Experimental and Numerical Study on the Effect of NO2 on n-Butanol/Biodiesel Dual-Fuel Combustion in a Compression Ignition Engine.

Nitrogen dioxide (NO2) is an active species of exhaust gas recirculation gas, and it has a significant impact on the autoignition and combustion processes of fuels. This study presented a comprehensive investigation of the effect of NO2 on the combustion characteristics of the n-butanol/biodiesel dual fuel. Experiments were conducted on a single-cylinder engine with 0, 100, 200, and 400 v/v ppm NO2 addition at two fuel injection ratios. The findings of the experiments indicated that adding NO2 resulted in an earlier start of heat release and an increase in peak in-cylinder pressure as compared to experiments where no NO2 was added. The evolutions of n-butanol, biodiesel, and OH radicals were evaluated using the computational fluid dynamics software coupled with the n-butanol-biodiesel-NO2 mechanism. The results revealed that when 400 v/v ppm NO2 was added, the consumption of n-butanol and biodiesel occurred earlier, and the formation of OH radicals was approximately an order of magnitude higher before the biodiesel was injected. Furthermore, reaction rate and flux analyses were performed to understand the effect of NO2 addition on the reaction process. When NO2 was added, 35% of the HO2 radicals reacted with NO which converted from NO2 via the reaction NO + HO2 ⇌ NO2 + OH, promoting the formation of OH radicals in the reaction system. The addition of NO2 can also enhance the consumption of CH3 radicals via the reaction CH3 + HO2 ⇌ CH3O + OH.

Introduction

Traditional spark and compression ignition (CI) engines have found it more challenging to fulfil increasingly stringent emission regulations and fuel efficiency demands. To address this, advanced combustion modes have been proposed, including homogeneous charge CI (HCCI),[1] premixed charge CI (PCCI),[2] partially premixed combustion (PPC),[3] and reactivity controlled CI (RCCI).[4] All these combustion modes utilized the exhaust gas recirculation (EGR) technique to reduce nitrogen oxides (NO) emissions. Furthermore, for various combustion modes, EGR was also used to manage the combustion phase in conjunction with other control techniques such as intake air temperature, fuel injection strategy, fuel reactivity, and so on. The effect of EGR on engine combustion can be classified into three categories:[5] dilution, thermal, and chemical. Early research concentrated on the dilution and thermal impact of EGR. The introduction of exhaust gas to reduce the concentration of O2 in the cylinder, as well as the higher specific heat capacity of CO2 and H2O, was thought to lower the maximum combustion temperature when the total heat release of the fuel in the cylinder remained constant, thereby inhibiting NO production. As the chemical kinetic mechanism of fuel combustion grew increasingly understood, the chemical effects of EGR were of considerable interest to researchers.

The active components in the EGR gas, such as NO, formaldehyde, sulfur oxides (SO), and others, could affect the oxidation process of fuels. With a large amount of EGR, the oxygen concentration decreased, and the influence of these active components on the ignition delay and combustion process of fuels became more apparent. Jansons et al.[6,7] introduced formaldehyde into the intake air of a direct injection CI (DICI) engine fueled with jet propulsion fuel (JP-8) and found that the addition of formaldehyde retarded the low-temperature heat release (LTHR), prolonged ignition delay, and suppressed HCHO* chemiluminescence. Kawasaki et al.[8] evaluated the impact of nitric oxide (NO) and nitrogen dioxide (NO2) on natural gas HCCI combustion in a rapid compression expansion machine (RCEM) that was modified from a single-cylinder diesel engine. The obtained findings demonstrated that by adding either NO or NO2, the autoignition was substantially accelerated, and the acute heat release was facilitated. Masurier et al.[9] investigated the impact of adding oxidizing species including ozone, NO, and NO2 on a single-cylinder HCCI engine fueled with iso-octane and found that all added species improved spontaneous ignition and advanced combustion phasing. Ozone was shown to have the most significant impact, whereas NO2 had the least. Kobashi et al.[10] found that adding NO to the intake air of a DICI engine fueled with diesel can reduce the ignition delay. The ignition delay decreased as the NO concentration increased. NO2 accelerated ignition more than NO. The addition of carbon monoxide (CO) or hydrocarbons (CH4, C2H4, C3H6, and C3H8) had little effect on the ignition delay.

