determined the droplet size and This makes difficult the determination of the et al the fuel-air ratio for a spray, Deadmore . (1979) air/fuel ratio in the entire range of ethanol studied the effect of fuel-to-air ratio on burner rig hot mole fractions. This method is useful to find corrosion, Desoky and Rabie (1983) showed the fuel experimentally the air/fuel ratio. economy benefits of using alcohols gasoline blends when experimental investigations were carried out 2. Experimentally, by two different ways: a) using to judge the performance of small spark ignition the Brettschneider’s equation Brettschneider engines running on alcohols, gasoline and alcohol- (1979, 1997), Schifter et al. (2011), Zhang et gasoline blends. Bardaie and Janius (1984) during al. (2013) that shows how the fuel is burned their experimental investigation found the power loss by means of the combustion exhaust gases of 3-4% when using ethanol in SI engine with a analysis. Results obtained from this equation, modified carburetor. Barwan (1985) studied blended using experimental data of exhaust gases, show fuel ranging from E10 to E70 and concluded that if the combustion is clean or not. Experiments the highest antiknock capability was obtained with need to be performed as many times as the blend E50. Palmer (1986) in his experimental investigation composition of ethanol-gasoline is changed. b) reported the engine power improved by 5% when using the following equation, Cengel and Boles 10% in gasoline was used as a fuel additive. Hamdam (2012): m and Jubran (1986) during their investigations using AF = a (1) 5% ethanol in gasoline under partial load found the m f thermal efficiency improved by 4-12%. Ohsuga and where: ma: air mass, measured using a Ohyama (1986) studied the oxygen-biased wide range flowmeter. air-fuel ratio sensor for rich and lean air-fuel ratios. m f : fuel mass, calculated from the heat supplied Additionally, according to Rigatos et al. (2014) the during combustion, Qs. control of the air/fuel ratio is very important in the operation of spark-ignition engines.To study the The heat supplied, Qs, is calculated as: performance of internal combustion engines using this Qs = m f LHV (2) mixture, the air/fuel ratio is required. This ratio has been determined in different ways for the ethanol- where:

1066 www.rmiq.org G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

LHV: lower heating value, which should be 2 Model development obtained for each ethanol-gasoline blend.

In the same way, the experiments should be repeated for each one of the blend compositions 2.1 Air/fuel ratio equation tested. The procedure used in this section, is described in a detailed way in Gomez´ (2014), it is used to obtain the equation for the air/fuel ratio as a function of the 3. Analytically, performing a chemical analysis ethanol mole fraction. of combustion for each ethanol-gasoline As previously commented, combustion products composition. One way to obtain the air/fuel only include CO ,H O and N , the excess air ratio of this mixture is shown in Cordeiro et 2 2 2 for tempering is not considered. Other products al. (2012), where the authors give and equation like CO and nitrogen oxides are neglected, because for this ratio proposed by Heywood Heywood an ideal analysis is performed. Also, nitrogen and (1988). However, in order to use this equation other gases of atmospheric air are considered to be the ratios H/C and O/C should be known, chemically neutral, as in Pulkrabek (2003). However, making the analysis complex. nitrogen affects the temperature and pressure during the combustion process. The maximum heat generated The chemical analysis of combustion can be by combustion depend on the LHV and on the ethanol developed supposing that combustion is instantaneous concentration in the blend, it can be estimated from and 100% efficient (ideal combustion), i.e., the equations 11 and 12 presented later. The higher reactants burn totally generating a clean combustion. temperature reached correspond to E0 (only gasoline) Further, products are considered to be: water vapor, in the entire range, the higher temperature generated carbon dioxide and gaseous nitrogen. Taking into by the stoichiometric reaction correspond to E0 or account these assumptions, a chemical equation can lower for any blend. The mechanical elements of be written including in the reactants the percentages combustion chamber are designed for this operation of ethanol and gasoline, reacting with atmospheric air. conditions, the used of ethanol/gasoline blends do This analysis can be performed for the whole range not need any modification in design or in mechanical of the ethanol-gasoline blend, from 0% to 100% of elements to be implemented. ethanol (E0 to E100, respectively). The ideal combustion equation for the blend ethanol-gasoline can be written as follows, This method simplifies considerably the considering that x is the mole fraction of ethanol and determination of the air/fuel ratio, because using a 1 − x the mole fraction of gasoline: single equation it is possible obtaining the ratio for the continuous range between E0 and E100. Additionally, xC2H5OH + (1 − x)C8H15 + a(O2 + 3.76N2) → this equation can be modified for more specific cases bCO2 + cH2O + dN2 (3) in a simple manner, for example if excess of air or a third fuel is present, etc. Taking into account the law of mass conservation, This last method is used in the present work to algebraic equations can be written for each chemical derive a general equation for the air/fuel ratio of the element, obtaining the following results for the ethanol-gasoline blend for any composition between coefficients of Eq. (3): 0% and 100% of ethanol. This equation is a function 47 − 35x a = (4) only of the mole fraction of ethanol in the blend. 4 This work is organized in four sections. The model b = 8 − 6x (5) development section presents: i) the chemical analysis 15 − 9x c = (6) for the whole range of the ethanol-gasoline blend, 2 ! ii) the analysis developed to the ideal Otto cycle for 47 − 35x the ethanol-gasoline blend, iii) a compression ratio d = 3.76 (7) 4 equation that maximizes the net work of the cycle, for each ethanol fraction in the blend. Finally, iv) the The number of moles of each chemical element in results and discussion section presents the principal the reactants and combustion products can be seen in findings of this research. Table 1.

www.rmiq.org 1067 G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

Table 1. Number of atoms in the reactants and combustion products. Number of atoms Element 8 − 6x C 15 − 9x H 47−35x 2 O 47−35x 3.76 2 N

Table 2. Molecular weights of the reactants and combustion products. Compound Molecular weight (g/mole) Fig. 1. Otto Cycle on a a) pressure-specific volume C H OH (Ethanol) 46 2 5 diagram, b) temperature-specific entropy diagram. C8 H15 (Gasoline) 111 O2 32 constant volume from state 2 to 3, the useful stroke is N2 28 developed during the isentropic expansion from state CO2 46 3 to 4 and, finally, heat is released to the cold reservoir H2 O 18 following a constant volume process from state 4 to 1, closing the cycle. All the processes are developed in Taking into account the molecular weights of a closed system by ideal gases with constant specific reactants and combustion products (Table 2), the mass heats. of each individual compound can be obtained using the Heat supplied during process 2-3 can be calculated number of moles of each compound calculated from as: Eqs. (4) to (7). Q = m LHV η = (m + m )C (T − T ) (10) Further, the total mass of fuel is: in f f c a f v 3 2

LHV f can be obtained from: m f = me + mg (8) − where: LHV f = ωLHVe + (1 ω)LHVg (11) m : mass of ethanol e The ethanol mass fraction in the blend, ω, is obtained m : mass of gasoline g from Eq. (12) considering the molecular weight, M, of In this way, the following equation is obtained for ethanol and gasoline: the air/fuel ratio of the ethanol-gasoline blend, using the definition given in Eq. (1): xM m ω = e = e (12) ! xM + (1 − x)M m 47 − 35x e g t AF = 34.32 (9) 111 − 65x Net work of the cycle is the difference between the work produced during expansion and the one Eq. (9) is only a function of the mole fraction of consumed during compression of the air-fuel mixture: ethanol in the blend. It expresses the ratio between the mass of atmospheric air and blend required to obtain Wn = W − W (13) an ideal combustion process. 34 12 where: 2.2 Analysis of the Otto-cycle W34 = (ma + m f )Cv(T3 − T4) (14) The analysis developed in this section applies to W = (m + m )C (T − T ) (15) the ideal air-standard conditions, Po= 101.325 kPa, 12 a f v 2 1 To=25°C of Otto cycle, represented in Fig. 1. The Thermodynamic efficiency of the Otto cycle is defined procedure followed is similar to that described by as: Pulkrabek (2003). The engine develops an isentropic Wn compression from state 1 to 2, heat is supplied at η = (16) Qin

1068 www.rmiq.org G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

Power produced by the engine, as a function of the Substituting Eq. (25) for T3 in Eq. (24), the optimum regime is (Wn in kJ): compression ratio is obtained:

Wn · n · Ncyl Nt = (in HP) (17) ! 1 ! 1 89.52 m f LHV f γ−1 LHV f γ−1 ropt = = Torque can be calculated from: (ma + m f )CvT1 (AF + 1)CvT1 (26) 60W τ = n (18) In Eq. (26), LHV f , AF, Cv and γ are a function of x. 2πn In this model, Cp and Cv are calculated as follows:

2.3 Optimum compression ratio for AF 1 C = C + C (27) maximum net work p AF + 1 pa AF + 1 p f AF 1 With the aim of searching for an optimal fraction C = C + C (28) v AF + 1 va AF + 1 v f of ethanol in the blend, in this section an equation is obtained that allows calculating the compression Finally, the efficiency of the cycle corresponding to the ratio that maximizes the net work of the cycle, for optimized compression ratio is: each ethanol fraction in the blend. In this way, the blends having optimum compression ratios achievable 1 (AF + 1)C T η = 1 − = 1 − v 1 (29) in practice can be identified. opt γ−1 LHV f The specific net work of the Otto cycles is: ropt

