Framework for Energy-Averaged Emission Mitigation Technique Adopting Gasoline-Methanol Blend Replacement and Piston Design Exchange

Abstract

Measurement to mitigate automotive emission varies from energy content modification of fuel to waste energy recovery through energy system upgradation. The proposed energy-averaged emission mitigation technique involves interfacing piston design exchange and gasoline–methanol blend replacement with traditional gasoline for low carbon high energy content creation. Here, we interlinked the CO, CO₂, NO_x, O₂, and HC to different design exchanges of coated pistons through the available brake power and speed of the engine. We assessed the relative effectiveness of various designs and coating thicknesses for different gasoline–methanol blends (0%,5%,10%, and 15%). The analysis shows the replacement of 5%, 10%, and 15% by volume of gasoline with methanol reduces the fuel carbon by 4.167%, 8.34%, and 12.5%, respectively. The fuel characteristics of blends are comparable to gasoline, hence there is no energy infrastructure modification required to develop the same amount of power. The CO and HC reduced significantly, while CO₂ and NO_x emissions are comparable. Increasing the coating thickness enhances the surface temperature retention and reduces heat transfer. The Type_C design of the steel piston and type_A design of the AlSi piston show temperature retention values of 582 °C and 598 °C, respectively. Type_A and type_B pistons are better compared to type_C and the type_D piston design for emission mitigation due to decarbonization of fuel through gasoline-methanol blend replacement. Surface response methodology predicts Delastic, σvon Mises, and Tsurface with percentage errors of 0.0042,0.35, and 0.9, respectively.

1. Introduction

An increase in the demand for emission control and stringent regulation [1] for energy usage forces energy investors to switch from high-carbon high-energy sources to low-carbon high-energy sources to meet day-to-day energy demands [2]. Energy-intensive carbon mitigation strategies, such as carbon avoidance, carbon embedding, and carbon removal policies are adopted for the faster control of climate degradation through intelligent energy usage [3]. The simultaneous restructuring of the energy infrastructure [4,5,6] and energy content modification are highly expensive reformation initiatives in the energy sector [7]. Such decoupling may lead to energy poverty due to the sudden termination of specialty sectors to be replaced with clean and green carbon-negative energy resources [8,9]. The intelligent interconnection of different energy resources in the sequential grid to deal with the high fluctuation of energy mode swing would be a better approach towards the restructuring of the energy infrastructure [10]. Energy decision making in the global transition to a low-carbon, high energy efficiency-based emission mitigation technique sometimes proves to be extremely costly energy-wise, and adds more expenditure on the global energy budget¹¹. An energy innovation strategy with minimum infrastructural reformation and optimum decarbonization [11,12] potential would not trigger much in the way of the social dynamics of energy behavior. It would be a sustainability energy alternative for the near future.

Through the carbon avoidance technique, carbon content at 4.167% (B₅), 8.34% (B₁₀), and 12.5% (B₁₅) could be removed from traditional gasoline fuel through blending to study the emission mitigation potential of the newly formulated fuel [7]. Carbon removal up to 15% through blending [13] hardly affects the energy content of fuel¹¹. Moreover, no significant reformation in the energy infrastructure is required. Toxicity, acid value, and corrosion-like redundant effects are not aggressive in these flexi-fuel formations. With higher heat content, a better-quality flash point, pour point, and viscosity are observed, which are comparable or superior to traditional fuels. However, sometimes, due to an increase in the % of methanol, NO_x emission (though it reduces carbonated emissions)formation is aggravated multiple times due to improved combustion [14] and enhanced exhaust muffler temperature [12,15]. Over the years, methanol as IC engine fuel is greatly evaluated [16] and methanol economy in China [17] is investigated with environmental implications. Cleaner production of methanol [18] is highly compatible with the existing industrial infrastructure and hydrogenation and reformation technology [19]. Methanol blending over the total replacement of fuel, especially with gasoline, is found to be successful due to insignificant changes in fuel properties. Without any compromise in energy and efficiency, the carbonated emissions [14], such CO, CO₂, and HC, reduced significantly due to the use of blends [20]. However, NO_x emission is a concern regarding the use of blends. Though there are techniques, such as the Miller cycle [21], muffler design modification [22], catalytic reduction [22,23], etc., which are suggested to reduce NO_x, an enhanced combustion chamber surface temperature [24,25] necessitates the thermal analysis of engine components with different coatings for different applications. For example, functionally graded coating in AlSi and steel pistons [26] or ceramic-coated pistons, especially for diesel engine [27] application [28], proved to be energy efficient. This is evident from thermal analysis of coated pistons using FEA [29].

Automotive Emission is controlled through multiple upgradation techniques without compromising energy and efficiency of the engine. Through appropriate experimentations, simulation, and optimization [15,30,31], this is proved. However, a simultaneous investigation of blend replacement [32,33,34] and coated piston design exchange analysis, as well as its effect on the energy and efficiency of the engine through a predicted and actual comparison using response surface methodology (RSM) [35,36,37,38],is a step forward andwasthe aim of this investigation. Though there are a limited number of works carried out in the blending of diesel with alcohols [39,40] gasoline–methanol blends and piston design exchange are a further step in this direction. Figure 1a,b show the energy dynamics of engine combustion and the steps for interfacing fuel replacement and piston design exchange.

2. Methods

Fuel carbon reduction through blending and fuel characterization. Gasoline and methanol were added in different proportions to prepare different blends (B₅, B₁₀, and B₁₅). Such a replacement ensures carbon avoidance. Equations (1)–(3) show the amount of carbon removed through blending.

$$\% \mathrm{carbon} \mathrm{removal} \left(B_5\right)=\frac{100\times 6-\left(95\times 6+1\times 5\right)}{100\times 6}=4.167\%$$

(1)

$$\% \mathrm{carbon} \mathrm{removal} \left(B_{10}\right)=\frac{100\times 6-\left(90\times 6+1\times 10\right)}{100\times 6}=8.34\%$$

(2)

$$\% \mathrm{carbon} \mathrm{removal} \left(B_{15}\right)=\frac{100\times 6-\left(85\times 6+1\times 15\right)}{100\times 6}=12.5\%$$

(3)

To evaluate the formulated fuel characteristics, the ASTM standard procedures were followed. Fuel density (by ASTM 4052), kinetic viscosity (using ASTM D445), acid value (using ASTM D664), flash point (by ASTM D93), calorific value (by ASTM D240), auto-ignition temperature (using ASTM E 659), Octane-MON (using ASTM D2700-MON), and Octane-RON (by ASTM D2700-RON) were measured and are explained in the Results and Discussion section. The modified fuel characteristics of the blends are very close to that of gasoline, hence this alternate arrangement is.

