Combine df1 and df2 side by side (columns from both, side by side), aligning rows by index label and keeping every index label that appears in either DataFrame, filling with NaN wherever one side has no value for a given label.

Input

df1 =
   x
a  1
b  2
c  3
df2 =
    y
b  10
d  20

Output

     x     y
a  1.0   NaN
b  2.0  10.0
c  3.0   NaN
d  NaN  20.0