Load-Dependent Efficiency and Emission Trade-Offs of n-Butanol–Diesel Blends in a Naturally Aspirated Diesel Engine

Abstract

This work systematically evaluates the combustion and emission characteristics of n-butanol–diesel blends to clarify load-dependent trade-offs. A single-cylinder diesel engine was operated under low (25%)- and high (75%)-load conditions using commercial diesel and n-butanol blends (5–15 vol%). The results indicate that n-butanol addition tends to improve brake thermal efficiency ($BTE$) and reduce brake-specific energy consumption ($BSEC$), particularly at high loads, likely due to enhanced premixed combustion and fuel oxygenation. Emission trends exhibited distinct load-dependent behaviors: nitrogen oxides (NO_x) emissions decreased at low loads, ostensibly because the charge-cooling effect of n-butanol’s high latent heat dominated, whereas they increased at high loads driven by elevated temperatures and oxygen availability. Smoke opacity, carbon monoxide (CO), and carbon dioxide (CO₂) emissions were consistently reduced across all operating conditions, suggesting benefits from improved oxidation and the lower carbon content. In contrast, unburned hydrocarbon (HC) emissions increased significantly, which is primarily attributed to prolonged ignition delay and local quenching arising from the fuel’s low cetane number and high latent heat. These findings demonstrate n-butanol’s potential to enhance efficiency and mitigate smoke, CO, and CO₂ emissions, though the trade-offs with HC and high-load NO_x necessitate optimized control strategies.

1. Introduction

Diesel engines have long been established as a primary power source in the transportation, agriculture, construction, and power generation sectors owing to their superior thermal efficiency and high reliability. However, recent global efforts to address the climate crisis are demanding significant structural changes in internal combustion engines. The International Energy Agency (IEA) has called for rapid decarbonization of the transport sector through its “Net Zero by 2050” roadmap [1], and the International Maritime Organization (IMO) has also adopted the 2023 IMO Strategy on Reduction of Greenhouse Gas (GHG) Emissions, setting a target to reach net-zero GHG emissions from international shipping by or around 2050 [2]. Furthermore, upcoming emission standards, such as Euro 7 in Europe, require reductions in nitrogen oxides (NO_x) and particulate matter (PM) to the lowest technically achievable levels [3], making it increasingly difficult to meet these environmental requirements with existing fossil-fuel-based combustion technologies alone.

To address these challenges, after-treatment technologies such as diesel oxidation catalysts (DOCs) [4], diesel particulate filters (DPFs) [5], and selective catalytic reduction (SCR) [6] have been commercialized. However, limitations, including increased equipment costs, efficiency penalties associated with elevated exhaust back pressure, and the need for complex control strategies, have been consistently identified. In this context, a growing body of recent studies has focused on the applicability of oxygenated alternative fuels as a viable approach to reduce reliance on exhaust after-treatment systems and to suppress pollutant formation at the source, i.e., during the combustion process itself [7,8]. In particular, alcohol-based oxygenated fuels have been reported to effectively reduce soot and unburned hydrocarbon (HC) emissions by mitigating local oxygen-deficient zones within the combustion chamber through fuel-borne oxygen, and by improving mixture formation and premixed combustion characteristics [7,9].

Among various alternative fuels, n-butanol is considered a promising candidate owing to its relatively high energy density, comparable to that of diesel, excellent blend stability, and potential for biomass-based production. From an energy-utilization perspective, these properties make n-butanol particularly suitable for low-level blending in compression ignition engines, where preserving fuel energy content and maintaining stable combustion behavior are critical. Recent reviews and comparative studies have reported that n-butanol exhibits a higher cetane number (CN) and calorific value than lower alcohols, along with superior compatibility with diesel, enabling its application without engine modification [10,11]. Collectively, these studies further indicate that n-butanol blends offer a relatively balanced trade-off among fuel stability, combustion performance, and emission characteristics compared to other alcohols, including ethanol and n-pentanol.

In particular, recent studies on spray and combustion behavior have demonstrated that the lower viscosity and surface tension of n-butanol physically enhance mixture formation and premixed combustion characteristics by promoting spray atomization and accelerating droplet breakup and evaporation processes [12,13,14]. Owing to these physical properties, numerous experimental studies have reported that n-butanol blends significantly reduce smoke opacity and PM emissions compared to diesel [15,16,17,18,19]. At the same time, NO_x emissions are reduced or mitigated under certain operating conditions due to the local reduction in combustion temperature associated with its high latent heat of vaporization [16,18]. Moreover, recent research has shown that n-butanol blending effectively suppresses the formation and toxicity of polycyclic aromatic hydrocarbons (PAHs), thereby simultaneously alleviating the environmental and human health impacts of exhaust emissions [20].

However, applying n-butanol does not always yield positive results across all operating conditions. In studies focusing on direct injection (DI) diesel engines, while smoke emissions were significantly reduced with butanol blending, the lower CN of the fuel increased the ignition delay. This prolonged ignition delay led to rapid increases in the heat release rate and pressure rise rate, resulting in higher NO_x emissions or a limited NO_x reduction effect [21,22,23].

On the other hand, in small diesel engines or under low-load conditions, a distinct trade-off has been observed in which HC and carbon monoxide (CO) emissions increase due to deteriorated combustion completeness caused by the charge cooling effect and combustion temperature reduction associated with the high latent heat of vaporization of n-butanol [21,24]. In addition, issues related to combustion instability and increased fuel consumption at low load have been repeatedly reported [25,26].

These conflicting outcomes suggest that the combustion and emission characteristics of n-butanol are susceptible to engine system configuration and combustion control strategies. A review of existing literature reveals that a significant proportion of studies have focused on heavy-duty or medium-duty diesel engines to analyze the effects of fuel property changes on combustion and emissions under high-load conditions [27], or have applied advanced combustion concepts such as low-temperature combustion (LTC) or reactivity controlled compression ignition (RCCI) to mitigate the soot–NO_x trade-off [28,29].

Moreover, recent research trends indicate a shift in focus toward coupling fuel properties with injection and control strategies rather than focusing on the isolated effects of the fuel itself. For example, strategies that apply post-injection to n-butanol blends to control spray and oxidation processes [30,31], or that regulate combustion reactivity through dual-fuel or ternary blends, are being actively pursued [32,33]. Furthermore, the use of computational fluid dynamics (CFD) to precisely analyze injection processes and combustion mechanisms is increasing, facilitating discussions on the optimization of fuel spray, mixing, and heat release characteristics [34,35,36].

While some research has been conducted on air-cooled diesel engines, these studies have primarily focused on ternary blends such as diesel–biodiesel–butanol, which limit clear separation and interpretation of the specific effects of n-butanol addition [37]. Meanwhile, although studies on binary blends containing n-butanol have been reported, they have predominantly been based on water-cooled or indirect injection (IDI) engines; thus, experimental quantitative analysis on mechanically injected air-cooled DI engines—where changes in fuel properties exert a direct physical influence on spray characteristics and injection timing—remains limited [38]. On such platforms, unlike electronically controlled engines that actively compensate for variations in fuel properties, changes in fuel properties exert a direct, physically transparent influence on spray characteristics and combustion phasing. In this context, the measured performance and emission data—obtained under fixed mechanical injection parameters—serve as a valuable experimental baseline that reveals the combustion process’s raw sensitivity to changes in fuel properties.

Therefore, this study aims to experimentally investigate the effects of n-butanol blending (5–15 vol%) on the performance and emission characteristics of a single-cylinder, air-cooled, mechanically injected diesel engine. Specifically, by establishing well-defined experimental conditions at low load (25%) and high load (75%) across a speed range of 1600–2800 rpm, this research provides an in-depth analysis of how the interacting effects of n-butanol—namely, the latent heat of vaporization, fuel-borne oxygen, and ignition delay—influence brake thermal efficiency ($BTE$), brake-specific energy consumption ($BSEC$), and mass-based exhaust emissions (NO_x, CO, HC, and carbon dioxide (CO₂)). By clarifying the behavior of n-butanol on a platform without an electronic control unit (ECU), this work provides a more direct view of intrinsic fuel-driven combustion responses that are often masked by active electronic control strategies. Consequently, these findings provide unique validation data for future numerical modeling and practical insights into the emission management of small-scale industrial engines, which are underrepresented in the literature.

2. Experimental Setup

2.1. Experimental Apparatus

All experimental tests were conducted on an ESSOM MT502E single-cylinder engine test rig (Bangkok, Thailand). The test engine was a Mitsuki MIT-178F model featuring an air-cooled, four-stroke, naturally aspirated, direct-injection diesel configuration with a displacement of 305 cm³ and a compression ratio of 21:1. The primary technical specifications of the engine are summarized in

Table 1. Specifications of the test engine.

Parameter	Specifications
Model	MITSUKI MIT-178F
Number of cylinders	1
Bore	78 mm
Stroke	62.5 mm
Cylinder volume	305 cm3
Compression ratio	21:1
Rated power	5.22 kW @ 3000 rpm
Injection type	Direct injection
Intake system	Naturally aspirated
Cooling system	Air-cooled
Ignition	Compression ignition
Type of loading	Eddy current dynamometer

.

Engine load was regulated by coupling the crankshaft to an eddy-current dynamometer. Brake torque was measured directly through the dynamometer’s in-line torque transducer, while engine speed was monitored using the integrated speed sensor on the test bench. The fuel mass flow rate was recorded by a high-precision flow meter installed in the fuel supply line. Intake air mass flow was evaluated from the pressure differential across a sharp-edged orifice plate housed in an air box, with the differential pressure read by an inclined-tube manometer and converted to mass flow using a pre-calibrated correlation. Exhaust gas temperature was measured by a K-type thermocouple positioned immediately downstream of the exhaust port. A schematic diagram of the overall experimental setup, including the fuel system, measurement instrumentation, and data-acquisition components, is presented in Figure 1.