Wn wn = (19) (ma + m f ) 3 Results and discussion Considering the isentropic processes 1-2 and 3-4: !γ−1 T V T 2 = 1 = 3 = rγ−1 (20) T1 V2 T4 3.1 Air/fuel ratio Combining Eqs. (13) to (15), (19) and (20), the following equation is obtained for the specific net Variation of the air/fuel ratio, as a function of the work of the cycle: ethanol mole fraction in the mixture, is calculated using Eq. (9) and represented in Fig. 2. As it is ! 1 γ−1 observed, this ratio is not a linear function of the wn = CvT3 1 − − CvT1 r − 1 (21) rγ−1 ethanol mole fraction. A similar tendency was shown by Yao et al. (2013) for the vaporization enthalpy of Deriving Eq. (21) respect to the compression ratio, r, the blend. Results found in this paper can be used to and making the result equal to zero: simulate the engine performance in the whole range of ∂wn concentrations. As commented before, air/fuel ratios = −C T (1 − γ)r−γ − C T (γ − 1)rγ−2 = 0 (22) ∂r v 3 v 1 can be found in the open literature for discrete values then: of ethanol mole fraction. T 3 = r2(γ−1) (23) Results found using the model developed in this T1 work has been compared with experimental and From Eq. (23), the optimum compression ratio that theoretical values found in the open literature. Fig. 2 maximizes the net work of the cycle is obtained: shows the comparison with theoretical results obtained by Kasseris (2011), Orbital (2002) and Mantilla ! 1 T3 2(γ−1) (2010). It is shown that, for ethanol mole fractions ropt = (24) T1 lower than 20% and higher than 95%, results are very similar to the ones obtained in the present work. Assuming that combustion efficiency is 100%, the However, the results in the intermediate range defer maximum temperature of the cycle can be obtained more significantly. As observed in Fig. 2, results from Eqs. (10) and (20): published in the open literature can be fitted using m f LHV f ) γ−1 a linear function, similar to a linear interpolation T3 = + T1r (25) (ma + m f )Cv between air/fuel ratios of pure components.

www.rmiq.org 1069

320 A comparison of Eq. 9 with experimental results reported in the literature was done to 321 verify the predictive capacity. Results calculated by the present model and experimental 322 values found in the open literature Costa and Sodré (2010), Szybist (2010) and Camarillo 323 (2011) can be observed in Fig. 4. The last work Camarillo (2011) gives results of the 324 air/fuel ratio for blends of ethanol-gasoline and hydrated ethanol-gasoline. Similarly, to 325 theoretical studies, experimental air/fuel ratios found in the open literature are very scarce 326 and do not cover the whole range of ethanol-gasoline blends. As observed, results predicted 327 are similar to those of Costa and Sodré (2010) and Szybist (2010), but highly differ from 328 values given by Camarillo (2011). This difference can be because of sensors errors in the 329 experimentation or probable not experimental stoichiometric conditions. The difference 330 between experimental data and values calculated using Eq. (9) is shown in Fig. 5. 331 Minimum difference in the whole range of ethanol concentration is about 0.7% respect to 332 results of Costa and Sodré (2010) and Szybist (2010).

333 The results here shown indicate that Eq. (9) could be utilized to calculate the air/fuel ratio G´omezet al./ Revista Mexicana deIngenier334 for´ıa Qtheu ethanol´ımica-gasoline Vol. 16, blends No. with 3 (2017) satisfactory 1065-1075 results in the interval of 0% to 100% of 335 ethanol mole fraction.

16 336 16 16 14 14 14 12 F A

, 12 12 F o 10 i F t A

A a ,

r ,

10 o 10 l i o

t 8 i e t a u r a f

r / l

8 r

8 l i e 6 e u A f u / f r /

i 6 6 r i

A 4 A This work Orbital (2002) 4 42 Mantilla (2010) Kasseris (2011) Interpolation This work Orbital (2002) This work Costa and Sodré (2010) 2 2 Szybist (2010) Camarillo (2011) Camarillo H (2011) 0 Mantilla 0.2(2010) 0.4Kasseris (2011)0.6 0.8Interpolation1

309 0 Ethanol0.2 mole fraction0.4 in the0.6 blend, x 0.8 1 0 0.2 0.4 0.6 0.8 1 310 Fig. 2. Comparison between air/fuel ratios obtained in the present work and theoretical337 data Ethanol mole fraction in the blend, x 309311 found in the open literature,Ethanol as amole function fraction of the in theethanol blend, mole x fraction in the blend. 310 Fig. 2. ComparisonFig. 2. Comparisonbetween air/fuel betweenratios obtained air /infuel the present ratios work obtained and theoretical338 in Fig. data 4 ComparisonFig. 4 Comparison between air/fuel between ratios obtained air/fuel in the ratios present obtainedwork and experimental 311312 found in the open literature, as a function of the ethanol mole fraction in the blend.data found in the open literature, as a function of the ethanol mole fraction in the blend. the present work and theoretical data found in the open339 in the present work and experimental data found in 312313 Relative errors between results predicted in this work and theoretical values340 found in the the open literature, as a function of the ethanol mole 314 open literatureliterature, can be observed as a function in Fig. 3. ofAs thecommented ethanol before, mole major fraction differences in exist in 313315 Relativethe range errorsthe between blend. between 20% results and 95% predicted of ethanol in this in thework blend. and theoretical For lower values and higher found values, in the fraction in the blend.

