E ﬀ ects of Diethyl Ether Introduction in Emissions and Performance of a Diesel Engine Fueled with Biodiesel-Ethanol Blends

: Biofuels provide high oxygen content for combustion and do modify properties that inﬂuence the engine operation process such as viscosity, enthalpy of vaporization, and cetane number. Some requirements of performance, fuel consumption, e ﬃ ciency, and exhaust emission are necessary for the validation of these biofuels for application in engines. This work studies the e ﬀ ects of the use of diethyl ether (DEE) in biodiesel-ethanol blends in a DI mechanical diesel engine. The blends used in the tests were B80E20 (biodiesel 80%-ethanol 20%) and B76E19DEE5 (biodiesel 76%-ethanol 19%-DEE 5%). Fossil diesel (D100) and biodiesel (B100) were evaluated as reference fuels. The results revealed similar engine e ﬃ ciencies among tested fuels at all loads. The use of B100 increased CO and NO x and decreased THC compared to D100 at the three loads tested. B80E20 fuel showed an increase in NO x emission in comparison with all fuels tested, which was attributed to higher oxygen content and lower cetane number. THC and CO were also increased for B80E20 compared to B100 and D100. The use of B76E19DEE5 fuel revealed reductions in NO x and CO emissions, while THC emissions increased. The engine e ﬃ ciency of B76E19DEE5 was also highlighted at intermediate and more elevated engine load conditions. Formal analysis, M.C., J.S., V.F., F.T., J.M. and E.T.; Funding acquisition, E.T.; Investigation, M.C.; Methodology, M.C., J.S., J.M. and V.F.; Project administration, M.C.; Resources, M.C. and E.T.; Supervision, J.M., V.F. and E.T.; Validation, M.C. and V.F.; Writing, M.C., J.M., V.F. and F.T.; Writing—review & editing, M.C.,


Introduction
The growing concern over climate change and fossil fuel dependency has increased visibility for renewable energy sources [1]. Biofuels have been identified as promising renewable fuels, in particular, when originated from waste feedstock and non-edible plant species [2].
The use of alternative fuels in compression ignition (CI) engines should be evaluated in many terms, as exhaust emissions, fuel stability, availability, distribution, and impacts on engine durability [3]. blend (5% of esterified cotton seed methyl ester and 95% of neat diesel) decreased NO x and unburned hydrocarbon emissions at lower to medium speed range and increased thereafter.
The research work with alternative fuels focused on optimizing the fuel blend based on fuel properties like kinematic viscosity, density, cloud point, and pour point.
DEE appears as a potential additive to diesel-biodiesel-ethanol blends among oxygenated biofuels due to its chemical properties such as high cetane number (>125), moderate energy density (similar to biodiesel), high oxygen content in its structure, low self-ignition temperature, prolonged flame duration, and adequate miscibility with diesel, biodiesel, and ethanol. In this context, DEE can be considered more suitable to be used for CI engines application than ethanol or methanol due to its higher CN and LHV [16]. Some research has been developed using DEE to improve ignition quality in blends composed by diesel and other feedstocks fuels. Lee and Kim [17] evaluated blends of diesel and DEE and results showed similar engine efficiency compared to diesel along with lower emissions of THC, CO, and PM. However, NO x emissions were higher. The authors attributed the results to the shorter ignition delay and high oxygen content of the blends. Qi et al. [18] investigated the effects of ethanol and DEE as additives in a diesel-biodiesel blend. The tested fuels were B30 (30% biodiesel and 70% diesel), BE-1 (5% diethyl ether, 25% biodiesel, and 70% diesel) and BE-2 (5% ethanol, 25% biodiesel, and 70% diesel). Reduction in smoke and CO were observed with BE-1 and BE-2, while NO x was higher for BE-2 and HC was higher for BE-1 and BE-2 when compared to B30. Jeevanantham et al. [19] used DEE in blends with diesel-biodiesel in volumetric proportion of 5% (D50B45DEE5) and 10% (D50B40DEE10). Results showed that NO x emissions were reduced in all test conditions concerning the other fuels. The authors attributed the results to the significant cooling effect caused by the high enthalpy of vaporization of DEE. This work approaches the application of DEE in biodiesel-ethanol blends. The DEE was used toward enhancing the CN, which was deteriorated as a result of ethanol introduction. In the literature, it is common to use blends with DEE, however, most part considering diesel in its composition. This work evaluates the performance and emissions of a light-duty naturally aspirated diesel engine with mechanical fuel injection using biodiesel, ethanol, and DEE. DEE is used as an additive to increase the CN, which is deteriorated as a result of ethanol introduction. The tests were developed in a light-duty naturally aspirated diesel engine with mechanical fuel injection, evaluate the performance, efficiency, and exhaust emissions.