The quantitative equations to estimate different fuel properties is given in Equations (4)–(11). The density of the gasoline and petroleum blend was numerically computed as per Equation (4).

$${\rho }_{blend}=\frac{{\rho }_g{\nu }_g+{\rho }_m{\nu }_m}{{\nu }_g+{\nu }_m}$$

(4)

There are three different methods to numerically estimate the kinetic viscosity of a gasoline and methanol blend: the Gambill method, the Refuta equation, and the Chevron formula, as given in Equation (5), Equation (6), and Equation (7), respectively.

$${\kappa }_{blend}=x_g{\kappa }_g{}^{\frac{1}{3}}+x_m{\kappa }_m{}^{\frac{1}{3}}$$

(5)

Refuta-mass fraction procedure:

$$VB{N}_i=14.534\times \ln \left(\ln \left({\kappa }_i+0.8\right)\right)+10.975$$

(6)

$$VB{N}_{blend}={\displaystyle {\sum }_{i=0}^{n=0}{x}_i}VB{N}_i$$

(7)

$${\kappa }_{blend}=\exp \left(\exp \left(\frac{VB{N}_{blend}-10.975}{14.534}\right)\right)-0.8$$

(8)

Chevron formula (volumetric basis):

$$VB{N}_i=\frac{\ln \left({\kappa }_i\right)}{\ln \left(1000\times {\kappa }_i\right)}$$

(9)

$$VB{N}_{blend}={\displaystyle {\sum }_{i=0}^{n=0}{v}_i}VB{N}_i$$

(10)

The octane number of a methanol–gasoline blend is estimated on the basis of the formula given in Equation (11):

$$OC{T}_{blend}=OC{T}_{gasoline}\times \left(v_{gasoline}\right)+OC{T}_{methanol}\times \left(v_{methanol}\right)$$

(11)

Now, there is the requirement of understanding the fuel characteristics of the newly formulated blended fuel, with a particular emphasis on the comparison to pure gasoline. Figure 2a–d show different piston designs under consideration.

3. Energy Conversion and Emission Measurement

The engine test rig used here was equipped with a single-cylinder four-stroke petrol engine mounted in a universal test bed and coupled with a water-cooled eddy current dynamometer [13]. The engine used here is air-cooled with a cylinder displacement of 105.6 cc, delivering maximum power of 5.59 kW at 7500 rpm and a maximum torque of 7.85 Nm at 6000 rpm at full load condition. Hence, the maximum operational testing torque was fixed to be 5 Nm. The testing targeted 500 rpm, 1000 rpm, and 1500 rpm, while the torque was set to 2 Nm, 3 Nm, 5 Nm, respectively [11]. The equivalent value of the brake mean effective pressure (BMEP) is 2.38 bar, 3.57 bar, and 5.95 bar, respectively. The details of the specification of the engine and dynamometer are given in the Supplementary Materials. The eddy current dynamometer equipped with a APPSYS WED 38S type magnetic water strainer was engaged in this study. It can monitor a maximum engine torque of 90 Nm and a speed of 7000 rpm. Other attachments include a torque calibration arm, water flow switch, magnetic pickup sensor, and reaction type torque sensor. The control panel contains a data acquisition system, computer hardware, and peripheral component interconnect (PCI) data card. The control panel monitors pressure (N/m²), temperature (°C), mechanical power (kW/HP), speed (rpm), and torque (Nm). The data generated from a running engine are saved in Excel form along with various graphical plots. The dynamometer has a control system for either manual or automatic mode. The testing arrangement also includes a six-gas emission analyzer of HORIBA make MEXA-584L, which can simultaneously sense CO, CO₂, NO_x, lambda, HC, and O₂. It uses the non-dispersive infrared (NDIR) technique. The measuring steps in this machine include a leak detection test and an HC hang up test to be passed prior to taking the reading of various combinations of torque (Nm) and speed (Nm) [11]. The data acquisition system attached to the dynamometer ensures the speed and torque at which emission from the engine is recorded.

Table 1. List of instruments with measuring range, accuracy, and uncertainty %.

Measured Quantity	Measurement Range	Accuracy	Type of Instrument	% Uncertainty
Load	±90 Nm	±0.1 Nm	Strain gauge type load cell	±0.1 Nm
Speed	0–7000 rpm	±1 rpm	Magnetic pick-up type speed sensor	±0.1 rpm
Time	-	±0.1 s	-	±0.2 s
CO	0–10% by vol	±0.001%	Non-dispersive infrared gas sensor	±1
CO2	0–20% by vol	±0.001%	Non-dispersive infrared gas sensor	±1
NOx	0–5000 ppm	±1 ppm	Electro chemical gas sensor	±0.6
EGT	0–900°C	±0.3 °C	K-type thermo-couple	±0.1

shows the list of instruments with a measuring range for experiments.

Through emission investigation, the effect of fuel carbon redundancy on various emission constituents at a constant rpm, coating thicknesses, and variable BMEP were investigated. Furthermore, for 1500 rpm, 5.95 bar (BMEP), and variable coating thicknesses, the various emission constituents were measured. Additionally, keeping the rpm to 1500, the [email protected] bar, and the [email protected] mm, emission behavior for the design exchange of the piston was considered to depict the fuel carbon redundant effect. The change in the value of fuel characteristic parameters did not deviate too much from that of the current gasoline. So, replacing B₅, B₁₀, and B₁₅ with gasoline is acceptable. Next, it was required to test the newly formed fuel for energy efficiency and emission formation. Engine testing, emission, and performance measurement procedures are discussed in the Method section. The effect of the piston design exchange and coating thickness is included in the emission, the elastic behavior of the coated and uncoated pistons were achieved through finite element analysis (FEA), and the temperature is validated by Buyukkyaet al. [29] at crown center and bowl rim locations for different coating thicknesses and piston materials (steel, AlSi).