For the design of the EGR system in many engines, the EGR gas was cooled using an EGR cooler and then blended with fresh air before entering the cylinder. During this procedure, NO in the EGR gas was converted to NO2.[11] Many studies have been conducted to investigate the influence of NO2 addition on the ignition behavior of various fuels such as hydrogen,[12] methane,[13−16] ethane,[14,15,17] methane/ethane mixture,[14,15,18] dimethyl ether,[19]n-butane,[20] and n-heptane.[21] Mathieu et al.[12] used a shock tube (ST) to evaluate the oxidation of hydrogen with three NO2 concentrations (100, 400, and 1600 v/v ppm) at 0.15, 1.3, and 3 MPa, 850–1700 K, and equivalency ratios of 0.3, 0.5, and 1.0. The addition of NO2 could promote the autoignition of H2. The ignition delays were shown to be highly dependent on pressure and NO2 concentration, although equivalency ratio variation had little effect on the ignition delay. Sahu et al.[13] studied the impact of NO2 addition (0, 200, and 400 v/v ppm) on the autoignition behavior of methane in a rapid compression machine (RCM) at 1.5 and 3.0 MPa, 900–1100 K, and equivalency ratios of 0.5, 1.0, and 2.0. The ignition delays were decreased when NO2 was added. The promoting impact of NO2 was discovered to increase with temperature, but the sensitizing effect was reduced at higher pressures. Deng et al.[17] explored the NO2 promoting effect on the ethane combustion in a shock tube at 0.12–2 MPa, 950–1700 K, and equivalency ratios of 0.5, 1.0, and 2.0. They found that adding NO2 increased reactivity at higher pressures and lower temperatures but had a negligible impact at low pressures and higher temperatures. Zhang et al.[14] compared the effect of NO2 addition on the autoignition behavior of methane, ethane, and methane/ethane mixtures in a shock tube. They discovered that adding NO2 to methane and methane/ethane mixtures reduced ignition delays significantly. However, for C2H6, the reduction in ignition delays was modest. The oxidation-enhancing effect of NO2 on dimethyl ether has been studied by Ye et al.[19] in a shock tube at 0.4 and 1 MPa, 987–1517 K, and equivalency ratios of 0.5, 1.0, and 2.0. The findings indicated that NO2 can considerably accelerate the autoignition of DME and that it performed better at low temperatures than at high temperatures. The influence of the equivalency ratio on ignition delays became more noticeable as NO2 concentrations increased. Wu et al.[20] studied the effect of NO2 on the ignition of n-butane in a shock tube at 1 and 2 MPa, 700–1200 K, and equivalency ratios of 1.0 and 2.0. The addition of trace amounts of NO2 (500 v/v ppm) enhanced the low-temperature reactivity of n-C4H10 and reduced the ignition delays while weakening the negative temperature coefficient (NTC) behavior of n-C4H10. Shi et al.[21] investigated the effect of NO2 (0.5, 1%) addition on n-heptane autoignition in a shock tube at 0.2 and 1 MPa, 700–1400 K, and an equivalency ratio of 1. The findings of the experiments showed that the NO2 effect was temperature- and NO2-concentration dependent. At high temperatures, NO2 can enhance n-heptane oxidation and shorten the ignition delay, but it had little impact on n-heptane autoignition around 700 K.

The use of biomass oxygenated fuels in engines can alleviate their reliance on petroleum resources while also lowering soot emissions. The most widely investigated biomass fuels included alcohols (methanol,[22] ethanol,[23] and n-butanol[24]), furans (2,5-dimethylfuran[25] and 2-methyl furan[26]), and biodiesel.[27] Compared to methanol and ethanol, n-butanol has a greater energy density and hydrophobicity. Biodiesel has similar physical and chemical properties as diesel; therefore, it can be utilized in compression-ignition engines without modification. Previously published work[28−31] investigated the combustion and emission characteristics of a CI engine fueled with n-butanol/biodiesel dual fuel using both experiments and numerical simulations. n-Butanol was port injection, and biodiesel was injected directly into the cylinder. The n-butanol/biodiesel dual-fuel combustion was regulated by both the chemical reaction kinetics of the fuel and the in-cylinder direct injection strategy, allowing for more precise ignition timing control than HCCI, PCCI, and PPC. To lower soot emissions of the engine, it was recommended to increase the n-butanol injection ratio; nevertheless, this would result in a greater influence of the air-fuel mixture reactivity on the ignition timing. As mentioned above, the active components in the EGR gas can also affect the ignition delays and fuel combustion process of the engine. Until recently, investigations on the influence of NO2 on fuel ignition characteristics have been limited to HCCI engines, shock tubes, and RCMs but rarely carried out on dual-fuel CI engines.

The present study aimed to investigate the effect of NO2 on n-butanol/biodiesel dual-fuel combustion in a CI engine. Various concentrations of NO2 were fed into the intake pipe of a single-cylinder CI engine to evaluate the changes in combustion characteristics such as pressure development and combustion phasing. Three-dimensional (3D) numerical calculation was carried out to determine the effect of NO2 on the evolutions of fuel consumption and OH radical generation as well as to clarify the in-cylinder distribution of OH radicals. Furthermore, kinetic chemical analyses were performed to gain further insight into the role of NO2 in fuel oxidation.