Experimental Setup
The engine used in the experiments was a four-stroke, two cylinders, direct fuel injected, and naturally aspirated diesel engine. The main characteristics of the engine are summarized in Table 1. The engine original specifications were kept constant throughout the tests. Figure 1 shows a schematic diagram of the experimental setup that consists of a diesel engine coupled with a hydraulic dynamometer (Shenck D210), a gas analyzer for exhaust emission, HC meter, fuel tank, scale, high-pressure fuel pump, and K-type thermocouples for temperature measurement. Experimental tests were performed under three loads: 2.7 kW, 5.4 kW, and 8.1 kW, and engine speed was fixed at 1700 rpm. These loads correspond to 25%, 50%, and 75% of the maximum brake power (BP) at the engine speed of 1700 rpm, using mineral diesel as a reference fuel. These loads were chosen since engine maximum load is not achievable with blends of reduced LHV.
The amount of fuel injected varied for each fuel tested due to the difference in LHV between the fuels (Table 2). This control was performed by varying the acceleration in the high-pressure fuel injection pump.

Fuels Characteristics
Evaluated blends were prepared using soybean biodiesel (B100), ethanol (purity of 99.3%), and diethyl-ether (DEE, purity of 99.7%). Diesel fossil pure (D100, maximum 10 ppm of sulfur) and pure biodiesel were tested as reference. The main characteristics of the fuels are depicted in Table 3. An IKA C2000 bomb calorimeter was employed to determine the lower heating value based on the standard ASTM-D240-87. The fuel density was measured using a DMA-5000 densimeter, while samples viscosity was measured by a model P capillarity viscometer. Table 2 shows the composition Experimental tests were performed under three loads: 2.7 kW, 5.4 kW, and 8.1 kW, and engine speed was fixed at 1700 rpm. These loads correspond to 25%, 50%, and 75% of the maximum brake power (BP) at the engine speed of 1700 rpm, using mineral diesel as a reference fuel. These loads were chosen since engine maximum load is not achievable with blends of reduced LHV.
The amount of fuel injected varied for each fuel tested due to the difference in LHV between the fuels (Table 2). This control was performed by varying the acceleration in the high-pressure fuel injection pump.

Fuels Characteristics
Evaluated blends were prepared using soybean biodiesel (B100), ethanol (purity of 99.3%), and diethyl-ether (DEE, purity of 99.7%). Diesel fossil pure (D100, maximum 10 ppm of sulfur) and pure biodiesel were tested as reference. The main characteristics of the fuels are depicted in Table 3. An IKA C2000 bomb calorimeter was employed to determine the lower heating value based on the standard ASTM-D240-87. The fuel density was measured using a DMA-5000 densimeter, while samples viscosity was measured by a model P capillarity viscometer. Table 2 shows the composition of single fuels, biodiesel-ethanol, and biodiesel-ethanol-DEE along with their volumetric compositions and LHV.
The LHV was predicted based on the volumetric fractions, density, and energy fractions of each blend component. This methodology was also used by Al-Esawi, Al Qubeissi, and Kolodnytska [9].
The stability of the B76E19DEE5 blend was evaluated under the test temperatures to certify that no fuel would be lost by evaporation.

Instrumentation
Fuel mass flow rate was obtained by gravimetric method using a digital scale while fuel consumption was evaluated under five cycles of measurement with a sampling time of 30 minutes. Exhaust gas emissions of CO, NO x , and THC were assessed using two gas analyzers under an average of 30 measurements for each fuel. The main characteristics and uncertainty of the employed instruments are shown in Table 4. In order to enhance the confidence level of the experiments, the tests were carried out under similar system conditions for all fuels, such as engine temperature and weather temperature between (29 ± 2 • C) and humidity (55 ± 8%).

