A data scientist is creating a histogram for a sample of $n = 512$ observations. The first quartile of the data is $Q_1 = 45$ and the third quartile is $Q_3 = 61$. Using the Freedman-Diaconis rule, $h = \\frac{2 \\cdot \\text{IQR}}{n^{1/3}}$, what is the optimal bin width $h$?