M-E and the thrusts from M-E devices are 2 different things. The thrusts from the current UFG design (Woodward design) noted in this paper scale with the cube of frequency, and the frequency is limited only by the ionic limits of the material or "phase angle" that describes how quickly the ceramic can move. It is limited to the low Ghz region, so we're looking to start with an ability to scale up this thrust efficiency many orders magnitude. The current test article operates at 38kHz, so at least 5 orders magnitude frequency. The simplest reason for not building test items in the microwave region is that the electrical engineering for such power systems is really "black magic" indulged in by the best PhD EE's. Without a real staff with such people onboard, Ghz UFG's are not an option.
The thrust efficiency also scales with the cube of the k of the ceramic used. The PZT in the current test items is cheap Steiner-Martin stuff available on EBay. It has a k of about 1,000. By contrast, the PMN-15 available from TRS has a k value of ~ 20,000-25,000. Simple substitution would yield an increase of thrust of about 8,000X.
The trouble of course is, there is no such thing as "simple substitution" when you are doing a pure research program. This work reported wasn't R&D. Future work with PMN-PT will be closer to R&D, but it's important to recognize that for instance, the PZT used in the past is a piezo-active material that possesses both a 1w piezo bulk acceleration response, and a 2w electrostrictive response. The current design makes use of both of these. PMN-PT is an electrostrictor with no characterization available about it's 1w piezo response, so using it may require a new design.
When you change something like this, it's not fair to say "all things equal" as they're not. What we can and should recognize is though, that these unfunded kHz investigations with cheap materials, low frequencies and no paid engineers on staff, do not limit what M-E physics is capable of. All indications are that this technology is capable of huge thrust efficiencies that would make propellantless inertial drives ubiquitous, as well as warp drives and wormhole generators possible. All this stuff is on the table, so don't be concerned about the thrust magnitudes seen thus far.
If the thrust magnitudes reported by Paul March are any indicator, M-E thrusters are already outperforming the best Ion thrusters which require propellant. These very low thrust magnitudes jim is reporting are very close to competitive with Ion engines and it would not take much scaling to make them obsolete. Just as you say, the first market to consider is sat station keeping, and especially the Hall Thrusters used on GEO sats. The next generation of pure test articles could easily sweep them aside if they work at all.
If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks