Given that you have asked such a broad question, one can hardly "prove" anything. However, let's just think through the question, shall we?
We are getting some kind of resonant effect at frequency f. Let's suppose it's a dielectric effect (given what you're described, that seems most appropriate). This means something in molecules A and B is polarizing in response to the oncoming wave and happens to have a natural response time that constructively interferes with the stimulating wave. All good so far?
So how does this natural response time arise? Did Zeus come down from Olympus with a lightning bolt and deem this molecule to have a response frequency of 10Hz? No, there is a physical reason for it. Let's suppose the mechanism is "polar bond reorientation." It responds with a given frequency because the bond has an associated charge q and has to move an associated mass m. Bust out your Newton's second law, and your Lorentz force equation, you an idea how much polarization you expect form the wave. Figured out what the mobility of charge is, and Bob's your auntie.
Now, imagine you double the frequency of the wave: Rather than toss your polarization perfectly from one extreme to another, you are catching halfway, slamming the brakes, and reversing course. Is there a resonance effect here? No. In fact, what you see is that in fairly short order (less than a decade) you see the resonance die out and the effect that you were observing utterly disappear. This is because you are applying the energy to the material faster than it can respond. There's a nice schematic curve of the concept here:
http://en.wikipedia.org/wiki/Dielectric_spectroscopy
So, you are observing some dielectric effect, and then you want to shift the frequency FIVE decades away. Imagine you have a friend by the arm, and you start jerking him back and forth at 1Hz, slinging him annoyingly about the place. Then, you start jerking him 20,000 times faster with the exact same force. At best, your are going to give him a hell of an Indian burn - you're certainly not going to turn him into chunky salsa. Granted, the analogy is sloppy but the effect holds.
Think of it a different way: Let's suppose you have green light. If you go to 20,000f of green light, you are putting yourself in the region between x-rays and gamma rays. Or, if you go from green DOWN to 1/20000f, you are into radio waves. Do you really expect to see some commonality of effects here? 20000f is a HUGE shift in the electromagnetic spectrum, and the practicalities change enormously.
If it's a resonant cavity effect, where you are just reflecting frequency back and forth, I can just tell you from an engineering perspective, that doesn't work either. At the 20,000th mode, you have too many other effects in the system causing loss and noise - resonance doesn't matter then. Practically speaking, if you get more than about 8-10 modes out of a cavity, you're doing pretty well.
Now, let me ask you: What saline solution are you proposing to do this on?
