AAA Abstract and Applied Analysis 1687-0409 1085-3375 Hindawi Publishing Corporation 121795 10.1155/2012/121795 121795 Research Article A Set of Mathematical Constants Arising Naturally in the Theory of the Multiple Gamma Functions Choi Junesang Kim Sung G. 1 Department of Mathematics Dongguk University Gyeongju 780-714 Republic of Korea dongguk.edu 2012 24 12 2012 2012 07 08 2012 10 09 2012 2012 Copyright © 2012 Junesang Choi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We introduce a set of mathematical constants which is involved naturally in the theory of multiple Gamma functions. Then we present general asymptotic inequalities for these constants whose special cases are seen to contain all results very recently given in Chen 2011.

1. Introduction and Preliminaries

The double Gamma function Γ2=1/G and the multiple Gamma functions Γn were defined and studied systematically by Barnes  in about 1900. Before their investigation by Barnes, these functions had been introduced in a different form by, for example, Hölder , Alexeiewsky  and Kinkelin . In about the middle of the 1980s, these functions were revived in the study of the determinants of the Laplacians on the n-dimensional unit sphere Sn (see ). Since then the multiple Gamma functions have attracted many authors' concern and have been used in various ways. It is seen that a set of constants {Aq  |  q:={1,2,3,}} given in (1.11) involves naturally in the theory of the multiple Gamma functions Γn (see  and references therein). For example, for sufficiently large real x and a, we have the Stirling formula for the G-function (see ; see also [21, page 26, equation (7)]): (1.1)logG(x+a+1)=x+a2log(2π)-logA+112-3x24-ax+(x22-112+a22+ax)logx+O(x-1)(x), where A is the Glaisher-Kinkelin constant (see [7, 2224]) given in (1.16) below. The Glaisher-Kinkelin constant A, the constants B and C below introduced by Choi and Srivastava have been used, among other things, in the closed-form evaluation of certain series involving zeta functions and in calculation of some integrals of multiple Gamma functions. So trying to give asymptotic formulas for these constants A, B, and C are significant. Very recently Chen  presented nice asymptotic inequalities for these constants A, B, and C by mainly using the Euler-Maclaurin summation formulas. Here, we aim at presenting asymptotic inequalities for a set of the mathematical constants Aq  (q) given in (1.11) some of whose special cases are seen to yield all results in .

For this purpose, we begin by summarizing some differential and integral formulas of the function f(x) in (1.2).

Lemma 1.1.

Differentiating the function (1.2)f(x):=xqlogx(q;x>0) times, we obtain (1.3)f()(x)=xq-  {j=1  (q-j+1)logx+P(q)}(;1q), where P(q) is a polynomial of degree -1 in q satisfying the following recurrence relation: (1.4)P(q)={(q-+1)P-1(q)+j=1-1  (q-j+1)({1};2q),1(=1). In fact, by mathematical induction on , we can give an explicit expression for P(q) as follows: (1.5)P(q)=j=1(q-j+1)·j=1  1q-j+1(;1q). Setting =q in (1.3) and (1.5), respectively, we get (1.6)f(q)(x)=q!(logx+Hq)(q), where Hn are the harmonic numbers defined by (1.7)Hn:=j=1n  1j(n). Differentiating f(q)(x) in (1.6) r times, we obtain (1.8)f(q+r)(x)=(-1)r+1q!(r-1)!  1xr(r). Integrating the function f(x) in (1.2) from 1 to n, we get (1.9)1n  f(x)dx=nq+1q+1logn+1-nq+1(q+1)2(q,n).

