TOPICS
Search

Madelung Constants


The quantities obtained from cubic, hexagonal, etc., lattice sums, evaluated at s=1, are called Madelung constants.

For cubic lattice sums

 b_n(2s)=sum^'_(k_1,...,k_n=-infty)^infty((-1)^(k_1+...+k_n))/((k_1^2+...+k_n^2)^s),
(1)

the Madelung constants expressible in closed form for even indices n, a few examples of which are summarized in the following table, where beta(n) is the Dirichlet beta function and eta(n) is the Dirichlet eta function.

nb_n(2)OEISconstant
2-4beta(1)eta(1)=-piln2A0860542.177586...
4-8eta(1)eta(0)=-4ln2A0166392.772588...

To obtain the closed form for b_2(s), break up the double sum into pieces that do not include i=j=0,

b_2(2s)=sum_(i=-1)^(infty)sum_(j=-infty)^(-1)((-1)^(i+j))/((i^2+j^2)^s)+sum_(j=0)((-1)^(i+j))/((i^2+j^2)^s)+sum_(j=1)^(infty)((-1)^(i+j))/((i^2+j^2)^s)+sum_(i=0)sum_(j=-infty)^(-1)((-1)^(i+j))/((i^2+j^2)^s)+sum_(j=1)^(infty)((-1)^(i+j))/((i^2+j^2)^s)+sum_(i=1)^(infty)sum_(j=-infty)^(-1)((-1)^(i+j))/((i^2+j^2)^s)+sum_(j=0)((-1)^(i+j))/((i^2+j^2)^s)+sum_(j=1)^(infty)((-1)^(i+j))/((i^2+j^2)^s)
(2)
=sum_(i,j=1)^(infty)((-1)^(-(i+j)))/((i^2+j^2)^s)+sum_(i=1)^(infty)((-1)^(-i))/(i^(2s))+sum_(i,j=1)^(infty)((-1)^(-i+j))/((i^2+j^2)^s)+sum_(j=1)^(infty)((-1)^(-j))/(j^(2s))+sum_(j=1)^(infty)((-1)^j)/(j^(2s))
(3)
=sum_(i,j=1)^(infty)((-1)^(i-j))/((i^2+j^2)^s)+sum_(i=1)^(infty)((-1)^i)/(i^(2s))+sum_(i,j=1)^(infty)((-1)^(i+j))/((i^2+j^2)^s),
(4)

where the negative sums have been reindexed to run over positive quantities. But (-1)^i=(-1)^(-i), so all the above terms can be combined into

 b_2(2s)=4[sum_(i,j=1)^infty((-1)^(i+j))/((i^2+j^2)^s)+sum_(i=1)^infty((-1)^i)/(i^(2s))].
(5)

The second of these sums can be done analytically as

 sum_(i=1)^infty((-1)^i)/(i^(2s))=-4^s(4^s-2)zeta(2s),
(6)

which in the case s=1 reduces to

 sum_(i=1)^infty((-1)^i)/(i^2)=-1/(12)pi^2.
(7)

The first sum is more difficult, but in the case s=1 can be written

 sum_(i,j=1)^infty((-1)^(i+j))/(i^2+j^2)=1/(12)pi(pi-3ln2).
(8)

Combining these then gives the original sum as

 b_2(2)=-piln2.
(9)

b_3(1) is given by Benson's formula (Borwein and Bailey 2003, p. 24)

b_3(1)=-sum^'_(i,j,k=-infty)^infty((-1)^(i+j+k+1))/(sqrt(i^2+j^2+k^2))
(10)
=-12pisum_(m,n=1,3,...)^(infty)sech^2(1/2pisqrt(m^2+n^2))
(11)
=-1.74756...
(12)

(OEIS A085469), where the prime indicates that summation over (0, 0, 0) is excluded.

b_3(1)=M is sometimes called "the" Madelung constant, corresponds to the Madelung constant for a three-dimensional NaCl crystal. Crandall (1999) gave the expression

