Over the decades, countries all over the world have been committed to adopting solutions to reduce dependence on fossil fuels and lessen their impact on the environment, especially the impact of internal combustion engines ^[1,2]. Biofuels, such as alcohols, vegetable oils, and biomass, can partially replace petroleum derivatives, thus attenuating environmental damage. From a socioeconomic perspective, the use of biofuels (e.g., biodiesel) has some advantages, such as their production in several areas, renewability, nontoxicity, and smaller amount of NOx emissions during the combustion process. Biofuels create jobs and income in their production chain and help reduce the use of fossil fuels ^[3].

Nevertheless, when only biodiesel is used, some physicochemical properties are affected, resulting in higher viscosity and higher fuel consumption, which reduces the thermal efficiency of engines, in addition to the high cost associated with biodiesel transesterification ^[4]. The solution proposed by several studies to improve the properties of diesel lies in its binary and ternary blend with other fuels, such as alcohols. In Brazil, ethanol is one of the most widely used alcohols in the industry because of the huge amount of feedstock and low production costs ^[2,5]. The low solubility of biodiesel, however, is a limitation for its use in compression ignition engines ^[6]. Nonetheless, the blends are advantageous, as they compensate for some properties such as the high viscosity of biodiesel, improving its flow rate and, consequently, fuel injection, thanks to the low viscosity of ethanol, thereby ensuring a more complete combustion reaction and reducing pollutant emissions ^[7].

To circumvent the problem with ethanol solubility, some studies have demonstrated different ways of blending ethanol into diesel or biodiesel, but some precautions and modifications in the engine are required ^[5]. The percentage of ethanol in the biodiesel-diesel blend should be kept to a minimum, with a purity level around 99% ^[8], not exceeding 5% by volume, given that higher percentages can cause loss of power. Other authors claim that the percentage can be kept to a maximum of 10% of the volume, unless solubility additives are used because of the miscibility between diesel-biodiesel and ethanol ^[9].

Regarding the different ratios of ethanol to diesel-ethanol-biodiesel blends, many authors have proposed to perform tests using different changes in operating parameters and blending ratios. Kim et al. ^[10] used diesel-biodiesel blends at concentrations of 3%, 5%, and 10% of anhydrous ethanol by volume in a four-cylinder compression ignition engine with direct and electronic fuel injection running at a slow load, which eventually decreased efficiency and increased specific fuel consumption (SFC) as the percentage of ethanol increased. According to those authors, when the percentage of ethanol increases, combustion occurs at a slower pace, increasing the ignition delay and affecting other operating parameters. Wu et al. ^[11] injected 10% and 30% of ethanol into a six-cylinder diesel engine equipped with an electronic injection system, causing a decrease in efficiency when compared to diesel. Ramachander et al. ^[12] blended 5% ethanol and 20% biodiesel into pure diesel, increasing SFC and reducing efficiency, thus affecting engine power. Alcohols have a low net calorific value (NCV) and lower viscosity as compared to diesel, which affects the blends, leading to an increase in SFC, because a larger fuel consumption is required to have the same power, and efficiency then tends to decrease.

Chockalingam et al. ^[13] blended 10% ethanol into diesel in a single-cylinder diesel engine and noticed that thermal efficiency was close to that of pure diesel. They also observed a reduction in exhaust fume output and an increase in Nox emissions when the engine was running at full load. However, the use of a stabilizer during the tests could explain their findings.

Some authors, such as Prbakaran and Viswanathan ^[14], found different results when higher percentages of ethanol were used in biodiesel blends. They observed a decrease in SFC and an increase in thermal efficiency for higher loads containing 50% ethanol and 50% biodiesel, in addition to an increase in the ignition delay and a high heat release rate. This resulted from the reduction of viscosity and density after the addition of ethanol, especially in the presence of high viscosity. Therefore, ethanol allows enhancing the saturation of the injected fuel, providing greater atomization and better air-fuel ratio and improving combustion, which ends up increasing thermal efficiency ^[2,14].

By drawing upon the studies cited hitherto, the present study aims to experimentally verify the consequences and feasibility of ethanol-biodiesel-diesel blend in a diesel generator set, using commercially available diesel blends (S-10) combined with 1%, 2% and 3% anhydrous ethanol, examining both energy and combustion parameters to demonstrate the pros and cons of using ethanol as an alternative energy source.