For each q, define a sequence {Aq(n)}n=1 by (1.10)logAq(n):=k=1n  kqlogk-(nq+1q+1+nq2+r=1[(q+1)/2]  B2r(2r)!  ·j=12r-1  (q-j+1)·nq+1-2r)logn+nq+1(q+1)2-r=1[(q+1)/2]+((-1)q-1)/2  B2r(2r)!P2r-1(q)nq+1-2r(n,q), where Br are Bernoulli numbers given in (1.12), Pr(q) are given in (1.5), and [x] denotes (as usual) the greatest integer ≦x. Define a set of mathematical constants Aq(q) by (1.11)logAq:=limn  logAq(n)(q). The Bernoulli numbers Br are defined by the generating function (see [21, Section 1.6]; see also, [26, Section 1.7]): (1.12)zez-1=r=0Brzrr!(|z|<2π). We introduce a well-known formula (see [21, Section 2.3]): (1.13)B2p=(-1)p+1  2(2p)!(2π)2p  ζ(2p)(p0:={0}), where ζ(s) is the Riemann Zeta function defined by (1.14)ζ(s):={n=11ns=11-2-sn=11(2n-1)s((s)>1)11-21-s  n=1(-1)n-1ns((s)>0;s1). It is easy to observe from (1.13) that (1.15)B4p<0,B4p+2>0(p0).

Remark 1.2.

We find that the constants A1, A2 and A3 correspond with the Glaisher-Kinkelin constant A, the constants B and C introduced by Choi and Srivastava, respectively: (1.16)logA1=limn{k=1nklogk-(n22+n2+112)logn+n24}=logA, where A denotes the Glaisher-Kinkelin constant whose numerical value is (1.17)A1.282427130,logA2=limn{k=1nk2logk-(n33+n22+n6)logn+n39-n12}=logB,logA3=limn{k=1nk3logk-(n44+n32+n24-1120)logn+n416-n212}=logC. Here B and C are constants whose approximate numerical values are given by (1.18)B1.03091  675  ,C0.97955  746  . The constants B and C were considered recently by Choi and Srivastava [16, 18]. See also Adamchik [27, page 199]. Bendersky  presented a set of constants including B and C.

2. Euler-Maclaurin Summation Formula

We begin by recalling the Euler-Maclaurin summation formula (cf. Hardy ([29, 30], page 318)): (2.1)k=1n  f(k)~C0+anf(x)dx+12  f(n)+r=1  B2r(2r)!f(2r-1)(n), where C0 is an arbitrary constant to be determined in each special case and Br are the Bernoulli numbers given in (1.12). For another useful summation formula, see Edwards [31, page 117].

Let f be a function of class C2p+2([a,b]), and let the interval [a,b] be partitioned into m subintervals of the same length h=(b-a)/m. Then we have another useful form of the Euler-Maclaurin summation formula (see, e.g., ): There exists 0<θ<1 such that (2.2)k=0mf(a+kh)=1habf(x)dx+f(a)+f(b)2+k=1ph2k-1(2k)!B2k(f(2k-1)(b)-f(2k-1)(a))+h2p+2(2p+2)!B2p+2k=0m-1f(2p+2)(a+kh+θh), where m,p. Under the same conditions in (2.2), setting m=n-1, a=1, b=n, and h=1 in (2.2), we obtain a simple summation formula (see ): (2.3)k=1m  f(k)=1nf(x)  dx+f(1)+f(n)2+k=1p  B2k(2k)!  (f(2k-1)(n)-f(2k-1)(1))+Rn(f,p), where, for convenience, the remainder term Rn(f,p) is given by (2.4)Rn(f,p):=B2p+2(2p+2)!  k=1n-1  f(2p+2)(k+θ) which is seen to be bounded by (2.5)|Rn(f,p)|2(2π)2p1n|f(2p+1)(x)|dx.

Zhu and Yang  established some useful formulas originated from the Euler-Maclaurin summation formula (2.1) (see also ) asserted by the following lemma.

Lemma 2.1.

Let and let f have its first 2p+2 derivatives on an interval [,) such that f(2p)(x)>0 and f(2p+2)(x)>0 (or f(2p)(x)<0 and f(2p+2)(x)<0) and f(2p-1)()=0. Then the following results hold true:

The sequence (2.6)an:=k=n  f(k)-nf(x)dx-12f(n)-k=1p-1  B2k(2k)!  f(2k-1)(n)(n) converges. Let a:=limn  an.

For f(2p)(x)>0 and f(2p+2)(x)>0, we have (2.7)0<(-1)p-1  (a-an)<(-1)p  B2p(2p)!  f(2p-1)(n)(n). For f(2p)(x)<0 and f(2p+2)(x)<0, we have (2.8)0>(-1)p-1(a-an)>(-1)p  B2p(2p)!  f(2p-1)(n)(n).

There exists θ(0,1) such that (2.9)k=n  f(k)=a+nf(x)dx+12  f(n)+k=1p-1  B2k(2k)!  f(2k-1)(n)+θ·B2p(2p)!  f(2p-1)(n).

3. Asymptotic Formulas and Inequalities for <inline-formula><mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M115"><mml:mrow><mml:msub><mml:mrow><mml:mi>A</mml:mi></mml:mrow><mml:mrow><mml:mi>q</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula>

Applying the function f(x) in (1.2) to the Euler-Maclaurin summation formula (2.1) with a=1 and using the results presented in Lemma 1.1, we obtain an asymptotic formula for the sequence Aq(n) as in the following theorem.

Theorem 3.1.

The following asymptotic formulas for the constants Aq(n) and Aq hold true: (3.1)logAq(n)~Cq+1(q+1)2+1-(-1)q2  Bq+1Hqq+1+(-1)qq!  r=[(q+1)/2]+1  B2r(2r)!  (2r-q-2)!n2r-q-1, where Cq's are constants dependent on each q and an empty sum is understood (as usual) to be nil. And (3.2)logAq=limnlogAq(n)=Cq+1(q+1)2+1-(-1)q2  Bq+1  Hqq+1.

Proof.

We only note that

1r[(q+1)/2](3.3)f(2r-1)(n)=nq+1-2r·j=12r-1(q-j+1)·logn+nq+1-2rP2r-1(q).

r[(q+1)/2]+1(3.4)f(2r-1)(n)=(-1)qq!(2r-q-2)!n2r-q-1.

Applying the function f(x) in (1.2) to the formula (2.9) with =1, and using the results presented in Lemma 1.1, we get two sided inequalities for the difference of logAq(n) and logAq asserted by Theorem 3.2.

Theorem 3.2.

The following inequalities hold true: (3.5)q!r=[(q+1)/2]+12p(2r-q-2)!(2r)!B2rn2r-q-1<(-1)q(logAq(n)-logAq)<q!r=[(q+1)/2]+12p+1(2r-q-2)!(2r)!B2rn2r-q-1(p,q,n).

Proof.

Setting the function f(x) in (1.2) in the formula (2.9) with =1, and using the results presented in Lemma 1.1, we get (3.6)logAq(n)=logAq+(-1)qq!r=[(q+1)/2]+1p-1(2r-q-2)!(2r)!B2rn2r-q-1+(-1)qq!(2p-q-2)!(2p)!B2pn2p-q-1θ, for some θ(0,1).

Replacing p by 2p+1 and 2p+2 in (3.6), respectively, we obtain (3.7)logAq(n)-logAq=(-1)qq!r=[(q+1)/2]+12p(2r-q-2)!(2r)!B2rn2r-q-1+(-1)qq!(4p-q)!(4p+2)!B4p+2n4p+1-qθ.logAq(n)-logAq=(-1)qq!r=[(q+1)/2]+12p+1(2r-q-2)!(2r)!B2rn2r-q-1+(-1)qq!(4p+2-q)!(4p+4)!B4p+4n4p+3-qθ. In view of (1.15), we find the following inequalities: (3.8)q!r=[(q+1)/2]+12p(2r-q-2)!(2r)!B2rn2r-q-1<logAq(n)-logAq<q!r=[(q+1)/2]+12p+1(2r-q-2)!(2r)!B2rn2r-q-1(q is  even ),q!r=[(q+1)/2]+12p(2r-q-2)!(2r)!B2rn2r-q-1<logAq-logAq(n)<q!r=[(q+1)/2]+12p+1(2r-q-2)!(2r)!B2rn2r-q-1(q is  odd ). Finally it is easily seen that the two-sided inequalities can be expressed in a single form (3.5).

Remark 3.3.

The special cases of (3.5) when q=1, q=2, and q=3 are easily seen to correspond with Equations (8), (31), and (32) in Chen's work , respectively.

Applying the function f(x) in (1.2) to the formula (2.3) and using the results presented in Lemma 1.1, we get two-sided inequalities for the logAq asserted by Theorem 3.4.