Experimental Setup and Methodology

Engine and Fuels

The experiments were conducted on a single-cylinder four-stroke CI engine. The engine was a single-cylinder version of a multi-cylinder marine diesel engine designed for fundamental research on high-efficiency and clean combustion. Table presents the specifications of the engine. The bore and stroke are 170 and 195 mm, respectively, and the displacement is 4.43 dm3. The compression ratio is 13.5. The biodiesel is injected directly into the cylinder using a common rail system. The injector nozzle has 8 holes and a 155° spray angle.

Table 1

Engine Specifications

items	specifications
number of cylinders	1
combustion chamber shape	stepped-lip
bore × stroke [mm]	170 × 195
displacement [dm3]	4.43
compression ratio	13.5
DI injector nozzle hole number	8
DI injector nozzle spray angle [°]	155

When conducting this n-butanol/biodiesel dual-fuel combustion experiment, n-butanol was utilized for port injection, and biodiesel was used for in-cylinder direct injection. Biodiesel derived from soybean oil was employed in this investigation. Both n-butanol and soybean biodiesel were available commercially. The physical and chemical properties of n-butanol, biodiesel, and diesel (for comparison) are presented in Table .

Table 2

Fuel Properties[32]

properties	n-butanol	soybean biodiesel	diesel
molecular formula	C4H10O	C18.76H34.48O2	C10–C20
CAS number	71-36-3	67784-80-9	68334-30-5
molecular weight	74	292	190–220
density [kg/dm3]	0.81	0.87–0.88	0.83–0.85
cetane number	25	48–52	52–55
viscosity at 40 °C [mm2/s]	2.22	4–4.3	3.35
lower heating value [MJ/kg]	33.2	38.4	42.8
boiling point [°C]	117	382	180–370

Experiment Platform

The engine test platform is shown in Figure . The NO in the EGR gas must be removed in the experiment in order to test the effect of NO2 on the n-butanol/biodiesel dual-fuel combustion, and then the required quantity of NO2 was fed into the intake pipe from a high-pressure gas cylinder. Consequently, a selective catalytic reduction (SCR) system was mounted on the EGR pipe to eliminate NO. The EGR gas was blended with the intake air and entered the pressure stabilization tank. The EGR valve controlled the flow of EGR gas. After the pressure stabilization tank, NO2 from the high-pressure cylinder was introduced into the inlet pipe. A high-precision flow meter regulated the amount of NO2. For thorough mixing of NO2 and engine intake gas, a mixing tank was installed after the NO2 input location. A gas analyzer was attached to the intake pipe to determine whether the NO2 concentration in the intake air met the target value. An n-butanol injection system assembly was installed on the intake pipe, and it was used to regulate the quantity of n-butanol injected. The intake and the exhaust pressure were controlled by the pressure regulators so as to imitate those of the real multi-cylinder engine. Combustion measurement equipment from the AVL company, which included a pressure sensor, a charge amplifier, data acquisition, and indicating software, was used to determine the pressure and heat release rate in the cylinder. The specifications for measuring instruments are shown in Table , including their measurement range and uncertainty. Considering the minimal uncertainty of measuring instruments, the impact of measurement error on test results is negligible. The measured value of in-cylinder pressure was set to the average of 200 cycles to eliminate pressure fluctuations.

Figure 1

Schematic representation of the experimental setup: (1) gas compressor; (2) buffer tank; (3) inlet control valves; (4) air flow meter; (5) mixing tank; (6) high-precision flow meter; (7) high-pressure cylinder; (8) gas analyzer; (9) engine; (10) dynamometer; (11) PFI injector; (12) DI injector; (13) cylinder pressure sensor; (14) DOC; (15) SCR; (16) exhaust back pressure valve; (17) combustion analyzer; (18) computer; (19) supercharger; (20) EGR cooler; (21) EGR valve.

Table 3

Specifications for Measuring Instruments

measured parameter	device	measuring range	uncertainty
intake concentration	AVL 415S	0–100%	≤0.5% of measured value
cylinder pressure	AVL GH15D	0–25 MPa	±0.3 bar
fuel mass flow rate	AVL 7355	0–125 kg/h	≤1% of measured value
air flow meter	ToCeiL20N100	0–1200 kg/h	≤1% of measured value
NO2 flow meter	ALICAT-LK2	10 g/min	≤0.6% of measured value

Experimental Methodology

The engine ran at 1500 rpm, with an EGR rate of 30%. n-Butanol was injected into the intake pipe, and biodiesel was injected directly into the cylinder with an injection timing of −10° CA ATDC. The impact of adding 0–400 v/v ppm NO2 on the ignition and combustion characteristics of the n-butanol/biodiesel dual fuel was investigated in this work. Simultaneously, two biodiesel injection ratios (denoted by RBD), that is, 20 and 40%, were chosen to evaluate the influence of NO2 on the ignition characteristics of the various fuel activities. The biodiesel injection ratio was defined as the ratio of biodiesel injection energy to the overall energy of the fuels, which included port injection of n-butanol and in-cylinder injection of biodiesel. The RBD can be expressed using the following equationwhere mBD is the mass of biodiesel and mNB is the mass of n-butanol.

