I have evaluated the following integration using Mathematica. I obtained a solution in terms of Meijer G function. I wonder if it can be simplified to be in terms of Bessel functions.

  x Exp[-θ x^2/(
     4 t) ] (BesselJ[0, λ x] BesselY[0, λ] - 
     BesselY[0, λ x] BesselJ[0, λ]), {x, 0, 
(2 t (E^(-((t λ^2)/θ)) BesselY[0, λ] - 
   BesselJ[0, λ] MeijerG[{{0}, {-(1/2)}}, {{0, 
       0}, {-(1/2)}}, (t λ^2)/θ]))/θ
  • $\begingroup$ Have you tried using FunctionExpand[] on it? $\endgroup$ Jan 12 '21 at 0:02
  • $\begingroup$ Yes. I have. But it doesn't simplify it. $\endgroup$ Jan 12 '21 at 0:10
  • $\begingroup$ Then it looks like Mathematica is unaware if the $G$-function has a further simplification, if it does have one. $\endgroup$ Jan 12 '21 at 0:17
  • $\begingroup$ What constraints/assumptions, if any, exist for parameters theta, lambda, and t? $\endgroup$
    – Bob Hanlon
    Jan 12 '21 at 1:19

I think that your Meijer G function is equal to $\frac{e^{-z}\operatorname{Ei}(z)}{\pi}$. I cannot prove it except plotting their difference to ridiculous precision, and referring to this similar-looking identity on the Wolfram Functions site:

g[z_] = MeijerG[{{0}, {-(1/2)}}, {{0, 0}, {-(1/2)}}, z];
h[z_] = E^(-z) * ExpIntegralEi[z] / π;

Plot[g[z] - h[z], {z, 0, 2}, WorkingPrecision -> 10^4]

enter image description here

  • 1
    $\begingroup$ Evaluating DifferentialRootReduce[MeijerG[{{0}, {-1/2}}, {{0, 0}, {-1/2}}, z] - Exp[-z] ExpIntegralEi[z]/Pi, z] yields a 4th-order ODE with all initial conditions equal to 0, so... $\endgroup$ Jan 12 '21 at 9:17
  • $\begingroup$ Thanks for your response. I reached to the same result by substituting numerical values for theta, lamda, and t. Thanks again. @Roman $\endgroup$ Jan 12 '21 at 15:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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