Exhaust gas composition was analyzed using a QRO-402 portable gas analyzer (QroTech, Bucheon, Republic of Korea). This instrument uses non-dispersive infrared spectroscopy for the detection of CO and HC, while NO_x concentrations are determined by an electrochemical sensor. The analyzer maintains a sampling flow rate of 4–6 L/min and offers measurement resolutions of 0.01% for CO, 0.1% for CO₂, and 1 ppm for both HC and NO_x. Smoke opacity was characterized using an OPA-102 opacimeter (QroTech, Bucheon, Republic of Korea), which operates on the principle of light absorption. This device provides opacity readings with a resolution of 0.1% and a rapid response time of 0.5 s, after a 3–6 min warm-up period. Detailed specifications regarding the measurement ranges, accuracy, and uncertainty of these instruments are summarized in

Table 2. Measurement range, accuracy, and uncertainty for the gas analyzer and smoke meter.

Emissions	Range	Accuracy	Uncertainty (%)
NOx	0–5000 ppm	±15 ppm	±3.6
CO	0–10%	±0.02%	±3.4
HC	0–9999 ppm	±20 ppm	±3.6
CO2	0–20%	±0.06%	±3.8
Smoke opacity	0–100%	±1%	±7.2

Note: At low HC concentration levels, the reported values are primarily discussed in terms of comparative trends rather than absolute magnitudes.

.

2.2. Test Fuels and Properties

Pure diesel (D100) was employed as the reference fuel. Three n-butanol–diesel blends were prepared on a volumetric basis: D95B05 (95% diesel + 5% n-butanol), D90B10 (90% diesel + 10% n-butanol), and D85B15 (85% diesel + 15% n-butanol). Compared with diesel, n-butanol has a lower heating value ($LHV$), higher oxygen content, and a lower CN, all of which are known to influence ignition characteristics, premixing behavior, and heat release processes.

As the physicochemical properties of the blended fuels were not directly measured in this study,

Table 3. Physicochemical properties of base fuels.

Properties	Diesel	n-Butanol
Lower heating value (MJ/kg)	43.2	33.1
Latent heat of vaporization (MJ/kg)	0.27	0.58
Cetane number	51	17
Self-ignition temperature (°C)	177	345
Density (kg/m3)	840	814
Kinematic viscosity at 40 °C (mm2/s)	3.75	2.69
Stoichiometric $AFR$ *	14.97	11.16
Oxygen (wt.%)	0	21.6

* $AFR$: air-fuel ratio.

summarizes the fundamental properties of the two base fuels—diesel and n-butanol. In addition, theoretical CO₂ emissions based on stoichiometric combustion were estimated for each blend to provide a fuel-chemistry-based reference. The calculated values were 3100 g/kg fuel for D100, 3065 g/kg for D95B05 (1.14% reduction relative to D100), 3030 g/kg for D90B10 (2.27% reduction), and 2995 g/kg for D85B15 (3.40% reduction). These reductions reflect the lower carbon content of n-butanol and serve as a theoretical baseline for contextualizing the experimentally observed, work-based CO₂ emission trends under engine operating conditions.

2.3. Test Protocol and Operating Conditions

To systematically evaluate the combustion and emission characteristics of the blended fuels, the engine was operated across a range of speed and load conditions. The test matrix consisted of four engine speeds—1600, 2000, 2400, and 2800 rpm—operated at two load levels corresponding to approximately 25% and 75% of full load.

Before data collection, the engine was warmed up using D100 at idle for about 15 min to establish stable thermal conditions. To avoid cross-contamination among test fuels, a strict fuel-replacement protocol was implemented: the fuel tank was drained and refilled with D100, and the engine was then operated for approximately 3 min to purge residual fuel from the supply lines. For each experimental set-point, the target load was first applied via the dynamometer, after which the engine speed was adjusted using the throttle to reach the specified rpm. Data acquisition began only after the operating condition had fully stabilized, ensuring steady-state measurements.

2.4. Calculation of Performance Parameters

The following key performance and combustion-related parameters were evaluated to characterize the engine’s operation. These parameters were calculated using standard definitions, as detailed below.

The $AFR$ is defined as the ratio of the intake air mass flow rate to the fuel mass flow rate:

$$AFR={\displaystyle \frac{{\dot{m}}_a}{{\dot{m}}_f}}$$

(1)

where ${\dot{m}}_a$ and ${\dot{m}}_f$ denote the mass flow rates of air and fuel (kg/h), respectively.

The air–fuel equivalence ratio ($\lambda$) was calculated as:

$$\lambda ={\displaystyle \frac{{AFR}_{actual}}{{AFR}_{stoi}}}$$

(2)

where ${AFR}_{stoi}$ is the stoichiometric $AFR$. For n-butanol–diesel blends, ${AFR}_{stoi}$ was determined on a mass basis by converting the volumetric blending ratios into mass fractions using the respective fuel densities.

$BTE$ was calculated as the ratio of brake power output to the chemical energy input of the fuel:

$$BTE={\displaystyle \frac{P_b}{0.278{\dot{m}}_f\cdot LHV}}$$

(3)

where $P_b$ is the brake power (kW).

Brake-specific fuel consumption ($BSFC$) represents the mass of fuel consumed per unit brake power output and is defined as:

$$BSFC={\displaystyle \frac{{\dot{m}}_f}{P_b}}\times {10}^3$$

(4)

where the unit is typically expressed in g/kWh.

To enable an energy-based comparison among fuels with different heating values, $BSEC$ was calculated as:

$$BSEC=BSFC\times LHV={\displaystyle \frac{{\dot{m}}_f\cdot LHV}{P_b}}$$

(5)

The resulting $BSEC$ values are expressed in MJ/kWh and represent the fuel energy input required per unit of brake power output.

3. Results and Discussion

3.1. Brake Thermal Efficiency

$BTE$, defined as the ratio of brake power output to the chemical energy input of the fuel, is widely used to evaluate overall engine performance. In this study, the diesel-only (D100) results are first used to establish a baseline, and the effects of n-butanol blending are discussed primarily in a comparative manner relative to this baseline. As shown in Figure 2, $BTE$ increased with engine speed under all fuel conditions.

At the 25% low-load condition (Figure 2a), the $BTE$ of D100 increased from 10.59% at 1600 rpm to 12.75% at 2800 rpm, reflecting improved fuel atomization and air–fuel mixing at higher engine speeds, together with a reduced relative contribution of mechanical friction losses. Compared to this baseline behavior, n-butanol–blended fuels exhibited slightly higher $BTE$ values, particularly at medium-to-high engine speeds. While D85Bu15 showed a comparable $BTE$ to D100 at 1600 rpm (10.53%), it achieved 12.64% at 2400 rpm and 12.91% at 2800 rpm, corresponding to improvements of approximately 0.14–0.16 percentage points relative to D100.

These efficiency gains are suggested to result from combustion characteristics commonly associated with n-butanol’s prolonged ignition delay, which promotes premixed combustion, as well as the fuel-borne oxygen that enhances combustion completeness. At 25% load, the $\lambda$ remained in the lean range of 1.9–2.1 for all fuels (

Table 4. Air-fuel equivalence ratio, $\lambda$.

Engine Load (%)	Engine Speed (rpm)	D100	D95Bu05	D90Bu10	D85Bu15
25	1600	1.98	2.02	2.03	1.97
	2000	2.08	2.08	2.07	2.08
	2400	1.97	1.97	1.98	1.98
	2800	1.91	1.93	1.93	1.94
75	1600	1.13	1.13	1.13	1.13
	2000	1.25	1.25	1.24	1.24
	2400	1.17	1.16	1.15	1.14
	2800	1.15	1.14	1.13	1.13

), indicating that the improved mixture formation induced by n-butanol blending outweighed the cooling losses associated with its high latent heat of vaporization [12,14,19].

At the 75% high-load condition (Figure 2b), $BTE$ values were approximately 7–10 percentage points higher than those observed at 25% load due to elevated in-cylinder temperatures and reduced pumping losses. For D100, $BTE$ increased from 17.49% at 1600 rpm to 22.80% at 2800 rpm. Under these conditions, the beneficial effect of n-butanol blending became more pronounced. In particular, D85Bu15 consistently outperformed D100 across the entire speed range, reaching 21.41% at 2000 rpm, which corresponds to a maximum improvement of approximately 0.36 percentage points relative to the diesel baseline.

Despite reducing $\lambda$ to approximately 1.1–1.2 at high load, the elevated combustion temperatures and intensified turbulence likely promoted rapid vaporization and mixing of n-butanol. This accelerated mixture formation presumably compensated for potential adverse combustion characteristics associated with ignition delay, thereby maximizing thermal efficiency [16]. Overall, while $BTE$ increased with engine speed and load across all fuels, blending 5–15 vol% n-butanol consistently enhanced thermal efficiency compared to D100, primarily through improved oxygen availability and mixture formation, with the most pronounced benefits observed under high-speed, high-load operating conditions [10].

3.2. Brake-Specific Energy Consumption

$BSEC$, defined as the fuel energy input required to produce a unit of brake power, provides a valuable metric for comparing fuels with different heating values. As shown in Figure 3, $BSEC$ decreased with increasing engine speed under both load conditions, indicating improved energy utilization at higher rotational speeds.

At the 25% low-load condition (Figure 3a), the $BSEC$ of D100 decreased from 33.95 MJ/kWh at 1600 rpm to 28.22 MJ/kWh at 2800 rpm, corresponding to a reduction of approximately 17%. This behavior reflects enhanced turbulence intensity and improved air–fuel mixing at higher engine speeds, which alleviates combustion inefficiencies associated with poor atomization and extended combustion duration at low speeds. Relative to this baseline trend, n-butanol–blended fuels exhibited comparable or slightly lower $BSEC$ values, particularly at higher engine speeds. Above 2400 rpm, D85Bu15 achieved a $BSEC$ reduction of approximately 0.8–1.2% compared to D100.