4. Emission Formation Mechanism

CO formation: CO is formed in burned gases due to the incomplete oxidation of fuel carbon and the subsequently insufficient oxygen available for the complete oxidation of fuel. If the A/F ratio decreases below the stochiometric A/F ratio, then the CO emission rises.

CO₂ formation: There are two steps involved inthe emission formation mechanism of CO₂. In the first step, HC is converted to CO. Intermediate molecules, such as HC, aldehyde, and ketones, are formed through many sequential oxidation processes, which is given in Equation (12):

$$\mathrm{RH}\to \mathrm{R}+{\mathrm{O}}_2\to \mathrm{RCHO}\to \mathrm{RCO}\to \mathrm{CO}$$

(12)

In the second step, CO is converted into CO₂ due to the presence of enough oxygen in the air. CO₂ formation is only possible due to complete combustion (sufficient time and sufficient oxygen to eliminate unburnt HC). The amount of CO₂ emission formed has a direct influence on the engine performance. In the recently developed engine, reduction in CO₂ was observed due to an improved combustion process.

$$2\mathrm{CO}+2\mathrm{OH}\to 2{\mathrm{CO}}_2+{\mathrm{H}}_2$$

(13)

NO_x formation: NO_x is formed due to the oxidation of molecular Nitrogen. With elevated flames during combustion, Nitrogen and Oxygen molecules in the induced air split into atomic species and combine to form NO. Along with NO, NO₂ is also partially formed. NO and NO₂ as a whole is known as NO_x.

$$Fuel Nitrogen\to Nitrogen Intermediate\to \mathrm{NO} or {\mathrm{N}}_2$$

(14)

HC formation: HC is formed due to the exhaust of unburnt fuel. HC escapes from a combustion chamber due to quenched flame in the crevices and the reduced temperature of the chamber walls. It is also formed due to the cylinder-oil layer fuel absorption on the wall of the cylinder and that of the combustion chamber during cold starting. HC reduces with an increase in A/F ratio. The lean mixture is due to the poor quality of combustion, leading to engine misfiring and causing an erratic engine operation and a sharp increase in HC emission.

5. Finite Element Analysis of Coated Piston Design Exchange

FEA uses a technique to convert the entire piston to the number of discrete elements through meshing. It uses the auto-mesh provision of the ANSYS workbench to perform this function. The tetrahedral element type was chosen for the analysis, which has better error convergence and accuracy of the result output. In this analysis, conductive heat transfer through the coated piston surface is the dominant mode of thermal energy transfer. This is based on the assumption that there is no heat transfer effect on the engine piston system dynamics. There is no cavitation in ring-skirt and liner contact. Additionally, the effect of the ring twist is absent.

The mode of heat transfer in the oil is neglected. The thermal boundaries of the piston include the ring land, skirt, underside, and combustion side. The aligned boundary conditions state that the piston crown has direct exposure to hot gases or flame at a temperature of 400 °C. The maximum gas pressure in the combustion chamber is 12 MPa in an engine cycle. To avoid thermal expansion, wear, and rapid heat transfer, the piston material is considered of AlSi and cast steel materials; the details of material properties are listed in

Table 2. Material properties of rigid elastic constituents.

Materials	Thermal Conductivity (W/m °C)	Density (kg/m3)	Specific Heat (J/kg °C)
Piston (AlSi)	155	2700	960
Piston (Steel)	79	7870	500
Bond coat (100%NiCrAl)	16.1	7870	764
100%MgZrO3	0.8	5600	650
70%MgZrO3+30%NiCrAl	4.6	6130.4	676.6
50%MgZrO3+50%NiCrAl	7.3	6543.7	697.39
30%MgZrO3+70%NiCrAl	10.4	7016.7	721.14
Rings (cast iron)	16	7200	460

. Such materials are stronger against compressive stress and possess suitablethermal conductivity. Furthermore, bond-coated AlSi or steel make pistons enhance thermal distortion and wear resistance. Figure 2 represents the CAD model of type_A, B, C, and D pistons. As given in the figure, type_A has no bowl provision in the crown and crown type_B, C, and D pistons have different bowl shapes. In this study, the thickness of the NiCrAl bond coat was considered to be 100µm, which was kept constant, and the thickness of the FGM was changed from 400µm to 1200µm at400µm increments, or the thickness of each inter-layer (70%MgZrO₃+30%NiCrAl, 50%MgZrO₃+ 50%NiCrAl, 30%MgZrO₃+70%NiCrAl) had been changed from 100µm to 400µm at100µm increment.

The finite element model developed here is validated with a similar type of work by Buyukkya [26]. The piston profile is thus modified to validate geometry. The temperature distribution thus obtained is compared for AlSi and of uncoated thin-coated steel (0.4 mm) pistons. The result shows an adequate agreement with the fringe pattern of the temperature. The highest temperature value in the current case is slightly higher as compared to work reported by Buyukkya [26] for AlSi pistons. The solid model, mesh information, and output parameters of the analysis are listed in Supplementary Materials.

Table 2 shows the validation of crown center and bowl rim temperature from the FEA model by Buyukkya et al. [29]. In the majority of cases of comparison, the difference is less than 5%, except for both crown and bowl rim locations of the AlSi (UC).

Table 3. Details of FEA model validation.

Sl. No.	Material	Temperature (°C) at Crown Centre [14]	Temperature (°C) at Crown Centre (Current)	Difference in Temperature	Temperature (°C) at Bowl Rim [14]	Temperature (°C) at Bowl Rim (Current)	Difference in Temperature
1	Steel (UC)	333	357	7.20%	357	372	4.20%
2	Steel (CT-0.4 mm)	388	379	−4.12%	418	422	0.96%
3	AlSi (UC)	285	325	14%	270	315	16.67%
4	AlSi (CT-0.4 mm)	342	358	4.6%	366	380	3.82%

shows the validation details of FEA model considering different piston materials.

