"Another contributing factor is that the equation for the thrust of an engine includes a term equal to nozzle-exit area times the amount by which exit pressure exceeds ambient."
I'm assuming you're talking about underexpansion? I can see how it would affect efficiency.
In part, that's what I'm talking about. In the simple one-dimensional-flow model of a rocket engine, thrust is equal to
F =
q ve +
Ae(
pe -
pa) ,
where
q is the mass flow rate (usually represented by m-dot, of course),
ve is the exit velocity,
Ae is the exit area,
pe is the exit pressure, and
pa is the ambient pressure. Right?
Now, let's do the thought experiment of starting up a rocket engine that's underexpanded at some non-zero ambient pressure. Underexpansion means
pe >
pa. So, the second term in the thrust equation is positive. The exhaust gas at the exit is pushing on the atmosphere, and that helps generate thrust.
Now let's lengthen the nozzle to the point that the exit pressure drops to ambient. The second term vanishes. However, the exit velocity increases. In fact, the increase in velocity more than offsets the the loss in the second term. Obviously, we need to consider equations other than just the thrust equation to determine this, but that's how it works out.
If we further extend the nozzle, the exit pressure drops below ambient, and the second term goes negative. The nozzle is now sucking on the atmosphere. Meanwhile, exhaust velocity has increased, but not enough to overcome the growth in the second term. The moral of the story is that the engine is most efficient when exit pressure matches ambient.
"One minor detail: even in the convergent part of the nozzle, enthalpy is being converted into kinetic energy; pressure and temperature both decrease as the throat is approached."
I was thinking about this last night, but I'm not sure I understand how pressure ratio affects efficiency in such cases (or if it even does). I can see how a higher PR would allow for higher exhaust velocities, which I believe I've heard but don't mentally grasp, that either it's necessary in order for an aircraft attain high velocities, or it's more efficient for such use cases.
I'm not sure I understand by what you mean by "in such cases." After all, every de Laval nozzle begins with a converging section.