Despite the increase in mass-based fuel consumption associated with the lower heating value of n-butanol (see Section 3.3), the promotion of premixed combustion by fuel-borne oxygen enhanced overall thermal efficiency, resulting in a net reduction in energy input. This improvement indicates that, under low-load conditions, the beneficial effects of improved mixture formation outweighed the penalties associated with reduced fuel calorific value [11].

At the 75% high-load condition (Figure 3b), $BSEC$ values were substantially lower than those at 25% load due to elevated in-cylinder temperatures and pressures. For D100, $BSEC$ decreased from 20.57 MJ/kWh at 1600 rpm to 15.77 MJ/kWh at 2800 rpm, reflecting more efficient conversion of chemical energy into practical work. Under these conditions, the energy efficiency improvement associated with n-butanol blending became more pronounced. In particular, at 2000 rpm, D85Bu15 recorded a $BSEC$ of 16.80 MJ/kWh, corresponding to a reduction of approximately 1.7% relative to the diesel baseline.

This behavior suggests that the high thermal environment under high-load operation effectively compensates for the cooling losses associated with the latent heat of vaporization of n-butanol. Consequently, enhanced vaporization, improved mixture formation, and self-oxygenation effects dominated the combustion process, leading to a measurable reduction in energy consumption compared to D100 [23].

3.3. Brake-Specific Fuel Consumption

$BSFC$, defined as the mass of fuel required to produce a unit of brake power, is a primary indicator of engine fuel economy. As shown in Figure 4, $BSFC$ decreased with increasing engine speed under both load conditions, reflecting improved combustion stability and reduced relative loss contributions at higher speeds.

At the 25% low-load condition (Figure 4a), the $BSFC$ of D100 decreased from 791.5 g/kWh at 1600 rpm to 657.8 g/kWh at 2800 rpm. Compared with D100, n–butanol–blended fuels consistently exhibited higher $BSFC$ values across the entire speed range, with the increase becoming more pronounced as the blending ratio increased. At 1600 rpm, D90Bu10 and D85Bu15 recorded $BSFC$ values of 812.7 g/kWh and 823.5 g/kWh, respectively, corresponding to increases of approximately 2.7% and 4.0% compared to D100.

This increase in $BSFC$ is primarily attributed to the $LHV$ of n-butanol (approximately 33.1 MJ/kg), which is about 23% lower than that of diesel (approximately 43.2 MJ/kg), thereby requiring a larger fuel mass to deliver the same shaft power output [16,17]. In addition, under lean operating conditions ($\lambda \approx 2.0$) at 25% load, the lower CN of n-butanol may have reduced ignition stability, thereby negatively affecting combustion efficiency, as reported in previous studies [25].

At the 75% high-load condition (Figure 4b), $BSFC$ values were substantially lower than those at 25% load for all fuels due to elevated in-cylinder temperatures that promote stable atomization and ignition, as well as a reduced relative contribution of mechanical losses. For D100, $BSFC$ decreased from 479.4 g/kWh at 1600 rpm to 367.7 g/kWh at 2800 rpm. Under these conditions, although $BSFC$ remained higher for n-butanol–blended fuels, the magnitude of the increase was noticeably lower than in low-load operation. At 2800 rpm, D85Bu15 exhibited a $BSFC$ of 375.0 g/kWh, corresponding to an increase of approximately 2.0% relative to the diesel baseline.

This behavior indicates that the high thermal environment at elevated load levels partially compensates for the cooling effect associated with n-butanol’s latent heat of vaporization and shortens the ignition delay. Consequently, while n-butanol blending inevitably increases $BSFC$ due to its lower energy density, this trend should be interpreted in conjunction with the $BSEC$ reduction discussed in Section 3.2. From an energy-based perspective, the observed increase in $BSFC$ primarily reflects differences in fuel properties rather than additional losses from incomplete combustion.

3.4. NO_x Emissions

NO_x emissions are primarily formed through high-temperature reactions between nitrogen and oxygen and are strongly influenced by combustion temperature, residence time, and local oxygen availability. In this study, NO_x emissions are discussed on a work-based basis (g/kWh) to enable objective comparison across operating conditions. As shown in Figure 5, the influence of engine speed and n-butanol blending on NO_x emissions differed markedly depending on engine load.

At the 25% low-load condition (Figure 5a), NO_x emissions for D100 increased from 5.54 g/kWh at 1600 rpm to a maximum of 6.90 g/kWh at 2000 rpm, followed by a gradual decrease to 6.15 g/kWh at 2800 rpm. This non-monotonic trend reflects the combined effects of increasing in-cylinder temperature with engine speed and the reduction in residence time at higher speeds. Compared to D100, n-butanol–blended fuels consistently exhibited lower NO_x emissions under low-load operation across the entire speed range. In particular, D85Bu15 recorded NO_x values of 4.93 g/kWh at 1600 rpm and 5.54 g/kWh at 2800 rpm, corresponding to reductions of approximately 10–11% relative to the diesel baseline. This reduction is attributed to the charge-cooling effect induced by n-butanol’s high latent heat of vaporization, which suppresses peak combustion temperatures under lean operating conditions [16,18].