314316 openrespectively, literature the can differences be observed are asin smallFig. 3. as As 0.7% commented in the blend before, E5. major differences exist in 315 the range between 20% and 95% of ethanol in the blend. For lower and higher values, 316 respectively, the20 differences are as small as 0.7% in the blend E5. 30 Orbital (2002) Mantilla (2010) Kasseris (2011) Costa and Sodré (2010) Szybist (2010) 20 Interpolation 25 Orbital (2002) Mantilla (2010) Kasseris (2011) ManuscritoCamarillo (2011) sometido a la Revista MexicanaCamarillo de H I(2011)ngeniería Química11 15 Interpolation 20 15 10 15

10 10 5 Relative error, e, [%]

Relative error, e, [%] 5 5

Relative error, e, [%] 0 0

0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1

Ethanol mole fraction in the blend, x 317 0 Ethanol0.2 mole 0.4fraction in the0.6 blend, x0.8 1 341 Fig. 5 Relative error between air/fuel ratios calculated in the present work and experimental 318 Fig. 3 RelativeFig. 3error Relative between errorair/fuel betweenratios calculated air/fuel in the ratios present calculated work and theoretical342 Fig. 5 Relative error between air/fuel ratios calculated 317 Ethanol mole fraction in the blend, x 343 data found in the open literature, as a function of the ethanol mole fraction in the blend. 319 data foundin in thethe open present literature, work as a andfunction theoretical of the ethanol data mole foundfraction in the the blend. in the present work and experimental data found in 318 Fig. 3 Relative error between air/fuel ratios calculated in the present work and theoretical344 319 data foundopen in the open literature,Manuscrito literature, asas sometido a afunction function a la of Revista the ethanol of Mexicana the mole ethanol defraction Ingeniería in mole the Q blend.uímica 10 the open literature, as a function of the ethanol mole 345 3.2. Analysis of Otto-cycle fraction in the blend. fraction in the blend. Manuscrito sometido a la Revista Mexicana de Ingeniería Q346uímica Previous10 studies found in the open literature about Otto cycles using mixtures of fuels can 347 be found in Pulkrabek (2003), Kasseris (2011) and Szybist (2010). In the present work, an 348 analysis ofSimilarly, a 4 times Otto cycle to using theoretical different concentrations studies, of the experimental ethanol-gasoline blend Relative errors between results predicted in this349 is performed. The characteristics of the engine used are shown in Table 3, similar to work and theoretical values found in the open350 Pulkrabekair/fuel (2003) ratios. Intake conditions found in considered the open are shown literature in Table 4, are corresponding very to 351 temperature and pressure at sea level. Thermodynamic properties are shown in Table 5. The literature can be observed in Fig. 3. As commented352 modelscarce previously and described do not was implemented cover the in a whole computer rangecode developed of ethanol- by the authors before, major differences exist in the range between353 usinggasoline Engineering Equation blends. Solver As software, observed, EES™ Klein (2015) results. predicted 20% and 95% of ethanol in the blend. For lower and354 are similar to those of Costa and Sodre´ (2010) higher values, respectively, the differences are as small355 and Szybist (2010), but highly differ from values as 0.7% in the blend E5. 356 given by Camarillo (2011). This difference can be A comparison of Eq. 9 with experimental results357 because of sensors errors in the experimentation or reported in the literature was done to verify the358 probable not experimental stoichiometric conditions. predictive capacity. Results calculated by the present359 The difference between experimental data and values model and experimental values found in the open calculated using Eq. (9) is shown in Fig. 5. Minimum literature Costa and Sodre´ (2010), Szybist (2010) and difference inManuscrito the whole sometido range a la Revista of ethanol Mexicana concentrationde Ingeniería Química12 Camarillo (2011) can be observed in Fig. 4. The last is about 0.7% respect to results of Costa and Sodre´ work Camarillo (2011) gives results of the air/fuel (2010) and Szybist (2010). ratio for blends of ethanol-gasoline and hydrated ethanol-gasoline.

