Impact of Adding Bioethanol and Dimethyl Carbonate on Gasoline Properties

Abstract

Bioethanol and dimetyl carbonate (DMC) are considered alternative fuels and additives to the synthesis compounds used now, since bioethanol is a biofuel and dimethyl carbonate (DMC) is non-toxic, biodegradable and can be produced in a cleaner way. In this study, the effect of adding dimethyl carbonate (DMC) and ethanol to gasoline on the volatility was investigated. The volatility was the main goal of this research but also, the effect on the antiknock properties was studied. Mixtures of gasoline with DMC or with bioethanol were prepared in different proportions of additive: 3%, 6% and 9% v/v. Additionally, mixtures with 3% v/v ethanol plus 3% or 6% v/v DMC, and3% DMC plus 6% v/v ethanol were prepared. For the volatility evaluation, the ASTM distillation curve and vapor pressure of these mixtures were determined experimentally in order to predict the performance of the resulting fuels. When adding oxygenated compounds, the increase in vapor pressure was proportional to the additive quantity. Additionally, modifications of the ASTM distillation curves were observed, with these indicating the formation of minimum boiling point azeotropes and the corresponding increase in volatility, with good effect on the ease of ignition in the engine. Based on the experimental results, the vapor lock index VLI, drivability index DI and vapor–liquid ratio temperature T_(V/L=20) were calculated to quantify the volatility. The experimental results showed that gasoline mixtures with these oxygenated compounds show a significant increase in antiknock properties. Thus, for mixtures with ethanol, the research octane number (RON) increases by up to 2.2 units and the motor octane number (MON) increases by up to 1.2 units. Gasoline mixtures with DMC have another behavior: RON increased by up to 1.5 units, while the MON value increased by up to 2.5 units. For an initial gasoline with RON = 94.7 and MON 84.7, these increases are important and make the difference by exceeding the RON = 95 limit. Adding dimethyl carbonate to gasoline–ethanol blends improves the sensitivity of the fuel.

1. Introduction

Gasoline is one of the main products in the oil processing industry. Different antiknock additives are used to increase its octane rate; in the past, this was achieved by using “ethylated” additives such as Pb(Et)₄. The ban of tetraethyl lead as an additive for gasoline came after evidence that lead is toxic even in low concentrations [1]. Nowadays, almost all countries have banned the use of lead-additive fuels. As a consequence, this required countries to find other solutions to increase the ignition performance of gasoline. So, lead-free additives containing oxygen are used; among them are alcohols and ethers.

Kivi J. et al. [2] studied the use of methyl tert-butyl ether (MTBE) and ethyl tert-butyl ether (ETBE) as components of gasoline, in terms of the operability and performance. The use of oxygenates reduced exhaust emissions at low ambient temperature; the cold starting was faster and cold handling improved when oxygenated fuels were used. However, water contamination with MTBE is a concern because MTBE is not easily detectable, and could have an impact on animals and human health. So, it has been classified as a human carcinogen, and its replacement became necessary [3,4].

Alcohol-based oxygenated fuels may be considered a solution for renewability. The use of renewable energy sources, such as biomass [5], is compulsory to limit the impact on global warming. Bioethanol, as an ecological variant, is frequently used in the gasoline formulation, but also as an additive in diesel fuels [6,7,8]. Its performance in mixtures with gasoline can be improved by adding some surfactants. Ahmed A. A. et al. [9] studied and discussed the impact of a nonionic surfactant in small concentrations on the distillation curve of the gasoline–ethanol mixture containing 90 vol% gasoline and 10 vol% anhydrous ethanol (named E10), due to the formation of azeotropes.

Dimethyl carbonate (DMC) is a carbonate ester. It is a flammable, colorless, transparent and dense liquid with a smell similar to methanol. DMC has been recommended by many researchers as an alternative fuel or oxygenated compound additive for diesel/gasoline fuel blends [10,11].