At the 75% high-load condition (Figure 5b), overall NO_x emission levels were higher than those at low load due to increased fuel injection and elevated in-cylinder temperatures associated with reduced $\lambda$. For D100, NO_x emissions decreased gradually from 6.51 g/kWh at 1600 rpm to 5.88 g/kWh at 2800 rpm. Under these conditions, n-butanol blending generally resulted in higher NO_x emissions than D100 at low-to-medium engine speeds. For example, at 2000 rpm, D90Bu10 recorded 6.89 g/kWh compared to 6.58 g/kWh for D100. This increase in NO_x emissions is attributed to two coupled physicochemical effects; first, the lower CN of n-butanol is widely reported to induce a longer ignition delay, which tends to increase the premixed combustion fraction and can lead to more rapid heat release, thereby elevating local flame temperatures [22]. Second, the presence of fuel-borne oxygen enhances local oxygen availability in high-temperature regions, accelerating nitrogen oxidation. This interpretation is further supported by higher exhaust gas temperatures (EGT) observed for n-butanol blends compared to D100 at high load, as shown in

Table 5. Exhaust gas temperature.

Engine Load (%)	Engine Speed (rpm)	D100 (°C)	D95Bu05 (°C)	D90Bu10 (°C)	D85Bu15 (°C)
25	1600	141	147	150	154
	2000	166	166	169	172
	2400	181	181	184	187
	2800	197	198	200	205
75	1600	335	339	344	352
	2000	362	370	372	382
	2400	385	395	398	411
	2800	394	413	420	436

.

The observed load-dependent behavior indicates that a balance between the charge cooling effect and the enhancement of local temperature and oxygen concentration governs the impact of n-butanol blending on NO_x formation. Under low-load conditions, the cooling effect associated with n-butanol’s high latent heat dominates, leading to a consistent reduction in work-based NO_x emissions. In contrast, at high load and moderate engine speeds, elevated ambient temperatures diminish this cooling effect, allowing premixed combustion characteristics and fuel-borne oxygen effects to prevail, thereby promoting NO_x formation. At higher engine speeds (2400–2800 rpm), however, the NO_x emissions of n-butanol blends remained comparable to those of D100, likely due to the combined influence of reduced residence time and power normalization effects.

3.5. CO Emissions

CO is an intermediate product of incomplete hydrocarbon oxidation, primarily influenced by local combustion temperature, oxygen availability, and residence time. As shown in Figure 6, while CO emissions were highly sensitive to engine load and speed, n-butanol blending consistently reduced CO emissions relative to the diesel baseline across all tested conditions. This trend confirms that the oxygenated nature of n-butanol (21.6 wt.%) effectively promotes carbon oxidation [15,40].

At the 25% low-load condition (Figure 6a), CO emissions for D100 increased from 7.17 g/kWh at 1600 rpm to 11.70 g/kWh at 2800 rpm. This increase is likely due to the combined effects of shortened oxidation residence time and enhanced wall quenching at higher engine speeds. In contrast, n-butanol blending effectively mitigated these trends. At 2800 rpm, D85Bu15 recorded a CO value of 9.06 g/kWh, a 22.5% reduction relative to D100. This improvement is attributed to the high oxygen content of n-butanol (21.6 wt.%) and improved mixture homogeneity, which together enhance oxidation completeness even under lean, low-temperature conditions [37].

At the 75% high-load condition (Figure 6b), overall CO emission levels were substantially lower than those at low load, reflecting improved combustion efficiency at elevated temperatures. For D100, CO emissions followed a non-monotonic trend, decreasing from 4.48 g/kWh at 1600 rpm to a minimum of 3.91 g/kWh at 2000 rpm before rising to 6.17 g/kWh at 2800 rpm. The higher emissions at low engine speeds are linked to limited turbulence and suboptimal atomization in the mechanical injection system. In contrast, the increase at high speeds is constrained by reduced residence time.

Under these high-load conditions, n-butanol blending demonstrated an even more pronounced suppression of CO. For instance, D85Bu15 achieved 2.96 g/kWh at 2000 rpm and 4.71 g/kWh at 2800 rpm, corresponding to significant reductions of 24.2% and 23.7%, respectively, compared to D100. This improvement suggests that fuel-borne oxygen and the intensified premixed combustion associated with n-butanol effectively counteract the formation of locally fuel-rich zones, promoting more complete oxidation, even under near-stoichiometric conditions [18].

Overall, while high engine speeds tend to elevate CO formation by limiting oxidation time, n-butanol blending consistently lowers work-based CO emissions. By enhancing oxygen availability and mixture quality, n-butanol offsets the mechanical injection system’s physical limitations across both load ranges.

