Assume that the uniformly distributed bound surface and bound volume current, $\vec{K}_{b}$,$\vec{J}_{b}$ respectively, are in the azimuthal direction $\phi$ within an infinitely long solenoid with a current passing through the wounded coils centered on the z-axis.
Clearly, the magnetic field field has no z and $\phi$ dependence due to the symmetry of current about the z-axis and the infinitely long solenoid. Hence,
$\vec{B}=\left \langle B_{s}\left ( s,\phi,\theta \right ),B_{\phi}\left ( s,\phi,\theta \right ),B_{z}\left ( s,\phi,\theta \right ) \right \rangle\rightarrow \vec{B}=\left ( B_{s}\left ( s \right ),B_{\phi}\left ( s \right ) ,B_{\theta}\left ( s \right )\right )$
Evidently, $B_{\phi}$ cannot exists for if it did it would violate the fact that a magnetic field cannot in a parallel to the current that generates it.
I know that using the Right hand rule in context of a current carrying solenoid, if the current runs in a direction 'towards' me, the thumb points in the positive z-direction. If the current runs in a direction 'away' from me, the thumb points in the negative z-direction. That thumb tells me the direction of the magnetic field. In both cases, it points outward/ away from the solenoid.
Here is where a bit of explanation would help:
How does this ties in with the fact that it violates the Biot-Savard law?