1070 www.rmiq.org

140 6000 rpm ] P

G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımicaH 120 Vol. 16, No. 3 (2017) 1065-1075 [

, i

W 4500 rpm

, 100 r e w o

Table 3. Characteristics of the 4T Otto cycle analyzed. p 80 140

d 30006000 rpm rpm e t ] a

P 60 c i

Parameter Value H 120 d [

n , i 3 I 1500 rpm W 40 4500 rpm Displacement volume 2400 cm , 100 r Number of cylinders 4 e w 20 o

p 80 Bore 72 mm

d 3000 rpm e

t 0

Stroke 74 mm a 60 0 0.2 0.4 0.6 0.8 1 c i

Compression ratio 8.6:1 d n Ethanol mole fraction in the blend, x 371 I 40 1500 rpm 372 Fig. 6. Indicated engine power as a function of the ethanol mole fraction in the blend for Table 4. Intake conditions in the 4T Otto373 cycledifferent regimes. 20 analyzed. 374 0 0 0.2 0.4 0.6 0.8 1 Parameter Value371375 In a similar way to the power, Ethanolthe torque mole decreases fraction slightlyin the blend, when x the ethanol mole fraction 376 augments (Fig. 7). In this case, the effect of concentration increases as the engine speed Intake temperature, T1 25 °C 372377 Fig.decreases. 6. Indicated The torque Fig.engine 6.corresponding power Indicated as a functionto engine pure gasoline of power the ethanol E0 as is aslightly mole function fractionhigher of than in thethat blend of pure for Intake pressure, P1 101.325373378 kPa differentethanol regimes.E100. Maximumthe ethanol difference mole reaches fraction 16.3% in the at 1500 blend rpm. for diﬀerent Volume percentage of 0% to 100% regimes.

ethanol in the blend 374379 Rotational speed 1500,375 3000, In a similar way to the power,1 the torque decreases slightly when the ethanol mole fraction 4500,376 6000 augments rpm (Fig. 7). In this case, the effect of concentration increases as the engine speed 377 decreases. The torque corresponding to pure gasoline E0 is slightly higher than that of pure 0.8 378 ethanol E100. Maximum] difference reaches 16.3% at 1500 rpm. m -

Table 5 Thermodynamic properties, taken from N k

[ 1500 rpm 379

0.6 Clements and Smy (1976). , t

e 1 u

Parameter Air Bio-ethanol Gasoline q 3000 rpm r

o 0.4 Speciﬁc heat, 1.005 2.3 2.22 T 0.8 4500 rpm Cp, kJ/kg K ] m

- 0.2

Speciﬁc heat, 0.718 2.3 2.22 N

k 6000 rpm [ 1500 rpm

Cv, kJ/kg K 0.6 , t

Lower heating value, - 26,900 44,300 , 0 e 0 0.2 0.4 0.6 0.8 1 u

kJ/kg q 3000 rpm r Ethanol mole fraction in the blend, x 380 o 0.4 T 381 Fig. 7. Torque ofFig. the 7.engine Torque as a of function the engine of the as ethanol a function mole of fraction the4500 ethanol rpmin the blend. mole0.2 fraction in the blend. The results here shown indicate that382 Eq. (9) could be utilized to calculate the air/fuel ratio for the ethanol- 6000 rpm gasoline blends with satisfactory results in383 the interval Thermodynamic0 properties are shown in Table 5. 0 0.2 0.4 0.6 0.8 1 of 0% to 100% of ethanol mole fraction. The model previously described was implemented 380 in a computerManuscritoEthanol code sometido mole developed afraction la Revista in by the the Mexicanablend, authors x de usingIngeniería Química14 TM 381 Fig. 7. Torque ofEngineering the engine as Equationa function Solverof the ethanol software, mole EES fractionKlein in the blend. (2015).

382 Fig. 6 shows the indicated power of the engine, 3.2 Analysis of Otto-cycle 383 as a function of the ethanol mole fraction, for different rotational speeds of the engine. It can be Previous studies found in the open literature about observedManuscrito that the sometido indicated a la powerRevista decreases Mexicana slightlyde Ingeniería Química14 Otto cycles using mixtures of fuels can be found in when the ethanol mole fraction augments and a more Pulkrabek (2003), Kasseris (2011) and Szybist (2010). pronounced reduction takes place for compositions In the present work, an analysis of a 4 times Otto cycle of ethanol higher than 80%. Moreover, the effect of using different concentrations of the ethanol-gasoline concentration increases as the engine speed rises. blend is performed. The characteristics of the engine In a similar way to the power, the torque used are shown in Table 3, similar to Pulkrabek (2003). decreases slightly when the ethanol mole fraction Intake conditions considered are shown in Table 4, augments (Fig. 7). In this case, the effect of corresponding to temperature and pressure at sea level. concentration increases as the engine speed decreases.

www.rmiq.org 1071

384 3.3. Optimum compression ratio

385 The optimum compression ratio of the cycle, corresponding to the maximum specific net 386 work, at T3=1340 K, is shown in Fig. 8 as a function of the blend composition. It can be 387 observed that the optimum compression ratio rises as the ethanol mole fraction in the blend 388 increases. As real compression ratios used in gasoline engines are lower than 15, all of the 389 cycles with ethanol-gasolineG´omezet blend can operate al./ Revista at their Mexicanaoptimum compression deIngenier ratio.´ıaQ u ´ımica Vol. 16, No. 3 (2017) 1065-1075

390 13.0 compression ratio of each blend represented in Fig.

