I'm trying to solve this function

$$z^2 + 2z + 2\exp(-z) = 0$$

In one solution I found out the exponential function has been approximated to $\exp(-z)=1-z$, which makes it and easy polynomial to solve.

But is it the complete solution of this equation? Or are there better methods to solve this equation?

  • $\begingroup$ Of course replacing $\exp(-z)$ with $1-z$ is only an approximation. Valid for $z$ near $0$. So you only expect an approximate solution to the original equation. And for solutions not close to $0$, this will not work at all. $\endgroup$
    – GEdgar
    Jul 19, 2017 at 17:59
  • $\begingroup$ google.com/… $\endgroup$
    – Bumblebee
    Jul 19, 2017 at 18:00
  • $\begingroup$ set $$z=x+iy$$ to solve your equation $\endgroup$ Jul 19, 2017 at 18:04
  • $\begingroup$ @Dr.SonnhardGraubner Huh? How in the world is this useful? It hardly does anything to simplify the problem. $\endgroup$ Jul 19, 2017 at 18:47
  • $\begingroup$ hm ok i will post it $\endgroup$ Jul 19, 2017 at 18:50

2 Answers 2


HINT: you will have $$(x+iy)^2+2(x+iy)+2e^{-x}(\cos(y)+i\sin(y))=0$$ and you must solve the System $$x^2-y^2+2x+2e^{-x}\cos(y)=0$$ $$2xy+2y+2e^{-x}\sin(y)=0$$ by a numerical method we obtain $$x \approx -3.293017589, y \approx 6.912667999$$

  • $\begingroup$ This seems to be wrong. $\endgroup$ Jul 20, 2017 at 9:41

HINT: Multiply both side of your equation $$ z^2+2z+2e^{-z}=0 $$ by $e^z$, from which it turns into $$ (e^{z/2}z)^2+4ze^{z/2}+4=-2(ze^z-2ze^{z/2}-1) $$ i.e. $$ \left(ze^{z/2}+2\right)^2=-2(ze^z-2ze^{z/2}-1) $$ and starting from this, you can try to manipulate RHS in order to get something useful.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .