Combustion Process of Canola Oil and n-Hexane Mixtures in Dynamic Diesel Engine Operating Conditions

: The article discusses the problem of using canola oil and n-hexane mixtures in diesel engines with storage tank fuel injection systems (common rail). The tests results of the combustion process in the dynamic operating conditions of an engine powered by these mixtures are presented. On the basis of the conducted considerations, it was found that the addition of n-hexane to canola oil does not change its energy properties and signiﬁcantly improves physicochemical properties such as the surface tension and viscosity. It contributes to the improvement of the ﬂammable mixture preparation process and inﬂuences the course of the


Introduction
In general use, the diesel engine will be used for many years to come (road, sea, rail, stationary engines of working machines) [1,2]. However, the sharpening of the standards concerning the emission of toxic compounds and the unstable situation on the market of petroleum-derivative fuels make it necessary to conduct research on the improvement of fuel supply systems and the construction of the internal combustion engine itself. This is done for obvious reasons, as it may have a positive impact on the improvement of the injection and combustion processes, which may have an influence on the reduction of fuel consumption and indirectly on the reduction of exhaust gas emissions. Another direction of research, leading to compliance with rigorous exhaust emission standards, is to improve the properties of fuels used in internal combustion engines. The physicochemical properties of fuels have a direct impact on the course of the injection and combustion processes [3]. Hydrocarbon alternative fuels not derived from crude oil processing, i.e., diesel, gasoline, etc., have an increasing share in the market for diesel fuels: CNG (compressed natural gas); alcohols (methanol, ethanol, and butanol); vegetable oil (canola oil, soya oil, sunflower oil, palm oil, and peanut oil); and methyl esters of vegetable oils such as canola oil and palm oil.
Additives to diesel or vegetable oils, which are intended to improve the combustion process and contribute to reducing the amount of toxic components in the exhaust gases, are also used. The main additives that have been tested are ethanol, methanol, ethers (methyl tertiary butyl, EMTB; ethyl tertiary butyl, EETB; dimethyl, DME; diethyl, DEE), n-butanol, and FAME additives (fatty acid methyl esters).
The density of the aqueous solution of studied canola oil+n-hexane mixtures was measured with a U-tube densitometer, DMA 5000 Anton Paar, with the precision of the density measurements equal to ±0.000005g cm −3 . The uncertainty was calculated to be equal to 0.01%. The viscosity measurements of the canola oil+n-hexane mixtures were performed with the Anton Paar viscometer, AMVn, with the precision of 0.0001mPas and an uncertainty of 0.3%. All density and viscosity measurements were made at 293 K. Table 1. Physicochemical properties of n-hexane [23].

Parameter Unit Value
Kinematic viscosity index in 20 •

Research Methodology and Research Station
The test object was a diesel engine with a common rail storage injection system installed in a Fiat Qubo vehicle, which met the Euro 5 emission standards. The vehicle was equipped with a five-stage gearbox. The test vehicle was equipped with an additional (external) independent fuel tank with an additional fuel pump, allowing for the quick replacement of the tested fuels. The modification of the fuel system of the vehicle concerned a low pressure system (about 3 bar). After switching to the auxiliary fuel system (low pressure), the main fuel system was automatically disconnected. The high-pressure fuel system was not modified, so the fuel pressure in the high-pressure system was the same for all the tests performed. The technical data of the engine and the view of the test vehicle are shown in Table 3 and Figure 1.  The Indimicro602 recording system of the AVL company with a built-in signal amplifier, cooperating with four analog input channels and two digital inputs, allowing for the recording of quick-change parameters in real time, was used to indicate the engine of the test vehicle. The connection diagram of the system is shown in Figure 2. The signals recorded by the AVL Indimicro system include the following. First, there is the pressure course inside the cylinder, which was recorded by the piezoelectric sensor AVL GH13P installed in the glow plug socket of the first cylinder by means of an adapter (Figure 2, position 3) and whose signal was processed in the amplifier-the AVL measuring module (Figure 2, position 6). Then, the engine crankshaft position signal informing about the crankshaft position was obtained from the induction sensor cooperating with the toothed flywheel by means of the AVL universal pulse conditioner 389Z01 analog-digital converter ( Figure  2 position 5).In addition, the injection parameters were analyzed on the basis of the analog control signal of the electromagnetic injector after conversion into a digital signal (Figure 2 item 7).
An indication of a compression ignition engine in conditions simulating vehicle motion in traction conditions was performed by simulating driving on a DF4FS-HLS chassis dynamometer. The diagram of the test stand is shown in Figure 1. The DF4FS-HLS chassis dynamometer was part of the system, which included a stationary (first) roller set with an electro-vacuum brake and hydraulic pump; a mobile (second) roller set with an electro-vacuum brake, hydraulic pump, and gear motor for drive; a control panel (dashboard);control of the hydraulic system; an axial fan for cooling the vehicle; and a PC with dynamometer software. The Indimicro602 recording system of the AVL company with a built-in signal amplifier, cooperating with four analog input channels and two digital inputs, allowing for the recording of quick-change parameters in real time, was used to indicate the engine of the test vehicle. The connection diagram of the system is shown in Figure 2. The signals recorded by the AVL Indimicro system include the following. First, there is the pressure course inside the cylinder, which was recorded by the piezoelectric sensor AVL GH13P installed in the glow plug socket of the first cylinder by means of an adapter ( Figure 2, position 3) and whose signal was processed in the amplifier-the AVL measuring module (Figure 2 An indication of a compression ignition engine in conditions simulating vehicle motion in traction conditions was performed by simulating driving on a DF4FS-HLS chassis dynamometer. The diagram of the test stand is shown in Figure 1. The DF4FS-HLS chassis dynamometer was part of the system, which included a stationary (first) roller set with an electro-vacuum brake and hydraulic pump; a mobile (second) roller set with an electro-vacuum brake, hydraulic pump, and gear motor for drive; a control panel (dashboard);control of the hydraulic system; an axial fan for cooling the vehicle; and a PC with dynamometer software.
The combustion engine of the traction unit is exploited mainly under dynamic conditions [25]. Vehicle driving tests are currently used to evaluate the energy and ecological parameters of engines. The most commonly used driving test is the Worldwide Harmonized Light Vehicles Test Procedure (WLTP). The characteristic points of the WLTP driving test corresponding to the dynamic conditions of engine operation have been selected.
Therefore, the tests consisted of a series of measurements for each of the fuels, which allowed the recording of fast-changing parameters during the acceleration of the engine with diesel. First of all, the torque (Mo) and power (Ne) values of the engine that were supplied with the tested fuels were determined. In the next stage of the research, a series of accelerations was performed, during which the following test methodology was applied (the measurement conditions result from the selection of characteristic points of the WLTP test running in dynamic conditions): gearbox ratio-fourth gear, initial vehicle speed 60 km/h, final speed-80 km/h, acceleration lever pitch during testing-40% constant for all fuels. The vehicle was loaded with rolling resistance forces. all, the torque (Mo) and power (Ne) values of the engine that were supplied with the tested fuels were determined. In the next stage of the research, a series of accelerations was performed, during which the following test methodology was applied (the measurement conditions result from the selection of characteristic points of the WLTP test running in dynamic conditions): gearbox ratiofourth gear, initial vehicle speed 60 km/h, final speed-80 km/h, acceleration lever pitch during testing-40% constant for all fuels. The vehicle was loaded with rolling resistance forces. Selected parameters of the combustion process, including the mean indexed pressure (pi), maximum combustion pressure (Pcmax), maximum rate of increase of combustion pressure (dp/dαmax), and the amount of heat produced, were calculated with the use of commonly known laws of thermodynamics, which were implemented in the AVL Indicom software. The calculations were performed at a frequency of 1 0 CA.
In the presented paper, the authors of each fuel performed 10 dynamic acceleration tests of the vehicle, during which the engine operating parameters were recorded. Table 4 presents the values of engine parameters averaged over 10 acceleration tests (at selected engine speed points).  show representative graphs from the conducted tests.

