The required vane geometry depends on the vane blade area, the length of the vane arm, the aspect ratio of the vane blade and very much on the position of the vane with respect to the wake of the rotor. As the main goal of a vane blade is to keep the rotor perpendicular to the wind, it must be able to overcome the friction moment of the head bearings. This friction moment depends very much on choice made for the head bearings. For all my VIRYA wind turbines I have used INA Permaglide bearings with a bronze bush and an inside Teflon coating running against a stainless steel shaft. These bearings have a low friction coefficient and can have a high pressure. They are also not sensible for water. So one can use a rather small diameter of the head pin and the friction moment is therefore rather low. The friction moment of ball bearings is even lower but these bearings have a much larger outer diameter and so the required head bearing housing is much larger. They also have to be protected against the penetration of water.
The moment of the vane blade increases about proportional to the angle of attack and increases also proportional to V^2. So the required angle of attack to get a certain moment increases strongly when the wind speed is lower. There is always a wind speed below which the rotor isn't turned in the wind. This wind speed should lay lower than the cut-in wind speed of the rotor. The wind speed in the rotor wake can be less than halve the undisturbed wind speed, so a vane blade which is placed outside the rotor wake can be much smaller than a vane blade which is placed in the rotor wake.
The normal coefficient Cn of a vane blade depends on the angle of attack alfa and on the aspect ratio. In my public report KD 551, Cn-alfa curves are given for five different aspect ratios (ratio in between the height and the width). The advantage of a square vane blade is that it stalls only at an angle of attack of more than 40° and that the maximum Cn value is rather high (about 1.6).