Theorem 3.4.

The following inequalities hold true: (3.9)αq+q!k=[(q+1)/2]+12p(2k-q-2)!(2k)!B2k<logAq<αq+q!k=[(q+1)/2]+12p+1(2k-q-2)!(2k)!B2k(q is  odd ),(3.10)αq-q!k=[(q+1)/2]+12p+1(2k-q-2)!(2k)!B2k<logAq<αq-q!k=[(q+1)/2]+12p(2k-q-2)!(2k)!B2k(q is  even ), where, for convenience, (3.11)αq:=1(q+1)2+1-(-1)q2Bq+1Hqq+1-k=1[(q+1)/2]B2k(2k)!P2k-1(q)(q).

Proof.

Setting the function f(x) in (1.2) in the formula (2.3), and using the results presented in Lemma 1.1, we have, for some θ(0,1), (3.12)logAq(n)=αq+(-1)qq!k=[(q+1)/2]+1p(2k-2-q)!(2k)!B2k(1n2k-1-q-1)+(-1)q+1q!(2p+1-q)!(2p+2)!B2p+2k=1n-11(k+θ)2p+2-q. Replacing p by 2p and 2p+1 in (3.12), respectively, we obtain (3.13)logAq(n)=αq+(-1)qq!k=[(q+1)/2]+12p(2k-2-q)!(2k)!B2k(1n2k-1-q-1)+(-1)q+1q!(4p+1-q)!(4p+2)!B4p+2k=1n-11(k+θ)4p+2-q,logAq(n)=αq+(-1)qq!k=[(q+1)/2]+12p+1(2k-2-q)!(2k)!B2k(1n2k-1-q-1)+(-1)q+1q!(4p+3-q)!(4p+4)!B4p+4k=1n-11(k+θ)4p+4-q. In view of (1.15), we find from that (3.14)αq-q!k=[(q+1)/2]+12p(2k-2-q)!(2k)!B2k(1n2k-1-q-1)<logAq(n)<αq-q!k=[(q+1)/2]+12p+1(2k-2-q)!(2k)!B2k(1n2k-1-q-1)(q is  odd ),αq+q!k=[(q+1)/2]+12p+1(2k-2-q)!(2k)!B2k(1n2k-1-q-1)<logAq(n)<αq+q!k=[(q+1)/2]+12p(2k-2-q)!(2k)!B2k(1n2k-1-q-1)(q is  even ). Now, taking the limit on each side of the inequalities in as n, we obtain the results in Theorem 3.4.

Remark 3.5.

It is easily seen that the specialized inequalities of (3.9) when q=1 and q=3 and (3.10) when q=2 correspond with those inequalities of Equations (9), (34), and (33) in Chen's work , respectively.

Acknowledgments

This research was supported by Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2012-0002957).