DMC has some advantages recommending it for use, as follows:

• High oxygen content, 53.28% by weight, with an important role in the complete combustion of the fuel, thus reducing greenhouse gas emissions, particularly carbon and hydrocarbon emissions [12];
• A very good miscibility with fossil fuels (diesel and gasoline) [10];
• It can be produced in a cleaner way, and is non-toxic, non-corrosive and environmentally friendly [13].

The authors of the study [14] investigated the preparation and application of the environmentally friendly, non-toxic substance, dimethyl carbonate, as an antiknock agent for gasoline. They concluded that mixtures of DMC with gasoline showed a significant increase in the octane number compared to other conventional but toxic additives. Not of the same opinion are Schifter and co-workers [15] who stated in 2017 that DMC is less effective than alcohols as a booster, and so its applicability in the gasoline market should be reduced. Additionally, they considered that DMC is less thermally efficient in blends with gasoline since they found, in their experiment, that the exhaust gases’ temperature is 8 °C higher than in the absence of DMC. From the present study, it was found that DMC increases the octane number of the gasoline even if not to the same extent as bioethanol, thus agreeing with Schifter et al. [15]. However, DMC can still be considered as an antiknocking agent.

A review of DMC synthesis procedures and the use of pure DMC and its blends with diesel/gasoline in internal combustion (IC) engines was performed in [16]. The influence on emissions and combustion performance is discussed. Pure DMC can be used directly in diesel engines with some minor modifications. The burning time becomes shorter due to the lower boiling point of DMC. Smoke emissions are reduced. Additionally, they present the effects of using DMC as an additive to diesel/gasoline fuels compared to the effects of using pure DMC.

Previous studies were conducted with gasoline of different qualities, and it is obvious that the results depend largely on the gasoline’s composition. This is why, in the present work, we studied a different gasoline, madeup with catalytic reforming and fluid catalytic cracking gasoline, the two main components makingup the majority of the pool gasoline in a refinery. By adding to the gasoline dimethyl carbonate or bioethanol (so-called binary mixtures), or bioethanol and DMC (ternary mixtures), the goal of our work was to explore the effects of these additives on the fuel properties.

2. Materials and Methods

2.1. Materials and Blends

Gasoline was madeup using two components from different processes in Khazmunay Gas Rompetrol Petromidia refinery: 60% v/v catalytic reforming gasoline and 40% v/v fluid catalytic cracking gasoline.

The composition of the gasoline was determined via gas chromatography, with the method described in Section 2.2, and the results are presented in

Table 1. The composition ofgasoline, expressed in % v/v and %wt.

Component	% v/v	%wt
n + iP4	0.35	0.27
n + iP5	6.85	5.53
n + iP6	6.06	5.12
n + iP7	11.35	10.01
n + iP8	7.03	6.37
n + iP9	3.17	2.94
n + iP10	1.02	0.96
n + iP11	1.32	1.26
Total Paraffins	37.15	32.46
OC4	0.27	0.22
n+ Cy O C5	2.96	2.48
n + Cy O C6	1.92	1.67
n + Cy O C7	1.12	1.01
n + Cy O C8	0.63	0.58
n + Cy O C9	0.18	0.17
n + Cy O 10	0.07	0.06
Total Olefins	7.15	6.19
N C5	0.09	0.090
N C6	0.96	0.965
N C7	1.64	1.692
N C8	1.13	1.130
N C9 + C10	0.74	0.743
Total Naphtens	4.56	4.62
A C6	0.73	0.82
A C7	10.8	12.13
A C8	16.72	18.55
A C9	12.63	14.04
A C10 + C11	10.08	11.18
Total Aromatics	50.96	56.72
Oxygenates	0.01	0.04
TOTAL	99.83	100.03

.

As determined by gas chromatography, the fuel contained (by weight) 32.46%wt normal and iso-paraffins (n + iP), 6.19%wt normal and cycloolefins (n +Cy O), 4.62%wt naphthens (N), 56.72%wt aromatics (A) and a residue of 0.04%wt oxygenated compounds, probably a contamination when stored.

The bioethanol used in the preparation of the samples was practically anhydrous (because it contained only 0.01% water, which was confirmed via coulometric titration using Karl Fisher apparatus from Methrom). The dimethyl carbonate with 99.7% purity was purchased from Merck, Germany. The gasoline and oxygenated compounds’ properties, ethanol and dimethyl carbonate are presented in

Table 2. Properties of gasoline, ethanol and dimethyl carbonate.

Properties	Gasoline	Ethanol	Dimethyl Carbonate
Molecular weight (g/mol)	103.07	46.07	90.08
Oxygen (%)	0	34.73	53.28
Density at 15 °C (g/cm3)	0.7892	0.7933	1.0754
Viscosity at 40 °C (mm2/s)	0.64	1.17	0.56
Flash point (°C)	-	17	17
Vapor pressure (mmHg) la 100 °F	285.09	115	143.75
RON	94.7	107 *	101–116 **
MON	84.7	89 *

* According to Ref. [17]. ** According to Refs. [18,19,20].

.

A total of 9 blends were prepared. All of the blends were homogeneous because the components were completely miscible. There were 3“binary” blends of gasoline + ethanol G+E (3% v/v, 6% v/v, 9% v/v ethanol), 3 blends of gasoline with DMC (3% v/v, 6% v/v, 9%v/v DMC) and 3“ternary” blends of G+E+DMC, with 3%E + 3% DMC, 3% E + 6% DMC and 6% E + 3%DMC, respectively. These percentages were established considering the usual oxygenated compound concentrations in fuels, thus providing a better comparison among products. The blending strategy was to share the final amount of oxygen in gasoline between ethanol and dimethyl carbonate, so that the oxygen content in the fuel could be kept below the accepted limit (3.7%wt oxygen). This limit, indicated in the fuel specification [21], is also accepted by the European Union.

The blends were prepared at 4 °C to minimize the vaporization of samples duringhandling, according to the laboratory procedure for vapor pressure determination. All samples were prepared accurately via the volumetric +/− 1%v/v. The blends were stored in well-stoppered containers to avoid volatile compound losses. In the present paper, tables and graphs are reported in % volume.

2.2. Analysis Methods

The analysis methods for testing the composition, the volatility and the antiknock properties are those applied to petroleum products. In order to obtain reliable results, the best equipment and most appropriate analysis methods were used. In a complementary way, the calculation of indexes and other values was performed to express the volatility and the octane rating.

2.2.1. Gas Chromatography

The gas chromatographic analysis used here is called PONA analysis, the abbreviation of names of hydrocarbon classes. It was carried out on the gasoline using the AC Analytical Control’s Reformulyzer@M4 from Agilent which has had its capabilities expanded in revised ASTM D6839-21 and ISO 22854 and can determine hydrocarbon per component, or per hydrocarbon class, and oxygenated compounds with high accuracy. The apparatus is in fact an aggregate containing multiple chromatographic columns: a capillary column for the separation of narrow fractions (C4, C5, C6, etc.) followed by four packed columns for the separation ofpure components of paraffins, olefins, naphthens and aromatics, respectively, each working in their own temperature program. The detector is FID. The carrier gas was nitrogen, and the sample volume was 1 μL.

2.2.2. Distillation Temperature Measurement Method

As a hint of volatility, the distillation curve was measured according to the ASTM-D86 test method for petroleum products and liquid fuels at atmospheric pressure. Distillation points of gasoline and of blends were measured using OPTIDIST^TM automatic apparatus. Samples of 100 mL of gasoline, binary blends and ternary blends were tested. Heating was performed at a rate of 5 mL /min.

To scrutinize the composition evolution during the distillation, the temperature was recorded in the following order: at first drop, then at 5, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90 and 95% volume distilled. Uncertainty was $\pm$0.015 mL.

The standard ASTM D86 (EN ISO 3405) sets landmarks on distillation characteristics as percent evaporated at 70 °C, 100 °C and 150 °C, denoted by E70, E100, E150 and final boiling point (FBP).For a gasoline proper for use in spark-ignition engines, the values are 22–50% v/v for E70, 46–71% v/v for E100 and min. 75% v/v for E150.

