Appendix/Mathematics/ToroidalConfusion

From JETohlineWiki
Jump to navigation Jump to search


Confusion Regarding Whipple Formulae[edit]

May, 2018 (J.E.Tohline): I am trying to figure out what the correct relationship is between half-integer degree, associated Legendre functions of the first and second kinds. In order to illustrate my current confusion, here I will restrict my presentation to expressions that give Qn−1/2m(cosh⁡η) in terms of Pm−1/2n(coth⁡η).

Published Expressions[edit]

From equation (34) of H. S. Cohl, J. E. Tohline, A. R. P. Rau, & H. M. Srivastiva (2000, Astronomische Nachrichten, 321, no. 5, 363 - 372) I find:

Expression #1

Qn−1/2m(cosh⁡η)

=

(−1)nπΓ(n−m+12)[π2sinh⁡η]1/2Pm−1/2n(coth⁡η).


From Howard Cohl's online overview of toroidal functions, I find:

Expression #2

Qm−1/2n(cosh⁡α)

=

(−1)nΓ(n−m+12)[π2sinh⁡α]1/2Pn−1/2m(coth⁡α),

Copying the Whipple's formula from §14.19 of DLMF,

Expression #3

Qn−12m(cosh⁡ξ)

=

Γ(m−n+12)Γ(m+n+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ).

So far, this gives me three similar but not identical formulae for the same function mapping! As per equation (8) in (yet another source!) A. Gil, J. Segura, & N. M. Temme (2000, JCP, 161, 204 - 217), the relationship is:

Qn−1/2m(λ)

=

(−1)nπ3/22Γ(n−m+1/2)(x2−1)1/4Pm−1/2n(x),

Gil, Segura, & Temme (2000):  eq. (8)

where:    

λ≡x/x2−1

This expression from Gil et al. (2000) means, for example, that by identifying x with coth⁡η, we have λ=cosh⁡η, and,

Qn−1/2m(cosh⁡η)

=

(−1)nπ3/22Γ(n−m+1/2)(coth2η−1)1/4Pm−1/2n(coth⁡η)

 

=

(−1)nπΓ(n−m+1/2)(π2)1/2[cosh2ηsinh2η−1]1/4Pm−1/2n(coth⁡η)

 

=

(−1)nπΓ(n−m+1/2)(π2)1/2[1sinh⁡η]1/2Pm−1/2n(coth⁡η).

That is, we have,

Expression #4

Qn−1/2m(cosh⁡η)

=

(−1)nπΓ(n−m+1/2)[π2sinh⁡η]1/2Pm−1/2n(coth⁡η),

which matches the above expression #1 drawn from Cohl et al. (2000), but which appears not to match either of the other two "published" (online) formulae, expressions #2 or #3.

Specific Application[edit]

I stumbled into this dilemma when I tried to explicitly demonstrate how Q−1/2(cosh⁡η) can be derived from P−1/2(z) where, from §8.13 of M. Abramowitz & I. A. Stegun (1995), we find,

Q−1/2(cosh⁡η)

=

2e−η/2K(e−η),

Abramowitz & Stegun (1995), eq. (8.13.4)

and,

P−1/2(z)

=

2π[2z+1]1/2K(z−1z+1).

Abramowitz & Stegun (1995), eq. (8.13.1)

When I used the Whipple formula as defined in §14.19 of DLMF (expression #3 reprinted above), the function mapping gave me the wrong result; I was off by a factor of Γ(12)=π. But, as demonstrated below, the Whipple formula provided by Cohl et al. (2000) and by Gil et al. (2000) — that is, expressions #1 and #4, above — does give the correct result.

Demonstration that Q−12 can be derived from P−12

Copying equation (34) from Cohl et al. (2000), we begin with,

Qn−1/2m(cosh⁡η)

=

(−1)nπΓ(n−m+12)[π2sinh⁡η]1/2Pm−1/2n(coth⁡η);

then setting m=n=0, we have,

Q−12(cosh⁡η)

=

πΓ(12)[π2sinh⁡η]1/2P−12(coth⁡η)

 

=

π2[1sinh⁡η]1/2P−12(coth⁡η).

Step #1:   Associate … z↔cosh⁡η. Then,

Q−12(cosh⁡η)

=

π2[1z2−1]1/2P−12(zz2−1).

Step #2:   Now making the association … Λ↔z/z2−1, and drawing on eq. (8.13.1) from Abramowitz & Stegun (1995), we can write,

P−12(Λ)

=

2π[2Λ+1]1/2K(Λ−1Λ+1)

 

=

2π[2z2−1z+z2−1]1/2K(z−z2−1z+z2−1).

Step #3:   Again, making the association … z↔cosh⁡η, means,

P−12(Λ)

=

2π[2sinh⁡ηcosh⁡η+sinh⁡η]1/2K(cosh⁡η−sinh⁡ηcosh⁡η+sinh⁡η)

⇒Q−12(cosh⁡η)

=

π2[1sinh⁡η]1/22π[2sinh⁡ηcosh⁡η+sinh⁡η]1/2K(cosh⁡η−sinh⁡ηcosh⁡η+sinh⁡η)

 

=

2[1cosh⁡η+sinh⁡η]1/2K(cosh2η−sinh2η[cosh⁡η+sinh⁡η]2)

 

=

2[1eη]1/2K(1e2η)

 

=

2e−η/2K(e−η).

This, indeed, matches eq. (8.13.4) from Abramowitz & Stegun (1995).

Cohl's Response to My (May 2018) Email Query[edit]

Proper Interpretation of DLMF Expression[edit]

Most of the confusion expressed above stems from the DLMF's use of bold fonts, such as the function on the left-hand side of expression #3, above — that is, the Whipple formula from §14.19 of DLMF,

Qn−12m(cosh⁡ξ)

=

Γ(m−n+12)Γ(m+n+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ).

What has been missing in my discussion is an appreciation of the following relationship between bold and plain-text function names,

Qνμ(x)=e−μπiQνμ(x)Γ(ν+μ+1).

After making the substitutions, μ→m and ν→(n−12), the Whipple formula displayed above as expression #3 becomes,

e−mπiQn−12m(cosh⁡ξ)Γ(n+m+12)

=

Γ(m−n+12)Γ(m+n+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ)

⇒Qn−12m(cosh⁡ξ)

=

emπiΓ(m−n+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ)

 

=

(−1)mΓ(m−n+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ),

which matches expression #2, above. But it does not appear to match expressions #1 or #4.

The standard "Euler reflection formula for gamma functions" is usually presented in the form,

Γ(z)Γ(1−z)

=

πsin⁡(πz)

|

for example, if
z→(m−n+12)

⇒Γ(m−n+12)Γ(n−m+12)

=

π{sin⁡[π2+π(m−n)]}−1

 

=

π(−1)m−n

DLMF §5.5(ii)

|
Valid for:

   z≠0,±1,±2, …

|

If we make the association,

z↔(m−n+12),

with m and n both being either zero or a positive integer, then, this Euler reflection formula becomes,

Γ(m−n+12)Γ(n−m+12)

=

π{sin⁡[π(m−n+12)]}−1

 

=

π(−1)m+n.

However, in our situation the so-called "Euler reflection formula for gamma functions" gives the relation,

π(−1)m+nΓ(n−m+12)

=

Γ(m−n+12).

Hence, we may also write,

Qn−12m(cosh⁡ξ)

=

(−1)m[π(−1)m+nΓ(n−m+12)](π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ)

 

=

(−1)nπΓ(n−m+12)(π2sinh⁡ξ)1/2Pm−12n(coth⁡ξ),

which matches expressions #1 and #4. So everything appears to be in agreement! Hooray!