12.5 8. As it is observed, a more pronounced decrease

r of the eﬃciency occurs for ethanol mole fractions

, o

i 12.0

t higher than 80% approximately. For this reason, it a r

n is recommended to use ethanol concentrations lower o

i 11.5 s

s than this value. e r

p 11.0 m o C 10.5 Conclusions

10.0

9.5 0 0.2 0.4 0.6 0.8 1 In this paper, an equation for the stoichiometric air/fuel 391 Ethanol mole fraction in the blend, x ratio of the ethanol-gasoline blend is derived and the performance of a 4 times Otto cycle using this mixture Fig. 8. CompressionFig. 8. ratio Compression that maximizes ratio the that net work maximizes of the Otto the cycle net workas a function of the 392 is evaluated. The following conclusions have been 393 ethanol mole offraction the Ottoin the blend. cycle as a function of the ethanol mole derived: fraction in the blend. 394 • The equation derived for the air/fuel ratio 395 According to Fig. 9,0.6 the efficiency of the cycle decreases when the ethanol mole fraction 396 grows. In this figure, the cycle operates at the optimum compression ratio of each blend of the ethanol-gasoline mixture can be used 397 represented in Fig. 8. As it is observed, a more pronounced decrease of the efficiency straightaway and without computational cost. It

398 occurs for ethanolo 0.59 mole fractions higher than 80% approximately. For this reason, it is h

depends only on the ethanol mole fraction in ,

399 recommended toe use ethanol concentrations lower than this value. l

c the blend and is valid for the whole range of y 0.58 c

o ethanol concentrations between 0% and 100%. t t O

The equation obtained oﬀers exact results for f 0.57 o

y the stoichiometric air/fuel ratio of the blend. A c n e i 0.56 comparison with experimental data reported in c i f f literature shows that results are very similar for E 0.55 low and high ethanol concentrations.

0.54 • Optimization of the compression ratio to 0 0.2 0.4 0.6 0.8 1 maximize the specific net work of the cycle, at 400 Ethanol mole fraction in the blend, x Fig. 9. ThermodynamicManuscrito sometido e ffia laciency Revista of Mexicana the Otto de cycle Ingeniería as Química15 T3 = 1340 K, provides ratios achievable in real 401 a function of ethanol mole fraction in the blend. gasoline engines. 402 Fig. 9. Thermodynamic efficiency of the Otto cycle as a function of ethanol mole fraction 403 in the blend. • The ethanol-gasoline blend offers suitable The torque corresponding to pure gasoline E0 404 results of power and torque. They reduce as is slightly higher than that of pure ethanol E100. maximum 6.4 and 6.44% (8.56 and 8.59% 405 Maximum difference reaches 16.3% at 1500 rpm. with E80) respectively, respect to pure gasoline, 406 4. Conclusions when the blend E70 is used and the rotational 407 3.3In this paperOptimum, an equation compression for the stoichiometric ratio air/fuel ratio of the ethanol-gasoline blendspeed of the engine is 6000 rpm. The lower 408 is derived and the performance of a 4 times Otto cycle using this mixture is evaluated.heating The value of ethanol (40% lower), and the following conclusions have been derived: 409 The optimum compression ratio of the cycle, fact that the fuel is oxygenated, are the reasons 410 corresponding• The equation to derived the maximum for the air/fuel specific ratio of net the work, ethanol- atgasoline mixture canof be the performance decrease respect to pure 411 T3=1340used K,straightaway is shown and without in Fig. computational 8 as a functioncost. It depends of only on the ethanol 412 mole fraction in the blend and is valid for the whole range of ethanol concentrationsgasoline. 413 the blendbetwee composition.n 0% and 100%. It can The beequation observed obtained that offers the exact results for the 414 optimumstoichiometric compression air/fuel ratio ratio rises of the as blend the. A ethanol comparison mole with experimental• The data efficiency of the cycle operating at the 415 fractionreported in the in blendliterature increases. shows that results As real are very compression similar for low and high ethanoloptimum compression ratio is evaluated. A 416 concentrations. ratios used in gasoline engines are lower than 15, all more pronounced decrease in the efficiency is 417 of the• Optimization cycles with of ethanol-gasoline the compression ratio blend to maximize can operate the specific net work ofobtained the for blends with ethanol mole fraction cycle, at T =1340 K, provides ratios achievable in real gasoline engines. 418 at their optimum3 compression ratio. higher than 80% approximately. The use of 419 •According The ethanol to-gasoline Fig. blend 9, the offers effi suitableciency results of theof power cycle and torque. They reduceethanol decreases the efficiency because, as 420 as maximum 6.4 and 6.44% (8.56 and 8.59% with E80) respectively, respect to pure 421 decreasesgasoline, when when thethe blend ethanol E70 is moleused and fraction the rotational grows. speed of the engine is commented6000 before, fuel is oxygenated and it has In this figure, the cycle operates at the optimum a lower heating value.

