\(L\)-values for conductor 32 (Q2071675)

The author investigates the special values of the \(L\)-function of the elliptic curve \(E : y^2 = x^3-x\) with conductor 32, and finds explicit representations for these numbers as periods. Here and in what follows, a period (in the sense of Kontsevich and Zagier) is a complex number whose real and imaginary parts are absolutely convergent integrals \(\int_D f\), where \(f \in \mathbb{Q}(x_1,\ldots,x_n)\) and \(D\) is a domain in \(\mathbb{R}^n\) given by polynomial inequalities with rational coefficients [\textit{M. Kontsevich} and \textit{D. Zagier}, in: Mathematics unlimited -- 2001 and beyond. Berlin: Springer. 771--808 (2001; Zbl 1039.11002)]. Conjecturally, all the special values (or leading terms) of motivic \(L\)-functions at integers are periods, up to a power of \(2\pi i\) [loc. cit., Section 3.2]. For a modular form \(f\) of weight \(k \geq 2\) and arbitrary level, with algebraic Fourier coefficients, a deep theorem of \textit{A. A. Beilinson} [Contemp. Math. 55, 1--34 (1986; Zbl 0609.14006)], \textit{C. Deninger} and \textit{A. J. Scholl} [Lond. Math. Soc. Lect. Note Ser. 153, 173--209 (1991; Zbl 0729.14002)] and \textit{M. T. Gealy} [On the Tamagawa number conjecture for motives attached to modular forms. Pasadena: California Institute of Technology (PhD thesis) (2006)] implies that all the values \(L(f,m)\) with \(m \geq k\) are periods. However, it seems difficult to use this theorem to obtain explicit expressions for these periods. A natural problem is then, for a given \(L\)-value, to write it explicitly as a period. This is what the author does for the \(L\)-function of the elliptic curve \(E: y^2 = x^3-x\). Instead of using Beilinson's theorem, he uses the method developed by \textit{M. Rogers} and \textit{W. Zudilin} in [Compos. Math. 148, No. 2, 385--414 (2012; Zbl 1260.11062)] and [Int. Math. Res. Not. 2014, No. 9, 2305--2326 (2014; Zbl 1378.11091)], by writing the associated cusp form of level 32 as a product of two Eisenstein series of weight 1. Previously, \textit{W. Zudilin} [Springer Proc. Math. Stat. 43, 381--395 (2013; Zbl 1316.11038)] had found explicit expressions for \(L(E,2)\) and \(L(E,3)\) as periods. The first result of the author is an expression of \(L(E,4)\) as the following explicit integral: \[ L(E,4) = \frac{\pi^3}{1536} \int_{[0,1]^4} \frac{(1-6y+y^2) \sqrt{1-y} \; \mathrm{d}y \, \mathrm{d}y_1 \, \mathrm{d}y_2 \, \mathrm{d}y_3}{\sqrt{y(1+y)} \bigl(y^4 + 4(1-y^2)(1-y_1^2)(1-y_2^2)(1-y_3^2)\bigr)}. \] The author then proceeds to give an explicit formula for the generating series of the values \(L(E,m)\), \(m \geq 2\), in terms of Jacobi elliptic functions. From there, an explicit representation of \(L(E,m)\) as a period can be derived for any given \(m \geq 2\).