Brake Specific Fuel Consumption (BSFC)
Brake specific fuel consumption may be defined as a ratio between fuel mass flow rate and engine brake power, which usually depends on the volumetric fuel injection system and fuel properties (e.g., density, calorific value, and viscosity). Figure 2 outline an increase in BSFC when the engine operates with biofuels compared to mineral diesel (D100), which may be explained by the lower LHV of biofuels blends, as shown in Table 2.
It is also observed that as loads increases, BSFC decreases for all fuels due to the higher engine efficiency at high engine loads [20].
When engine operated with B100 there was an increase in BSFC of about 14% on average, under the three loads, in comparison with D100, while the difference in LHV was only 12% lower for B100 in comparison with D100. Thus, it is possible to infer a slight reduction in engine efficiency when B100 is used. Besides, the higher viscosity of B100 leads to an increase in BSFC. Higher viscosity alters fuel jet atomization, increasing the average diameter and density of fuel droplets while reduces combustion efficiency [21]. When compared to B100, the biodiesel-ethanol (B80E20) blend has shown an increase in BSFC of 5.3%, 4.2%, and 3% at loads of 2.7 kW, 5.4 kW, and 8.1 kW, respectively. The further increase of fuel consumption of B80E20 is due to a reduction in LHV of about 4.4% for the blend, which occurred as a result of the introduction of ethanol, which explains the increase in BSFC. Furthermore, a greater difference can be observed at lower loads due to lower combustion temperatures [22], whereas the high ethanol enthalpy of vaporization induces a reduction in combustion temperature and thus combustion efficiency decreases. A slight decrease in BSFC was observed in the blend with diethyl ether (B76E19DEE5) in concerning to B80E20 blend, under all tested conditions. This reduction can be attributed to the short increment in LHV of about 1.2%. However, other factors also contribute to improving combustion quality for this blend, such as higher CN and the high oxygen content [8]. It is also observed that as loads increases, BSFC decreases for all fuels due to the higher engine efficiency at high engine loads [20].
When engine operated with B100 there was an increase in BSFC of about 14% on average, under the three loads, in comparison with D100, while the difference in LHV was only 12% lower for B100 in comparison with D100. Thus, it is possible to infer a slight reduction in engine efficiency when B100 is used. Besides, the higher viscosity of B100 leads to an increase in BSFC. Higher viscosity alters fuel jet atomization, increasing the average diameter and density of fuel droplets while reduces combustion efficiency [21]. When compared to B100, the biodiesel-ethanol (B80E20) blend has shown an increase in BSFC of 5.3%, 4.2%, and 3% at loads of 2.7 kW, 5.4 kW, and 8.1 kW, respectively. The further increase of fuel consumption of B80E20 is due to a reduction in LHV of about 4.4% for the blend, which occurred as a result of the introduction of ethanol, which explains the increase in BSFC. Furthermore, a greater difference can be observed at lower loads due to lower combustion temperatures [22], whereas the high ethanol enthalpy of vaporization induces a reduction in combustion temperature and thus combustion efficiency decreases. A slight decrease in BSFC was observed in the blend with diethyl ether (B76E19DEE5) in concerning to B80E20 blend, under all tested conditions. This reduction can be attributed to the short increment in LHV of about 1.2%. However, other factors also contribute to improving combustion quality for this blend, such as higher CN and the high oxygen content [8].

Brake Specific Energy Consumption (BSEC)
Another way of evaluating the performance of the fuels tested is by using the BSEC. While BSFC represents a measure of fuel mass consumption, BSEC allows a view of the energy consumption of the fuels, in accordance with the engine brake power. The results of the BSEC obtained in the experiments are presented in Figure 3.

