Elliptic Functions: Genesis of Theta Functions

Be the first to leave a comment!
We have seen in the last post that the elliptic functions can be expressed in the form of infinite product and these products look actually like ratio of two infinite products. We wish to consider these products (which are more commonly known as theta functions) in more detail in this post.

Theta Functions θ3 and θ4

Let us consider the infinite product expansion of dn(u,k) dn(u,k)=kn=11+2q2n1cos(πuK)+q4n212q2n1cos(πuK)+q4n2 We put τ=iK/K so that q=exp(πiτ)=exp(πK/K) and z=πu/2K. The denominator of the above expansion can be taken as a function h(z)=Gn=1(12q2n1cos2z+q4n2)=Gn=1(1q2n1e2iz)(1q2n1e2iz) where G is some constant independent of z to be determined later.

It is easily seen that the function h(z) has zeroes for z=±(2n1)(πτ/2)+mπ where m,n are integers. We express the same by saying that the zeroes of h(z) are z=πτ/2(modπ,πτ).

Then clearly h(z) is periodic with period π. Also we have h(z+πτ)=Gn=1(1q2q2n1e2iz)(1q2q2n1e2iz)=1q1e2iz1qe2izh(z)=q1e2izh(z) Note that the function h(z) we have defined by means of an infinite product is defined only when |q|<1 and when that is the case the defining product is absolutely convergent. Since h(z) is periodic with period π we can express it as a Fourier series h(z)=n=ane2niz and since h(z) is even we have an=an. Since h(z+πτ)=q1e2izh(z) we have n=ane2ni(z+πτ)=q1n=ane2(n1)izn=anq2n+1e2niz=n=an+1e2niz Since the Fourier series is unique we must have an+1=q2n+1an. By repeated application of this relation it is easy to see that an=(1)nqn2a0.

Thus we have h(z)=Gn=1(1q2n1e2iz)(1q2n1e2iz)=a0n=(1)nqn2e2niz To simplify things we choose the constant G such that we have a0=1. This determines (although still indirectly) G and we have the relation h(z)=Gn=1(1q2n1e2iz)(1q2n1e2iz)=n=(1)nqn2e2niz The above function h(z) as we have defined is one of the four theta functions and historically it has been denoted by θ4(z,q)=θ4(zτ)

We thus have θ4(z,q)=Gn=1(12q2n1cos2z+q4n2)=n=(1)nqn2e2niz=1+2n=1(1)nqn2cos(2nz) We can handle the numerator in the infinite product expansion of dn(u,k) to get another theta function θ3(z,q)=θ3(zτ) and get the following θ3(z,q)=Gn=1(1+2q2n1cos2z+q4n2)=n=qn2e2niz=1+2n=1qn2cos(2nz) It is quite obvious that θ3(z,q)=θ4(z±π/2,q) and hence the zeroes of θ3(z,q) are given by z=(π/2)+(πτ)/2(modπ,πτ).

Like the case of θ4(z,q) it is easy to see that θ3(z+πτ)=q1e2izθ3(z,q) and therefore if we consider the function F(z)=θ3(z,q)/θ4(z,q), we can easily see that F(z+π)=F(z) and F(z+πτ)=F(z) so that F(z+2πτ)=F(z). It is now clear that the function F(z) is doubly periodic with periods π and 2πτ. This corresponds exactly with the periods 2K and 4iK of dn(u,k) if we notice that z=πu/2K. Moreover the poles of the function θ4(z,q) coincide with the those of dn(u,k). It follows that the function dn(u,k)/F(z) is doubly periodic with no poles and hence is a constant.

Thus we have dn(u,k)=Dθ3(z,q)θ4(z,q) where z=πu/2K and D is a constant independent of u. If we put u=0 and u=K we can evaluate D=k and this justifies the product expansions obtained in last post. We also obtain the value of k in the form of theta functions as k=θ24(0,q)θ23(0,q) It is customary to denote θi(0,q) by θi(q) or even by θi when there is no confusion. Thus we get k=θ24/θ23

Theta Functions θ1 and θ2

We now focus on obtaining the product expansions of sn(u,k) and cn(u,k). It is clear that the poles of the sn(u,k) match with those of the zeroes of θ4(z,q) (noting that z=πu/2K). Hence we expect that there could be a relation of the form sn(u,k)=g(z)θ4(z,q) Here we also expect g(z) to be periodic with period 2π and the zeroes of g(z) should match that of sn(u,k). This means that the functions g(z) should vanish at z=0(modπ,πτ). This latter property can be achieved by setting g(z)=θ4(z+12πτ) But this will not make g(z) periodic with 2π. Moreover we have another problem to solve, namely to make the ratio g(z)/θ4(z,q) doubly periodic with the imaginary period πτ. This would require us to have the periodicity factor g(z+πτ)/g(z) to match the periodicity factor θ4(z+πτ,q)/θ4(z,q).