1072 Manuscrito sometido a la Revista Mexicanawww.rmiq.org de Ingeniería Química16

G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

This work can be improved considering a better References kinetic scheme and secondary products, resulting from an inefficient combustion. Badwan M.S. (1985). Performance and Knock limits Nomenclature of ethanol-gasoline blend in spark ignited. SAE Paper 850213. AF air/fuel ratio Bardaie MZ, Jaius R. (1984). Conversion of spark- Cp specific heat at constant pressure, Ignition engine for alcohol usage comparative kJ/kg K performance. Agricultural Mechanization in Cv specific heat at constant volume, kJ/kg Asia-Africa and Latin America 15, 31-40. K LHV lower heating value, kJ/kg Brettschneider, J. (1979). Berechung des m mass, kg luftverhaeltnisses λ von Luft-Kraftstoff- M molecular weight, g/mole Gemsichen und des Einflusses on Messfehlern n rotational speed, rpm auf λ, Bosh Technische Berichte, Band 6, Heft Nt theoretical power, W 4, Seite 177-186, Stuttgat Ncyl cylinder number Q heat supplied, J Brettschneider, J. (1997). Extension of the r compression ratio Equation for Calculation of the Air-Fuel R ideal gas constant, kJ/kg K Equivalence Ratio, SAE Technical Paper T temperature, K 972989, doi:10.4271/972989. w specific work, J/kg Camarillo, J.A. (2011). Estudio de la combustion´ de W work, J un motor monocil´ındrico de ignicion´ alimentado x mole fraction of ethanol in the flexible con mezclas gasolina-etanol anhidro e hidratado fuel a distintas concentraciones, Master Thesis, Greek symbols Universidad Veracruzana, Mexico.´ γ ratio between Cp and Cv η efficiency Cengel, Y.A. and Boles, M.A. (2012). τ torque, N m Termodinámica, 7th ed., editorial Mc Graw Hill, ω mass fraction of ethanol in the flexible Nevada, USA, 491. fuel Subscripts: Chavez-Guerrero,´ L., Hinojosa-Rivera, M., Medina- 1 − 4 thermodynamic states of the cycle Lott, B., Vannier, R.N., Ringuede,´ A. and Cassir, a air M. (2017). Synthesis and characterization of c combustion Co-doped Lanthanum Nickelate perovskites for e ethanol soid oxide fuel cell cathode material. Revista f fuel Mexicana de F´ısica63, 6-11. g gasoline Chen, L., Stone, R. and Richardson, D. (2012). A n net study of mixture preparation and PM emissions opt optimum using a direct injection engine fueled with t total stoichiometric gasoline/ethanol blends. Fuel 96, 120-130. Acknowledgment Clements, R.M. and Smy, P.R. (1976). The variation of ionization with air/fuel ratio for a spark- The financial support of this study by the Science ignition engine. Journal of Applied Physics 47, research grant SIP20121350, 20130558 and Project 505-509. SIP20160812, by the National Polytechnic Institute of Mexico (IPN), is greatly appreciated. A. Gomez´ also Cordeiro de Melo, T.C., Machado, G.B., Belchior, acknowledges the financial support from the National C.R.P., Colaco, M.J., Barros, J.E.M., de Council for Science and Technology (CONACYT) of Oliveira, E.J. and de Oliveira, D.G. (2012). Mexico. Hydrous ethanol-gasoline blends-Combustion

