Improvement of Fusel Oil Features and Effect of Its Use in Different Compression Ratios for an SI Engine on Performance and Emission

: In this study, the effects of the use of improved fusel oil on engine performance and on exhaust emissions in a spark-ignition engine were investigated experimentally in consideration of the water, gum, and moisture content at high compression ratios according to TS EN 228 standards. In the study, a four-stroke, single-cylinder, air-cooled, spark plug ignition engine with an 8/1 compression ratio was used at three different compression ratios (8/1, 8.5/1, 9.12/1). Experiments were performed for six different ratios of fuel blends (F0, F10, F20, F30, F40, and F50) at a constant speed and different loads. The data obtained from the experiments were compared with the original operating parameters of the engine while using gasoline. According to the test results, the optimal engine performance was at a 9.12/1 compression ratio and with a F30 fuel blend. With the increase from an 8/1 to 9.12/1 compression ratio for the F30 fuel blend, the overall efficiency increased by 6.91%, and the specific fuel consumption decreased by 2.35%. The effect of the optimum fusel blend on the emissions was also examined and CO emissions were reduced by 36.82%, HC emissions were reduced by 23.07%, and NO x emissions were reduced by 15.42%, while CO 2 emissions were increased by 13.88%.


Introduction
Due to the increasing energy demand in the world and the fact that the petroleum fuels used in motor vehicles will not be able to meet the needs in the near future, recently, many studies have been carried out on alternative energy sources and the development of existing systems [1].
In addition to fossil fuels, alternative fuels are used in internal combustion engines. Fusel oil can be used as an alternative fuel in internal combustion engines due to the amount of alcohol it contains. As alternative fuels are advantageous over fossil fuels, due to the lower exhaust emissions, and because they are renewable and can be produced from waste materials, they are more economical. Alcohols can be used in engines as standalone fuel or as gasoline-alcohol blends [1]. Alcohols may have a cooling effect on the fresh mixture due to their lower thermal values and higher vaporization temperatures, which may result in an increased volumetric efficiency of the engine [2][3][4]. Since alcohols are higher in octane number than gasoline, the engine can operate at higher compression ratios, which in turn can improve engine efficiency and reduce fuel consumption [5][6][7][8].
Fusel oil is a distillation by-product from ethyl alcohol fermentation [9]. Normal and branched chain monohydric alcohols with lower carbon atoms (2-5 carbon atoms) are the only natural sources of fusel oil [9][10][11]. Fusel oil may be colorless, yellow, brown, or green, depending on the type of

Preparation of a Proper Fuel Blend
The amount of H2O contained in fusel oil as a factory waste has adverse effects on engine performance and combustion [15]. Even though olefins increase the octane number in gasoline, the unstable olefins formed during cracking cause the formation of gum when they come into contact with air to oxidize. Figure 1 shows the amount of gum in the fusel-gasoline blend. The amount of gum and H2O in the fusel oil prevents a homogeneous blend with gasoline [19]. In order to provide an alternative fuel, the amount of gum must be reduced, and it must be refined to reduce the water content. For this purpose, the fusel oil, having been subjected to a series of processes, was made available for use in engine tests. Distillation efficiency of fusel oil was found to be 96.5% at the final boiling point, 139.9 °C. The amount of gum present is at the limit determined by TS EN 228, which is a maximum of 5 mg/100 ml. In the mixture of a fusel-gasoline fuel blend (10% fusel + 90% gasoline), this value was found to be 26.6 mg/100 ml, which was reduced to 13.4 mg/100 ml using a gum solver and to 0.8 mg/100 ml using a molecular sieve Z4-01 (2.5-5 mm) moisture scavenger, and was refined of H2O approximately by 96.9%. The reduction in gum and H2O was observed on a GC/MS chromatography mass spectroscopy device. The fuel blend was accorded to TS EN 228 standards, and a homogeneous fuel blend was obtained. Removal of water content from fusel oil leads to a distillation of alcohols in lower temperatures but does not cause a significant change in the amount of alcohol [24]. Figure 2 shows the Sieve Z4-01 (2.5-5 mm) moisture scavenger and the improved fusel-gasoline blends. Five blends of unleaded gasoline-fusel oil were prepared, and their characteristics are shown in Table 2.