3.6. HC Emissions

HC emissions originate primarily from incomplete oxidation or flame interruption in low-temperature regions, such as wall-quenching zones and crevice volumes. As shown in Figure 7, HC emissions increased systematically with both engine speed and n-butanol blending ratio under all load conditions, indicating a strong sensitivity of HC formation to fuel properties and combustion characteristics.

At the 25% low-load condition (Figure 7a), HC emissions for D100 increased from 0.37 g/kWh at 1600 rpm to 0.62 g/kWh at 2800 rpm. This increase is mainly attributed to the shortened residence time at higher engine speeds, which limits fuel oxidation under lean combustion conditions. Compared to D100, n-butanol–blended fuels exhibited substantially higher HC emissions across the entire speed range. At 2800 rpm, D85Bu15 recorded an HC emission of 1.33 g/kWh, which represents more than a twofold increase relative to the diesel baseline. This behavior is primarily linked to the lower CN of n-butanol, which is associated with prolonged ignition delay. Such behavior allows a greater fraction of the injected fuel to diffuse toward the peripheral regions of the combustion chamber, forming excessively lean mixtures. In addition, the high latent heat of vaporization of n-butanol induces localized charge cooling, promoting flame quenching and incomplete oxidation [25].

At the 75% high-load condition (Figure 7b), overall HC emissions were lower than at low load due to elevated in-cylinder temperatures that favor oxidation; however, the increasing trend with n-butanol blending remained pronounced. For D100, HC emissions increased from 0.27 g/kWh at 1600 rpm to 0.34 g/kWh at 2800 rpm. Under these conditions, the impact of n-butanol was even more significant; at 2800 rpm, D85Bu15 recorded 0.62 g/kWh, corresponding to an increase of approximately 80% relative to D100. As discussed in Section 3.4, this increase occurred despite the high thermal environment that favors NO_x formation. This apparent contradiction is likely due to the specific combustion characteristics of n-butanol, in which a prolonged ignition delay promotes fuel diffusion toward the chamber periphery. Consequently, although the bulk gas temperature is high, local cooling near the spray tip and walls inhibits oxidation reactions and enhances HC survival [18,22].

Overall, HC emissions increased monotonically with n-butanol blending regardless of engine load. This behavior confirms that the formation of over-lean mixtures and localized flame quenching—driven by the low ignitability and high latent heat of vaporization of n-butanol—are the dominant mechanisms governing HC emissions in this mechanically injected engine platform.

3.7. CO₂ Emissions

The complete oxidation of hydrocarbon fuels produces CO₂ and is closely linked to fuel consumption and overall combustion efficiency. As shown in Figure 8, CO₂ emissions showed strong dependence on engine load and speed; however, a systematic reduction in CO₂ was observed with increasing n-butanol blending ratio across both load conditions.

At the 25% low-load condition (Figure 8a), CO₂ emissions for D100 decreased from 2061 g/kWh at 1600 rpm to 1801 g/kWh at 2800 rpm. This reduction with increasing engine speed reflects the improved thermal efficiency and reduced energy consumption discussed in Section 3.1 and Section 3.2. Compared to D100, n-butanol–blended fuels consistently exhibited lower CO₂ emissions across the entire speed range. At 2800 rpm, CO₂ emissions decreased to 1733 g/kWh for D85Bu15, a 3.8% reduction relative to the diesel baseline. This trend is primarily attributed to the lower carbon-to-hydrogen (C/H) ratio of n-butanol, which results in reduced CO₂ formation per unit of brake power output, consistent with prior literature [17]. Additionally, as discussed in Section 3.6, the significant increase in HC emissions with n-butanol blending under low-load conditions may have contributed secondarily to the observed CO₂ reduction.

At the 75% high-load condition (Figure 8b), CO₂ emissions were substantially lower on a g/kWh basis than those at low load, reflecting the significantly higher $BTE$ achieved at elevated loads. For D100, CO₂ followed a non-monotonic trend, reaching a minimum of 1100 g/kWh at 2000 rpm before increasing to 1164 g/kWh at 2800 rpm. This behavior is associated with the combined influence of optimal energy utilization at intermediate speeds and increased fuel consumption required at higher speeds under near-stoichiometric operation.

Even under high-load conditions, n-butanol blending maintained a consistent reduction in CO₂ emissions. At 2800 rpm, CO₂ emissions decreased from 1164 g/kWh for D100 to 1070 g/kWh for D85Bu15, representing an 8.1% reduction. The magnitude of CO₂ reduction was even more pronounced at high load, confirming that the lower carbon content of n-butanol plays a dominant role in governing CO₂ formation even under severe operating conditions [15].

For reference, theoretical CO₂ emissions based solely on stoichiometric combustion and fuel composition predict reductions of approximately 1.1–3.4% per unit mass of fuel for 5–15 vol% n-butanol blends relative to D100 (see Section 2.2). The larger reductions observed on a work-based (g/kWh) basis—reaching up to 8.1% at high load—indicate that, beyond fuel carbon content, the improved $BTE$ under engine operating conditions contributes significantly to the measured CO₂ mitigation.