6. Response Surface Methodology

The approach of response surface methodology (RSM) is generally used for the optimization of the process parameter and to find out the optimal conditions to improve the responses. Response surface methodology combines the test design parameters and output parameter optimization. RSM includes the input parameter design that gives a range of input variables in which the test is being conducted. By using a mathematical model, the relationship between the input and output variables are investigated and an optimal value of the input parameter is provided to build an ideal response [36]. Let us assume that all controlled variables (a₁, a₂, …, a_k) and all the experiments are continuous, controllable, and measurable. The human error is ignored, so the linear output response r may be given as

$$r=f\left(a_1,a_2,\dots \dots ,a_k\right)+\delta$$

(15)

where δ represents the different sources of the variable, which are not considered in $f$, or is also considered as the statistical error; the mean value of this statistical error is 0. The polynomials that are used in RSM is a second order, given as

$$r={\beta }_0+{\displaystyle \sum }_{i=1}^k{\beta }_i{a}_i+{\displaystyle \sum }_{i=1}^k{\beta }_{ii}{a}_i^2+{\displaystyle \sum }_{i<j}{\displaystyle \sum }{\beta }_{ij}{a}_i{r}_j+\epsilon$$

(16)

where the parameter ${\beta }_i$, ${\beta }_j$ = 0, 1, 2 … k is the regression coefficient [37].

7. Results and Discussion

In our current investigation, we aimed to replace part of heavy gasoline with light methanol and to observe the carbon-redundant effects on fuel characteristics. Due to such exchange, fuel properties, such as calorific value, density, acid value, kinetic viscosity, flash point, auto-ignition temperature, and Octane number (RON and MON) must not deviate much from pure gasoline. Figure 3a–h show details of such fuel characteristic reformation due to carbon redundancy. For a 4.167% reduction in fuel carbon, replacement of B₅ (95% by volume of gasoline and 5% by volume of methanol) with gasoline yielded a 1.33% increase in calorific value, 0.077% increase in density, 6.6% increase in acid value, 1.15% increase in kinetic viscosity, 6% increase in flash point, 1.38% increase in auto-ignition temperature, and a 2.17% increase in Octane^RON and 0.1% increase in Octane^MON.

Similarly, to achieve an 8.34% reduction in fuel carbon, B₁₀ (90% by vol. gasoline and 10% by vol. of methanol) is replaced with gasoline. This led to a 5.2% increase in calorific value, 0.15% increase in density, 7.8% increase in acid value, 2.6% increase in kinetic viscosity,1.2% increase in flash point, 2.76% increase in auto-ignition temperature, and a 4.13% increase in Octane^RON and 0.2% increase in Octane^MON. Finally, to achieve a 12.5% reduction in fuel carbon, B₁₅ (85% by vol. gasoline and 15% by vol. of methanol) is replaced with gasoline. This led to a 0.154% increase in density, 3.79% increase in kinetic viscosity, 9.3% increase in acid value, 18% increase in flash point, 7.8% increase in calorific value, 4.1% increase in auto-ignition temperature, and a 6.2% increase in Octane^RON and 0.3% increase in Octane^MON observed. Changes in fuel characteristics due to fuel carbon redundancy is insignificant. In addition, corrosion effect and attack on the metal surface is negligible. Hence, these three different blends are expected to perform similarly to gasoline. Emission analysis can predict whether these are energy efficient or not. The numerical procedure as well as the experimental procedure (ASTM-based) are presented in the Methods subsection.

Figure 4a–d show the fuel carbon redundancy effect on emission formation at 1500 rpm for pistons with [email protected] mm and variable BMEP. For [email protected] bar, a 4% reduction in fuel carbon (B₅) reduced the CO emission by 147.23 gm/kWh. Again, an 8%reduction in fuel carbon(B₁₀), reduced the CO emission by 197.14 gm/kWh. Again, a 12% reduction in fuel carbon at [email protected] bar reduced the CO emission by 200.02 gm/kWh. For pure gasoline, as BMEP increased from 2.38 bar to 3.57 bar to 5.95 bar, the CO emission reduced from 232.34 gm/kWh to 109.16 gm/kWh to 162.67 gm/kWh, respectively. However, for all blends (B₅, B₁₀, and B₁₅), the CO emission increased with an increase in BMEP. The highest CO emission reduction was achieved at [email protected] bar for a 12% reduction in fuel carbon. The CO decreased with the increase in BMEP for gasoline, but increased for all the blended fuels. Figure 3b shows the variation of CO₂ due to fuel carbon reduction (−4%C, −8%C, and −12%C) through blend replacement by gasoline. In all cases (gasoline, B₅, B₁₀, and B₁₅), CO₂ emission decreased with an increase in BMEP (from 2.38 barto3.57 barto5.95 bar). It was observed that a 4% reduction in fuel carbon reduced CO₂ by 15.8 gm/kWh, 6.46 gm/kWh, and 17.96 gm/kWh for [email protected] bar,3.57 bar, and 5.95 bar, respectively. Furthermore, an 8% reduction in fuel carbon at these engine BMEP levels yielded CO₂ emissions at 0.8 gm/kWh, 7.18 gm/kWh, and 0.5 gm/kWh, respectively. Figure 3c shows the variation of NO_x due to fuel carbon reduction (−4%C, −8%C, and −12%C). The maximum reduction of NO_x occurred for B5(−4%C) (0.4114 gm/kWh) at [email protected] bar. Figure 3d shows the effect of lowering fuel carbon on the HC emission of the engine. With BMEP increased, the HC emission increased for gasoline and B₅. However, for B₁₀ and B₁₅, HC decreased. The lowest HC emission occurred at B₁₀(−8%C) for [email protected] bar and was found to be 0.02002 gm/kWh.

