Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I was wondering if there is a known closed form solution for the zeros of the spherical Bessel functions. While doing a quantum assignment, I came across them as a solution for the spherical infinite potential well. However, I only read about them as just a sequence of numbers, but no generating function or closed form expression. Any suggestions would be great.

Thanks,

share|cite|improve this question
    
Have you already had a look at Watson's Treatise (cfr. math.stackexchange.com/questions/98885/…)? – Pacciu Feb 3 '12 at 1:01
    
@Pacciu Thanks, I just picked up a copy, and made a list of some books from that thread ;) – kηives Feb 3 '12 at 14:07
up vote 4 down vote accepted

For $n=-1,0$, finding the roots of the spherical Bessel functions $j_n(x)$ and $y_n(x)$ is somewhat easy, since:

$$\begin{array}{ll} j_{-1}(x)&=&\frac{\cos\,x}{x}&\quad&y_{-1}(x)&=&\mathrm{sinc}(x)\\ j_0(x)&=&\mathrm{sinc}(x)&\quad&y_0(x)&=&-\frac{\cos\,x}{x}\\ \end{array}$$

where $\mathrm{sinc}(x)=\dfrac{\sin\,x}{x}$ is the sine cardinal. Solving for zeros of other orders results in rather complicated transcendental equations, which I doubt have closed-form solutions. However, you will want to see these DLMF entries for some more information that can help you in numerically determining the zeros (e.g., asymptotic expansions); approximations derived from formulae there can then be subsequently polished with Newton-Raphson or some other iterative method of choice.

share|cite|improve this answer

Cheat Code: spherical Bessel roots are their respective half-integer Bessel roots

According to Abramowitz, 1964, Ch9, pp 440,"Zeros and Their Asymptotic Expansions"

The zeros of $j_{n}(x)$ and $y_{n}(x)$ [i.e.: Spherical Bessel] are the same as $J_{n+1/2}(x)$ and $N_{n+1/2}(x) $ [i.e.: 'Normal' Bessel]

Mathematica Implementation

I used, for the spherical Bessel of the first kind:

nRoots = 4;
nBessel = 3;
SphBesselRoot[l_, k_] := N[BesselJZero[l + 1/2, k]];
Grid[Table[SphBesselRoot[l, i], {l, 0, nBessel}, {i, 1, nRoots}]]

This prints a table of the first four roots (columns) for the four spherical Bessel functions (rows) :

3.14159 6.28319 9.42478 12.5664
4.49341 7.72525 10.9041 14.0662
5.76346 9.09501 12.3229 15.5146
6.98793 10.4171 13.698  16.9236

Comparing to tables of the roots online, this table is in good agreement.

I found this method easier than root finding or asymptotic expressions.

share|cite|improve this answer
1  
...and how exactly did you think were these computed internally by Mathematica, by the swish of a unicorn's tail? – J. M. Dec 10 '15 at 3:11
    
Nope, I didn't have enough money to add unicorns to my computer! I posted this because I could find lots of tables, but few generating functions / ways to compute the tables easily and quickly. – heisenBug Feb 19 at 23:35

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.