www.rmiq.org 1073 G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

and emission investigations on a Flex-Fuel Mendoza, D., Santiago, P. and Reyes-Perez,´ E. engine. Fuel 97, 796-804. (2006). Carbon nanotubes produced from hexane and ethanol. Revista Mexicana de F´ısica Costa, R.C. and Sodre,´ J.R. (2010). Hydrous ethanol 52, 1-5. vs. gasoline-ethanol blend: Engine performance and emissions. Fuel 89, 287-293. Miranda, E.N. (2003). What fuel for a rocket? Revista Mexicana de F´ısica49, 191-193. Deadmore D. L., Lowell C. E. and Kohl F. J. (1979). The effect of fuel to air ratio on burner rig hot Murali, M.V.S.,Kishor,K., Gupta,A.V.S.,Murthy,P.V.K. corrosion. Corrosion Science 19, 371-378. and Narasimha, N. (2012). Performance of copper coated two stroke spark ignition engine Desoky A.A, Rabie L.H. (1983). Comparative with methanol-blended gasoline with catalytic performance study of spark ignition engine converter. API Journal of Renewable and burning alcohols, gasoline and alcohol-gasoline Sustainable Energy 4, 013102. blend, in: 6th Miami International Conference on Alternative Energy Sources of Condensed Ohsuga M. and Ohyama Y. (1986). Astudy on the Papers, 12-14 December 1983, Miami Beach, oxygen-biased wide range air-fuel ratio sensor FL, USA, pp.331. for rich and lean air-fuel ratios. Sensors and Actuators 9, 287-300 Gomez,´ A. (2014). Determinacion´ y analisis´ de la ecuacion´ para la relacion´ aire/combustible de la Orbital Engine Company. (2002). A Literature mezcla etanol-gasolina. Master Thesis, Instituto Review Based Assessment on the Impacts of Politecnico´ Nacional, Mexico.´ a 20% Ethanol Gasoline Fuel Blend on the Australian Vehicle Fleet. Report to Environment Gonca, G. and Dobrucali, E. (2016). The effects Australia 1. of engine design and operating parameters on Palmer F.H. (1986). Vehicle performance of gasoline the performance of a diesel engine fueled containing oxygenates. MI MechE C319/86:33- with diesel-biodiesel blends. AIP Journal of 46 Renewable and Sustainable Energy 8, 025702. Polymeropoulos C. E. and Sernas V. (1977). Hamdan M. A. and Jubran B. A. (1986). The effect Measurement of droplet size and Fuel-air ratio of ethanol addition on the performance of diesel in sprays. Combustion and Flame 29, 123-131. and gasoline engines. Dirasat 13, 229-244 Pulkrabek, W.W. (2003). Engineering Fundamentals Hernandez,´ M., Menchaca, L. and Mendoza, A. of the Internal Combustion Engine, 2nd ed. Pp (2014). Fuel economy and emissions of light- 35-103.Prentice Hall, England. duty vehicles fueled with ethanol-gasoline blends in a Mexican City. Renew Energy 72, Rigatos, G., Siano, P. and Arsie, I. (2014). 236-242. Flatness-based Embedded Control of Air-fuel Ratio in Combustion Engines. AIP Conference Heywood, J.B. (1988). Internal Combustion Proceedings 1618, 251-259. Engine Fundamentals, 1st ed., (Mc Graw-Hill, Massachusetts, USA. 62-92. Rocha-Mart´ınez, J.A., Navarrete Gonzalez,T.D.,Pav´ ´ıa Miller,C.G.,Paez-Hern´ andez,´ R. and Angulo- Kasseris, E.P. (2011). Knock limits in spark ignited Brown, F. (2002). Otto and Diesel engine direct injected engines using gasoline/ethanol models with cyclic variability. Revista Mexicana blends, PhD Thesis, MIT. de F´ısica48, 228-234. Klein, S.A. (2015). Engineering Equation Solver Schifter, I., Diaz, L., Rodriguez, R., Gomez,´ J.P. and © 1992-2015, Professional V9.939 (9/02/15), Gonzalez, U. (2011). Combustion and emissions License #3751. behavior for ethanol-gasoline blends in a single cylinder engine. Fuel 90, 3582-3592. Mantilla, J.M. (2010). Modelado de la combustion´ de mezclas gasolina-etanol en motores de Szybist, J. (2010). Investigation of Knock Limited combustion´ interna. Ph.D. Thesis. Universidad Compression Ratio of Ethanol Gasoline Blends. Nacional de Colombia, Colombia. SAE Technical Paper 1, 619.

1074 www.rmiq.org G´omezet al./ Revista Mexicana deIngenier ´ıaQu ´ımica Vol. 16, No. 3 (2017) 1065-1075

Yao, Y. Ch., Tsai, J.H. and Wang, I.T. (2013). compression ratios. (2006). Applied Thermal Emissions of gaseous pollutant from motorcycle Engineering 26, 2272-2278. powered by ethanol-gasoline blend. Applied Energy 102, 93-100. Zhang H. F., Seo K. and Zhao H. (2013), Combustion and emission analysis of the direct DME Yucesu,¨ H.S., Topgul,T.,¨ Cinar, V. and Okur, injection enabled and controlled auto-ignition M.Eﬀect of ethanol-gasoline blends on engine gasoline combustion engine operation. Fuel performance and exhaust emissions in diﬀerent 107, 800-814.