I'm having difficulty proving the LHS equals the RHS. Both the limit and integral is giving me problems.
$$ 4\pi\,\mathrm{rad}^2 ≟ \lim_{n\to\infty}n\int_{-\pi/2}^{\pi/2} \cos^{-1}(\sin(φ)² + \cos(φ)² \cdot \cos(\frac{2\pi\,\mathrm{rad}}{n}))\,dφ \qquad ①$$
The RHS is formed by trying to show when angular displacement in one plane (θ longitude) is integrated with angular displacement in an orthogonal plane (φ latitude) that the maximum 3d angle is 4π rad².
If δ is the angular distance (great circle angle) for two points at same latitude φ given a rotation of Δθ, then a 3d longitudinal strip of angles can be approximated by integrating longitude angular distance δ across the full range of latitude angles for small Δθ.
$$ δ = cos^{-1}(\sin(φ)² + \cos(φ)²\cos(Δθ)) $$
$$ \text{longitudinal slice} ≈\int_{-\pi/2}^{\pi/2} δ\,dφ $$
This can then be multiplied by the number of strips $n$ required for a full rotation, thus $n=\frac{2π\,\mathrm{rad}}{Δθ}$
$$ \text{full spherical angle} = \lim_{n\to\infty}\int_{-\pi/2}^{\pi/2} δ\,dφ $$
The expected value is 4π rad²= 4π sr, and from numerical integration it seems to converge on 4π. However, I am having a hard time proving ①.