Chemical Engineering Department, UCLA, Los Angeles, CA 90095

Because of the Federal and State regulatory requirements (1990 Clean Air Act Amendments, 1990 California Air Resources Board Regulations), [1,2] there is a growing need to understand the nature of engine exhaust emissions and their impact on health and environment. Combustion engine emissions have been shown to be major contributors to air pollution in urban areas. Vehicle emissions are divided into two groups; regulated and unregulated pollutants. Regulated pollutants are carbon monoxide (CO), nitrogen oxides (NO_x), and unburned fuel or partly oxidized hydrocarbons (HC). The levels of emissions of these pollutants are specified by law. Unregulated pollutants include carbon deposits and polycyclic aromatic hydrocarbons (PAHs). Carbon deposits increase engine wear and tear, while some of the PAH isomers are known to be carcinogenic and mutagenic [3].

In a study conducted by South Coast Air Quality Management District (SCAQMD), mobile source emissions were shown to contribute about 98 % of CO, 84 % of NO_x, and 62 % of volatile organic compounds in the urban atmosphere [4].

In addition to the possible carcinogenic role of engine exhaust emissions, acute health effects from exposure to exhaust emissions have been well established. The possible connection between cancer and exposure to diesel engine exhaust has been investigated in occupationally exposed people [5]. Based on animal studies, it has been postulated that the main culprits in cancer formation in humans are the PAHs, and their substituted derivatives (methyl-PAHs, nitro-PAHs, oxygenated nitro-PAHs, and oxy-PAHs) and the particulate matter from the exhaust on which the PAHs are adsorbed [5,6,7].

The objective of this study is to explore the effects of a water-based fuel additive developed recently on engine out emissions and on the amount of carbon deposited in a gasoline engine.

A sketch of the experimental set-up is given in Figure 1. Experiments were done using a single cylinder, spark ignited Mark III Transparent Combustion Engine (Megatech Corp., Billerica, MA) without an exhaust emission control catalyst. Engine specifications are given in Table 1. The dynamometer was used to start and load the engine. It was also possible to monitor engine speed, torque, cylinder pressure, manifold pressure, cooling air pressure, and power output of the engine by the dynamometer. The cylinder or combustion chamber in the engine is made of a special heat resistant glass. This enables one to monitor carbon deposition during engine operation. High accuracy rotometers were used to measure both the flow rates of fuel and air. An electronic fuel pump and surge-tank have been used to establish reliable fuel and air delivery, respectively. Engine system had two carburetor controls. A needle valve controlled the amount of fuel that flows through the lines, and a throttle valve that controlled the amount of air in the air:fuel mixture to establish the equivalence ratio (f , defined as the actual air:fuel ratio to the stoichiometric air:fuel ratio). A commercially available gasoline with 87 RON was used as the fuel. In order to avoid fuel composition changes, we used the same batch of gasoline for all the experiments.

The additive was acquired from ATG Inc. (Monrovia, CA), and it was homogeneously dispersed into the gasoline. At the 0.1 % (v/v) loading used in the experiments, gasoline-additive mixture was clear; that is no clouding was observed initially and over an extented period of time.

Engine-out emissions for NO_x, total hydrocarbons (THC), and CO were measured by an on-line digital gas analyzer (OTC RG240 Digital Gas Analyzer, Owatonna, MN) connected to the exhaust pipe by a sample line.

Carbon deposits in the engine were measured gravimetrically using a novel technique. In this approach, the engine was operated under sufficiently fuel-rich conditions that lead to measurable amounts of carbon deposits on the piston head in a 30 minute run. After each experiment, the engine was completely dismantled, and the carbon deposited on the piston head was carefully scraped off, and weighed using a sensitive analytical balance. The engine was then cleaned, re-assembled and tuned-up to the same operating conditions for the undertaking of the next experiment.

Engine out emissions were determined as a function of engine speed at 1500, 1750, 2000, 2250, and 2500 RPM and at different equivalence ratios, both for the base case and in the presence of the fuel additive.

Engine out emissions are presented in Figures 2, 3, and 4 for the baseline conditions. These results show that CO and NO_x emissions were strongly dependent on the equivalence ratio while THC was not. As seen in Figure 2, THC emissions showed a slight minimum around f =0.96-1.0 depending on the engine speed. In addition, an increase in engine RPM decreased THC emissions. Minimum THC emissions was obtained at 2500 RPM the highest speed studied. As seen in Figure 3, CO concentration uniformly increased with decreasing equivalence ratio as expected, at a constant RPM. The production of some CO is inevitable when fuel is burned with insufficient air. However, some CO will be emitted under a broad range of conditions, because of mixing and reaction rate limitations.

The importance of NO_x emissions from combustion sources lie in their contribution to the formation of secondary pollutants. As seen in Figure 4, NO_x emissions increased with increasing equivalence ratio within the range investigated as expected from flame temperature considerations. For engine speeds of 2000, 2250, and 2500 RPM, NO_x concentration showed maxima around equivalence ratio of 1.05. Since the combination of temperature and oxygen concentration determines the amount of NO_x formation, NO_x emissions peak occurred on the fuel lean side (f > 1.0) rather than the fuel rich side (f < 1.0). These results are similar to observations made in earlier investigations [10, 11]. It is also interesting to note that minimum NO_x emissions were observed at engine speed of 2250 RPM.

In the second set of experiments, i.e. in the presence of additive, the engine out emissions showed no systematic departure from the trends observed in the absence of the additive. In addition, the levels of NO_x, THC, and CO were well within the limits of accuracy of the measurements made in the absence of additives. Consequently, these results are not presented.

As noted before, carbon formation studies were performed at sufficiently fuel rich conditions under which measurable carbon deposition on the piston head occurred over a 30 minute operating period. In this study we report carbon deposition rates measured at an equivalence ratio of 0.72 (fuel rich). Lower equivalence ratios lead to excessive carbon formation and result in the early termination of the runs. At higher equivalence ratios, carbon formation rate was too slow, and longer operating times become necessary to accumulate measurable quantities of deposits. In Figure 5, the amount of carbon deposited on the piston head is plotted as a function of engine speed both for the base case and in the presence of the additive. The experiments corresponding to each data point were repeated three times to asses the repeatability of the results. Average standard deviations for the base case and in the presence of additive were 0.16 and 0.11 mg, respectively. As seen in Figure 5, carbon deposition rates decreased with increasing RPM, a result which is consistent with previous measurements [12]. Decrease in residence times in the engine at high RPM may be a possible explanation for this finding. However, it is particularly important to note that the presence of 0.1 % (v/v) additive in the fuel significantly decreased the rate of carbon deposition at a given RPM (Figure 5). In addition, this effect was observed consistently over the entire engine speed range investigated. As can be seen in Figure 5, carbon deposition rates decreased by about 32 %, 20 %, 44 % at 1750, 2000 and 2250 RPM respectively.

The presence of 0.1 % (v/v) additive has been shown to decrease carbon deposition rate on the piston head of a SI-gasoline engine by as much as 44 %. The additive, however, did not significantly alter the gaseous engine out emissions.

This research was funded, in part, by the UCLA Center for Clean Technology and ATG Inc., Monrovia,CA. One of the author, Fikret Inal, also would like to thank Izmir Institute of Technology, Turkey for fellowship.