[freebeard]

...the air will separate from the plasma...

The [ionized] air is the plasma. It doesn't separate, it accelerates. Locally, under control.

Either a direct current (DC) or an alternating current (AC) power supply or a microwave microdischarge can be used for different configurations of plasma actuators.... The performance of plasma actuators is determined by dielectric materials and power inputs, later is limited by the qualities of MOSFET or IGBT.
[snip]
Manipulation of the encapsulated electrode and distributing the encapsulated electrode throughout the dielectric layer has been shown to alter the performance of the dielectric barrier discharge (DBD) plasma actuator. Locating the initial encapsulated electrode closer to the dielectric surface results in induced velocities higher than the baseline case for a given voltage. In addition, Actuators with a shallow initial electrode are able to more efficiently impart momentum and mechanical power into the flow.[4]

No matter how much funding has been invested and the number of various private claims of a high induced speed, the maximum, average speed induced by plasma actuators on an atmospheric pressure [conviction[?], without any assistant of mechanical amplifier (chamber, cavity etc.), is still less than 10 m/s.[5]

Influence of temperature
The surface temperature plays an important role in limiting the usefulness of a dielectric barrier discharge plasma actuator. The thrust produced by an actuator in quiescent air increases with a power law of the applied voltage. For voltages greater than a threshold, the exponent of the power-law reduces limiting the thrust increase, and the actuator is said to have “saturated,” limiting the actuator’s performance.

en.wikipedia.org/wiki/Plasma_actuator