The total energy of n-butanol and biodiesel injected per cycle was maintained constant throughout all test conditions, which was 240 mg of equivalent biodiesel. Equivalent biodiesel was defined as the injection mass of n-butanol converted to that of biodiesel with the same low heating value. The total mass (denoted by mtotal) of equivalent biodiesel mass can be expressed using the following equation

The intake NO2 concentrations were set at 0, 100, 200, and 400 v/v ppm. Although the EGR gas normally delivered less than 400 v/v ppm of NO2 when the engine was operating in actual work applications, the condition of 400 v/v ppm NO2 in the intake gas was set in this study to make the effect of NO2 on combustion characteristics of the engine more obvious and appreciable. Table lists the engine operating conditions.

Table 4

Experimental Conditions

items	set value
engine speed [rpm]	1500
EGR rate [%]	30
injection timing of biodiesel [°CA ATDC]	–10
biodiesel injection ratios (RBD) [%]	20, 40
intake NO2 concentration [v/v ppm]	0, 100, 200, 400
intake temperature [°C]	35
intake pressure [MPa]	0.17

Numerical Models

Reduced Mechanism of n-Butanol–Biodiesel–NO2

The combustion reaction kinetics of biodiesel was difficult to investigate directly due to the complexity of its components and high molecular weight. To simplify the biodiesel chemical kinetics mechanism, methyl decanoate (MD) was used as the surrogate of biodiesel. In our previous work,[32] it has been demonstrated through engine bench tests that the ignition delays, combustion characteristics, and soot emissions of MD were similar to those of soybean oil methyl ester under a wide range of operating conditions. Based on the detailed chemical kinetic mechanisms of MD developed by Herbinet et al.[33,34] and n-butanol developed by Sarathy et al.,[35] a reduced n-butanol–biodiesel dual-fuel mechanism with 157 species and 641 reactions[29] was obtained using the directed relation graphs (DRG) method,[36] sensitivity analysis, and reaction path analysis.

The C0–C1/NO sub-mechanism developed by Glarborg et al.[37] was merged into the reduced n-butanol–biodiesel mechanism. In our previous study[38] on the effect of NO2 on the ignition delays of methanol using a shock tube, it was demonstrated that this sub-mechanism can accurately predict the autoignition of methanol with various NO2 added concentrations. The final n-butanol–biodiesel–NO2 mechanism consisted of 186 species and 723 reactions (see the Supporting Information). In order to verify the mechanism, the tests of NO2 sensitization for the autoignition of n-butanol and biodiesel were conducted in a shock tube at the initial pressure of 0.6 MPa, temperatures of 1250–1550 K, and the equivalence ratio of 1. The detailed setup of the shock tube test platform is described in detail in our prior study.[38,39] The SENKIN code,[40] which is part of the CHEMKIN-II package,[41] was used to calculate the ignition delays and species consumption rates. Figure depicts a comparison of ignition delays between measurements and simulation results. As can be observed, the ignition delays calculated using this mechanism were closer to the experimental measurements.

Figure 2

Comparison of measured ignition delays (symbols) and predictions (lines) for (a) n-butanol and (b) MD with and without NO2 addition.

CFD Model

The 3D CFD software AVL FIRE, coupled with the n-butanol–biodiesel–NO2 reduction mechanism, was utilized to simulate the in-cylinder combustion process of the engine. The k−ζ–f model[42] was used to determine the turbulent motion in the combustion chamber since it was shown to be appropriate for highly compressed flows with moving boundaries. Because the Webber number was substantially larger for high-pressure injection, the Wave model[43] was the most appropriate breakup model. The Dukowicz model[44] was used to describe droplet heating and evaporation. For spray wall interactions, the Walljet1 model was employed, which was based on Naber’s spray/wall impingement model.[45] Since the engine had a centrally mounted injector with eight injection holes, a grid of 1/8 combustion chamber was used for the computation to reduce the time required for the numerical simulation. To ensure grid independency, two grids were generated for evaluation. Figure shows the two grid densities with the piston at the top dead center. The basic computational grids are shown in Figure a, which has 5,21,380 grids. Figure b shows the computational grids generated by refining the spray block, containing 19,62,160 grids. The calculations started when the intake valve closed (−139° CA ATDC) and finished until the exhaust valve opened (123° CA ATDC). The piston wall and cylinder head temperatures were set to 540 K, while the cylinder liner temperature was set to 450 K. To determine the effect of the integration time step on the calculated result, when the calculation interval was −20–60° CA, the integration time step for basic grids was 0.05° CA, whereas for refined grids, it was 0.02° CA. For both grids, the integration time step in the other calculation intervals was 0.5° CA.

Figure 3

Computational grids: (a) 5,21,380 grids and (b) 19,62,160 grids.

To begin, the numerical results using the two grids were validated against the experimental data for cylinder pressure and heat release rate at RBD = 20% with 400 v/v ppm NO2 addition, as shown in Figure a. It can be seen that the calculated results were almost identical for both grids and agree well with the experimental data. Then, the evolutions of n-butanol and biodiesel were compared using the two grids under the same condition, as illustrated in Figure . As can be seen, the computed curves for both grids were rather close. Therefore, utilizing 5,21,380 computational grids with an appropriate integration time step was quite acceptable for modeling n-butanol/biodiesel dual-fuel combustion.

Figure 4

Validation of the model for simulation: (a) comparison of test and simulated in-cylinder pressure and heat release rate curves at RBD = 20% with 400 v/v ppm NO2 addition and (b) comparison of n-butanol and biodiesel evolutions calculated using the two grids at RBD = 20% with 400 v/v ppm NO2 addition.

Results and Discussion

Experimental Study on the Effect of NO2 Addition on the Combustion Characteristics of the n-Butanol/Biodiesel Dual Fuel

This study investigated the combustion characteristics of the n-butanol/biodiesel dual-fuel engine with 0, 100, 200, and 400 v/v ppm NO2 addition at RBD = 20 and 40%. Figure depicts a comparison of in-cylinder pressure and heat release rate profiles for various NO2 concentrations. As shown in Figure a, for RBD = 20%, increasing the NO2 concentration caused the beginning moment of heat release to be earlier as well as an increase in the peak heat release rate and in-cylinder pressure. The maximum cylinder pressure was 8.07 MPa without NO2 addition; however, when the NO2 addition concentration was increased to 400 v/v ppm, the peak in-cylinder pressure climbed to 8.74 MPa. The heat release rate curve was bimodal for RBD = 40%, as illustrated in Figure b. As the NO2 concentration increased, the first heat release occurred marginally earlier, whereas the second heat release was clearly advanced, and the heat release rate peak was obviously higher. The peak in-cylinder pressure increased with increasing NO2 concentration, as it did for RBD = 20%.

Figure 5

Effect of NO2 addition on in-cylinder pressure and heat release rate: (a) RBD = 20% and (b) RBD = 40%.

It can be seen that for RBD = 20%, the heat release rate had a single peak, while for RBD = 40%, the heat release rate showed two peaks. This was because when RBD = 20%, the tiny quantity of biodiesel injected into the cylinder rapidly atomized and evaporated, and when the biodiesel ignited, it was essentially consumed as premixed combustion, resulting in a single peak heat release rate. When RBD = 40%, because of the increased injection amount of biodiesel, part of the mixture produced during the ignition delay of biodiesel spontaneously ignited to form the first heat release, and the remaining biodiesel was burnt as diffusion combustion, resulting in the second heat release in conjunction with the combustion of n-butanol.

The crankshaft angle that corresponded to 10% of the cumulative heat release was regarded as the moment of ignition and was referred to as CA10. Figure depicts the CA10s of various NO2 addition concentrations of RBD = 20 and 40%. The CA10 advanced in response to an increase in the concentration of NO2. For RBD = 20%, the CA10 was 2.3° CA ATDC without NO2 addition, while the CA10 was advanced by 0.3, 0.7, and 1.5° crank angles with NO2 addition concentrations of 100, 200, and 400 v/v ppm, respectively. For RBD = 40%, the extent of CA10 advancement was noticeably less than it was for RBD = 20% when NO2 was added. CA10 exhibited less advancement when the NO2 concentrations were 100 and 200 v/v ppm, but it showed a more obvious advancement when the NO2 concentrations were increased to 400 v/v ppm. Adding 400 v/v ppm NO2 could advance CA10 by 0.4° as compared to no NO2 addition. This demonstrated that when a higher proportion of biodiesel was directly injected into the cylinder, a lower NO2 concentration (less than 200 v/v ppm) was not sensitive to CA10. CA10 appeared earlier with RBD = 40% than with RBD = 20% for the same NO2 addition concentration. This was due to the fact that biodiesel had a high low-temperature reaction activity, and when the in-cylinder direct injection ratio increased, biodiesel created more reactive radicals through the low-temperature reaction process, causing the autoignition to occur sooner.

Figure 6

Effect of various NO2 concentrations on CA10s.

Figure illustrates the combustion durations with various NO2 added concentrations. As can be seen, for RBD = 20 and 40%, the combustion duration decreased as the NO2 addition concentration increased, indicating that NO2 accelerated the combustion process. For RBD = 20%, the combustion duration was 16.2° crank angle without NO2 and 13.3° crank angle with 400 v/v ppm NO2, a reduction of 18%. For RBD = 40%, when no NO2 was added, the combustion duration was 19.2° crankshaft angle and practically unchanged with 100 v/v ppm NO2 addition. However, when the NO2 addition concentration was increased to 400 v/v ppm, the combustion duration was reduced to 16.6° crankshaft angle, a 13% decrease. In comparison to RBD = 40%, RBD = 20% had a shorter combustion duration. This was because the mixture was more homogeneous and the heat release rate was faster when the biodiesel injection volume was lower for RBD = 20%. For RBD = 40%, the biodiesel injection volume was larger, and the biodiesel diffusion combustion resulted in a longer combustion duration.

Figure 7

Effect of various NO2 concentrations on combustion durations.

Numerical Simulations of the Effect of NO2 Addition on n-Butanol/Biodiesel Dual-Fuel Combustion

The previous section indicated that adding NO2 can considerably influence the combustion characteristics of the engine, notably at 400 v/v ppm NO2 addition. Hence, the effect of 400 v/v ppm NO2 addition on the processes of fuel consumption and OH radical production was evaluated employing 3D CFD simulation for RBD = 20 and 40%.

Figure presented the evolutions of n-butanol, biodiesel, and OH radicals of n-butanol/biodiesel dual-fuel combustion without and with 400 v/v ppm NO2 addition at RBD = 20 and 40%. It was found that when NO2 was added, the beginning of biodiesel and n-butanol consumption as well as the completion of the reaction occurred earlier than when no NO2 was added. The evolutions of the mole fraction of OH radicals in the cylinder showed that even before the biodiesel was injected, there was a modest amount of OH radical generation. The amount of OH radicals produced increased significantly when the biodiesel was injected. When biodiesel autoignited, the amount of OH radicals increased dramatically. With 400 v/v ppm NO2 addition, the OH radical production was about an order of magnitude more than that when no NO2 was added before −10° CA ATDC. For RBD = 20%, the crank angle corresponding to the sharp increase in OH radicals occurred 1.5° crank angle earlier with 400 v/v ppm NO2 addition than without. For RBD = 40%, this value is 0.5° crank angle.

Figure 8

Effect of 400 v/v ppm NO2 addition on evolutions of n-butanol, biodiesel, and OH radicals: (a) RBD = 20% and (b) RBD = 40%.

Hydroxyl radicals, which were highly reactive, can accelerate the evolution of reactants. Figure shows the distribution of OH radicals in the cylinder without and with 400 v/v ppm NO2 addition at RBD = 20 and 40%. As can be seen from Figure a, for RBD = 20%, only minimal amounts of OH radicals were found in the biodiesel spray region at −7° CA ATDC when no NO2 was added; however, with 400 v/v ppm of NO2 addition, a few amounts of OH radicals were generated in all areas of the combustion chamber. Minor amounts of OH radicals were produced around the biodiesel spray region without NO2 at −5° CA ATDC, but when 400 v/v ppm NO2 was added, the OH radical concentration increased significantly. It can also be shown that with 400 v/v ppm NO2 addition, the OH radical concentration in the biodiesel spray region was higher than that without NO2. As the piston moved upward, the cylinder temperature increased, and the OH radical concentration in the combustion chamber further increased around the biodiesel spray region, accelerating the oxidation of n-butanol. Thus, at −3° CA ATDC, the high concentration of OH radicals filled the bulk of the combustion chamber volume when 400 v/v ppm of NO2 was added; however, without NO2, the high concentration of OH radicals stayed localized in and around the spray region. When the cases of RBD = 20% and RBD = 40% were compared, it was found that the amount of in-cylinder biodiesel injection increased for RBD = 40%, which greatly increased the amount of OH radicals generated by biodiesel during the low-temperature reaction, causing CA10 to occur earlier than in the case of RBD = 20%, as shown in Figure .

Figure 9

Effect of NO2 on the in-cylinder distribution of OH radicals: (a) RBD = 20% and (b) RBD = 40%.

Chemical Kinetic Analysis

To provide further insight into the influence of NO2 addition on the reaction process, the reaction rate and flux analysis for ST simulations at the initial pressure of 3 MPa and the initial temperature of 800 K were performed using the SENKIN code. The condition of the in-cylinder mixture composition with RBD = 20% (Table ) was analyzed.