The periodicity factor for g(z) can be found by noting that g(z+πτ)=θ4(z+12πτ+πτ)=q1e2i(z+πτ/2)θ4(z+12πτ)=q1e2izeπiτg(z) It is now easy to see that we can get our job done if we change the definition of g(z) to g(z)=Aeizθ4(z+12πτ) where A is a constant to be determined later. The above function g(z) is better known historically as θ1(z,q). Thus we have θ1(z,q)=Aeizθ4(z+12πτ) It is then clear that θ1(z,q) is periodic with period 2π and the periodicity factor θ1(z+πτ,q)/θ1(z,q)=q1e2iz. Therefore the function F(z)=θ1(z,q)/θ4(z,q) is doubly periodic with periods 2π,πτ and has poles at z=0(modπ,πτ). It thus follows that the functions sn(u,k)/F(z) is doubly periodic entire function and thus is a constant. We have now established that
sn(u,k)=Sθ1(z,q)θ4(z,q) for some constant S depending upon A and k.

Similarly we can handle the elliptic function cn(u,k). Clearly we can take cn(u,k)=g(z)/θ4(z,q) where g(z) is having a period 2π and the periodicity factor should be such that the function g(z)/θ4(z,q) is periodic with period 2πτ. Hence we must have the periodicity factor of g(z) the same as that of θ4(z,q) but opposite in sign. Both these goals can be achieved by setting g(z)=θ1(z+π/2,q).

Thus we come to our last theta function θ2 defined by θ2(z,q)=θ1(z+π2,q) And like before, we have the relation cn(u,k)=Cθ2(z,q)θ4(z,q) where C is a constant dependent on A and k.

It now remains to evaluate the constants A,C,S. To begin with we have θ1(z,q)=Aeizθ4(z+12πτ)=Aeizn=(1)nqn2e2ni(z+πτ/2)=Aeizn=(1)nqn2qne2niz=Aq1/4n=(1)nq(n+1/2)2e(2n+1)iz=2Aiq1/4n=0(1)nq(n+1/2)2sin(2n+1)z A natural choice for the constant A is A=iq1/4 and then θ1(z,q)=2n=0(1)nq(n+1/2)2sin(2n+1)z Also it is worth noticing the infinite product expansion of θ1(z,q). We have θ1(z,q)=iq1/4θ4(z+12πτ)eiz=iq1/4Gn=1(1q2n2e2iz)(1q2ne2iz)eiz=2Gq1/4sinzn=1(1q2ne2iz)(1q2ne2iz) Thus we finally have θ1(z,q)=2n=0(1)nq(n+1/2)2sin(2n+1)z=2Gq1/4sinzn=1(12q2ncos2z+q4n)  and since θ2(z,q)=θ1(z+π/2,q) it follows that θ2(z,q)=2n=0q(n+1/2)2cos(2n+1)z=2Gq1/4coszn=1(1+2q2ncos2z+q4n) Again note that since θ1(z,q)=iq1/4eizθ4(z+12πτ) we have θ4(z+12πτ,q)=iq1/4eizθ1(z,q) Replacing z with z+πτ/2 in (A) we get θ1(z+12πτ,q)=iq3/4eizθ4(z+πτ)θ1(z+12πτ,q)=iq1/4eizθ4(z,q) Dividing (C) by (B) we get θ1(z+12πτ,q)θ4(z+12πτ,q)=θ4(z,q)θ1(z,q) This is an identity which will be used in the evaluation of the constant S and C. To evaluate them we just need to put u=K in the formula for sn(u,k) to get 1=Sθ1(π2,q)θ4(π2,q)=Sθ2(0,q)θ3(0,q)S=θ3(0,q)θ2(0,q)=θ3θ2 Again putting u=K+iK in the formula for sn(u,k) we get 1k=Sθ1(π2+πτ2,q)θ4(π2+πτ2,q)=Sθ4(π2,q)θ1(π2,q)=Sθ3(0,q)θ2(0,q)=S2 and thus we get S=k1/2,k=θ22θ23 To evaluate C we put u=0 in the formula for cn(u,k) to get 1=Cθ2(0,q)θ4(0,q)C=θ4θ2=θ4/θ3θ2/θ3=kk

Summary

To summarize we have the following results θ1(z,q)=2n=0(1)nq(n+1/2)2sin(2n+1)z=2Gq1/4sinzn=1(12q2ncos2z+q4n)θ2(z,q)=2n=0q(n+1/2)2cos(2n+1)z=2Gq1/4coszn=1(1+2q2ncos2z+q4n)θ3(z,q)=Gn=1(1+2q2n1cos2z+q4n2)=n=qn2e2niz=1+2n=1qn2cos(2nz)θ4(z,q)=Gn=1(12q2n1cos2z+q4n2)=n=(1)nqn2e2niz=1+2n=1(1)nqn2cos(2nz)sn(u,k)=θ3θ2θ1(z,q)θ4(z,q)=2q1/4ksin(πu2K)n=112q2ncos(πuK)+q4n12q2n1cos(πuK)+q4n2cn(u,k)=θ4θ2θ2(z,q)θ4(z,q)=2q1/4kkcos(πu2K)n=11+2q2ncos(πuK)+q4n12q2n1cos(πuK)+q4n2dn(u,k)=θ4θ3θ3(z,q)θ4(z,q)=kn=11+2q2n1cos(πuK)+q4n212q2n1cos(πuK)+q4n2 where z=πu2K,q=exp(πKK)k=θ22θ23=4qn=1(1+q2n1+q2n1)4k=θ24θ23=n=1(1q2n11+q2n1)4 and since we have k2+k2=1 we get the fundamental identity θ42+θ44=θ43 To complete the story we have yet to evaluate the constant G and this will be done in the next post.

Print/PDF Version

0 comments :: Elliptic Functions: Genesis of Theta Functions

Post a Comment