The physical 'meaning' of the imaginary part of the impedance is that it represents the energy storage part of the circuit element.
To see this, let the sinusoidal current $i = I\cos(\omega t)$ be the current through a series RL circuit.
The voltage across the combination is
$$v = Ri + L\frac{di}{dt} = RI\cos(\omega t) - \omega LI\sin(\omega t)$$
The instantaneous power is the product of the voltage and current
$$p(t) = v \cdot i = RI^2\cos^2(\omega t) - \omega LI^2\sin(\omega t)\cos(\omega t) $$
Using the well known trigonometric formulas, the power is
$$p(t) = \frac{RI^2}{2}[1 + \cos(2\omega t)] - \frac{\omega LI^2}{2}\sin(2\omega t) $$
Note that the first term on the RHS is never less than zero - power is always delivered to the resistor.
However, the power for the second term has zero average value and alternates symmetrically positive and negative - the inductor stores energy half the time and releases the energy the other half.
But note that $\omega L$ is the imaginary part of the impedance of the series RL circuit:
$$Z = R + j\omega L$$
Indeed, via the complex power S, we see that the imaginary part of the impedance is related the reactive power Q
$$S = P + jQ = \tilde I^2Z = \frac{I^2}{2}Z = \frac{RI^2}{2} + j\frac{\omega L I^2}{2} $$
Thus, as promised, the imaginary part of the impedance is the energy storage part while the real part of the impedance is the dissipative part.
However, since in real-life circuits both voltages and currents are real numbers...
I really don't want to sound like Wikipedia, but I think that this is a great place to say "citation needed". $\endgroup$