Results
Maximum torque (Nm) and maximum power (KW) were determined during engine tests. For diesel fuel (Df), the maximum torque and power values were 191.3 Nm at 2123 rpm and 57 kW at 3916 rpm, respectively. For vegetable fuels, lower torque values were obtained of about 11% (10%Hex90%Co) and 10% (15%Hex85%Co) and power of about 17% (10%Hex90%Co) and 14% Selected parameters of the combustion process, including the mean indexed pressure (p i ), maximum combustion pressure (P cmax ), maximum rate of increase of combustion pressure (dp/dα max ), and the amount of heat produced, were calculated with the use of commonly known laws of thermodynamics, which were implemented in the AVL Indicom software. The calculations were performed at a frequency of 1 0 CA.
In the presented paper, the authors of each fuel performed 10 dynamic acceleration tests of the vehicle, during which the engine operating parameters were recorded. Table 4 presents the values of engine parameters averaged over 10 acceleration tests (at selected engine speed points).  show representative graphs from the conducted tests.   During the tests conducted on the diesel engine fueled with the tested fuels, the average acceleration of the test vehicle was obtained, the course of which is shown in Figure 5

Results
Maximum torque (Nm) and maximum power (KW) were determined during engine tests. For diesel fuel (Df), the maximum torque and power values were 191.3 Nm at 2123 rpm and 57 kW at 3916 rpm, respectively. For vegetable fuels, lower torque values were obtained of about 11% (10%Hex90%Co) and 10% (15%Hex85%Co) and power of about 17% (10%Hex90%Co) and 14% (15%Hex85%Co). The accelerations obtained were the lowest for the canola oil with a 10% addition of n-hexane. The effect of the amount of n-hexane addition in Co on the obtained values of torque and engine power as well as acceleration, which may be related to the influence of physicochemical properties of fuels, was observed. Detailed values obtained during tests of the test vehicle powered by the tested fuels are presented in Table 5.
During the tests conducted on the diesel engine fueled with the tested fuels, the average acceleration of the test vehicle was obtained, the course of which is shown in Figure 5:Df-0.55 m/s 2 , 10%Hex90%Co-0.39 m/s 2 , and 15%Hex85%Co-0.50 m/s 2 .
(a) (b) Figure 5. The speed of the test vehicle (a) and acceleration of the vehicle (b) loaded with resistance to movement of the vehicle, powered by the tested fuels, i.e., Df, 10%Hex90%Co, and 15%Hex85%Co. Initial conditions: gearbox ratio-IV gear, initial vehicle speed-60 km/h, final speed-80 km/h, acceleration lever pitch-constant for all fuels-40%. Sampling frequency in 1s.   During the performed accelerations of the vehicle, points were selected (shown in Figure 5a) at which combustion process parameters were determined, i.e., the mean index pressure (p i ), maximum combustion pressure (P cmax ), and maximum rate of increase of combustion pressure (dp/dα max ). Detailed values are shown in Table 4. It was found that the average indexed pressure at each of the measurement points (I, II, III, and IV) was the highest for diesel oil. The biggest difference was observed in relation to 15%Hex85%Co and reacheda maximum of 13%. It was observed that increasing the content of n-hexane in the mixture with canola oil led in most of the cases to a decrease in the mean index pressure and maximum combustion pressure and to an increase in the maximum rate of increase of combustion pressure. Similar trends in the observed parameters were observed for other engine speeds analyzed. Figure 6 shows the course of the main parameters of the combustion process of an engine fueled with diesel oil (position a), canola oil with 10% n-hexane (position b), and canola oil with 15% n-hexane (position c). It was found that under dynamic conditions, the self-ignition delay angle (αID) was the lowest for Df in the whole engine speed range. For 15% of the n-hexane content in Co, the angle of auto-ignition delay was higher by approximately 13% in relation to Df. Rapeseed fuels were characterized by comparable self-ignition delay values. other engine speeds analyzed. Figure 6 shows the course of the main parameters of the combustion process of an engine fueled with diesel oil (position a), canola oil with 10% n-hexane (position b), and canola oil with 15% nhexane (position c). It was found that under dynamic conditions, the self-ignition delay angle (αID) was the lowest for Df in the whole engine speed range. For 15% of the n-hexane content in Co, the angle of auto-ignition delay was higher by approximately 13% in relation to Df. Rapeseed fuels were characterized by comparable self-ignition delay values.

Discussion
The different physicochemical properties of mixtures of canola oil and n-hexane in comparison with diesel oil, as shown in Table 2, cause different coursesin the injection and combustion processesof these fuels. In diesel engines, one of the most important parameters affecting the

Discussion
The different physicochemical properties of mixtures of canola oil and n-hexane in comparison with diesel oil, as shown in Table 2, cause different coursesin the injection and combustion processesof these fuels. In diesel engines, one of the most important parameters affecting the combustion process is the self-ignition delay [26]. Differences in the calorific value and the heat of combustion of mixtures cause changes in the operational indicators of engine operation. In the dynamic conditions of engine operation supplied with the tested fuels, the injection time was similar (the differences reached a maximum of 5%). This resulted in the delivery of a volume-comparable amount of fuel to the combustion chamber.
The reason for the lower values of the average indicated pressure ( Table 4) that were obtained when supplying canola oil with an addition of n-hexane to Df was the generation of less heat during combustion, which is shown, among others, in Figure 7. During combustion processes of diesel oil on the heat-excitation rate curves, the classical kinetic and diffusion phase could be distinguished, whereas during the combustion of canola oil with the addition of n-hexane, the kinetic phase disappeared with increased n-hexane content, which resulted in higher pressure rise velocities-see Figure 8. This also significantly affected the beginning of combustion, whose angle for diesel oil appeared the earliest (before the kinetic phase of combustion). For canola oil with added n-hexane, the start of combustion appeared later by approximately 4-12 • CA in terms of Df(depending on engine speed), while an increase in the amount of n-hexane caused further delay in the start of combustion. A similar tendency was observed forthe angle of self-ignition delay. Figure 6 illustrates the differences in the occurrence of the angle of the beginning of combustion and other parameters of the combustion process when feeding the fuels under study. Another reason for the observed changes could have been a different course of the flammable mixture formation process, due to the different viscosity of the fuels tested. The occurrence of the kinetic phase of combustion resulted in a decrease in the maximum rate of increase of combustion pressure (dp/dα) in the engine fueled with Df. For canola oil, with the addition of n-hexane, the maximum rate of pressure rise was higher than for Df, and brewing increased with increase of n-hexane content in the mixture. The addition of n-hexane to canola oil results in a reduction of surface tension, slight changes in density, and significant changes in viscosity (Figures 9 and 10). The surface tension value of the oil mixture and 5% n-hexane is only 0.8 mN/m higher than the surface tension of the diesel oil. In turn, for the mixture of oil and 10% n-hexane, it is 0.8 mN/m lower. Changes in surface tension as a function of n-hexane concentration in the mixture with canola oil indicate that the mixture does not behave as an ideal mixture (Figure 9) [27], i.e., changes in this tension are not directly proportional to the composition of the mixture. According to the suggestion of Fowkes and van Oss, the surface tension results from Lifshitz-van der Waals interactions, hydrogen bonds, and electrostatic interactions [28,29]. Therefore, it depends on the type of functional groups occurring in the liquid molecules at the phase boundaries. In the case of a mixture of canola oil and n-hexane, the dominant functional groups that may occur at the liquid-air interface are -CH 3 , =CH 2 , ≡CH, =CO, and -COOH. The surface tension of the canola oil/n-hexane mixture depends on the density of the mentioned functional groups. Since the main components of canola oil are unsaturated higher fatty acids (>90%), the contribution of the =CO and -COOH groups to the surface tension of the oil is significant. Therefore, the orientation of oleic, linoleic, and linoleic acid molecules at the phase boundary has a major influence on the surface tension of canola oil. The n-hexane molecules present in the mixture with canola oil due to the strong hydrophobic interactions between the n-hexane molecules and the apolar part of unsaturated fatty acids increase the likelihood of orientation of the formed acid-n-hexane complexes with the hydrophobic part directed toward the gas phase, which in turn lowers the surface tension of the mixture. This may explain the nonlinear relationship between the surface tension of the mixture and its composition (Figure 9). [28,29]. Therefore, it depends on the type of functional groups occurring in the liquid molecules at the phase boundaries. In the case of a mixture of canola oil and n-hexane, the dominant functional groups that may occur at the liquid-air interface are -CH3, =CH2, ≡CH, =CO, and -COOH. The surface tension of the canola oil/n-hexane mixture depends on the density of the mentioned functional groups. Since the main components of canola oil are unsaturated higher fatty acids (>90%), the contribution of the =CO and -COOH groups to the surface tension of the oil is significant. Therefore, the orientation of oleic, linoleic, and linoleic acid molecules at the phase boundary has a major influence on the surface tension of canola oil. The n-hexane molecules present in the mixture with canola oil due to the strong hydrophobic interactions between the n-hexane molecules and the apolar part of unsaturated fatty acids increase the likelihood of orientation of the formed acid-n-hexane complexes with the hydrophobic part directed toward the gas phase, which in turn lowers the surface tension of the mixture. This may explain the nonlinear relationship between the surface tension of the mixture and its composition (Figure 9).  The value of surface tension, which is closely related to the interactions between functional groups in the surface layer, can be a decisive factor for the vapor pressure of individual components of a mixture. Since the changes in the surface tension of the tested mixtures as a function of n-hexane concentration indicate non-ideal behaviour of the mixture, it can be expected that the vapor pressure of n-hexane and canola oil components does not comply with the Raoult law [27]. This, in turn, may have a considerable influence on the obtained flash temperature values of the tested blends (<40°C). Moreover, the decrease in surface tension of the canola oil/n-hexane mixture as a function of the nhexane volume concentration results in a decrease in the volume of the mixture droplets flowing out of the injector tip, which may have a significant impact on the injection process.
