You guys misunderstand O2 sensors and air to fuel ratio. In the old days with carburetors, you would have been correct. Those devices metered a certain volume of fuel per cubic foot or air, so the literal air to fuel ratio would have remained constant at all times. Thus, when you used fuel with ethanol in a carbureted car, since there was less energy per gallon of fuel, the carburetor was delivering less potential energy. In a stoichiometric sense, the fuel/air mix was getting leaner, meaning less combustion energy per cubic foot of delivered air. That is actually how the story got started that ethanol reduced emissions. In the carburetor days, ethanol really did reduce emissions. The reason was the leaner stoichiometric mixture. But then came fuel injection systems with O2 sensors.

The O2 sensor measures the excess air in the exhaust gas. When you add ethanol to the fuel, if you did nothing else (like in the old carburetor days), then since there is less energy in the fuel, there would be more excess oxygen in the exhaust gas. The O2 sensor senses this, and it commands the fuel injectors to inject a bit more fuel to bring the excess exhaust oxygen back to the initial value. Thus, a fuel injected car with oxygen sensor does not keep a fixed literal air to fuel ratio like a carburetor did, it keeps a constant stoichiometry between fuel and air. When the fuel contains ethanol, the literal fuel to air ratio is commanded to be higher so that the stoichiometric ratio remains fixed. Thus, mileage drops, but power remains the same. The engine puts out the same horsepower as before, because the fuel injection system has been commanded to send more fuel than before, and that balances things out. More fuel, but fuel of less energy per gallon, equals the same energy as before.

The above is why ethanol no longer reduces exhaust emissions. In the carburetor days, at fixed literal air to fuel ratio, ethanol caused leaner operation, which reduced emissions. Now, the feedback loop of exhaust gas O2 sensor back to fuel injectors does not keep a fixed literal air to fuel ratio, it keeps a fixed stoichiometric ratio, hence ethanol no longer makes a car run leaner, and thus no longer reduces emissions.

 

O2 sensors continually adjust the A/F mixture but carbs did adjust also, accelerator pumps, venturis, metering rods, etc. A/F mix changed with carbs, not as accurately or as fast as with injection but the A/F mixture wasn't "fixed" as you claim.

It's simple; Alcohol has less energy available than gas, it takes more than a liter of alcohol to produce the same amount of energy as a liter of gas. Same A/F mixture, same engine, the alcohol mix will produce less energy.

 

Sorry, I was guilty of the classic mistake of using technical terms like stoichiometry to explain a technical topic to a non-technical person. Let me try again. First, I’ll concede that air to fuel ratio delivered by a carburetor was not exactly constant, but it was very close to constant, influenced only very slightly by the variable density of the fuel through the metering jets. In contrast, the air to fuel ratio delivered by a modern fuel injection system varies widely in response to the fuel that is being used. So let me get on with what is hopefully a clearer description. If you don’t believe me, Google it on the web. There are plenty of similar descriptions out there.

To burn a pound of pure gasoline, about 14.7 to 15 pounds of air are required, depending on the exact composition of the gasoline. So let’s call it 15 pounds in round numbers, and thus the air to fuel ratio on pure gasoline needs to be about 15 to 1. To burn a pound of ethanol, you only need 9 pounds of air, so if you are running pure ethanol, the air to fuel ratio needs to be 9 to 1. A modern engine recognizes this difference by means of the exhaust gas O2 sensor.

If you are running pure gasoline, to get to zero excess air in the exhaust gas, the air to fuel ratio would have to be 15 to 1. If you then start introducing ethanol, but leave the air to fuel ratio at 15 to 1, then excess air will start showing up in the exhaust gas because the ethanol doesn’t need as much air to burn. The exhaust gas O2 sensor senses this, and sends a signal to the engine computer that it is getting too much air. The engine computer then commands the air to fuel ratio to go down until the O2 sensor no longer senses excess air in the exhaust gas. With 10% ethanol in the gas, air to fuel ratio must drop from about 15 to about 14.4. The way it accomplishes this is to command that the fuel injectors deliver a bit more fuel into the engine. That is why mileage drops. More fuel injected means more fuel used, which in turn means less mileage. But it is also why power does not drop. You are injecting more fuel, but each gallon of fuel has less energy, and those two factors cancel each other out, such that you are injecting the same amount of energy as you were before.

If you want to relate that description to the earlier one about stoichiometry, what stoichiometry means is the required ratio of reactants in a chemical reaction. So if you want 1 to 1 stoichiometry of air to fuel, then the actual air to fuel ratio must be about 15 to 1 if you are burning gasoline, or 9 to 1 if you are burning pure ethanol. So the combination of O2 sensor, fuel injectors, and engine computer in a modern engine maintain constant stoichiometry of 1 to 1, they do not maintain constant air to fuel ratio. The air to fuel ratio is varied to whatever it needs to be to achieve 1 to 1 stoichiometry. I should hasten to point out that I just told you a small fib. The computer actually maintains a stoichiometry of very slightly greater than 1 to 1, for a very small amount of excess air to remain in the exhaust gas. But the point is, it’s the stoichiometry that’s being held constant, not the air to fuel ratio.