As the initial step, commercial diesel oil (S10) was used to obtain reference data (baseline). S10 contains 10% biodiesel, with a maximum sulfur content of 10 ppm. After obtaining the reference data, anhydrous ethanol (with a purity level of 99.90%) was directly blended with S-10 in 1 L beakers at the concentrations of 1%, 2% and 3% by volume, referred to as B10E1, B10E2, and B10E3, respectively. The chosen anhydrous ethanol ratios were employed to verify how the single-cylinder compression ignition generator set (Table 1 and Fig. 1) would behave at minimum concentrations when combined with commercial diesel fuel, considering that, according to the literature, previous tests had been conducted with blends containing percentages greater than 5% v/v ^[15-17]. Table 2 shows the properties of the fuels used.

Table 1

Manufacturer/Model	Branco/BD-6500 CFE
Engine	Single-cylinder, four-stroke, and air-cooled diesel engine
Type of fuel injection	Direct
Ignition system	Electronic
Fuel injection angle	22° APMS
Diameter (mm) x stroke (mm)	86 x 70
Engine capacity (cm3)	406
Compression ratio (-)	19:1
Revolutions per minute (rpm)	3600

Source: Manufacturer’s Manual.

Figure 1 Source: Figure produced by the authors.

Table 2

Properties	S-10 diesel	Anhydrous ethanol
Specific weight at 25 °C (kg/m3)	791	747
GCV (kJ/kg)	43,133.0	31,120.0
NCV (kJ/kg)	39,705.7	28,260.3
Viscosity at 40 °C (cP)	2.0 - 4.5	1.08
Cetane number	40 at least	8
Purity (%)	-	99
Latent heat of vaporization (kJ/kg)	250*	903*

Source: Table built by the authors; *Guedes, 2017.

The tests with B10E1, B10E2, and B10E3 fuels were performed without any modification to the engine and under the same operating conditions. A test using the 3% ethanol blend (B10E3*) was carried out to verify and compare performance after changes in the engine parameters, adjusting the injection pressure of the nozzle from 18 bar to 22 bar. Other fuel injection pressures were tested but, for operational reasons, only the 22 bar pressure was used.

The test time for each fuel amounted to 60 minutes, split into two parts: 1) stabilization and 2) data collection. The averages of 500 cycles were computed every 10 minutes, always checking the exhaust gas and lubricating oil temperature stabilization, fuel consumption, and airflow for approximately 30 minutes. Triplicates of the data were then obtained and used to calculate the averages.

The electrical quantities at the generator’s output, including effective electrical power (P_elec), were measured using a MULT-K Grafic device (Kron®, São Paulo/SP, Brazil). A resistor bank was used to generate different load values in relation to the total engine load. A test was conducted with S-10 diesel, applying different loads to verify the most efficient one, which would then be used in the blends. A load of approximately 79% of the maximum power (3.5 kW) was chosen for testing the blends.

An AVL Encoder 365C rotation sensor was installed on the crankshaft to obtain the crank angle (θ), top dead center (TDC), and engine speed. A GU21D piezoelectric pressure sensor was installed in the engine cylinder head near the intake valve, in addition to an Indimicro 602 charge amplifier, both manufactured by AVL.

The AVL indicom program was used to obtain the in-cylinder pressure curve for each tested fuel, and the start of combustion (SOC) was then determined using the pressure derivative method, given that this method contains fewer reading errors. The mass fraction burned (MFB) is calculated in relation to the net heat release rate. This rate is calculated from the integral heat release fraction, which relates the amount of net heat released at the crank angle to a percent mass of fuel burned ^[18,19], as shown in Eq. (1)

γ is the ratio between specific heats, c _p and c _v , which change as a function of temperature; V is the volume of the piston stroke; and p is the pressure of the cylinder gas.

The end of combustion (EOC) was determined by the angle at which 90% of the fuel mass was burned. The ignition delay time (IDT) was obtained by the time span between the start of injection (SOI) and the start of combustion (SOC) ^[18].

The temperature was read by using type K thermocouples installed in the lubricating oil drain nozzle (T_lubric), in the exhaust gas outlet (T_gas), and in the air intake manifold (T_air) after the air box (expander), in order to verify the moment in which the engine reached the permanent state of combustion and then collect the data. Fuel consumption ( ) was directly determined by a DIGIMED precision digital scale, model DG-15WT, which calculates the variation in fuel mass fraction ( ) consumed by the engine at a given time (∆t), as shown in Eq. (2). A Bosch mass airflow (MAF) sensor was used to measure the mass airflow rate ( ) consumed by the engine.