The density of the mixture canola oil and 10% n-hexane is slightly lower than the density of diesel oil (Figure 10). Although the addition of n-hexane significantly reduces the viscosity of canola oil, even with its content in the mixture of 20%, the viscosity of the blend is slightly higher than that of diesel oil. On the other hand, the decrease in viscosity affects the change in the angle at the The value of surface tension, which is closely related to the interactions between functional groups in the surface layer, can be a decisive factor for the vapor pressure of individual components of a mixture. Since the changes in the surface tension of the tested mixtures as a function of n-hexane concentration indicate non-ideal behaviour of the mixture, it can be expected that the vapor pressure of n-hexane and canola oil components does not comply with the Raoult law [27]. This, in turn, may have a considerable influence on the obtained flash temperature values of the tested blends (<40 • C). Moreover, the decrease in surface tension of the canola oil/n-hexane mixture as a function of the n-hexane volume concentration results in a decrease in the volume of the mixture droplets flowing out of the injector tip, which may have a significant impact on the injection process.
The density of the mixture canola oil and 10% n-hexane is slightly lower than the density of diesel oil ( Figure 10). Although the addition of n-hexane significantly reduces the viscosity of canola oil, even with its content in the mixture of 20%, the viscosity of the blend is slightly higher than that of diesel oil. On the other hand, the decrease in viscosity affects the change in the angle at the beginning of combustion.
From the point of view of using a mixture of canola oil and n-hexane for diesel engines, it is important that the combustion heat of canola oil under the influence of n-hexane and the oxygen consumption in this process are changed. It is known that canola oil is composed of many chemical compounds; however, the content of oleic, linoleic, and linolenic acid is over 90% [30]. The process of combustion of these acids can be represented by a reaction (1)-(9): The following chemical reactions (10)- (12) were used to compare the amount of oxygen needed to burn n-hexane: C 6 H 14 + 9.5O 2 = 6CO 2 + 7H 2 O (10) Based on the combustion reaction of 1 mole of n-hexane and the three acids mentioned above, it can be concluded that the amount of oxygen needed to burn n-hexane is much lower than in the case of acids, but such a comparison does not reflect the amount of oxygen needed to burn equal volumes of these substances. Therefore, the number of oxygen moles needed to burn each of these compounds has been calculated by taking into account the density values. If the product of combustion of all compounds is carbon dioxide then the number of oxygen moles needed to burn 1 dm 3 of the compound is 80.34, 80.25, 80.43, 80.43, and 72.75, respectively, for oleic acid, linoleic acid, linoleic acid, linolenic acid, and n-hexane. This comparison shows that the number of oxygen moles necessary to burn the same amount of n-hexane is significantly lower than for the three acids mentioned above. These differences should be reflected in the values of heat of combustion of n-hexane and tested acids, and thus canola oil. The heat value of canola oil combustion per 1 dm 3 is 35284.32 kJ. It turned out that this value did not differ much from the value for oleic acid (35147.60 kJ). However, the combustion heat of n-hexane, as expected from the amount of oxygen needed for its combustion, is lower and amounts to 31911 kJ. Taking these values into consideration, the heat of combustion of the canola oil/n-hexane mixture as a function of its composition was calculated. The calculations show that in the range of concentrations (from 0 to 20% of n-hexane in the mixture) the heat of combustion slightly decreases from 35284.32 to 34609.66 kJ/dm 3 ( Figure 11). Changes in the combustion heat of a canola oil and n-hexane mixture are related to the amount of oxygen needed to burn the mixture. Considering that more than 90% of the oil composition is made up of the acids tested, the average value of oxygen