Abstract
Fossil fuels such as diesel have consistently played a key role in energy production. Owing to the nonrenewable nature of fossil fuel resources and the pollution resulting from their combustion, numerous environmental problems have emerged. Therefore, the use of clean, renewable, safe, and cost-effective energy sources capable of meeting global energy demands is of vital importance. Research has shown that renewable fuels alone are not suitable for use in diesel engines; consequently, the blending of multiple fuels has attracted increasing attention. In the present study, the simultaneous effects of a quaternary fuel blend (diesel, biodiesel, bioethanol, and ammonia) on the performance and emissions of a single-cylinder direct-injection diesel engine were investigated, and the response surface methodology (RSM) was employed as an advanced statistical approach to optimize fuel and engine operating parameters. The results indicated that the optimal blend, consisting of 40% diesel, 20% biodiesel, 20% bioethanol, and 20% ammonia, was able to reduce CO and HC emissions by 65% and 60%, respectively, although NOx emissions increased by 21%. Furthermore, engine performance decreased by 23% as the diesel fraction in the fuel blend was reduced from 100% to 40%; however, the addition of ammonia partially compensated for this reduction. This study highlights the importance of utilizing various combinations of alternative fuels for application in internal combustion engines.
EXTENDED ABSTRACT
Introduction 
Energy is a fundamental need of modern society, traditionally supplied by fossil fuels such as diesel. However, the non-renewable nature of fossil resources and their environmental impacts, including greenhouse gas emissions and particulate matter, necessitate the development of sustainable alternatives. Biofuels, such as biodiesel and bioethanol, have been widely investigated as renewable and oxygenated fuel sources. Biodiesel improves combustion and reduces CO and HC emissions, but suffers from high viscosity and poor low-temperature properties. In contrast, bioethanol enhances cold flow properties and fuel stability, making their blends with diesel more attractive. Ammonia has recently emerged as a promising carbon-free fuel due to its high hydrogen content, storability, and potential to reduce CO₂ emissions significantly. Despite challenges such as toxicity, slow combustion, and increased NOx emissions, ammonia can be used in dual- or multi-fuel strategies to optimize performance and emissions. 
Several studies have demonstrated that blending biodiesel, bioethanol, and ammonia with diesel can improve combustion efficiency, lower particulate and CO₂ emissions, and reduce dependence on fossil fuels. Additives such as TiO₂ nanoparticles further enhance combustion and catalytic activity, contributing to lower fuel consumption and pollutant formation.Nevertheless, most prior research has focused on binary or ternary blends, while limited work has explored quaternary mixtures of diesel, biodiesel, bioethanol, and ammonia in compression ignition engines. In this context, this study applies response surface methodology (RSM) to evaluate and optimize the combined effects of these fuels on engine performance and emissions, addressing both energy security and environmental sustainability.
Material and Methods 
Tests were conducted on a single-cylinder, naturally aspirated Mitsubishi NM45 diesel engine (operating at 3000 rpm, water-cooled, with direct injection). The fuels included commercial diesel, biodiesel from sunflower oil via transesterification, bioethanol, and liquefied ammonia synthesized by the Haber process. Blends of diesel (40–100%), biodiesel (0–40%), bioethanol (0–20%), and ammonia (0–20%) were examined at engine loads of 25, 50, and 75%. The engine was coupled to a hydraulic dynamometer for torque and power measurement, while a digital tachometer controlled speed.
 A Testo 330 gas analyzer measured exhaust emissions, and ammonia flow was regulated by a Krohne DK 800 R flowmeter. Data acquisition was performed through a PLC-based monitoring unit (Atech/Delta 10SXR). Before each test, the system was flushed with the target fuel, and measurements were taken under steady-state conditions. Each experiment was repeated three times. Response surface methodology (RSM) with a mixture design was employed to model and optimize performance and emissions.