Figure 5a–d show CO, CO₂, NO_x, and HC emission variations due to fuel carbon removal through blending at 1500 rpm, [email protected] bar, and variable coating thicknesses. CO emission decreased with an increase in coating thickness. Fuel carbon removal of −4%C through blending (B₅) at the same condition lowered the CO emission (59.95 gm/kWh, 20.1 g/kWh, 3.95 gm/kWh, and 11.13 gm/kWh) for CT@uncoated at 0.4 mm, 0.8 mm, and 1.2 mm, respectively. Again, an 8% fuel carbon removal lowered the CO emission by 176.77 gm/kWh, 177.29 g/kWh, 112.76 gm/kWh, and 105.13 gm/kWh, respectively. Similarly, a 12% fuel carbon removal (B₁₅) in the same condition lowered the CO emission by 215.45 gm/kWh, 133.22 g/kWh,129.64 gm/kWh, and 102.34 gm/kWh, respectively. In all cases except −12%C, the CO emission decreased with an increasing coating thickness, while in such an excepted case, the maximum emission occurred at CT@4 mm (64.2 gm/kWh). Figure 4b shows the carbon dioxide emission for different coating thicknesses. The CO₂ emission increased with an increase in coating thickness. Moreover, at a particular coating thickness, enhanced carbon removal elevates the CO₂ emission. In the uncoated piston condition, a 12% reduction in fuel carbon enhanced the CO₂ emission by 62.49 gm/kWh. Further to this analysis, for [email protected] mm,0.8 mm, and 1.2 mm, a maximum fuel carbon removal led to 34.48 g/kWh, 4.84 g/kWh, and 5.96 gm/kWh, respectively. The CO₂ emission was highest at −12 °C with [email protected] mm and a value of 219.87 g/kWh. The details of the data related to all emissions are given in Supplementary Materials.

Figure 5c shows the NO_x emission for different coating thicknesses. NO_x follows the same trend as CO₂. The maximum NO_x level is 0.756 gm/kWh at B15 (−12%C) with a 1.2 mm coating thickness. The lowest NO_x emission is 0.305 gm/kWh at (−8%C) for the uncoated condition. Figure 4d shows the HC emission for the same condition as described in Figure 5a–c. With an increase in carbon removal (−4%C,−8%C, and −12%C) for all coating thicknesses, HC decreases. Furthermore, in increasing, the HC emission decreases. A highest HC emission is 0.476 gm/kWh for pure gasoline in the uncoated condition, while the lowest is0.026 gm/kWh at (−12%C) with a 1.2 mm thickness.

Figure 6a–d show the fuel carbon redundancy effect on emission formation at 1500 rpm, 5.95 bar BMEP, [email protected] mm, and variable piston design exchange. Figure 6a shows that piston_D exhibiteda maximum CO emission (302 gm/kWh), while piston_C exhibited a minimum CO emission (103.42 gm/kWh). Similarly, at −4%C, a maximum (242 gm/kWh, piston_C) and minimum (99.83 gm/kWh at piston_D)CO emission occurred. Again, for −8%C, a maximum (295.89 gm/kWh, piston_D) and minimum (48.47 gm/kWh, type_B)CO emission occurred. Finally, a 12% fuel carbon removal led to a maximum (280.45 gm/kWh, piston_D) and minimum (33 gm/kWh, piston_A)CO emission. For piston_A, CO emission decreased with carbon removal (Δ_4%@3.959 gm/kWh, Δ_8%@112.77 gm/kWh, and Δ_12%@129.67 gm/kWh). For the piston_B design (Δ_4%@98.13 gm/kWh, Δ_8%@149.49 gm/kWh, and Δ_12%@155.1 gm/kWh),a reduction in CO emission was observed. For the type_C design (Δ_4%@138.6 gm/kWh, Δ_8%@134.3 gm/kWh, and Δ_12%@155.85 gm/kWh),an increase in CO emission was observed. For the type_D piston design (Δ_4%@73.92 gm/kWh, Δ_8%@6.46 gm/kWh, and Δ_12%@21.91 gm/kWh),a reduction in CO emission wasobserved.

Figure 6b shows the effect of fuel carbon reduction on CO₂ emission considered for different piston design exchange. Maximum CO₂ emission occurred at −4%C for type_D piston and was found to be 316.7 gm/kWh, and the lowest was 177.4 gm/kWh for piston_B. A higher carbon removal reduced the CO₂ emission in the type_C piston design while, in case of type_D, it increased. There is a mixed trend in the case of piston_A and piston_B. For type_A piston, the CO₂ (Δ_4%@17.93 gm/kWh, Δ_8%@0.6 gm/kWh, and Δ_12%@-4.34 gm/kWh) decreased with the corresponding fuel carbon reduction. Similar was the case for the CO₂ emission of piston_B(Δ_4%@25.13 gm/kWh, Δ_8%@5.75 gm/kWh, and Δ_12%@23.79 gm/kWh). For piston design exchange type_C, CO₂ emission (Δ_4%@21.61 gm/kWh, Δ_8%@25.61 gm/kWh, and Δ_12%@32.31 gm/kWh) reduced. Finally, for piston_D (Δ_4%@-19.38 gm/kWh, Δ_8%@-8.61 gm/kWh, and Δ_12%@-17.95 gm/kWh) reduction was achieved.

Figure 6c shows the NO_x emission variation due to piston design exchange at 1500 rpm, [email protected] bar, and [email protected] mm. The maximum NO_x emission occurred at piston_D using gasoline (1.97 gm/kWh). The lowest emission occurred in the case of piston design_A (0.35 gm/kWh), In the type_A piston, NO_x formation changes (Δ_4%@-0.285 gm/kWh, Δ_8%@0.165 gm/kWh, and Δ_12%@-0.186 gm/kWh) were observed. For design_B, NO_xvaries (Δ_4%@-0.372 gm/kWh, Δ_8%@-0.2 gm/kWh, and Δ_12%@-0.79 gm/kWh). Again, for piston_C, Δ_4%@0.15 gm/kWh, Δ_8%@-0.34 gm/kWh, and Δ_12%@-0.48 gm/kWh was observed. Finally, for piston_D, Δ_4%@0.16 gm/kWh, Δ_8%@0.21 gm/kWh, and Δ_12%@0.3 gm/kWh was observed. Figure 6d shows the HC variation due to piston design exchange. The minimum HC emission is 0.03 gm/kWh due to a 12% reduction in fuel carbon and design_A implementation. The maximum HC emission observed in this study is 0.29 gm/kWh in the case of design exchange_C due to a 12% fuel carbon reduction. In case of piston_A (Δ_4%@0.05 gm/kWh, Δ_8%@0.106 gm/kWh, and Δ_12%@0.15 gm/kWh), a reduction in HC emission was observed compared to gasoline. Similarly, for design exchange B, C, and D, Δ_4%@0.018 gm/kWh, Δ_8%@0.51 gm/kWh, and Δ_12%@0.082 gm/kWh; Δ_4%@0.002 gm/kWh, Δ_8%@0.006 gm/kWh, and Δ_12%@-0.062 gm/kWh; and Δ_4%@0.034 gm/kWh, Δ_8%@-0.06 gm/kWh, and Δ_12%@0.014 gm/kWh was observed, respectively.