The fractional distillation results can also be expressed as the distillation temperature (°C) at 10, 50, 90 and 97% v/v share of sample fractions.

2.2.3. Vapor Pressure Measurement Method

Another way to express volatility is through the vapor pressure of the blends which, in our work, was measured according to the standard test method for the liquid fuels ASTM D5191 [22], using the vapor pressure analyzer HVP970 AMETEK MINIVAP VPX pert.

For gasoline and its blends, vapor pressure determinations were made at 37.8 °C (100 °F) and in the 4:1 vapor/liquid ratio. The MINIVAP VPX pert performs fully automatically. A sample of 1 mL was inserted with a glass syringe. Previously, the rinsing of the measuring cell was performed to avoid contamination from the previous sample. Uncertainty was $\pm$0.6 kPa.

2.2.4. Research Octane Number (RON) and Motor Octane Number (MON) Measurement Method

The most usual indications of the antiknock properties are RON and MON; the antiknock index (AKI) and sensitivity (S) were therefore calculated.

Octane number measurements were performed using ASTM tests for the research octane number (ASTM D2699) and for the motor octane number (ASTM D2700) of spark-ignition engine fuel. The octane meter was WAUKESCHA CFR and the standard materials were iso-octane (RON = 100) and n-heptane (RON = 0). Uncertainty was $\pm$0.06 ON for RON and $\pm$0.04 ON for MON.

3. Results and Discussion

3.1. Inquiry into Volatility

3.1.1. ASTM Distillation

The distillation curve is a diagram drawn for a mixture of compounds correlating boiling temperature with the distilled fraction expressed in % volume. The peak fraction (0% v/v to 20% v/v) provides information on cold start, engine warm-up, evaporative emissions and vapor lock. The average fraction (20% v/v to 90% v/v) can be used to interpret the fuel’s capability for warm-up, acceleration and cold weather performance. Results for the final distillate fraction (from 90% v/v to endpoint) are used to estimate the fuel’s tendency to scale [23,24,25].

ASTM distillation was carried out for gasoline and all studied blends with bioethanol and DMC.

The distillation parameters for samples were observed at initial boiling point, 5%, 10%,15%,20%, 30%,40%,50%,60%, 70%,80%, 90% and 95% volume of evaporation and final boiling point. The range of boiling points for gasoline falls in the range of 30–210 °C (EN ISO 3405). It should be mentioned that the boiling points of pure ethanol and dimethyl carbonate are 78.4 °C and 90 °C, respectively.

The distillation curves are represented in Figure 1a–c. The distillation curves of gasoline with different volumes of ethanol (0%, 3%, 6% and 9%) are shown in Figure 1a; the curves for gasoline added with dimethyl carbonate (0% 3%, 6% and 9% v/v) are shown in Figure 1b. Ternary blends of gasoline + ethanol + dimethyl carbonate are shown in Figure 1c. Each figure includes four distillation curves; the upper is for pure gasoline and the others are for blends. In Figure 1c, the distillation characteristics for all samples comply with the limits in the standard ASTM D86 (EN ISO 3405), so gasoline and its blends are prone to being used in the spark-ignition engine.

The volatility of these blends differ from each other based on the additive content. As can be seen from Figure 1, the distillation profile of the gasoline is smooth and steadily increases in the range of 43 °C to 210 °C.It suggests that the distillation curve of gasoline influenced by its hydrocarbon composition has a specific boiling point for each of its components [9]. GC analysis of the gasoline revealed high concentrations of heavy aromatics with higher boiling points that alkanes with a similar number of carbon atoms in molecules.