The signals obtained by the thermocouples, digital scale, and MAF sensor were sent to the A202 data collection and processing system (Contemp Ind.) The system then sent the data to the DAQFactory-Pro software program, v16.2, which analyzed and monitored the variables in real time (Fig. 1).

Specific fuel consumption (SFC), expressed in g/kW-h (see Eq. 3), takes into account the total mass flow rate of all fuels used during the tests ( ) (g/h) and P_elec (kW), which is the electrical power of the engine.

i = B10, B10E1, B10E2, B10E3, and B10E3* (pressure: 22 bar)

Moreover, the heat release of fuel in terms of energy was quantified, taking into account and NCV (Eq. 4) ^[20]:

where,

The NCV of the blends used during the test was determined by applying the law of conservation of mass to the fuels, considering the energy available in the fuel ( ) when ethanol is introduced (Eq. 4) ^[20].

By considering as a function of the volume flow rate given by and isolating and substituting it in Eq. 5., we have:

By isolating the NCV _i , one obtains Eq. (5) as a function of the volume fraction of the fuels (x _i ):

Calorific value plays an important role in the increase or decrease of SFC, aiding in the analysis of engine performance. Table 2.1 shows the results obtained for the specific mass and NCV of the blends, using Eq. 5.

The mass flow rate ( ) is used to assess consumption when different fuels are utilized. The data provided by the precision scale during the tests (Table 4) show a decrease in fuel consumption with 1% ethanol (1,407 g/h) when compared to S-10 (1,485 g/h), followed by an increase in consumption for the other blends. This increase occurs because the fuel injection system tends to add a larger amount of fuel to accomplish the same displacement, considering that energy is lost when ethanol is added. Fuel consumption increased as ethanol was added to S-10 diesel, accounting for 1,504 g/h (B10E2) and 1,541 g/h (B10E3).

Table 3

	B10E1	B10E2	B10E3
ρmix (25°C) kg/m3	791	790	790
LHVmx (kJ/kg)	39.593	39.480	39.367

Source: Manufacturer’s Manual

Table 4

	Diesel S-10	B10E1	B10E2	B10E3	B10E3*
Mass flow rate [g/h]	1,485	1,407	1,504	1,541	1,666
Power[kW]	3.54	3.30	3.04	2.64	2.92
Energy [kW]	16.4	15.5	16.5	17	18.2

Source: Manufacturer’s Manual

When the fuel injection pressure was changed from 18 bar to 22 bar, the fuel consumption observed for B10E3* was higher than that of the other blends.

Note that power decreased as ethanol was added to the S-10 diesel blend. This explains the findings obtained for the fuel mass flow rate (Table 4). Power also increased when the injection pressure, 22 bar, was altered in B10E3*.

The thermal energy of each fuel, shown in Eq. 4, can be calculated based on the consumption and NCV of each fuel. Table 4 shows a reduction in thermal energy when B10E1 rather than diesel was used, as there was a small increase in energy when a larger percentage of ethanol was added to the S-10 blend.

The tests performed in this study showed that the addition of anhydrous ethanol to S-10 diesel increased SFC, triggered by the speed of combustion in the blends, which is reflected in the shaft power, causing a decrease in effective electric power and also in effective thermal efficiency. As a result, the in-cylinder pressure decreases as ethanol is added. Also, the higher the percentage of ethanol added to the blend, the lower the heat release rate and the lower the temperature inside the cylinder, but SOC and ignition delay increase and the combustion duration is shortened.

The rise in fuel injection pressure from 18 bar to 22 bar attenuated the effects of ethanol on the generator set, with a decrease in the SFC of B10E3*, but power remained high. Similar heat release rates were observed in B10E3 and B10E3*. This change delayed SOC, affected the ignition delay, and increased combustion duration.

In conclusion, the smaller percentage of ethanol affects the combustion process and engine performance under predefined operating conditions, causing changes in various parameters. The factors that contributed to this phenomenon were the low cetane number and high latent heat of ethanol, affecting mainly the ignition delay. Based on the findings obtained in this paper, it is essential that further studies be carried out, making other modifications in compression ignition engines, such as variations in injection pressure and injection angle, in order to find the best alternatives for the use of ethanol as fuel. Ethanol, as an alternative source of renewable energy, can replace petroleum-based fuels.