Brake Specific Energy Consumption (BSEC)
Another way of evaluating the performance of the fuels tested is by using the BSEC. While BSFC represents a measure of fuel mass consumption, BSEC allows a view of the energy consumption of the fuels, in accordance with the engine brake power. The results of the BSEC obtained in the experiments are presented in Figure 3.   Figure 3 shows relatively small differences between the fuel samples tested. This similarity of results shows that the performance of the proposed fuel blends is consistent with the conventional diesel of the engine design. Considering the test conditions, in the lowest load, 2.7 kW, the lowest BSEC value can be observed with the D100 and the highest value for the B80E20 fuel. This can be attributed to the cooling effect of the chamber and, therefore, to the increase of the delay of the start of the ignition using B80E20. These factors are due to the high enthalpy of evaporation and low CN of the blend B80E20, which has more influence at the low load condition due to the lower temperatures in the chamber wall and in the residual gases of the combustion [12,15]. Regarding the D100, its better results can be explained by adequate properties of the D100 in accordance with the design parameters of the diesel engine, providing adequate conditions for burning [20].  Figure 3 shows relatively small differences between the fuel samples tested. This similarity of results shows that the performance of the proposed fuel blends is consistent with the conventional diesel of the engine design. Considering the test conditions, in the lowest load, 2.7 kW, the lowest BSEC value can be observed with the D100 and the highest value for the B80E20 fuel. This can be attributed to the cooling effect of the chamber and, therefore, to the increase of the delay of the start of the ignition using B80E20. These factors are due to the high enthalpy of evaporation and low CN of the blend B80E20, which has more influence at the low load condition due to the lower temperatures in the chamber wall and in the residual gases of the combustion [12,15]. Regarding the D100, its better results can be explained by adequate properties of the D100 in accordance with the design parameters of the diesel engine, providing adequate conditions for burning [20].
At the highest load, 8.1 kW, an improvement of the BSEC was observed using B80E20 with respect to D100. This can be explained due to the higher temperatures in the chamber and in the residual gases of combustion in that condition. In addition, the fuel oxygen content improves the fuel-burning quality [8,12,15].
Considering B76E19DEE5, the presence of DEE reduced the BSEC compared to B80E20 in all test conditions. At the 8.1 kW load, B76E19DEE5 presented the lowest BSEC with respect to the other fuels tested. These results can be attributed to the improvement of the burning process due to the oxygen content and the elevation of the CN of the blend with DEE [16][17][18].

Engine Efficiency
Engine efficiency may be defined as the useful energy output of the engine as a function of fuel heat energy. As shown in Figure 4, higher loads present increased engine efficiency as a result of a more adequate air/fuel ratio and enhanced mechanical efficiency [20].  A slight reduction of about 0.7% on average, under all loads, was observed in engine efficiency while using B100 with respect to D100. These variations can be attributed to differences in viscosity and density, which influence atomization and decrease combustion efficiency [21]. This can be confirmed by a more significant reduction in BSFC in comparison to that observed for the LHVs when B100 replaces D100. The results are consistent with some studies in the literature [22][23][24][25] despite other studies show increased efficiency while the engine operates with blends of biodiesel and diesel. It can be discussed that when biodiesel raw material presents low viscosity, the oxygen in the biodiesel molecule and the higher cetane number than diesel can improve the combustion characteristic [26][27][28]. On the other hand, addition of ethanol to the blend decreases viscosity, LHV, and CN in comparison with B100, although oxygen content increases. When related to B100, the blend B80E20 presents an increase in efficiency of 0.4% and 1.6% at 5.4 kW and 8.1 kW loads, respectively. The lower viscosity of B80E20 improves atomization and provides better evaporation and air-fuel mixing, hence improving combustion efficiency. In addition, higher oxygen content causes faster combustion [7,25,29]. At 2.7 kW, similar efficiencies values were observed. Introduction of diethyl ether (B76E19DEE5) revealed a slight increase of 0.5% in engine efficiency under 2.7 kW and 5.4 kW of engine loads in comparison with B80E20 fuel, whereas under 8.1 kW the increase was about 2%. DEE A slight reduction of about 0.7% on average, under all loads, was observed in engine efficiency while using B100 with respect to D100. These variations can be attributed to differences in viscosity and density, which influence atomization and decrease combustion efficiency [21]. This can be confirmed by a more significant reduction in BSFC in comparison to that observed for the LHVs when B100 replaces D100. The results are consistent with some studies in the literature [22][23][24][25] despite other studies show increased efficiency while the engine operates with blends of biodiesel and diesel. It can be discussed that when biodiesel raw material presents low viscosity, the oxygen in the biodiesel molecule and the higher cetane number than diesel can improve the combustion characteristic [26][27][28]. On the other hand, addition of ethanol to the blend decreases viscosity, LHV, and CN in comparison with B100, although oxygen content increases. When related to B100, the blend B80E20 presents an increase in efficiency of 0.4% and 1.6% at 5.4 kW and 8.1 kW loads, respectively. The lower viscosity of B80E20 improves atomization and provides better evaporation and air-fuel mixing, hence improving combustion efficiency. In addition, higher oxygen content causes faster combustion [7,25,29]. At 2.7 kW, similar efficiencies values were observed. Introduction of diethyl ether (B76E19DEE5) revealed a slight increase of 0.5% in engine efficiency under 2.7 kW and 5.4 kW of engine loads in comparison with B80E20 fuel, whereas under 8.1 kW the increase was about 2%. DEE addition to the blend promotes a reduction in viscosity, which facilitates fuel atomization, and due to the higher CN of diethyl ether, it decreases the ignition delay [18].

