The overwhelming majority of current gasoline engine must run at or near stoichiometric, and are regulated to a very tight AFR band near stoichiometric via an O2 sensor, so a discussion about lean burn is not germane to the benefits of hydrogen injection devices retrofitted to current engines as these products do.
However, that being said, hydrogen injection in useful quantities can and will aid gasoline engine combustion near the lean limit because hydrogen has an exceptionally high laminar flame velocity, and as already stated, a very wide flammability range. This will shorten the end-of-combustion timing from TDC and therefore improve the indicated efficiency, because the cycle operates closer to a constant-volume cycle than would otherwise be the case of slow, lean combustion. It would also extend the misfire limit, but only IF the engine is running close to that limit.
An engine that is running stoichiometric or rich, I don't believe, will see any noticeable improvement.
For Diesel applications, the danger is that hydrogen has a low-autoignition temperature, and could pre-ignite before the piston reaches TDC. This greatly increases cylinder pressures and produces negative work. There is very little fuel economy improvement to be gained from increased combustion efficiency, one of the biggest claims of these hydrogen devices, because with Diesel combustion it is already in the order of 99%. However, some gain can be realised if, like in the gasoline engine illustrated above, the combustion duration could be shortened, such that the cycle operates as close as possible to a constant-volume process. How much could be gained? -- maybe 5%.
BUT, this cannot happen given the rate of hydrogen generated and burned in the engine, as has been calculated from some claims (one gallon of water consumed per 6000 miles). In order to have any effect whatsover, the hydrogen must account for at least about 1% of the total fuelling quantity. I have calculated
here that these devices that consume one gallon of water per 6000 miles, produce hydrogen at a rate of the order of parts per million (0.0001%). In other words, unless these devices produce and burn a whole lot more hydrogen, they ain't worth and don't make a hill of beans of difference.