mole needed to burn 1 dm 3 of oil was used to calculate the number of oxygen needed to burn 1 dm 3 of oil. The calculated amount of oxygen mole needed to burn 1 dm 3 of the mixture, similarly to the heat of combustion, changes slightly (from 80.34 to 78.82). On the basis of the conducted considerations, it can be stated that the addition of n-hexane to canola oil slightly changed the heat of combustion and significantly improved the physicochemical properties such as the surface tension and viscosity. that in the range of concentrations (from 0 to 20% of n-hexane in the mixture) the heat of combustion slightly decreases from 35284.32 to 34609.66 kJ/dm 3 (Figure 11). Changes in the combustion heat of a canola oil and n-hexane mixture are related to the amount of oxygen needed to burn the mixture. Considering that more than 90% of the oil composition is made up of the acids tested, the average value of oxygen mole needed to burn 1 dm 3 of oil was used to calculate the number of oxygen needed to burn 1 dm 3 of oil. The calculated amount of oxygen mole needed to burn 1 dm 3 of the mixture, similarly to the heat of combustion, changes slightly (from 80.34 to 78.82). On the basis of the conducted considerations, it can be stated that the addition of n-hexane to canola oil slightly changed the heat of combustion and significantly improved the physicochemical properties such as the surface tension and viscosity. Figure 11. A plot of heat of combustion (curve 1) and amount of oxygen mole needed to burn off 1 dm 3 of canola oil-n-hexane mixture (curve 2) vs. n-hexane concentration. Figure 11. A plot of heat of combustion (curve 1) and amount of oxygen mole needed to burn off 1 dm 3 of canola oil-n-hexane mixture (curve 2) vs. n-hexane concentration.
The addition of n-hexane to canola oil caused a significant decrease in its viscosity and surface tension. Changes in the surface tension of canola oil with the addition of n-hexane as a function of composition are not linear, and synergy in the reduction of surface tension is observed. It was found that the mixture of canola oil and n-hexane is not ideal. A significant decrease in the viscosity and surface tension of oleic acid, the main component of canola oil, by adding n-hexane, indicates that such a mixture can be used in diesel engines.

Conclusions
The conducted research and its analysis allow us to formulate a few final conclusions: • The vapor pressure of n-hexane and canola oil components does not comply with Raoult law, and this in turn may have a significant influence on the obtained values of ignition temperature of the tested blends (<40 • C); the reduction of surface tension of the canola oil mixture with n-hexane as a function of n-hexane volume concentration causes a decrease in the volume of mixture drops flowing out of the injector tip, which may have a significant influence on the injection process; • Based on the calculations carried out, it can be concluded that the addition of n-hexane to canola oil slightly changed the heat of combustion of the same volume of the prepared mixture and significantly improved physicochemical properties such as the surface tension and viscosity; • Under dynamic conditions of engine operation supplied with the tested fuels, the injection time was similar (the differences reached a maximum of 5%); this resulted in supplying the combustion chamber with a volume comparable amount of fuel; • Obtaining lower values of the average pressure indexed when supplying canola oil with n-hexane in relation to Df was caused by lower heat generation during combustion; during diesel combustion, the classical kinetic and diffusion phase could be distinguished on the heat transfer rate curves, whereas during the combustion of canola oil with n-hexane, the kinetic phase disappeared with increased n-hexane content, which resulted in higher pressure increase rates; for canola oil with n-hexane addition, the beginning of combustion occurred later by approximately 4-12 • CA with respect to Df (depending on engine speed), while an increase in the share of n-hexane caused further delay in the beginning of combustion-a similar tendency was observed for the angle of self-ignition delay.