3

I want to find the zeros of the solution to this ODE

DSolve[{y''[x] - (I x - 2) y[x] == 0, y[0]==0,y'[0]==1}, y[x],x]

I use

ResourceFunction["Intercepts"]

and get a product of several hypergeometric functions. Are there alternative ways of getting the zeros?`For instance, such as the Bessel zeros, or the sine cosine zeros at $n\pi$ and $n\pi+\frac{\pi}{2}$?

Thanks

J. M.'s missing motivation
  • 124,525
  • 11
  • 401
  • 574
Vangsnes
  • 591
  • 2
  • 9
  • 1
    You could use something like the the technique in this answer to get zeroes within a range. – J. M.'s missing motivation May 28 '22 at 16:48
  • There is AiryAiZero just like BesselJZero, but Mma cannot solve Solve[BesselJ[1, x] == BesselJ[2, x], x, Method -> Reduce] or even Solve[BesselJ[1, x] == 1/10, x, Method -> Reduce], even though it can solve Solve[BesselJ[1, x] == 0 x, Method -> Reduce]. On a bounded interval, it can solve Solve[BesselJ[1, x] == BesselJ[2, x] && 0 < x < 10, x, Method -> Reduce]. – Michael E2 May 28 '22 at 17:16

1 Answers1

6

Maybe something like this, with a bounded domain:

func = y[x] /. 
   First@
    DSolve[{y''[x] - (I x - 2) y[x] == 0, y[0] == 0, y'[0] == 1}, 
     y[x], x];

Solve[func == 0 && -1/10 < Re[x] < 4 && -1 < Im[x] < 2, x] (* {{x -> 0}, {x -> Root[{AiryAi[-(-1)^(2/3) (-2 + I #1)] AiryBi[2 (-1)^(2/3)] - AiryAi[2 (-1)^(2/3)] AiryBi[-(-1)^(2/3) (-2 + I #1)] &, 1.9742593 + 0.4276746 I}]}, {x -> Root[{AiryAi[-(-1)^(2/3) (-2 + I #1)] AiryBi[2 (-1)^(2/3)] - AiryAi[2 (-1)^(2/3)] AiryBi[-(-1)^(2/3) (-2 + I #1)] &, 3.4529450 + 1.0382391 I}]}} *)

Root objects are an exact representation of algebraic or, in this case, transcendental roots. They are numeric and may be calculated to arbitrary precision.

Michael E2
  • 235,386
  • 17
  • 334
  • 747