Due to some limitations, I am interested to exert an electric field in a strange way in long ranges; something like using electromagnetic rays or varying magnetic fields to induce a current in a flask of cultured cells.
Sorry Amir, your question is not completely clear to me. I try and guess a first answer.
You want to apply an electric field to a "flask" (in vitro cells), but you mention also magnetic field. You don't specify the frequency: for low frequency they are completely decoupled, the electric field behaves with a large wave impedance (it is "capacitive"), while the magnetic field is low impedance(it is "inductive"). At high frequency they are related, not necessarily by a TEM propagation, but any field structure that is decided by the means of propagation and the way you generate and apply the field.
Let's simplify things and assume you want to expose cells to low frequency electric/magnetic field:
1) electric field: it is created by a large potential applied between two electrodes, or plates, like in a big capacitor; you have a preferential direction, so that you should move this system around your flask to change the direction, or create three orthogonal big capacitors.
2) magnetic field: it is similar but you have to use coils; a simple coil creates a magnetic field that is very unsatisfactory from the point of view of distribution and uniformity, so two coils sharing the same axis and separated by a convenient distance are normally used (Helmholtz coils, I wrote a paper on these things, they are very useful also to calibrate probes). Again three orthogonal systems get the job done.
-- for both you need amplifiers, whether for large voltage (electric field) or large current (magnetic field).
Ok, if we go to high frequency, or your problem is more complex, then you have to accurately state it, describing geometries, purpose, etc. Otherwise it will be difficult you get some information.
Just a hint: if you want to apply some electromagnetic field at a conveniently high frequency (let's say hundreds of MHz) [you talk about "rays"], then a logperiodic, or Yagi, antenna is probably what you are thinking of. Yagi is quite cheap because it's the old TV antenna, but is narrowband; log-p is broadband. Don't try to build them, both have complicated equations. And you need a RF amplifier, of course.