Results and Discussion  
The statistical analysis of variance (ANOVA) demonstrated that the combined effects of fuel type, engine load, and ammonia addition significantly influenced emissions of CO, NOx, and HC. Most interactions among diesel, biodiesel, bioethanol, ammonia, and engine load were significant at the 1% or 5% levels, confirming the strong dependence of emissions on both fuel composition and operating conditions. The regression models developed for each pollutant exhibited excellent predictive performance, with R² values of 0.9961 for CO, 0.9959 for NOx, and 0.9635 for HC, indicating close agreement between predicted and experimental data. Lack-of-fit was not significant, further supporting the adequacy of the models.
For CO emissions, the lowest level (146 ppm) was observed at 25% load when the engine operated on a mixture of 40% diesel, 40% biodiesel, 20% bioethanol, and 20% ammonia in the intake air. In contrast, the highest CO level (423 ppm) was recorded with pure diesel and no ammonia. The reduction of CO by up to 65% under blended fuel conditions was attributed to the oxygenated nature of biodiesel and bioethanol, as well as the carbon-free structure of ammonia, which collectively promoted more complete combustion. At higher loads (75%), CO emissions further decreased, reaching a minimum of 86 ppm under conditions of high ammonia and low diesel fractions, due to improved combustion completeness.
In the case of NOx, emissions increased with higher shares of biodiesel, bioethanol, and ammonia. The maximum value (233 ppm) occurred at a 75% load with a blend of 40% diesel, 40% biodiesel, 20% bioethanol, and 20% ammonia. These increases were attributed to higher combustion temperatures, oxygen availability, and the nitrogen content of ammonia. Load was the most dominant factor, as increasing the load from 25% to 75% resulted in a 49% increase in NOx.
For HC emissions, the lowest concentration (50 ppm) was achieved with a mixture of 40% diesel, 40% biodiesel, 20% bioethanol, and 20% ammonia at a 75% load. Oxygenated fuels enhanced combustion efficiency and reduced HC emissions by 50–60% compared with pure diesel. However, load variations also played a key role, with higher loads significantly reducing HC due to improved combustion conditions.
Regarding engine performance, the maximum torque (14.34 Nm) and power (3.3 kW) were observed at a 75% load with an 80% diesel, 20% biodiesel, and 20% ammonia blend. While replacing diesel with renewable fuels reduced performance by ~23%, ammonia addition slightly improved it (~5%). These findings highlight that optimized fuel blends can reduce emissions without significantly compromising performance.
Conclusions
This study investigated the effects of diesel, biodiesel, bioethanol, and ammonia blends on the performance and emissions of a single-cylinder diesel engine. Using Response Surface Methodology (RSM), the influence of fuel composition and engine load was optimized and analyzed. Results showed that the blend of 40% diesel, 40% biodiesel, 20% bioethanol, and 20% ammonia achieved the lowest HC (50 ppm) at 75% load and reduced CO emissions by 65% at 25% load compared with pure diesel. However, this blend produced the highest NOx (233 ppm), while pure diesel at 25% load yielded the lowest (98 ppm). Replacing diesel with biofuels decreased power by approximately 23%, although the addition of ammonia improved performance by nearly 5%. The maximum power (3.3 kW) and torque (14.3 N·m) were achieved with a 80% diesel, 20% biodiesel, and 20% ammonia blend at 75% load. Higher loads increased NOx but reduced HC and CO, indicating more complete combustion.
Acknowledgement
The authors would like to express their gratitude to the Research Council of Shahrekord University for financially supporting this study under grant number 96GRN31M1007.
Author Contributions
Abolfazl Ziaadini: Methodology
Sajjad Rostami: Supervisor
Bahram Hossinzade Samani: Validation
Mostafa Kiani: Validation
Data Availability Statement
"Not applicable"
Ethical Considerations
This section states ethical approval details (e.g., Ethics Committee, ethical code) and confirms adherence to ethical standards, including avoidance of data fabrication, falsification, plagiarism, and misconduct.
Conflict of Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
Funding Statement
This work was supported by Research Council of Shahrekord University under grant number 96GRN31M100