Mod p structure of alternating and non-alternating multiple harmonic sums
Journal de Théorie des Nombres de Bordeaux, Tome 23 (2011) no. 1, pp. 299-308.

Le théorème bien connu de Wolstenholme affirme que, pour tout premier p>3, la (p-1)-ième somme partielle de la série harmonique est congrue à 0 modulo p 2 . Si on remplace la série harmonique par k1 1/n k pour k pair alors la congruence est vraie seulement modulo p au lieu de modulo p 2 . On peut considérer des généralisations aux sommes harmoniques multiples (SHM) et aux aux sommes harmoniques multiples alternées (SHMA) qui sont des sommes partielles de séries zêta multiples et, respectivement, de sommes d’Euler alternées. Beaucoup de résultats dans cette direction ont été obtenus dans les articles récents [6, 7, 8, 10, 11, 12], que nous récapitulerons dans ce papier. Il apparait que, pour un premier p, la (p-1)-ième somme partielle d’une SHM ou SHMA générale n’est plus congrue à 0 modulo p ; cependant elle peut souvent être exprimée en terme de nombres de Bernoulli. Donc c’est un problème assez intéressant de trouver qui ils sont. Dans cet article, nous fournirons un cadre théorique dans lequel ce genre de résulats peut s’exprimer et être étudié plus longuement. Nous calculerons aussi quelques SHM supplémentaires modulo un premier p quand le poids est inférieur à 13.

The well-known Wolstenholme’s Theorem says that for every prime p>3 the (p-1)-st partial sum of the harmonic series is congruent to 0 modulo p 2 . If one replaces the harmonic series by k1 1/n k for k even, then the modulus has to be changed from p 2 to just p. One may consider generalizations of this to multiple harmonic sums (MHS) and alternating multiple harmonic sums (AMHS) which are partial sums of multiple zeta value series and the alternating Euler sums, respectively. A lot of results along this direction have been obtained in the recent articles [6, 7, 8, 10, 11, 12], which we shall summarize in this paper. It turns out that for a prime p the (p-1)-st sum of the general MHS and AMHS modulo p is not congruent to 0 anymore; however, it often can be expressed by Bernoulli numbers. So it is a quite interesting problem to find out exactly what they are. In this paper we will provide a theoretical framework in which this kind of results can be organized and further investigated. We shall also compute some more MHS modulo a prime p when the weight is less than 13.

Publié le :
DOI : https://doi.org/10.5802/jtnb.762
Classification : 11M41,  11B50,  11A07
Mots clés : Multiple harmonic sums, alternating multiple harmonic sums, duality, shuffle relations.
@article{JTNB_2011__23_1_299_0,
     author = {Jianqiang Zhao},
     title = {Mod $p$ structure of alternating and non-alternating multiple harmonic sums},
     journal = {Journal de Th\'eorie des Nombres de Bordeaux},
     pages = {299--308},
     publisher = {Soci\'et\'e Arithm\'etique de Bordeaux},
     volume = {23},
     number = {1},
     year = {2011},
     doi = {10.5802/jtnb.762},
     zbl = {1269.11086},
     mrnumber = {2780631},
     language = {en},
     url = {https://jtnb.centre-mersenne.org/articles/10.5802/jtnb.762/}
}
Jianqiang Zhao. Mod $p$ structure of alternating and non-alternating multiple harmonic sums. Journal de Théorie des Nombres de Bordeaux, Tome 23 (2011) no. 1, pp. 299-308. doi : 10.5802/jtnb.762. https://jtnb.centre-mersenne.org/articles/10.5802/jtnb.762/

[1] J. Blümlein, Algebraic relations between harmonic sums and associated quantities. Comput. Phys. Commun. 159 (2004), 19–54. arXiv: hep-ph/0311046. | MR 2105551 | Zbl 1097.11063

[2] J. Blümlein and S. Kurth, Harmonic sums and Mellin transforms up to two-loop order. Phys. Rev. D 60 (1999), art. 01418. arXiv: hep-ph/9810241.

[3] M.E. Hoffman, The Hopf algebra structure of multiple harmonic sums. Nuclear Phys. B (Proc. Suppl.) 135 (2004), 214–219. arXiv: math.QA/0406589. | MR 2111842

[4] M.E. Hoffman, Quasi-Shuffle Products. J. Algebraic Combin. 11 (2000), 49–68. | MR 1747062 | Zbl 0959.16021

[5] M.E. Hoffman, Algebraic aspects of multiple zeta values. In: Zeta Functions, Topology and Quantum Physics, Developments in Mathematics 14, T. Aoki et. al. (eds.), Springer, 2005, New York, pp. 51–74. arXiv: math.QA/0309425. | MR 2179272 | Zbl 1170.11324

[6] M.E. Hoffman, Quasi-symmetric functions and mod p multiple harmonic sums. arXiv: math.NT/0401319

[7] R. Tauraso, Congruences involving alternating multiple harmonic sum. Elec. J. Combinatorics, 17(1) (2010), R16. | MR 2587747 | Zbl pre05686998

[8] R. Tauraso and J. Zhao, Congruences of alternating multiple harmonic sums. In press: J. Comb. and Number Theory. arXiv: math/0909.0670

[9] J.A.M. Vermaseren, Harmonic sums, Mellin transforms and integrals. Int. J. Modern Phys. A 14 (1999), 2037–2076. arXiv: hep-ph/9806280. | MR 1693541 | Zbl 0939.65032

[10] J. Zhao, Wolstenholme type Theorem for multiple harmonic sums. Intl. J. of Number Theory, 4(1) (2008), 73–106. arXiv: math/0301252. | MR 2387917 | Zbl pre05275145

[11] J. Zhao, Finiteness of p-divisible sets of multiple harmonic sums. In press: Annales des Sciences Mathématiques du Québec. arXiv: math/0303043. | MR 3113307 | Zbl 1295.11093

[12] X. Zhou and T. Cai, A generalization of a curious congruence on harmonic sums. Proc. Amer. Math. Soc. 135 (2007), 1329-1333 | MR 2276641 | Zbl 1115.11006