Barnes E. W. The theory of the G-function Quarterly Journal of Mathematics 1899 31 264 314 Barnes E. W. The genesis of the double gamma functions Proceedings of the London Mathematical Society 1899 31 1 358 381 10.1112/plms/s1-31.1.358 1576719 Barnes E. W. The theory of the double Gamma function Philosophical Transactions of the Royal Society A 1901 196 265 388 Barnes E. W. On the theory of the multiple Gamma functions Transactions of the Cambridge Philosophical Society 1904 19 374 439 Hölder O. Uber Eine Transcendente Funktion 1886 1886 Göttingen, Germany Dieterichsche Alexeiewsky W. P. Uber Eine Classe von Funktionen, die der Gammafunktion Analog Sind 1894 46 Leipzig Weidmannsche Buchhandlung Kinkelin V. H. Uber eine mit der Gamma Funktion verwandte transcendente und deren Anwendung auf die integralrechnung Journal für Die Reine und Angewandte Mathematik 1860 57 122 158 Choi J. Determinant of Laplacian on S3 Mathematica Japonica 1994 40 1 155 166 1288030 ZBL0806.58053 Kumagai H. The determinant of the Laplacian on the n-sphere Acta Arithmetica 1999 91 3 199 208 1735678 Osgood B. Phillips R. Sarnak P. Extremals of determinants of Laplacians Journal of Functional Analysis 1988 80 1 148 211 10.1016/0022-1236(88)90070-5 960228 ZBL0653.53022 Quine J. R. Choi J. Zeta regularized products and functional determinants on spheres The Rocky Mountain Journal of Mathematics 1996 26 2 719 729 10.1216/rmjm/1181072081 1406503 ZBL0864.47024 Vardi I. Determinants of Laplacians and multiple gamma functions SIAM Journal on Mathematical Analysis 1988 19 2 493 507 10.1137/0519035 930041 ZBL0641.33003 Voros A. Spectral functions, special functions and the Selberg zeta function Communications in Mathematical Physics 1987 110 3 439 465 891947 10.1007/BF01212422 ZBL0631.10025 Choi J. Some mathematical constants Applied Mathematics and Computation 2007 187 1 122 140 10.1016/j.amc.2006.08.091 2323562 ZBL1117.33003 Choi J. Cho Y. J. Srivastava H. M. Series involving the zeta function and multiple gamma functions Applied Mathematics and Computation 2004 159 2 509 537 10.1016/j.amc.2003.08.134 2096722 ZBL1061.33001 Choi J. Srivastava H. M. Certain classes of series involving the zeta function Journal of Mathematical Analysis and Applications 1999 231 1 91 117 10.1006/jmaa.1998.6216 1676729 ZBL0932.11054 Choi J. Srivastava H. M. An application of the theory of the double gamma function Kyushu Journal of Mathematics 1999 53 1 209 222 10.2206/kyushujm.53.209 1678018 ZBL1013.11053 Choi J. Srivastava H. M. Certain classes of series associated with the zeta function and multiple gamma functions Journal of Computational and Applied Mathematics 2000 118 1-2 87 109 Higher transcendental functions and their applications 10.1016/S0377-0427(00)00311-3 1765942 ZBL0969.11030 Choi J. Srivastava H. M. Adamchik V. S. Multiple gamma and related functions Applied Mathematics and Computation 2003 134 2-3 515 533 10.1016/S0096-3003(01)00301-0 1931554 ZBL1026.33003 Choi J. Srivastava H. M. Quine J. R. Some series involving the zeta function Bulletin of the Australian Mathematical Society 1995 51 3 383 393 10.1017/S0004972700014210 1331431 ZBL0830.11030 Srivastava H. M. Choi J. Series Associated with the Zeta and Related Functions 2001 Dordrecht, The Netherlands Kluwer Academic Publishers 1849375 Glaisher J. W. L. On the product 1122nn Messenger of Mathematics 1877 7 43 47 Glaisher J. W. L. On the constant which occurs in the formula for 1122nn Messenger of Mathematics 1894 24 1 16 http://mathworld.wolfram.com/Glaisher-KinkelinConstant.html Chen C.-P. Glaisher-Kinkelin constant Integral Transforms and Special Functions, IFirst 2011 1 8 Srivastava H. M. Choi J. Zeta and Q-Zeta Functions and Associated Series and Integrals 2012 Amsterdam, The Netherlands Elsevier Science Publishers Adamchik V. S. Polygamma functions of negative order Journal of Computational and Applied Mathematics 1998 100 2 191 199 10.1016/S0377-0427(98)00192-7 1659113 ZBL0936.33001 Bendersky L. Sur la fonction gamma généralisée Acta Mathematica 1933 61 1 263 322 10.1007/BF02547794 1555378 ZBL0008.07001 Hardy G. H. Divergent Series 1949 Oxford, UK Clarendon Press , Oxford University Press Hardy G. H. Divergent Series 1991 2nd New York, NY, USA Chelsea PublishingCompany Edwards J. A Treatise on the Integral Calculus with Applications: Examples and Problems 1954 1-2 New York, NY, USA Chelsea Publishing Company Graham R. L. Knuth D. E. Patashnik O. Concrete Mathematics: A Foundation for Computer Science 1994 2nd Reading, Mass, USA Addison-Wesley Publishing Company 1397498 Zhu Y.-H. Yang B.-C. Accurate inequalities for partial sums of a type of divergent series Acta Scientiarum Naturalium Universitatis Sunyatseni 1998 37 4 33 37 1675492