Table 5

Composition of the Mixture for Chemical Kinetic Analysis

	concentration [v/v ppm]
species	w/o NO2	w/NO2
MD	896	896
n-butanol	9904	9904
NO2	0	400
N2	7,71,400	7,71,000
O2	1,71,200	1,71,200
CO2	21,200	21,200
H2O	25,400	25,400

Figure a shows the comparison with the rate of NO2 consumption. It was apparent that the added NO2 was consumed mainly by three types of reaction pathways: (1) NO2 reacted with HO2 radicals to generate HNO2 via the reaction NO2 + HO2 ⇌ HNO2 + O2 and HONO via the reaction NO2 + HO2 ⇌ HONO + O2. HNO2 was converted to HONO via the reaction HNO2(+M) ⇌ HONO(+M), and HONO was broken to NO via the reaction NO + OH(+M) ⇌ HONO(+M). (2) By undergoing the reaction CH3 + NO2 ⇌ CH3O + NO, the conversion of NO2 to NO resulted. (3) NO2 was converted to NO via the reaction NO2 + H ⇌ NO + OH, while H radicals were changed to OH radicals as a result of the reaction.

Figure 10

Rate of consumption of (a) NO2, (b) NO, (c) CH3, and (d) HO2 and rate of production of (e) OH radicals. (Solid lines: w/400 v/v ppm NO2; dotted lines: w/o NO2).

Figure a illustrates that the reaction of NO2 with CH3 radicals proceeded at a fast rate. The rates of CH3 radical consumption were compared to understand the extent of the influence of NO2 on CH3 radical consumption, as shown in Figure b. Without NO2, two reactions dominated CH3 radical consumption, CH3 + O2(+M) ⇌ CH3O2(+M) and CH3 + HO2 ⇌ CH3O + OH. When the reaction time was shorter than 5.5 ms, the CH3 radical consumption rate via the reaction CH3 + O2(+M) ⇌ CH3O2(+M) was about 1 order of magnitude more than that of CH3 + HO2 ⇌ CH3O + OH. When NO2 was present, CH3 radicals were not only consumed via the two processes mentioned above but also reacted with NO2 to produce CH3O radicals via CH3 + NO2 ⇌ CH3O + NO. As the reaction time exceeded 4.8 ms, the rate of consumption of CH3 radicals via the reaction CH3 + NO2 ⇌ CH3O + NO surpassed that of the other two reactions. This demonstrated that the addition of NO2 had a significant effect on the rate of CH3 radical consumption.

The reaction of NO2 with HO2 radicals likewise had a high reaction rate, as shown in Figure a. To determine the influence of NO2 on HO2 radical consumption, the rates of HO2 radical consumption were compared, as can be seen in Figure c. Without NO2, HO2 radical consumption was dominated by HO2 + HO2 ⇌ H2O2 + O2, which produced H2O2. When NO2 was added, HO2 radicals were consumed not only by the above reaction but also by the reaction NO + HO2 ⇌ NO2 + OH to create OH radicals. Furthermore, as noted previously, HO2 radicals were consumed by the reactions NO2 + HO2 ⇌ HNO2 + O2 and NO2 + HO2 ⇌ HONO + O2. As a result, it appeared that the addition of NO2 accelerated the rate of consumption of HO2 radicals by NO2 and NO converted from NO2.

The above analysis demonstrated that NO2 was converted to NO through a series of reactions and that NO was also involved in key reactions. Figure d showed a comparison with the rate of NO consumption. The vast majority of NO reacted with HO2 radicals to form NO2 through the reaction NO + HO2 ⇌ NO2 + OH, which converted HO2 radicals to OH.

Figure e shows the comparison with the rate of OH radical production radicals. Without NO2, two reactions dominated the formation of OH radicals, H2O2(+M) ⇌ OH + OH(+M) and CH2CHO + O2 ⇌ CH2O + CO + OH. With the addition of NO2, the two reactions indicated above remain dominant in the production of OH radicals, while NO converted from NO2 promoted the generation of OH radicals via the reaction NO + HO2 ⇌ NO2 + OH.

The flow analysis of the n-butanol/MD mixture without and with 400 v/v ppm NO2 addition was performed during 50% n-butanol consumption, as shown in Figure . The n-butanol molecule underwent dehydrogenation at the beginning of the reaction, yielding C4H8OH-1 radicals and C4H8OH-3 radicals. The majority of the C4H8OH-1 radicals then reacted with O2 to produce nC3H7CHO and HO2 radicals via C4H8OH-1+O2 ⇌ nC3H7CHO + HO2. A small amount of C4H8OH-1 radicals and all of C4H8OH-3 radicals underwent the low-temperature reaction process, that is, RH → R• → ROO• → •QOOH → •OOQOOH → •U(OOH)2 → ketohydroperoxides, during which HO2 radicals were produced. In a similar vein, HO2 radicals were produced during the low-temperature reaction process of biodiesel. Without NO2, 91% of HO2 radicals was consumed by the reaction HO2 + HO2 ⇌ H2O2 + O2 to produce H2O2, while the remainder was consumed in reactions involving intermediate components. When NO2 was added, 35% of the HO2 radicals was consumed by the reaction NO + HO2 ⇌ NO2 + OH, which resulted in the formation of OH radicals, while 55% was consumed by the reaction HO2 + HO2 ⇌ H2O2 + O2 to produce H2O2. This indicated that the addition of NO2 during the low-temperature reaction process could promote the formation of OH radicals in the reaction system.

