Deisigning a wind turbine was actually my senior capstone for aerospace engineering at UMCP (I was on a team that competed in the USDOE Collegiate Wind Competition). We used an open source program called QBlade to help inform our blade design (I actually took point on this). What we found was quite interesting. We tested a couple standard wing airfoils including the Clark-Y and NACA 0012. These airfoils produced very poor power curves for a conventional horizontal axis wind turbine (HAWT).
Separately, there are airfoils out there designed specifically for HAWT applications. NREL has designed the S807, which is reasonable, but for small scale applications actually doesn't work very well. Then there's the SG6043, which my team ultimately settled on. This airfoil is very good in small scale applications. In our design work, the SG6043 produced a pretty stable power curve across a wide range of TSRs and windspeeds.
Overall, you'll notice that one key difference between the conventional wing airfoils and the wind turbine airfoils is that the L/D curves in the wind turbine foils is more symmetric about alpha == 0.
The 2021 UMCP design report is available from the USDOE, and goes into more detail about how we came to these conclusions.