Increase of Compression Rate for the Test Engine
In the experiments, three different cylinder heads of the spark-ignition engine used were removed from the grounding plates and grinded, and the compression ratio was altered. Cylinder heads were grinded at ateliers of Mercedes Benz (Hadımköy, Istanbul-Turkey) through measurements. Figure 3 shows the grinding operations; 0.40 and 0.80 mm of sawdust were removed from the cylinder heads. In order to calculate the compression ratios, the volumetric capacity of the cylinder was measured.
. The geometric compression ratio, defined as the ratio of the cylinder volume (V1) at initial compression to the end-compression volume (V2), is shown in Equation (1), where Vh is the piston displacement (swept) volume and Vc is clearance volume.
In the actual engine, the compression does not start at bottom dead center (BDC) but after the intake valve is closed. Therefore, the compression ratio (εg) of the actual motor can be expressed as in Equation (2). Vk is the volume that the piston covers until the valves close after BDC. Table 3 shows the altered compression ratios.

Experiment Procedure
In the experiments, a Honda HK 550 MS, which is a four-stroke, single cylinder, carburetted, and spark-ignition generator, was used. The technical specifications of the generator and the engine of the generator used in the experiments are given in Table 4. A Bilsa MOD 2210 WINXP-K exhaust gas analyzer was used in the experiments to measure the CO, HC, CO2, and NOx emissions, and lambda (λ) values. The precisions are shown in Table 5. The experimental setup is shown in Figure 4. In order to measure fuel consumption, an Ender SWOCK YP20002 electronic scale that can measure up to 2 kg with a precision of 0.01 g was used. A Delta SW 305 digital chronometer was used to determine the fuel consumption per unit time. A K/pt100 type thermocouple was used to measure the exhaust gas temperature. The PCE-FOT 10 brand digital instrument was used to measure the engine oil temperature. The numerical data from the sensors were read from the motor and monitored on the test computer. Since the lower thermal value of fusel oil is lower than that of gasoline, the main nozzle on the carburettor was widened to adjust to λ = 1. The nozzle cross-section was adjusted to λ = 1 in all experiments by means of the conical-tipped fuel adjustment screw. The air excess coefficient was monitored on the computer during the experiments. Six 1000 W halogen projector lamps were used to load the fixed speed generator at different operating loads for dynamometer purposes. A system consisting of a piezoelectric pressure transducer, inductive pick-up, charge amplifier, oscilloscope, and personal computer (PC) was set up to measure the in-cylinder pressure of the test engine. To collect data for the in-cylinder pressure, a Kistler model 611C piezoelectric transducer mounted on the spark plug was used, where the in-cylinder pressure values were transferred to a Rigol digital oscilloscope (DS2202E) via a Kistler model 5018A charge amplifier and recorded on a PC. The data regarding the crank angles and the position of the top dead center were transmitted to an oscilloscope via an inductive pick-up.
Before starting the data collection for the experimental tests, the engine was operated until it reached the operating temperature. Engine tests were conducted at constant speed using six different volumetric fuel blends (F0, F10, F20, F30, F40, and F50) at the original cylinder head (CR = 8:1); the overall efficiency at 1000-6000 W load ranges, specific fuel consumption, in-cylinder pressure data, and the CO, HC, CO2, and NOx exhaust emission values were examined. As a result, an optimal fuel blend and the most efficient operation range were determined at different compression ratios (8.5/1-9.12/1).

Brake Thermal Efficiency and in-Cylinder Pressure at Different Compression Ratios
Brake thermal efficiency is the ratio of the useful mechanical power obtained from the engine to the energy released by the fuel consumed per unit time [33]. Since the octane numbers of alcohols are higher than those of gasoline, they can work without knocking at high compression rates. This, as it increases the combustion efficiency and, thus, the end-compression temperature and pressure, can provide a higher overall efficiency than does the gasoline [34].
The most important parameters having an effect on engine efficiency are the physical and chemical characteristics and the compression ratio of the fuel used [6,34,35]. Figure 5 shows the brake thermal efficiency graph based on a CR of 8/1 engine load. At an 8/1 compression ratio, we found that for all loads, the brake thermal efficiency for the F10, F20, and F30 fuel blends, when compared to that of the F0 fuel, increased by an average of 1.41%, 2.77%, and 4.29% respectively, whereas for the F40 and F50 fuel blends, it decreased by an average of 5.33% and 10.54%, respectively. The abundance of oxygen content in the molecular structure of alcohols in the fusel oil and the higher octane number in comparison to gasoline improved combustion and increased efficiency. For higher rates of fusel oil in fuel blends, as the H2O content increases, the temperature inside the cylinder and the efficiency decrease. This shows that the H2O content of fusel oil decreases efficiency, as shown in some studies in the literature [18,27,36,37]. Awad et al. [13] conducted an RSM analysis to examine the effect of different H2O content of fusel oil-gasoline blends on engine performance. It was stated that fusel oil with water content within fusel oil-gasoline blends, whether water is reduced or not, has a significant effect on the brake thermal efficiency (BTE) engine load.  Figure 6 shows the in-cylinder pressure graph of six different fuel blends (F0, F10, F20, F30, F40, F50) at a CR of 8:1. The vaporization temperature of fusel oil is higher than that of gasoline. The increase in the internal combustion pressure (ICP) ratio, the decreased H2O content of the fusel oil, the higher octane number of fusel oil in comparison to gasoline and fusel oil's oxygen content improved the combustion. The review of graph showed that the ICP of F30 fuel increased by the fusel ratio in the fuel blend, and it was found to be 33.02 bars. The overall efficiency increased in parallel with the increase in the amount of load. According to the experimental results, among the blends tested, the F30 fuel has the highest overall efficiency. Figure 7 shows the overall efficiency changes at all loads studied with the F30 fuel blend at three different compression ratios. When the compression ratio was increased from 8:1 to 8.5:1 and then to 9.12:1 for all the loads tested using the F30 fuel blend, on average, there was an increase by 4.29%, 8.07%, and 17.20%, respectively. It was found that the overall efficiency of the F30 fuel blend at the compression ratio of 9.12 was the highest. An increase in the cylinder temperature due to the increase in compression ratio and the compression of the air-fuel mixture in a smaller area resulted in a better combustion and evaporation of the high-octane fusel oil. This result is similar to some studies in the literature [38,39]. The in-cylinder pressures of the F30 and F0 fuels at a compression ratio of 8.5:1 and 9.12:1 are given in Figure 8. In Figure 8A, at a CR of 8.5:1, the maximum pressure of the fuel F30 after the initiation of combustion cycle was measured to be 40.62 bars with a 6.8% increase in pressure in comparison to the F0 fuel. In Figure 8B, at a CR of 9.12:1, the maximum pressure of the fuel F30 after the initiation of the combustion cycle was measured to be 51.98 bars with a 6.12% increase in pressure in comparison to the fuel F0.

Brake Specific Fuel Consumption
Specific fuel consumption is the value that shows how much of the chemical energy of the fuel used in the engine to produce heat is converted to power in the crankshaft [40]. In other words, specific fuel consumption refers to the amount of fuel consumed per unit power. Specific fuel consumption decreased as the engine load and the CR increased. When the specific fuel consumption of F30 fuel for all engine loads at 8:1 compression ratio increased to 8.5:1 and then to 9.12:1, the specific fuel consumption increased by an average of 3.56% and decreased by an average of 1.27% and 9.08%, respectively, in comparison to the F0 fuel. This is shown in Figure 9. Since CR increases engine efficiency and power, specific fuel consumption decreases with a CR increase. The lower thermal value of fusel oil is lower than that of gasoline. The lower thermal value of fusel oil increased the fuel consumption in order to generate the same power at a CR of 8:1. This is similar to some studies in the literature [39,41].

Effect of Fusel Blends on the Amount of Carbon Monoxide Emissions at Different Compression Ratios
CO emissions occur in the cases of incomplete combustion and insufficient oxygen in the cylinder to achieve a full combustion in the rich air/fuel mixtures when there is not enough time for combustion in the cylinder [41]. The CO emission concentration depends largely on the engine operating condition and the air/fuel ratio. Figure 10 shows the CO emission changes based on the compression ratios of the F30 fuel at all engine loads and at constant speed. For all the engine loads for the F30 fuel blend, in comparison to the F0 fuel, the volumetric CO emission values decreased by 20.88% at CR 8:1, 40.02% at CR 8.5:1, and 49.81% at CR 9:1. This is similar in some studies in the literature [7,26,42,43]. As the CR increases, the chemical reactions increase, and CO emissions may decrease as combustion temperature increases.

Effect of Fusel Blends on the Amount of Hydrocarbon Emissions at Different Compression Ratios
Hydrocarbon (HC) emission occurs in the case of incomplete combustion due to lack of air and oxygen in the cylinder [41,43]. HC emission is the fuel that is exhausted unburned. HC emission changes for the F30 fuel are shown in Figure 11. Turbulence in the cylinder increases due to the increase in engine load. Since the turbulence near the exhaust duct enables post-exhaust oxidation, HC emissions decrease as the load increases. The F30 fuel used in the experiments showed a decrease in HC emission values due to increased engine loads. For all the engine loads for the F30 blend, in comparison to the F0 fuel, the HC emissions at CRs of 8:1, 8.5:1, and 9.12:1 decreased by an average of 30.19%, 45.38%, and 52.07%, respectively. The highest HC emission was at a CR of 8:1, and the lowest HC emission was at a CR of 9.12:1. This is similar to some studies in the literature [41].