The trend of the distillation temperatures is similar for all the blends with ethanol content, as shown in Figure 1a. As the ethanol was added to the gasoline, a reduction in boiling temperatures in the range of 5–50% v/v on the distillation curve occurred, thus indicating the increase in the volatility and the formation of positive azeotropes [26,27,28,29]. The fact that when more ethanol is added there is more of a decrease in boiling points is caused in the region of 5–50% v/v distillate, and not only due to the azeotrope formation but also because ethanol (b.p. 78.4 °C) exceeding the azeotrope composition will distill at its own boiling point; the biggest depression in boiling point of −25 °C was observed for the mixture of G 91% +E 9%, at 30% distillate. In the region of 10–30% v/v, the slope of the distillation curve decreased progressively with the ethanol concentration from 1.33 °C/% for the original gasoline to 0.6 °C/% for the mixture G 91% + E 9%. Once the ethanol evaporates, the boiling temperature rises relatively sharply and approaches the original gasoline distillation curve. The final boiling point of the gasoline blends never exceeds the final boiling point of the pure gasoline: 210 °C.

The azeotrope formation is a complex phenomenon implying ethanol and light hydrocarbons (e.g., pentane isomers, hexane isomers, heptane isomers, cyclohexane, benzene and toluene). The azeotropes can be binary, ternary or multicomponent. A few examples of binary azeotropes data, as they follow from vapor–liquid equilibria calculated using the CHEMCAD8 simulation program via the UNIFAC method, are presented in

Table 3. Binary azeotropes of ethanol at atmospheric pressure (boiling point b.p. = 78.4 °C).

Second Component	b.p.of Second Component, °C	b.p. of the Azeotropic Mixture, °C	Second Component Concentration, %wt.
n-Pentane	36.2	34.3	95
n-Hexane	68.9	58.7	79
n-Heptane	98.5	70.9	51
Cyclohexane	80.7	64.9	69.5
Benzene	80.2	68.2	67.6
Toluene	110.8	76.7	82

.

From Figure 1b, it is noted that distillation curves for dimethyl carbonate blends are very similar to those of the base gasoline, in the range of 5–20% v/v. This is because dimethyl carbonate does not form azeotropes with the gasoline’s components, so up to 90 °C, only gasoline’s hydrocarbons distill. In the range of 30–60% v/v, it can be observed that there is a reduction in the boiling points compared to the gasoline used as a raw material; this reduction is bigger when the % v/v of added DMC increases: −2 °C depression for the mixture G 97% + DMC 3%, −4–5 °C for G 94% + DMC 6% and −6 °C for G 91% + DMC 9% at 60% v/v distilled. The maximum boiling point depression (−10 °C) was recorded at 40% v/v distillate for the mixture G 91% + DMC 9%. This decrease in boiling points intervenes when the hydrocarbon boiling points exceed the boiling point of DMC (90 °C).

In Figure 1c, the distillation curves of the ternary mixtures containing ethanol and dimethyl carbonate are shown. As expected, the presence of ethanol mostly influences the distillation curve. The depression in the boiling point in the range 5–50% v/v, caused by the formation of azeotropic mixtures of ethanol with light hydrocarbon in gasoline, increases as the ethanol content increases. The curves for the ternary mixtures resemble those for ethanol mixtures, but the boiling point depression is smaller for ternary mixtures (−16 °C at 30% v/v distilled for the mixture G 91% + E 6%+ DMC 3%) compared with the binary mixture G 91% +E 9% at 30% v/v (−25 °C) due to the combined effect of ethanol (the strong depressant) and DMC (the mild depressant).

To conclude, the ethanol shows the greatest boiling point depression while dimethyl carbonate shows the smallest depression in the distillation curve when compared to ethanol, meaning that ethanol has the strongest effect of increasing the volatility of the gasoline.

3.1.2. Reid Vapor Pressure

Reid vapor pressure (RVP) addresses the problems with fuels for spark-ignition engines. Fuels with too-low vapor pressure may cause cold start problems and poor warm-up performance in vehicles while a fuel with too-high vapor pressure may contribute to hot drivability/hot restart problems [30].

Figure 2 illustrates the effect of adding oxygenate compounds in gasoline on the vapor pressure of the blend. The measured RVP of all blends is higher than that of pure gasoline (38 kPa). The gasoline–ethanol blends have the highest RVP of all values around 44 kPa, within the uncertainty interval of$\left(\pm 0.6 \mathrm{kPa}\right),$ regardless of the concentration being in the 3–9% v/v range. Christensen et al. [31] suggested that the addition of low concentrations of ethanol to gasoline substantially increases vapor pressure, which was confirmed in the present study.

The tendency for the DMC blends is to steadily increase the RVP when the concentration of DMC increases, but the values of increase are small and lower (up to 1.2 kPa) compared to ethanol blends (6 kPa). The values of RVP for the additive gasoline with DMC are also within the uncertainty interval. When gasoline is very volatile, maintaining the RVP of gasoline to a certain point is generally desired [30]. Abdalla et al. [16] concluded that dimethyl carbonate has excellent characteristics as a gasoline oxygenated additive, such as high octane number and moderate Reid vapor pressure (RVP). Therefore, the use of the dimethyl carbonate can be advantageous in some environmental conditions.

The ternary blends have a similar behavior to the ethanol blends, and the RVP values are around 41.3 kPa. It is an ascertainment that ethanol is the determinant for the volatility of binary and ternary blends with gasoline and has the strongest effect on increasing the RVP [32], due to the formation of positive azeotropes with C5–C8 hydrocarbons (alkanes, olefins and aromatics) having boiling points in the range from 30 °C to 120 °C [29,33]. The RVPs of ethanol are far less than that of gasoline, but in blends with it, RVP values are increased significantly [32]. The formation of the azeotropic mixture is a function of pressure, so that with increasing pressure in the system, volatility lowers and vice versa [33].

It is worth noting that the RVP of blends depends on gasoline composition and non-linearity may occur at blending for the RVP of the mixture, so the behavior is probably different when the base gasoline differs from our case.

3.1.3. Calculation of Volatility Indexes

Volatility indexes combine the experimental data of the ASTM distillation with those of the RVP, giving more information about the behavior of gasoline and its blends in the spark-ignition engines from the volatility point of view such as the tendency to form vapor lock, the easiness of cold start and warm-up and the smoothness and steadiness at acceleration.

Calculation of Vapor Lock Index (VLI)

Using the experimental data for vapor pressure and distillation, the vapor lock index (VLI) was calculated. It is one of the volatility parameters, expressing the tendency of the gasoline to form a vapor lock, and it is calculated by Formula (1), mentioned in the standard EN 228 [34].

$$VLI=10\ast VP+7\ast E70$$

(1)

where VP is vapor pressure and E70 is the volume % evaporated at 70 °C (read on the distillation curve).

The calculated VLI is found in

Table 4. The effect of adding ethanol and DMC to gasoline on the vapor lock index (VLI), drivability index (DI) and vapor–liquid ratio.

Gasoline and Blends,% v/v	VLI	DI	T(V/L = 20), °C
100% G	454.9	632.5	74.1
97% G + 3% E	564.5	619.0	70.3
94% G + 6% E	591.3	613.0	70.1
91% G + 9% E	635.0	603.5	69.5
97% G + 3% DMC	467.2	618.0	73.1
94% G + 6% DMC	477.7	613.5	72.5
91% G + 9% DMC	480.5	604.5	71.9
94% G + 3% E + 3% DMC	540.0	612.5	70.2
91% G + 3% E + 6% DMC	524.0	611.5	71.3
91% G + 6% E + 3% DMC	583.0	606.0	70.5

.

The index for the base gasoline was 454.9 and increased when adding oxygenate compounds up to 635. A more important increase was provided by ethanol (e.g., VLI is 591.3 when adding 6% v/v ethanol, compared with 477.7 for adding 6% v/v DMC) and ternary mixtures have intermediary values, observing, for example, that the blend containing 6% v/v ethanol and 3% v/v DMC has a VLI of 583.

The VLI typical values are in the range of 800 to 1250 according to ASTM D4814-08 Standard Specification for Automotive Spark-Ignition Engine Fuel. They vary with seasons and lower values provide greater protection against vapor lock and against hot fuel handling problems [9,30].