Derivation From Scratch[edit]

Whenever he deals with these types of relations, Cohl usually begins with,

Expression #5

Qνμ(cosh⁡η)

=

π2Γ(ν+μ+1)eiμπ[1sinh2η]1/4P−μ−12−ν−12(coth⁡η)

Making the pair of substitutions,

ν

=

n−12,

     

n∈

ℕ0={0,1,2,⋯},

μ

=

m,

     

m∈

ℕ0={0,1,2,⋯},

we also have,

ν+μ+1

=

n−12+m+1

=

n+m+12,

−μ−12

=

−m−12,

 

 

−ν−12

=

−(n−12)−12

=

−n,

eiμπ

=

eimπ

=

(−1)m,

in which case,

Qn−12m(cosh⁡η)

=

π2Γ(n+m+12)(−1)m[1sinh⁡η]P−m−12−n(coth⁡η).


Now, since,

Pνμ(z)

=

P−ν−1μ(z),

if we make the substitution,

−(ν+1)

→

−(m+12)

    ⇒   

ν

→

m−12,

we also know that,

Pm−12μ(z)

=

P−m−12μ(z).

Hence, we can write,

Qn−12m(cosh⁡η)

=

π2Γ(n+m+12)(−1)m[1sinh⁡η]Pm−12−n(coth⁡η).


Finally, another relation states that, for n∈ℕ0,

Pm−12−n(z)

=

[Γ(m−n+12)Γ(m+n+12)]Pm−12n(z).

So, we obtain,

Qn−12m(cosh⁡η)

=

(−1)mπ2Γ(n+m+12)[1sinh⁡η][Γ(m−n+12)Γ(m+n+12)]Pm−12n(coth⁡η).

 

=

(−1)mπ2Γ(m−n+12)[1sinh⁡η]Pm−12n(coth⁡η).

This matches expressions #2 and #3, above.

Index Values of Zero[edit]

Setting n=m=0 gives the following sought-for relationship:

Q−120(cosh⁡η)

=

π2Γ(12)[1sinh⁡η]P−120(coth⁡η).

 

=

π2[1sinh⁡η]P−120(coth⁡η).

Joel's Additional Manipulations[edit]

From §14.19.6 of DLMF, we find the following summation expression:

Q−12μ(cosh⁡ξ)+2∑n=1∞Γ(μ+n+12)Γ(μ+12)Qn−12μ(cosh⁡ξ)cos⁡(nϕ)=(12π)1/2(sinh⁡ξ)μ(cosh⁡ξ−cos⁡ϕ)μ+(1/2)

Then, if we again employ the DLMF relationship between bold and plain-text function names, namely,

Qn−12μ(x)=e−μπiQn−12μ(x)Γ(μ+n+12),

where we have made the substitution, ν→(n−12), the Sums expression becomes,

e−μπiQ−12μ(cosh⁡ξ)Γ(μ+12)

=

(12π)1/2(sinh⁡ξ)μ(cosh⁡ξ−cos⁡ϕ)μ+(1/2)−2∑n=1∞Γ(μ+n+12)Γ(μ+12)[e−μπiQn−12μ(cosh⁡ξ)Γ(μ+n+12)]cos⁡(nϕ)

⇒Q−12μ(cosh⁡ξ)

=

eμπiΓ(μ+12)[(12π)1/2(sinh⁡ξ)μ(cosh⁡ξ−cos⁡ϕ)μ+(1/2)]−2∑n=1∞Qn−12μ(cosh⁡ξ)cos⁡(nϕ).

When dealing with Dyson-Wong tori, we will set μ=0, in which case the Sums expression becomes,

Q−12(cosh⁡ξ)

=

[π/2(cosh⁡ξ−cos⁡ϕ)12]−2∑n=1∞Qn−12(cosh⁡ξ)cos⁡(nϕ).

But this can be rewritten in the form,

∑n=0∞ϵnQn−12(cosh⁡ξ)cos⁡(nϕ)

=

[π/2(cosh⁡ξ−cos⁡ϕ)12]

See Also[edit]

Tiled Menu

Appendices: | VisTrailsEquations | VisTrailsVariables | References | Ramblings | VisTrailsImages | myphys.lsu | ADS |