Is the dilogarithm
differentiable at ? Justify your answer.
Check that the function is even defined at . The series has radius of convergence , and at the endpoint it becomes the convergent -series with :
So the function is defined there, and by Abel's theorem it is also continuous from the left. Differentiability is a strictly stronger requirement.
Differentiate the series inside the disc. For term-by-term differentiation is valid:
The inner sum is the Mercator series , so
Numerical check at : a central difference of the series gives and ✓.
Examine the derivative as . As approaches from below, so , hence
The slope grows without bound: at it is , at it is , at it is — increasing like , slowly but without limit.
Conclude with the definition of the derivative. Since is differentiable on with derivative tending to , the mean value theorem forces the difference quotient to diverge as well. So the left derivative is and no finite derivative exists:
Geometrically the graph has a vertical tangent there.
See it in the local expansion. Landen's identity gives, with ,
Numerically at : the expansion gives against the true ✓. The term is what does the damage — its derivative blows up, confirming the failure of differentiability from a second direction.
Need to solve a different problem like this? Open the solver →