Calculation of Drivability Index (DI)

According to the definition, the drivability index measures the smoothness and steadiness of acceleration for an automotive vehicle. The DI gives information about the cold start and engine’s warm-up.

Using the experimental data, the drivability index (DI) was calculated entirely from ASTM distillation data, using Formula (2), from [35].

$$DI=1.5\left(T10\right)+3.0\left(T50\right)+1.0T\left(90\right)$$

(2)

where DI is drivability index, T10, T50 and T90 are temperatures (°C) at 10, 50 and 90% v/v distilled (read on the distillation curves).

The DI typical value in the US, according to ASTM-D4814, is in the range of 375–610 (°C-derived). The data obtained in the present study are generally over 610, but this is due to the base gasoline which had a high DI (=632.5).

The effect of adding oxygenate compounds is to decrease the DI proportionally with the volume added. The decrease is comparable for both additives in binary blends and values in ternary blends.

Calculation of Temperature for a Vapor–Liquid Ratio of 20 (T_{V/L =20})

The vapor lock protection potential T_V/L=20 is considered to be the best index to assess hot fuel handling problems.

The temperature for the vapor–liquid ratio of 20 (T_V/L=20) (ASTM-D4814, 2008) is the temperature at which the fuel forms a volumetric vapor–liquid ratio of 20, at atmospheric pressure.

The temperature for theV/L ratio of 20 is calculated via Relationship (3):

$$T_{(\mathrm{V}/\mathrm{L}=20)}=\left[52.47-0.33\ast \left(RVP\right)\right]+0.2\ast \left(T10\right)+0.17\ast \left(T50\right)$$

(3)

where RVP is in kPa andT10 and T50 are temperatures (°C) at 10 and 50% volume distilled (read on the distillation curve).

The temperature for V/L = 20 varies with climate and season; for the US, the normal range is 35–60 °C.

The data presented in Table 4 demonstrate that G+E and G+DMC blends at the studied concentrations slightly decrease the T_(V/L=20) index. Ethanol provokes more of a decrease in the index value than DMC added in the same proportion, and ternary blends have intermediate values of the index.

Ethanol and DMC can serve to control the volatility in the case of gasoline with low RVP by modifying the volatility indexes (VLI, DI and T_(V/L=20)), but results are not spectacular. The volatility of the base gasoline is a more important factor. In the present case, the base gasoline has an initial RVP of 38 kPa, which is under the low limit for the temperate climate in summer season (45 kPa) [21]. Adding oxygenate compounds may be insufficient for bringing the indexes in the limits, so adding a lighter hydrocarbon fraction would also be necessary.

3.2. Inquiry into the Antiknock Properties

An appropriate fuel for SI engines must be resistant to auto-ignition to avoid knocking. The indexes used for fuel ignition quality are referred to as octane rating and octane number.

In the present study, the research octane number (RON) and motor octane number (MON) of the gasoline and its blends were determined experimentally. The RON offers information about the fuel’s behavior in mild operation conditions, closer to driving in a town (low temperature in the engine and speed) and MON offers information in harsher conditions (high temperature and speed), closer to highway driving.

The octane rating is better expressed by the antiknock index (AKI), which is the arithmetic average of RON and MON. This index is recommended by ASTM D4814 to be 87, adjusted with season, and it is used at petrol station dispensers, being labeled with (R+M)/2. The AKI gives more precise information of the fuel’s quality used both in milder and harsher conditions.

The sensitivity (S) of gasoline is the difference between the RON and MON. This should be as small as possible to have a fuel which is robust to changes in driving. It is recommendable at S < 10 octane units.

The experimental values of RON and MON and the calculated AKI and S are shown in

Table 5. Antiknock properties of gasoline and its blends.