The energy and efficiency of an engine is greatly influenced by the exhaust performance parameters, such as temperature, back pressure, exhaust gas density, muffler noise, etc. In the context of piston design exchange and fuel carbon reduction through blending, Figure 7a–d show the effect on such parameters. Figure 7a shows the temperature variation. The maximum temperature variation is 360 °C when using gasoline. Replacing gasoline with B₅(−4%C) reduced the exhaust temperature by 5.5 °C. Similarly, replacing gasoline with B₁₀(−8%C) and B₁₅(−12% C) reduced the exhaust temperature by 3 °C and 3.5 °C, respectively. An increase in the exhaust temperature beyond some limit is not energy-efficient, or engine-friendly. Figure 7b shows the effect of a low-carbon fuel option on back pressure. Piston C and D designs developed more back pressure compared to the A and B designs. Replacing gasoline with B₅, B₁₀, andB₁₅, the back pressure increased by Δ_4%@0.0095 MPa, Δ_8%@0.02 MPa, and Δ_12%@0.0344 MPa for design A. Similarly, the back pressure increment was observed to be Δ_4%@0.0095 MPa, Δ_8%@0.0233 MPa, and Δ_12%@0.032 MPa; Δ_4%@0.088 MPa, Δ_8%@0.02 MPa, and Δ_12%@0.0109 MPa; and Δ_4%@0.0199 MPa, Δ_8%@0.002 MPa, and Δ_12%@0.029 MPa for B, C, and D, respectively. Figure 7c shows the density variation of exhaust gas. The gas density indirectly affects the energy efficiency as it interlinks combustion and operation. The maximum gas density of 1.228 kg/m³ was observed for piston_D at a 12% fuel carbon reduction. The lowest density variation was1.221 kg/m³ for piston_A when using pure gasoline. Figure 7d shows the bar chart of the noise level at exhaust measured in dB. In all other cases except type_C, the maximum noise level is at −4%C fuel carbon removal. Piston_Dhas a higher noise level compared to other designs. In almost all cases, an increase in the % of carbon removal reduces the noise level. The combined effect of design exchange and fuel carbon removal can reduce noise level up to 2.32% for piston_A,3.18% for piston_B, 2.46% for piston_C, and 1.1% for piston_D.

The operational efficiency of the engine system depends on the thermal conductivity of the piston material. For better energy efficiency, the temperature retention of a metallic piston plays an important role. The life expectancy, frictional behavior, and decarbonization potential are largely affected by this tendency. The selected suitable material and coating can be altered. The gas pressure is one of the load considerations for piston design exchange. For a high strength-to-weight ratio and better heat rejection, two different piston materials (AlSi and Steel) are considered here along with NiCrAl as a bond coat: 70% MgZrO₃ + 30% NiCrAl as a main coating. Detailed properties of the piston, bond coat, and coating material are given in an above table, which includes thermal conductivity, density, and specific heat. Finite element analysis is carried out, the details of which is given in the methods section. Figure 8a–h show the FEA simulated temperature for steel and AlSi pistons for four different designs (A, B, C, D). For the steel piston, the maximum temperature increases with an increase in coating thickness. The highest temperature of 391 °C developed for the uncoated type_A piston. For the coated piston, 422 °C@[email protected] mm, 582 °C@[email protected] mm, and 598 °C@[email protected] mm was observed. For the AlSi piston, the maximum temperature rise occurred at 340 °C for the uncoated type_B, at 392 °C with a [email protected] mm for type_B, at 516 °C fortype_C with a [email protected] mm, and 593 °C with a [email protected] mm for piston_A, for the same coating thickness and piston type. AlSi developed a lower temperature compared to steel because of more conductivity in the former case. A maximum temperature reduction of 13% is possible by replacing AlSi with steel.

Figure 9a–f show the strain energy and interlinked parameters (von-Mises stress and elastic deformation) in response to piston design exchange and variable coating thicknesses. As part of the total energy, strain energy plays a considerable role in energy efficiency. The lower the strain energy due to resilience, the higher the back power available to the wheel. Figure 9a,b show the resilience for pistons of different designs (A, B, C, D) with variable coating thicknesses. For the AlSi piston, a better strain energy for the type_B piston with [email protected] mm occurs (240 kJ/kg). With the increase in CT for all designs, strain energy increases up to [email protected] mm and then decreases. For both AlSi and steel, in all cases of coating, the order of strain energy is type_B > type_C > type_D >type_A. Figure 8c,d show the von-Mises response to the above stated conditions. The highest is 1332 MPa for the type_BAlSi piston with [email protected] mm. Figure 8e,f show the elastic deformation in the stated load and boundary conditions. For all the contemporary cases, the AlSi piston shows more deformation than the steel piston.

8. Evolution of Mathematical Model—Design of Experiment for Response Surface Methodology

Here, we demonstrate the effect of A₁(δ), B₁(CT), and C₁(E) and A₂(Blend), B₂(BMEP), and C₂(N) at six independent variables with three levels. The variables shown in

Table 4. Independent variables used for Box–Behnken design for optimization.

Variables	Symbol	Unit	Level
			−1	0	1
δ	A1	mm	2	3.5	5
CT	B1	mm	0	0.4	0.8
E	C1	GPa	200	215	230
Blend	A2	%	5	3.5	15
BMEP	B2	bar	2.98	3.57	5.95
N	C2	rpm	500	1000	1500

were independent variables selected for optimization.

Independent variables used for the Box–Behnken design to optimize the effect of A₁(δ), B₁(CT), and C₁(E) and A₂(Blend), B₂(BMEP), and C₂(N) at three levels is shown in Table 4 and

Table 5. Details of actual vs. predicted responses of output parameters (elastic).

Run	δ	CT	E	Delastic		σvon-Mises		Tsurface
				Actual	Predicted	Actual	Predicted	Actual	Predicted
1	5	0.8	215	0.3935	0.3891	306	158.38	379	404.63
2	3.5	0.4	215	0.3855	0.3855	317.7	317.7	341	341
3	2	0.4	230	0.3872	0.3812	872	755.75	375	361.75
4	5	0.4	200	0.211	0.217	247	363.25	379	392.25
5	5	0.4	230	0.4345	0.424	232	309.5	324	323.5
6	2	0.8	215	0.3615	0.3526	678	724.13	415	453.38
7	3.5	0.8	200	0.1964	0.1948	470	501.38	582	543.13
8	3.5	0	230	0.3874	0.389	234	202.62	315	353.88
9	3.5	0.4	215	0.3855	0.3855	317.7	317.7	341	341
10	5	0	215	0.3738	0.3827	79	32.88	333	294.63
11	2	0	215	0.3684	0.3728	95	242.62	340	314.38
12	3.5	0.4	215	0.3855	0.3855	317.7	317.7	341	341
13	3.5	0.4	215	0.3855	0.3855	317.7	317.7	341	341
14	3.5	0	200	0.2278	0.2129	133	62.87	379	404.13
15	2	0.4	200	0.2031	0.2136	770	692.5	422	422.5
16	3.5	0.4	215	0.3855	0.3855	317.7	317.7	341	341
17	3.5	0.8	230	0.3785	0.3934	301	371.13	489	463.88

. A total of 17 datapoints were taken separately to get the response for D_elastic, σ_von-Mises, and T_surface, and CO, CO₂, NO_x, and HC. The variables shown in Table 5 and

Table 6. Details of actual vs. predicted responses of output parameters (emission).

Run	Blend	BMEP	N	CO		CO2		NOx		HC
				Actual	Predicted	Actual	Predicted	Actual	Predicted	Actual	Predicted
1	10	2.38	500	59.8401	67.97	264.035	256.31	264.035	0.1196	0.584848	0.4668
2	5	3.57	500	311.061	272.37	247.533	260.61	247.533	0.2748	1.05007	0.4668
3	5	2.38	1000	149.87	193.32	257.688	247.97	257.688	0.3231	0.618078	0.4668
4	10	3.57	1000	53.91	53.91	302.117	302.12	302.117	0.2856	0.212672	0.4668
5	10	3.57	1000	53.91	53.91	302.117	302.12	302.117	0.2856	0.212672	0.4668
6	5	5.95	1000	210.788	204.62	265.305	282.94	265.305	0.4997	1.09659	0.4668
7	15	5.95	1000	231.274	200.25	272.286	284.64	272.286	0.3069	0.578202	0.4668
8	15	3.57	500	53.3709	51.97	299.578	320.57	299.578	0.0821	0.219318	0.4668
9	10	5.95	1500	55.5273	60.76	341.469	337.83	341.469	0.5105	0.16615	0.4668
10	10	2.38	1500	15.6339	−29.69	265.305	303.01	265.305	0.3339	1.47541	0.4668
11	15	3.57	1500	44.7453	83.43	338.93	325.85	338.93	0.2964	0.186088	0.4668
12	10	5.95	500	107.82	139.78	300.848	274.5	300.848	0.2961	0.252548	0.4668
13	5	3.57	1500	56.6055	58.01	380.82	359.82	380.82	0.4891	0.259194	0.4668
14	15	2.38	1000	55.5273	49.27	286.884	266.61	286.884	0.1304	0.385468	0.4668
15	10	3.57	1000	53.91	53.91	302.117	302.12	302.117	0.2856	0.212672	0.4668
16	10	3.57	1000	53.91	53.91	302.117	302.12	302.117	0.2856	0.212672	0.4668
17	10	3.57	1000	53.91	53.91	302.117	302.12	302.117	0.2856	0.212672	0.4668

were independent variables selected for optimization.

For the determination of the predicted value of all responses, the coded and actual regression equations are shown. For regression equation, design software v8.0 was used.

For D_elastic:

In terms of the coded factor:

Coded Equation = 0.3855 + 0.011575 A − 0.0034375 B + 0.0936625 C + 0.00665 AB + 0.00985 AC + 0.005625 BC + 0.0001125 A² − 0.0113125 B² − 0.0766625 C²

(17)

In terms of the actual factor:

Actual Equation = −16.3156 − 0.0911889 A − 0.192385 B + 0.150848 C + 0.0110833 AB + 0.000437778 AC + 0.0009375 BC + 0.00005 A² − 0.0707031 B² − 0.000340722 C²

(18)

For σ_von-Mises:

In terms of the coded factor:

Coded equation = 317.7 − 193.875 A + 151.75 B + 2.375 C − 89 AB − 29.25 AC − 67.5 BC + 108.775 A² − 136.975 B² + 103.775 C²

(19)

In terms of the actual factor:

Actual Equation = 20,206.1 − 128.828 * A + 4002.17 B − 189.117 C − 148.333 AB − 1.3 AC − 11.25 BC + 48.3444 A² − 856.094 B² + 0.461222 C²

(20)

For T_surface:

In terms of the coded factor:

Coded Equation = 341 − 17.125 A + 62.25 B − 32.375 C − 7.25 AB − 2 AC − 7.25 BC − 20.25 A² + 46 * B² + 54.25 C²

(21)

In terms of the actual factor:

Actual Equation = 11,676.1 + 75.5278 A + 227.708 B − 105.042 C − 12.0833 AB − 0.0888889 AC − 1.20833 * BC − 9 A² + 287.5 B² + 0.241111 C²

(22)

For CO:

In terms of the coded factor:

Coded equation = 65.9218 + −37.1057 * A + 40.5673 * B + −44.1707 * C + 34.9209 * AB + 61.4574 * AC + 4.66038 * BC + 82.3475 * A² + 13.5954 * B² + −19.8119 * C²

(23)

In terms of the actual factor:

Actual Equation = 888.522 + −114.179 * A + −57.1656 * B + −0.197424 * C + 3.91271 * AB + 0.024583 * AC + 0.00522171* BC + 3.2939 * A² + 4.26694 * B² + −7.92477 × 10⁻⁵ * C²