NO x Emissions
NO x emissions for the studied blends compared to D100 and B100 are shown in Figure 5. It has been widely discussed that as loads increases, NO x exhaust emission increases due to higher temperatures in the combustion chamber that increases the NO x formation by thermal mechanism [30][31][32]. B100 showed minor increases in NOx emissions in comparison with fossil diesel. This increase can be explained primarily by the oxygen content in the biodiesel molecule, which increases oxygen availability and combustion temperatures [33]. Besides, the higher bulk modulus of biodiesel can advance its injection increasing the preparation time of the fuel-air mixture, which increases combustion pressure and temperature, thus, contributing to NOx formation [33]. B80E20 fuel resulted in an increase in NOx emissions compared to the other tested fuels. These results can be attributed to the lower CN, which prolongs the premix formation time and increases heat release ratio (HRR) in the initial phase of combustion, thus higher temperatures and NOx emission are obtained [34,35]. Furthermore, the high oxygen content of ethanol (Table 2) contributes to NOx formation [8]. The high oxygen content of the fuel has a positive effect on the combustion since the oxygen presented in the fuel is more active when compared to the molecular oxygen contained in the air [36]. Even though higher oxygen content produces greater combustion efficiency, higher temperatures increase NOx formation [37,38].
The use of DEE (B76E19DEE5) revealed a significant reduction in NOx emission. In comparison with the B80E20 blend, the reduction was approximately 6%, 13%, and 3%, at 2.7 kW, 5.4 kW, and 8.1 kW engine loads, respectively. Besides, under 5.4 kW and 8.1 kW, the B76E19DEE5 blend has shown the lowest NOx emission among all fuels tested. These results can be attributed to two main factors that are mainly related to fuel characteristics. Firstly, the higher CN, which decreases the ignition delay and the premix formation time, thus reducing the combustion pressure and temperature peaks. Furthermore, the enthalpy of vaporization of the blend is also elevated, which reduces combustion temperature, thus, reducing NOx emissions [8]. B100 showed minor increases in NO x emissions in comparison with fossil diesel. This increase can be explained primarily by the oxygen content in the biodiesel molecule, which increases oxygen availability and combustion temperatures [33]. Besides, the higher bulk modulus of biodiesel can advance its injection increasing the preparation time of the fuel-air mixture, which increases combustion pressure and temperature, thus, contributing to NO x formation [33]. B80E20 fuel resulted in an increase in NO x emissions compared to the other tested fuels. These results can be attributed to the lower CN, which prolongs the premix formation time and increases heat release ratio (HRR) in the initial phase of combustion, thus higher temperatures and NO x emission are obtained [34,35]. Furthermore, the high oxygen content of ethanol (Table 2) contributes to NO x formation [8]. The high oxygen content of the fuel has a positive effect on the combustion since the oxygen presented in the fuel is more active when compared to the molecular oxygen contained in the air [36]. Even though higher oxygen content produces greater combustion efficiency, higher temperatures increase NO x formation [37,38].
The use of DEE (B76E19DEE5) revealed a significant reduction in NO x emission. In comparison with the B80E20 blend, the reduction was approximately 6%, 13%, and 3%, at 2.7 kW, 5.4 kW, and 8.1 kW engine loads, respectively. Besides, under 5.4 kW and 8.1 kW, the B76E19DEE5 blend has shown the lowest NO x emission among all fuels tested. These results can be attributed to two main factors that are mainly related to fuel characteristics. Firstly, the higher CN, which decreases the ignition delay and the premix formation time, thus reducing the combustion pressure and temperature peaks. Furthermore, the enthalpy of vaporization of the blend is also elevated, which reduces combustion temperature, thus, reducing NO x emissions [8]. Figure 6 shows the CO emission for the fuels tested. CO emissions increased for higher engine load due to lower air-fuel ratios presented in the combustion phase [37]. The use of B100 increases CO emissions compared to D100 in about 20%. This may be attributed to the lower air-fuel ratio and poor atomization that results due to biodiesel higher viscosity [37,38]. Some studies also indicate a reduction in CO emissions when the engine operates with biodiesel [27,31] as a result of the higher oxygen concentration. Blending ethanol with biodiesel (B80E20) considerably increased CO emission in comparison with B100 under all loads up to 35%. This is attributed to the reduction in combustion temperature due to the high enthalpy of vaporization of ethanol [8]. This result is consistent with the result found in Ç elik et al. [7]. The addition of DEE to biodiesel-ethanol blend produced a slight decrease in CO emission in comparison with B80E20, particularly under medium and high engine load. The higher cetane number and increased oxygen content of the B76E19DEE5 blend improve combustion efficiencies and decrease CO emission [39][40][41]. At 2.7 kW, however, a slight increase was verified and has been attributed due to the decrease in combustion temperature as a result of both alcohol and DEE cooling effect into the blend.