Gasoline and Blends, % v/v	RON	MON	AKI	S
100%G	94.7	84.7	89.7	10.0
97%G + 3%E	95.2	85.0	90.1	10.2
94% G + 6%E	96.2	85.5	90.8	10.7
91%G + 9%E	96.9	85.9	91.4	11.0
97%G + 3%DMC	94.9	85.0	89.9	9.9
94%G + 6%DMC	95.4	85.5	90.4	9.9
91%G + 9%DMC	96.2	87.0	91.6	9.2
94%G + 3%E + 3%DMC	96.0	85.2	90.6	10.8
91%G + 3%E + 6%DMC	96.4	86.3	91.3	10.1
91%G + 6%E + 3%DMC	96.7	85.7	91.2	11.0

, for gasoline and its blends.

It is well known that adding oxygenated components to gasoline will increase the octane number of the blends. Therefore, blending them with gasoline can decrease the risk of knock in SI engines [15,36]. In a previous study, Schifter et al. [10] showed that DMC and ethanol have a stronger tendency to increase the octane number in the gasoline blend compared to the ethyl tert-butyl ether (ETBE) blend.

The gasoline with ethanol shows a higher increase in RON then adding the same quantity of DMC, e.g., when the mixture contains 9% v/v ethanol, RON increases by 2.2 units, since gasoline with dimethyl carbonate shows an increase of only 1.5 units in the RON. In general, at the same concentration of the added compound, dimethyl carbonate provides a higher increase in the MON, as can be seen in Table 5. Therefore, the difference between the RON and MON is smaller when adding DMC (S = 9.2–9.9), meaning that sensitivity improves compared with adding the same quantity of ethanol (S=10.2–11). This is the reason for some authors considering dimethyl carbonate as a superior octane appreciator [14].

For the ternary blends, an increase in the RON and MON is consistent; furthermore, AKI values are very good (all over 90), but blends are more sensitive than those with DMC only.

Generally, the octane number increases when a fuel contains molecules with double bonds and aromatic rings [17], as is the case for our base gasoline. At a higher octane number, it is possible to use a high compression ratio to increase thermal efficiency, but the fuel with an octane number that is too high can also affect the engine [37,38].

4. Conclusions

Ethanol E and dimethyl carbonate DMC are two oxygenated components that can be used in blends with gasoline.

The objective of this study was to examine the impact of E and DMC addition on volatility and other properties to reach an acceptable solution. For this purpose, nine samples of binary and ternary blends were prepared. E and DMC were added in concentrations from 3 to 9% v/v. A higher share of oxygenated compounds added to gasoline led to higher oxygen content in the blend, thus improving combustion and reducing the amount of carbon monoxide and unburned fuel in the exhaust gas, thus reducing smog [38]. Oxygenated compounds were selected based on the oxygen share in the molecule; therefore, mixtures with small quantities of E and DMC were selected. Among the studied blends, only the binary 91% v/v G + 9% v/v DMC and the ternary 91% v/v G+3% v/v E + 6% v/v DMC had oxygen content over the 3.7%wt limit [21], so these blends should be avoided according to the specific data sheet. Too much oxygen in gasoline leads to increasing the temperature at exhaust, so NO_x emissions will also increase, favored by a higher temperature.

Adding these oxygenated compounds in gasoline blends leads to an increase in volatility, more in the case of ethanol, and less for DMC use. This has a good effect for gasoline with low volatility, as demonstrated by experimental data and calculated indexes. However, in regions with higher environmental temperatures, these blends can face volatility problems. Due to the relatively smaller increase in vapor pressure, the blends containing DMC may be more attractive blending components than ethanol in these regions. In ternary blends (G+E+DMC), it is possible to offset the volatility according to requirements.

The antiknock properties (RON, MON, AKI) improve by increasing E and DMC content in gasoline blends. Adding ethanol increases the RON more, and adding DMC increases the MON more and decreases sensitivity. The ternary mixtures can balance these properties, according to the requirements.

In conclusion, using one or another of these mix recipes depends on what exactly the manufacturer needs. The most important thing is the gasoline composition which varies from one pool to another; so, it is important that these tests should be conducted in every refinery laboratory to meet the exigencies of the market.