 M=-2pi+(Gamma(1/8)Gamma(3/8)sqrt(2))/(pi^(3/2)) 
 +2sum^'_(m, n, p=-infty)^infty((-1)^(m+n+p))/(sqrt(m^2+n^2+p^2)[exp(4pisqrt(m^2+n^2+p^2))+1]).
(13)

Similar results were found by Tyagi (2004),

M=-1/2-(ln2)/pi-pi/3-2sum^'_(m,n,p=-infty)^infty((-1)^(m+n+p))/(sqrt(m^2+n^2+p^2)[exp(2pisqrt(m^2+n^2+p^2))-1])
(14)
M=sqrt(2)-pi+2sum^'_(m,n,p=-infty)^infty((-1)^(m+n+p))/(sqrt(m^2+n^2+p^2)[exp(2pisqrt(m^2+n^2+p^2))+1])
(15)
M=-1/4-(ln2)/(2pi)-(2pi)/3+1/(sqrt(2))-2sum^'_(m,n,p=-infty)^infty((-1)^(m+n+p))/(sqrt(m^2+n^2+p^2)[exp(4pisqrt(m^2+n^2+p^2))-1])
(16)

the last of which converges rapidly. Averaging (16) and (13) then gives the beautiful equation

 M=-1/8-(ln2)/(4pi)-(4pi)/3+1/(2sqrt(2))+(Gamma(1/8)Gamma(3/8))/(pi^(3/2)sqrt(2)) 
 -2sum^'_(m,n,p=-infty)^infty((-1)^(m+n+p))/(sqrt(m^2+n^2+p^2)[exp(8pisqrt(m^2+n^2+p^2))-1]),
(17)

which is correct to 10 decimal digits even if the sum is completely omitted (Tyagi 2004).

However, no closed form for b_3(1) is known (Bailey et al. 2006).

For hexagonal lattice sums, h_2(2) is expressible in closed form as

h_2(2)=piln3sqrt(3)
(18)
=5.9779868...
(19)

(OEIS A086055).


See also

Benson's Formula, Harmonic Series, Lattice Sum

Explore with Wolfram|Alpha

References

Bailey, D. H.; Borwein, J. M.; Crandall, R. E.; and Zucker, I. J. "Lattice Sums Arising from the Poisson Equation." J. Phys. A 46, 115201, 31 pp., 2013.Bailey, D. H.; Borwein, J. M.; Kapoor, V.; and Weisstein, E. W. "Ten Problems in Experimental Mathematics." Amer. Math. Monthly 113, 481-509, 2006.Borwein, J. and Bailey, D. Mathematics by Experiment: Plausible Reasoning in the 21st Century. Wellesley, MA: A K Peters, 2003.Borwein, J. M. and Borwein, P. B. Pi & the AGM: A Study in Analytic Number Theory and Computational Complexity. New York: Wiley, 1987.Buhler, J. and Wagon, S. "Secrets of the Madelung Constant." Mathematica in Education and Research 5, 49-55, Spring 1996.Crandall, R. E. "New Representations for the Madelung Constant." Exp. Math. 8, 367-379, 1999.Crandall, R. E. and Buhler, J. P. "Elementary Function Expansions for Madelung Constants." J. Phys. Ser. A: Math. and Gen. 20, 5497-5510, 1987.Finch, S. R. Mathematical Constants. Cambridge, England: Cambridge University Press, 2003.Havil, J. "Madelung's Constant." §3.4 in Gamma: Exploring Euler's Constant. Princeton, NJ: Princeton University Press, pp. 33-35, 2003.Sloane, N. J. A. Sequences A016639, A085469, A086054, and A086055 in "The On-Line Encyclopedia of Integer Sequences."Tyagi, S. "New Series Representation for Madelung Constant." Oct. 17, 2004. http://arxiv.org/abs/cond-mat/0410424.

Referenced on Wolfram|Alpha

Madelung Constants

Cite this as:

Weisstein, Eric W. "Madelung Constants." From MathWorld--A Wolfram Web Resource. https://mathworld.wolfram.com/MadelungConstants.html

Subject classifications