(24)

For CO₂:

Coded equation = 310.547 + 5.08595 * A + 13.2494 * B + 27.5092 * C + −4.23412 * AB + −23.4839 * AC + 4.15896 * BC + 3.88754 * A² + −36.1184 * B² + 18.4856 * C²

(25)

Actual Equation = −18.0973 + 15.4967 * A + 101.934 * B + −0.0183395 * C + −0.474411 * AB + −0.00939356 * AC + 0.00465989 * BC + −0.155502 * A² + −11.3358 * B² + 7.39426 × 10⁻⁵ * C²

(26)

For NO_x:

Coded Equation = 0.315039 + −0.096367 * A + 0.0882749 * B + 0.107167 * C

(27)

Actual Equation = 0.0874646 + −0.0192734 * A + 0.0494537 * B + 0.000214333 * C

(28)

For HC:

Coded equation = 0.466784

(29)

Actual Equation = 0.466784

(30)

The plot between the predicted and actual value of all the responses is shown in Figure 10a,b.The predicted value is reasonably near to the experimental value so the correlation between the predicted and experimental value is validated.

The effects of the input process variable on the different output parameter can be seen. It can be observed that E has substantial effects on all output parameters among all reaction parameters. Figure 11a,b show the Box–Cox plot of elastic response and emission response, respectively. It transforms the elastic and emission data to close to normal distribution. Figure 12a–b show the perturbation plot for the output variable, which is from a mathematical method, to find the approximate values of elastic and emission parameters. Here it is observed that there is a single point on the perturbation plot of HC emission due to zero influence on currently considered input parameters.

Figure 13, Figure 14 and Figure 15 show the 3D plot for comparative analysis of the maximum value of elastic deformation, von Mises, and surface temperature in response to CT-delta, E-delta, and E-CT, respectively. Similarly, Figure 16, Figure 17 and Figure 18 show the same 3D plot for emission constituents like CO, CO₂, NO_x, and HC, respectively.

8.1. Effect on Reaction Parameter

The three-dimensional curves were drawn to investigate the effect of the independent process parameter on the dependent process parameter. These graphs were drawn by taking two other parameters at the zero level.

8.2. Effect on D_elastic

It can be seen in the 3D counter plot in Figure 13a that when E is held constant, the variation of D_elastic is constant with respect to the other two parameters. Figure 13b shows that when CT is held constant, D_elastic decreases with respect to E and δ. Figure 13c shows that when δ is held constant, D_elastic decreases with respect to E and CT.

8.3. Effect on Surface Temperature

Figure 15 shows the effect of various input parameters on surface temperature. It can be observed in Figure 15a that when E is held constant, the surface temperature rapidly decreases and then gradually increases with respect to CT and δ. Figure 15b shows that when CT is held constant, the surface temperature gradually decreases and then rapidly increases with respect to the other two parameters E and δ. Figure 15c shows that when δ is held constant, the surface temperature increases with respect to E and CT.

8.4. Optimization of Parameter

The optimization of individual input parameters was performed to achieve the minimum of D_elastic, maximum of von Mises stress, and minimum surface temperature. The optimal value of all input parameters is given in

Table 7. Optimum process parameters.

Parameter	Aim	Lower Limit	Upper Limit
δ	Within range	2 mm	5 mm
CT	Within range	0 mm	0.8 mm
E	Within range	200 GPa	230 GPa
Delastic	Minimize	0.1964 mm	-
Σvon-Mises	Maximize	-	872 GPa
Tsurface	Minimize	315 °C	-

. The predicted value of all input variables, providing positive agreement with the simulation result, is shown in

Table 8. Optimum model validation under optimized condition.

Sl.	Output Parameter	δ	CT	E	Predicted Value	Simulation Value	% Error
1	Delastic	2.16	0.15	216.64	0.1964 mm	0.1922 mm	0.0042
2	σvon-Mises	3.79	0.18	214.64	872 GPa	871.65 GPa	0.35
3	Tsurface	4.27	0.37	218.16	315 °C	314.1 °C	0.9

.

9. Conclusions

The energy-intensive engine system upgradation and emission mitigation technique discussed here informs us that reducing the fuel carbon % through a gasoline methanol blend does not enhance carbonated emissions (CO, CO₂, and HC). However, NO_x emissions enhanced due to an increase in the exhaust muffler temperature. Such an incremental change in heat transfer is due to the improved combustion due to the lightness of fuel. Such aggravation can be controlled through exhaust after treatment. Carbon removal up to 12% does not much alter the energy conversion process. The system energy output, fuel characteristics, and emission in the engine system is as compatible as that with the gasoline in the energy system. In the scenario of a global energy mode swing, such an energy alternative gives hope to reforming energy policy through the emergence of methanol economy. The piston design exchange and PVD coating indirectly improve energy and efficiency by retaining a reasonable amount of heat energy. Additionally, NO_x and other emission constituents were found to be under control due to piston improvement through coating. Our main conclusions observed are as follows:

• Piston type_D has a maximum CO emission of 300 gm/kWh at −8% decarbonization, while type_A and type_B have the lowest CO emission level of 50 gm/kWh at −12% decarbonization using the B₁₅ blend.
• At −12% decarbonization, the type_B design has thelowestCO₂ emission of 175 gm/kWh among all designs considered.
• For both piston designs, type_A and type_B, the NO_x emission is lowest at −8% decarbonization through blending.
• Overall, it is observed that the type_A and type_B designs are more efficient compared to other designs in mitigation emission through blend replacement and design exchange.

The immediate implication of this research can be applied to newly produced automotive engines as it would not have any significant infrastructural modification. Only existing piston design modifications to type_A and type_B type would be necessary. Interfacing gasoline and methanol helps the decarbonization of power production and improves air quality. This energy innovation alternative ensures the sustainable clean energy transition. It benefits greater energy access through the inclusion of methanol as potential global energy alternatives for the ongoing energy transition to low-carbon energy resources. In addition, methanol production lies within the gasoline supply chain and its manufacturing can be aligned with petrochemical complexes that ensure almost zero additional networking. Future studies in this research may address fuel injection pressure, vaporization, and surface tension of the blend while in use.