THC Emissions
Total hydrocarbons emissions of the three engine operating loads (Figure 7) with blends of B80E20 and B76E19DEE5 were higher than those of D100 and B100. The use of B100 increases CO emissions compared to D100 in about 20%. This may be attributed to the lower air-fuel ratio and poor atomization that results due to biodiesel higher viscosity [37,38]. Some studies also indicate a reduction in CO emissions when the engine operates with biodiesel [27,31] as a result of the higher oxygen concentration. Blending ethanol with biodiesel (B80E20) considerably increased CO emission in comparison with B100 under all loads up to 35%. This is attributed to the reduction in combustion temperature due to the high enthalpy of vaporization of ethanol [8]. This result is consistent with the result found in Çelik et al. [7]. The addition of DEE to biodiesel-ethanol blend produced a slight decrease in CO emission in comparison with B80E20, particularly under medium and high engine load. The higher cetane number and increased oxygen content of the B76E19DEE5 blend improve combustion efficiencies and decrease CO emission [39][40][41]. At 2.7 kW, however, a slight increase was verified and has been attributed due to the decrease in combustion temperature as a result of both alcohol and DEE cooling effect into the blend.

THC Emissions
Total hydrocarbons emissions of the three engine operating loads (Figure 7) with blends of B80E20 and B76E19DEE5 were higher than those of D100 and B100.
The increase of THC emissions when ethanol is blended with diesel and biodiesel has been previously reported in the literature [35,42]. Emissions decreased with increasing engine loads for all tested fuels. This is due to the higher combustion temperature at higher loads that enhances combustion efficiency and thus reduces unburnt THC [43]. The increase of THC emissions when ethanol is blended with diesel and biodiesel has been previously reported in the literature [35,42]. Emissions decreased with increasing engine loads for all tested fuels. This is due to the higher combustion temperature at higher loads that enhances combustion efficiency and thus reduces unburnt THC [43].
The use of B100 shows significant reductions in THC emissions by 45%, 31%, and 38% at 2.7 kW, 5.4 kW, and 8.1 kW loads, respectively, in comparison with D100. Among the factors responsible for the decrease in hydrocarbon emission is the higher CN and oxygen content of B100 since the oxygen contained in the fuel provides cleaner and more complete combustion related to D100. Besides, biodiesel combustion starts earlier in the combustion chamber due to its higher bulk modulus providing more time for mixture formation [44]. Similar results were previously reported in the literature [45,46]. B80E20 fuel shows an increase in THC emission compared to B100 and D100. Although the oxygen content in ethanol could increase the combustion quality, other properties have the opposite influence, such as lower CN, lower LHV, and higher enthalpy of vaporization. The higher enthalpy of vaporization reduces combustion temperature, which influences the oxidation rate and THC formation [8,47]. The lower CN increases the ignition delay and decreases the total combustion time, which can increase the THC formation [45]. A reduction in LHV due to ethanol addition leads a higher amount of fuel injected per cycle, which also may favor THC formation [8].
The addition of diethyl ether to biodiesel-ethanol blend (B76E19DEE5) increased THC emission in comparison with the B80E20 blend. This result may be attributed to the high enthalpy of vaporization and high volatility of the blend when DEE was introduced, which reduces combustion temperature related to B80E20 [35,48].

