On an article by Euler, premature though posthumous (Q2165306)

There are many gems still hidden in the collected works of Leonhard Euler. This article is about Euler's article \textit{De novo genere serierum rationalium} \dots\ written by Euler at the end of his life, and dealing with formulas for computing \(\pi\). Although \textit{C. E. Sandifer}, in his beautiful book [How Euler did even more. Washington, DC: Mathematical Association of America (MAA) (2015; Zbl 1314.01005)], discussed the results of Euler's article in question, he did not notice that the formula \[ \pi = \sum_{n=0}^\infty \frac{(-1)^n}{4^n} \Big(\frac2{4n+1} + \frac2{4n+2} + \frac1{4n+3}\Big), \] discovered (along with several others) in 1997 by \textit{D. Bailey} et al. [Math. Comput. 66, No. 218, 903--913 (1997; Zbl 0879.11073)] occurs explicitly in Euler's article. Euler derives this series from the simple observation that \(x^4 + 4 = (x^2 + 2x + 2)(x^2 - 2x + 2)\), which implies that \[ \arctan \frac{x}{2-x} = \int_0^x \frac{dt}{2-2t + t^2} = \int_0^x \frac{2\,dt}{4+t^4} + \int_0^x \frac{2t\,dt}{4+t^4} + \int_0^x \frac{t^2\,dt}{4+t^4}. \] The formula follows (with \(x = 1\)) by developing the denominator into a geometric series and integrating. The author also shows how Euler could have derived the famous BBP-formula with a simple modification of the reasoning above.