The Madelung constant of rock salt
Find a closed form for the Madelung constant of rock salt, M ≈ 1.7475645946, the alternating lattice sum of unit charges on Z^3.
State the problem, from HorizonMath
Browse files+45 −0 README.md added View file
| @@ -0,0 +1,45 @@ | ||
| 1 | +# The Madelung constant of rock salt | |
| 2 | + | |
| 3 | +The electrostatic energy of an ion in a sodium chloride crystal is set by one number, the Madelung constant $M \approx 1.7476$. It is a conditionally convergent lattice sum of alternating charges. The two-dimensional analogue has a closed form in terms of Dirichlet series values, but whether the three-dimensional constant can be written in terms of standard functions is unknown. | |
| 4 | + | |
| 5 | +## Problem | |
| 6 | + | |
| 7 | +Put a unit charge of sign $(-1)^{i+j+k}$ at every point $(i,j,k) \in \mathbb{Z}^3 \setminus \{0\}$, so the nearest-neighbour distance is 1. The Madelung constant of the rock-salt (NaCl) structure is | |
| 8 | + | |
| 9 | +$$ | |
| 10 | +M = -\sum_{(i,j,k) \neq (0,0,0)} \frac{(-1)^{i+j+k}}{\sqrt{i^2 + j^2 + k^2}} = 1.74756459463318219063\ldots | |
| 11 | +$$ | |
| 12 | + | |
| 13 | +The series converges only conditionally. The accepted value comes from summing over expanding cubes $\max(|i|,|j|,|k|) \le N$ with $N \to \infty$, not over expanding spheres [3, 4]. A rapidly convergent form is Benson's formula [1]: | |
| 14 | + | |
| 15 | +$$ | |
| 16 | +M = 12\pi \sum_{\substack{i,j \ge 1 \\ i,j \text{ odd}}} \operatorname{sech}^2\Bigl(\frac{\pi}{2}\sqrt{i^2 + j^2}\Bigr). | |
| 17 | +$$ | |
| 18 | + | |
| 19 | +Find a closed-form expression for $M$ in terms of standard constants and special functions. | |
| 20 | + | |
| 21 | +## Background | |
| 22 | + | |
| 23 | +Madelung introduced such sums in 1918 to compute the electrostatic energy of ionic crystals. The Born–Landé lattice energy of a crystal such as NaCl is proportional to its Madelung constant [4]. In 1974 Zucker evaluated the analogous alternating sums $M_N(s) = \sum' (-1)^{i_1+\cdots+i_N} (i_1^2 + \cdots + i_N^2)^{-s}$ in dimensions $N = 2, 4, 6, 8$ in terms of the Riemann zeta function and the Dirichlet eta and beta functions [2]. For example, $M_2(s) = -4\beta(s)\eta(s)$, so the two-dimensional Madelung constant is $4\beta(\tfrac12)\eta(\tfrac12) \approx 1.6155$ in magnitude [4]. No such evaluation is known in three dimensions. Lattice sums of this kind are surveyed in [5]. | |
| 24 | + | |
| 25 | +## What is known | |
| 26 | + | |
| 27 | +- $M = 1.7475645946331821906362120355443974034851614366247417581528\ldots$. HorizonMath gives these 59 significant digits as its reference, and [OEIS A085469](https://oeis.org/A085469) lists the same expansion. Benson's formula and other rapidly convergent series yield many more digits [1, 4]. | |
| 28 | +- Closed forms exist for the alternating sums in dimensions 1, 2, 4, 6 and 8 [2, 4]. For the three-dimensional constant, Burrows, Cooper and Schwerdtfeger wrote in 2022 that "it is currently not known if the Madelung constant can be expressed in terms of standard functions" [4]. | |
| 29 | + | |
| 30 | +## What would count as progress | |
| 31 | + | |
| 32 | +- A closed-form expression that matches $M$ to 20 significant digits, HorizonMath's acceptance threshold. Its evaluator checks this at 100-digit precision and rejects numerical shortcuts such as quadrature, root-finding, infinite series and fitted constants. Benson's formula converges fast enough to test a candidate to hundreds of digits. | |
| 33 | +- A proof that a candidate expression equals $M$, for example by a reduction of the three-dimensional sum to known Dirichlet series. | |
| 34 | + | |
| 35 | +## References | |
| 36 | + | |
| 37 | +1. G. C. Benson, "A simple formula for evaluating the Madelung constant of a NaCl-type crystal", *Can. J. Phys.* 34 (1956), 888–890. [doi:10.1139/p56-095](https://doi.org/10.1139/p56-095) | |
| 38 | +2. I. J. Zucker, "Exact results for some lattice sums in 2, 4, 6 and 8 dimensions", *J. Phys. A* 7 (1974), 1568–1575. [doi:10.1088/0305-4470/7/13/011](https://doi.org/10.1088/0305-4470/7/13/011) | |
| 39 | +3. D. Borwein, J. M. Borwein and K. F. Taylor, "Convergence of lattice sums and Madelung's constant", *J. Math. Phys.* 26 (1985), 2999–3009. [doi:10.1063/1.526675](https://doi.org/10.1063/1.526675) | |
| 40 | +4. A. Burrows, S. Cooper and P. Schwerdtfeger, "The Madelung constant in N dimensions", *Proc. R. Soc. A* 478 (2022), 20220334. [doi:10.1098/rspa.2022.0334](https://doi.org/10.1098/rspa.2022.0334), [arXiv:2202.01392](https://arxiv.org/abs/2202.01392) | |
| 41 | +5. J. M. Borwein, M. L. Glasser, R. C. McPhedran, J. G. Wan and I. J. Zucker, *Lattice Sums Then and Now*, Cambridge University Press, 2013. [doi:10.1017/CBO9781139626804](https://doi.org/10.1017/CBO9781139626804) | |
| 42 | + | |
| 43 | +## Source | |
| 44 | + | |
| 45 | +Imported from [HorizonMath madelung_nacl](https://github.com/ewang26/HorizonMath/blob/4ef0b61a8a6d4c5ded661fab7c2e896af84abe6e/data/problems_full.json#L748-L758), curated by Erik Y. Wang, Sumeet R. Motwani et al. Open according to the source as of 2026-10-20. The statement above is written in our own words; see the source for the original wording, history and discussion. | |