Conclusions
Emissions, fuel consumption, and efficiency of a DI mechanical diesel engine coupled to a hydraulic dynamometer were measured using fossil diesel, biodiesel, biodiesel-ethanol (B80E20), and biodiesel-ethanol-diethyl ether (B76E19DEE5). Although BSFC increased with the use of biofuels due to the reduction in LHV, engine efficiencies increased with the use of B80E20 and B76E19DEE5, especially at medium and high loads. The higher oxygen content was appointed as the main reason for that improvement. In the case of the B76E19DEE5 fuel, the CN increased in comparison with B80E20, which further enhance the engine efficiency. Considering BSEC, the similarity of results The use of B100 shows significant reductions in THC emissions by 45%, 31%, and 38% at 2.7 kW, 5.4 kW, and 8.1 kW loads, respectively, in comparison with D100. Among the factors responsible for the decrease in hydrocarbon emission is the higher CN and oxygen content of B100 since the oxygen contained in the fuel provides cleaner and more complete combustion related to D100. Besides, biodiesel combustion starts earlier in the combustion chamber due to its higher bulk modulus providing more time for mixture formation [44]. Similar results were previously reported in the literature [45,46]. B80E20 fuel shows an increase in THC emission compared to B100 and D100. Although the oxygen content in ethanol could increase the combustion quality, other properties have the opposite influence, such as lower CN, lower LHV, and higher enthalpy of vaporization. The higher enthalpy of vaporization reduces combustion temperature, which influences the oxidation rate and THC formation [8,47]. The lower CN increases the ignition delay and decreases the total combustion time, which can increase the THC formation [45]. A reduction in LHV due to ethanol addition leads a higher amount of fuel injected per cycle, which also may favor THC formation [8].
The addition of diethyl ether to biodiesel-ethanol blend (B76E19DEE5) increased THC emission in comparison with the B80E20 blend. This result may be attributed to the high enthalpy of vaporization and high volatility of the blend when DEE was introduced, which reduces combustion temperature related to B80E20 [35,48].

Conclusions
Emissions, fuel consumption, and efficiency of a DI mechanical diesel engine coupled to a hydraulic dynamometer were measured using fossil diesel, biodiesel, biodiesel-ethanol (B80E20), and biodiesel-ethanol-diethyl ether (B76E19DEE5). Although BSFC increased with the use of biofuels due to the reduction in LHV, engine efficiencies increased with the use of B80E20 and B76E19DEE5, especially at medium and high loads. The higher oxygen content was appointed as the main reason for that improvement. In the case of the B76E19DEE5 fuel, the CN increased in comparison with B80E20, which further enhance the engine efficiency. Considering BSEC, the similarity of results shows that the performance of the proposed fuel blend was similar with the conventional diesel fuel, used in the original engine design. However, at the 8.1 kW load, B76E19DEE5 presented the lowest BSEC in relation to the other fuels tested. This lower BSEC was consistent with the highest engine efficiency in the same condition.
The use of B100 increased NO x and CO emissions compared to D100 (1% and 20%, respectively, on average), which was attributed to the higher viscosity of biodiesel. On the other hand, THC emissions decreased with the use of B100 due to the oxygen content in biodiesel, which provides a cleaner and more complete combustion when compared with D100. Addition of ethanol (B80E20) showed an increase in NO x emissions compared to B100 as well as compared to the other fuels. The results were attributed to the higher oxygen content and lower CN in the mixture. Considerable increases in CO and THC emissions were attributed to the high enthalpy of vaporization and lower LHV of ethanol. Finally, addition of DEE to biodiesel-ethanol blend (B76E19DEE5) generated significant reductions in NO x emissions. In comparison with B80E20 fuel, the reductions (3% to 13%) were attributed to the higher CN and the higher enthalpy of vaporization of the blend. At 5.4 kW and 8.1 kW loads, NO x emissions were the lowest among all tested fuels. Regarding CO emissions, the results showed reductions (3% to 14%) in 5.4 kW and 8.1 kW loads, respectively, compared to B80E20 fuel. It was attributed to the higher CN and oxygen content of the blend. At 2.7 kW load, however, only a slight increase was verified. The THC emissions raised in comparison with B80E20 fuel up to 37% at the high engine load. The results were attributed to the higher enthalpy of vaporization and by the elevated oxygen content of the B76E19DEE5 blend. No problems with engine stability were noticed using the blends proposed. This could be verified by low variation in the engine speed and at the values of torque acquisition.
It was clear in this work the positive effects of adding DEE in moderate concentration to the ethanol biodiesel mixture, in which there were slight increases in engine efficiency and reductions in NO x emissions.