**3. Results and Discussion**

#### *3.1. Rheological Properties of Gasoline*/*Oil Double Blends, and Diesel*/*Gasoline*/*Oil Triple Blends.*

The major difference between diesel fuel and vegetable oils is the viscosity. In general, vegetable oils exhibit viscosity values in the range of 30–45 mm<sup>2</sup>/s or cSt, although castor oil has a much higher value of 227.0 cSt. For its part, fossil diesel exhibits values in the range of 2.5–6 cSt. The viscosity of the biofuel for being employed in a conventional diesel engine has to be in the range of 2.0 to 4.5 cSt (UNE EN ISO 3104). Therefore, viscosity is the essential parameter to modify in any vegetable oil, by mixing it with gasoline, for being employed in current diesel engines. In this respect, due to the importance of viscosity, its correct determination is critical to evaluate the quality of the biofuel. Thus, the viscosity, Cloud Point and Pour Point values of the different gasoline/oil blends studied are collected in Tables 1 and 2. As can be seen, an increase in the gasoline content in the blends promotes a decrease in their viscosity values, as well as a decrease in the Cloud and Pour Point of the double blends.

**Table 1.** Viscosity values at 40 ◦C (ASTM D2270-79), cloud point and pour point of the gasoline/sunflower oil blends. Errors are always calculated as the average of three measurements.



**Table 2.** Viscosity at 40 ◦C (ASTM D2270-79), cloud point and pour point values of the gasoline/castor oil blends. Errors are always calculated as the average of three measurements.

When sunflower oil is employed, optimum viscosity values for operating in compression ignition engines are achieved with the addition of 35–45% gasoline. In the case of castor oil, given its higher viscosity, it is necessary to incorporate a greater amount of gasoline in the blends (60–70%) to achieve an adequate viscosity value. Anyway, the gasoline/oil blends, in variable proportions, can be employed as a biofuel, even with oils which exhibit higher viscosity values, as is the case of castor oil. It must be highlighted that the presence of gasoline allows pure castor oil to be used without any chemical treatment, i.e., eliminating the transesterification to convert this oil in biodiesel and, therefore, considerably reducing the cost of the process. Be that as it may, the viscosity measurements indicate that it is possible to obtain biofuels with suitable rheological properties for being employed in diesel engines, complying with the European regulations EN 590, which stablish that viscosity at 40 ◦C must be in the range of 2.0–4.5 cSt.

Regarding the behavior of the double blends at low temperatures, the gasoline/oil blends began to solidify (Pour Point) at temperatures around −12 ◦C and they were completely frozen (Cloud Point) at −20 ◦C. This behavior is very similar to that exhibited by conventional fossil diesel, with temperatures around −10.0 ◦C and −18.5 ◦C for Pour Point and Cloud Point, respectively. However, the standard EN 590 classify the diesel fuel into two groups (with several subclasses) destined for specific climatic environments, depending on the geographical areas where it will be used. Therefore, the values of cloud point and pour point are in a wide range, between +5 and −34 ◦C.

As abovementioned, triple blends have been obtained by mixing different proportions of diesel with both a gasoline/sunflower oil blend with 40% of gasoline (Table 3) and also with a gasoline/castor oil blend with 60% of gasoline (Table 4). These triple blends are designated with the percentage that represents the corresponding gasoline/oil mixture, considering its biofuel character.

Hence, the B20 triple blend contains 80% of diesel, whereas the remaining 20% is a mix of gasoline and oil, in the proportions of the double blend previously studied. Therefore, with sunflower oil, since it is used in a double mixture containing 40% gasoline, the resulting triple blend corresponding to 80/8/12 diesel/gasoline/sunflower oil. In the case of castor oil, as the double mixture selected contains 60% of gasoline, the resulting triple blend corresponds to the proportion 80/12/8, diesel/gasoline/castor oil. The gasoline/oil double blends are represented as B100.


**Table 3.** Viscosity values of the triple blends, diesel/gasoline/sunflower oil, obtained by adding different amounts of fossil diesel to a double mixture of gasoline/sunflower oil, containing 40% gasoline.

**Table 4.** Viscosity values of the triple blends, diesel/gasoline/castor oil, obtained by adding different amounts of fossil diesel to a double mixture of gasoline/castor oil, containing 60% gasoline.


*3.2. Energy Performance and Pollutant Emissions Generated in a Diesel Engine Electric Generator, Fueled with Di*ff*erent Biofuel Blends*

As was expected, the triple blends of diesel/gasoline/sunflower oil and diesel/gasoline/castor oil exhibited adequate viscosity values to be employed as biofuels in a conventional diesel engine. Therefore, those blends were tested to ensure whether they present enough energy to guarantee the adequate performance of the engine. Furthermore, for comparative purposes, conventional diesel fuel was also employed as reference in the same conditions.

Thus, Figure 2 shows the power generated (Figure 2a) and the opacity in the smokes (Figure 2b) at different power demanded for the triple blends of diesel/gasoline/sunflower oil. In general, a stabilization of the power generated occurs between 3000 and 4000 W. In all the cases, at the highest value of power demanded (5000 W), a decrease in the power generated was observed. This generalized behavior could be explained by considering that the engine responds to the energy density of the fuel used, i.e., to the cetane index of the blend. Thus, the engine has a nominal capacity of 5000 W of electrical power, but this could only be achieved when the cetane number is 100, that is, when pure cetane is employed as fuel. In fact, if we apply the formula (generated power/power demanded) × 100 for the fossil diesel at the higher power-generated value (5000 W), we can estimate that its cetane number is around 52. This value is very close to the nominal value of the cetane number for fossil diesel (51), as it is collected in EN 590. Consequently, this could be a useful and simple procedure to calculate the cetane number of a fuel in a very approximate way.

**Figure 2.** Power generated (in Watts), based on the power demanded (in Watts) (**a**) and opacity (in Bacharach) as a function of the power demanded (**b**) by the triple blends of diesel/gasoline/sunflower oil.

Regarding the behavior of the triple blends with sunflower oil (Figure 2a), the B20, B40 and B60 blends exhibited a similar performance than fossil diesel. However, the blends with lower or no amount of diesel in the blend, i.e., B80 and B100, generate lower power values than fossil diesel, which is in agreemen<sup>t</sup> with their lower cetane number. Therefore, it seems that, at least, a minimum amount of diesel (40%) is required for a good performance of the triple blends.

According to the contamination results obtained for the blends of diesel/gasoline/sunflower oil (Figure 2b), it can be seen that for all the triple blends, the opacity generated from 2000 W onwards was lower than with fossil diesel. In fact, the higher the amount of oil, the lower the opacity generated, achieving a considerably reduced opacity with the blend B100. It must be taken into account that all the units have been expressed in units according to ASTM D 2156-94, Standard Test Method for Smoke Density in Flue Gases from Burning Distillate Fuels. The reduction in soot opacities, considerably lower than diesel for the B80 and B100 triple blends, could be partly attributed to the power loss exhibited by them. However, for the blends B20, B40 and B60, these reductions can be mainly attributed to the biofuel chemical properties, as the power generated is similar to fossil diesel.

Considering the blends composed by diesel/gasoline/castor oil, similar results of power generated than those obtained with their counterparts using sunflower oil were obtained for the blends B80 and B100, see Figures 2a and 3a. However, the good results of power generated obtained with B40 and B60, even better than fossil diesel, has to be highlighted, whereas the B20 blend performed in a similar way than diesel. Regarding the opacity generated for the blends with castor oil, see Figure 3b, independently on the power demanded from 1000 W onward, all the blends performed better than fossil diesel. Furthermore, in comparison to the behavior observed with the blends of diesel/gasoline/sunflower oil, their counterparts using castor oil exhibited lower opacity values. Be that as it may, independently of the vegetable oil employed, a significant reduction in opacity values has been obtained, mainly at medium and high demand (from 2000 W onward), being a reduction in the range of 20–50% less than that obtained with diesel.

On the other hand, when a biofuel is considered for being employed in a diesel engine, another important factor to be taken into account is the consumption at di fferent power demands. In this sense, the consumption of the di fferent biofuels employed have also been evaluated and the results are plotted in Figure 4.

**Figure 3.** Power generated (in Watts), based on the power demanded (in Watts) (**a**) and opacity (in Bacharach) as a function of the power demanded (**b**) by the triple blends of diesel/gasoline/castor oil.

As can be seen, at low power demand (1000 W), the consumption of the blends is always higher than that obtained with fossil diesel, independently of the oil employed in the blend. However, at the highest power demanded (5000 W), the opposite behavior is observed. In fact, at 5000 W of power demanded, only the double blends of gasoline/oil exhibited higher consumption than diesel (20% higher with sunflower and 10% with castor oil). Anyway, it can be observed that the blends obtained using castor oil exhibited a lower consumption than their counterparts employing sunflower oil.

**Figure 4.** *Cont.*

**Figure 4.** Consumption values as a function of the power demanded of the engine for the blends of diesel/gasoline/sunflower oil (**a**) and diesel/gasoline/castor oil (**b**).