Effect of Fusel Blends on the Amount of Carbon Dioxide Emissions at Different Compression Ratios
Carbon dioxide (CO2) is an end combustion product of any fuel containing a carbon molecule in its structure. In gasoline engines, the emission of carbon dioxide is related to the complete combustion of the fuel and occurs due to the high combustion temperature. The presence of sufficient oxygen (O2) for complete combustion increases the amount of carbon dioxide (CO2) emissions. Since fusel oil contains oxygen, the hydroxyl radical OH (one of the major oxidizing agents) converts carbon monoxide to carbon dioxide with the presence of sufficient oxygen O2. Figure 12 shows the variation of CO2 emissions at different compression rates and at all loads. For all engine loads for the F30 blend, in comparison to the F0 fuel, the volumetric CO2 emissions increased by an average of 12.56% at CR 8:1, by 17.65% at CR 8.5:1, and by 24.99% at CR 9.16:1. The abundance of O2 content in fusel oil improved the combustion. This shows similarities with some studies in the literature, where alcohol fuels reduce the amount of CO2 due to the high O2 content [41].

Effect of Fusel Blends on the Amount of Nitrogen Oxides Emissions at Different Compression Ratios
NOx is an exhaust emission product due to cylinder temperatures. NOx emissions are particularly noticeable at temperatures above 1500 °C [19]. The second most important parameter having effect on NOx formation is the oxygen concentration in the cylinder [13]. Figure 13 shows the NOx graph for the different compression ratios of F0 and the F30 blend. For all engine loads tested with the F30 blend in comparison to F0, NOx emission decreased by an average of 58.23%, 39.88%, and 15.42% at CRs of 8:1, 8.5:1, and 9.12:1, respectively. As CR increases, NOx emissions increase as the combustion temperature increases. The reason why NOx emission is lower for fusel oil at different compression ratios compared to that of gasoline is due to the lower thermal value and alcohol and H2O content of fusel oil. This situation is similar to some studies in the literature [6,41,44].

Results
The conversion of waste fusel oil into an alternative fuel source to gasoline is important in terms of preventing environmental pollution and the economic use of natural resources. In this study, water and gum content of fusel-gasoline blends were improved upon; the optimal fuel blend was determined, and the effect of increasing compression ratio was investigated. The test was carried out in a single cylinder, four-stroke, SI engine at different working loads, at λ = 1, and at different compression ratios. The CO, HC, CO2, and NOx values were measured as BTE, brake specific fuel consumptions (BSFC), ICP, and the emission values of the improved fusel oil.
Gum and H2O content of fusel oil was reduced by 0.8 mg/100 ml, and a homogenous fuel blend was obtained. The presence of oxygen in the fusel oil and the higher octane number of it, in comparison to gasoline, increased the combustion efficiency of the improved fusel oil fuel blend. For all engine loads, F30 was found to be the ideal fuel blend with a BTE increase by 4.29% and with a maximum pressure value of 33.2 bars for ICP at a CR of 8:1.
Increasing the compression ratio increased the post-combustion pressure and temperature. F0 and the F30 fuel blend were compared at all loads and with different compression ratios. BTE improved by 17.20%, and ICP increased by 52.1% for the improved fusel oil blend at a CR of 9.12:1. BSFC was found to be the most efficient compression ratio by a reduction of 9.08%. Due to the higher latent vaporization heat and abundant oxygen content of the improved fusel oil, for the ideal blend, namely, F30, the CO, HC, and NOx decreased by 49.81%, 52.07%, and 15.42% respectively, whereas CO2 increased by 24.99% at a CR of 9.12:1.
Fusel oil cannot be blended homogeneously due to the gum and H2O content in it; this not only causes knocking by worsening the combustion in the engine, but it also causes corrosion by leaving soot residue in the cylinder and intake valve. Experiment results show that waste fusel oil, obtained as a by-product, can be used as an alternative fuel to a certain extent in the gasoline engines after a series of processes.
Author Contributions: This article is derived from a doctoral study. This doctoral study was supported by Karabuk University Scientific Research Projects Department (BAP). During the thesis, the article was completed under the leadership of the co-authors. Hasan Saygın supported the supply of waste fusel oil from sugar factories and was responsible for improving the structural feature of the oil as well as providing an alternative fuel feature and fuel analysis. Bülent Özdalyan supported the establishment of laboratory infrastructure and conducted experiments. They have always been supportive and provided guidance in writing this article; they interpreted the results of the analysis and corrected the format accordingly. All authors have read and agreed to the published version of the manuscript.