Figure 11

Flux analysis for the n-butanol/MD mixture during 50% n-butanol consumption. (Roman font: w/400 v/v ppm NO2; italic font: w/o NO2).

The macromolecular hydrocarbon intermediates were cleaved to CH3 radicals as well as various light alkyl, alkenyl, olefin, aldehyde, and ketone intermediates as the reaction proceeded.[33−35] As shown in Figure b, NO2 had a significant impact on CH3 radical consumption. Figure also illustrates the reaction flow analysis of CH3 radicals. Without NO2, 31% of CH3 radicals was converted to CH3O2 via the reaction CH3 + O2(+M) ⇌ CH3O2(+M) and 51% of CH3 radicals was converted to CH3O via the reaction CH3 + HO2 ⇌ CH3O + OH. With the addition of NO2, the reaction CH3 + O2(+M) ⇌ CH3O2(+M) consumed 31% of CH3 radicals, CH3 + HO2 ⇌ CH3O + OH consumed 50%, and 21% of CH3 radicals was converted to CH3O by the reaction CH3 + NO2 ⇌ CH3O + NO. Figure b shows that the rate of CH3 radical consumption via the reaction CH3 + NO2 ⇌ CH3O + NO was higher than that via the reaction CH3 + HO2 ⇌ CH3O + OH, indicating that the addition of NO2 can promote the consumption of CH3 radicals.

The reaction flow of NO2 and NO interconversion is also depicted in Figure . According to Figure a, at the beginning of the reaction process, the added NO2 reacted with HO2 radicals to produce HONO via the reaction NO2 + HO2 ⇌ HONO + O2 or HNO2 via the reaction NO2 + HO2 ⇌ HNO2 + O2, and HNO2 can be converted to HONO via the reaction HNO2(+M) ⇌ HONO(+M). The HONO was subsequently converted to NO via the reaction NO + OH(+M) ⇌ HONO(+M). As the reaction progressed, an increasing amount of NO2 was converted directly to NO by reacting with CH3 radicals via the reaction CH3 + NO2 = CH3O + NO. The generated NO was converted to NO2 mainly through the reaction NO + HO2 ⇌ NO2 + OH.

Conclusions

To investigate the effects of NO2 on the combustion characteristics of a CI engine fueled with n-butanol/biodiesel dual-fuel, a series of experiments and chemical kinetic simulations were carried out in this work. Experiments were conducted in a single-cylinder engine with two biodiesel injection ratios (RBD = 20 and 40%). NO2 was added to the engine intake pipe at concentrations of 0, 100, 200, and 400 v/v ppm. Using the 3D CFD software coupled with the n-butanol–biodiesel–NO2 mechanism, the influence of 400 v/v ppm NO2 addition on the evolutions of n-butanol, biodiesel, and OH radicals as well as the in-cylinder distribution of OH radicals was evaluated. Moreover, the reaction rate and flux analysis were performed to get further insight into the effect of NO2 addition on the reaction process. Major conclusions of this work are as follows:

The presence of increasing NO2 concentrations resulted in an earlier start of heat release, a shorter duration of combustion, and an increase in peak heat release rate and in-cylinder pressure. For RBD = 20%, the CA10 gradually advanced as the NO2 concentration increased. For RBD = 40%, CA10 exhibited less advancement when the NO2 concentrations were 100 and 200 v/v ppm, but it showed a more obvious advancement when the NO2 concentrations were increased to 400 v/v ppm. The extent of CA10 advancement for RBD = 40% was noticeably less than for RBD = 20% with NO2 addition.

When 400 v/v ppm NO2 was added, the beginning of n-butanol and biodiesel consumption as well as the completion of the reaction occurred earlier than when no NO2 was added. Before the biodiesel was injected, the formation of OH radicals was about an order of magnitude higher with 400 v/v ppm NO2 addition than without NO2.

The added NO2 was mainly consumed via three reaction pathways: NO2 reacted with HO2 radicals to form HNO2 and HONO, with CH3 to promote CH3 consumption, and with H radicals to produce OH radicals.

The addition of NO2 during the low-temperature reaction process can promote the formation of OH radicals in the reaction system. Flux analysis revealed that without NO2, 91% of HO2 radicals was consumed by the reaction HO2 + HO2 ⇌ H2O2 + O2 to produce H2O2. When NO2 was added, 35% of the HO2 radicals was consumed by the reaction NO + HO2 ⇌ NO2 + OH, which resulted in the formation of OH radicals. Moreover, the rate of CH3 radical consumption via the reaction CH3 + NO2 ⇌ CH3O + NO was higher than that via the reaction CH3 + HO2 ⇌ CH3O + OH, indicating that the addition of NO2 can promote the consumption of CH3 radicals.