Overall, increasing the n-butanol blending ratio consistently reduced work-based CO₂ emissions relative to diesel operation. This result demonstrates that oxygenated fuels with lower carbon content can effectively mitigate tailpipe CO₂ emissions on an energy-normalized basis. However, a comprehensive assessment of carbon footprint would require a complete life-cycle analysis.

3.8. Smoke Opacity

Smoke opacity is an indicator of the relative concentration of PM emitted during combustion and is influenced by fuel molecular structure, oxygen content, and in-cylinder mixing characteristics. As shown in Figure 9, smoke opacity varied with engine load and speed; however, compared to the diesel baseline, a consistent reduction was observed with increasing n-butanol blending ratio under all tested conditions.

At the 25% low-load condition (Figure 9a), smoke opacity for D100 increased with engine speed, rising from 5.0% at 1600 rpm to 6.8% at 2800 rpm. This behavior is attributed to the formation of locally rich zones at the spray tip or within dead volumes, despite overall lean conditions, and to shortened combustion duration at higher speeds, which limits the time available for soot oxidation.

Compared with D100, n–butanol–blended fuels exhibited lower smoke opacity across the entire speed range. At 2800 rpm, D90Bu10 and D85Bu15 recorded smoke opacity values of 5.80% and 4.90%, corresponding to reductions of 14.7% and 27.9%, respectively. This reduction is primarily attributed to the high oxygen content of n-butanol, which promotes oxidation reactions within the diffusion flame and inhibits the formation pathways of PAHs, known soot precursors [20]. In addition, combustion behavior often linked to the prolonged ignition delay of n-butanol blends is suggested to increase the premixed combustion fraction; this enhances mixture homogeneity during the early combustion phase, thereby suppressing soot formation [40].

At the 75% high-load condition (Figure 9b), absolute smoke opacity increased markedly due to higher fuel injection and the expansion of rich combustion zones as $\lambda$ decreased to approximately 1.1–1.2. For D100, smoke opacity rose from 13.0% at 1600 rpm to 23.2% at 2800 rpm. Although elevated in-cylinder temperatures and pressures under high-load operation enhance soot oxidation, the soot formation rate associated with excessive fuel supply outweighs the oxidation capability, resulting in an overall increase in smoke emissions.

Despite these conditions, n-butanol blending substantially reduced smoke opacity relative to diesel operation. At 2800 rpm, D90Bu10 and D85Bu15 recorded smoke opacity values of 16.1% and 12.8%, corresponding to significant reductions of 30.6% and 44.8%, respectively. This improvement is attributed to two complementary mechanisms: the chemical effect of fuel-borne oxygen, which enhances soot oxidation under sufficiently high-temperature conditions [17], and the physical impact of reduced fuel viscosity, which improves atomization and expands the premixed combustion phase, thereby reducing the extent of rich diffusion flames [12,14].

4. Conclusions

In this study, the effects of blending 5–15 vol% n-butanol on combustion performance and emission characteristics were experimentally investigated in a single-cylinder, air-cooled diesel engine under 25% low-load and 75% high-load conditions. The main findings are summarized as follows:

• Despite its lower calorific value than D100, n-butanol blending improved $BTE$, particularly at high speed and load. This improvement is attributed to the oxygenated nature of n-butanol and a likely increase in the premixed combustion fraction—a behavior commonly associated with n-butanol blends—which enhances in-cylinder mixture formation and combustion. Consequently, although mass-based $BSFC$ increased, $BSEC$ decreased, indicating improved energy conversion efficiency.
• NO_x emissions exhibited a strong dependence on engine load. At 25% low load, the charge-cooling effect induced by n-butanol’s high latent heat of vaporization dominated, resulting in reduced NO_x emissions. In contrast, at 75% high load, elevated combustion temperatures offset the cooling effect, while intensified premixed combustion and increased local oxygen availability became dominant, leading to a moderate increase in NO_x emissions.
• n-Butanol blending consistently reduced smoke and GHG emissions. Smoke opacity was significantly reduced across the entire operating range due to the combined effects of fuel-borne oxygen and suppression of soot precursor formation. CO₂ emissions also decreased relative to D100, primarily due to the lower C/H ratio of n-butanol, while CO emissions decreased due to enhanced oxidation reactions.
• In contrast, HC emissions increased with increasing n-butanol blending ratio under all operating conditions. This behavior is attributed to combustion characteristics often associated with n-butanol blends, including potential ignition-delay-related effects, local flame quenching, and the formation of over-lean mixtures. Accordingly, future work should focus on tailoring established mitigation strategies—such as after-treatment and combustion control—to this specific mechanically injected platform. To support this optimization, further studies will incorporate in-cylinder pressure analysis to quantitatively assess combustion phasing and ignition delay, together with detailed PM morphology analysis to clarify particulate formation mechanisms. These efforts will enable a more precise quantitative evaluation of the trade-offs among combustion efficiency, NO_x, HC, and smoke emissions for n-butanol blends.