2
$\begingroup$

1) Proof that $z=i^i$ is a real number

Euler's identity;

$$e^{i\pi} + 1 = 0$$

can be manipulated in order to obtain the result:

$$e^{i\pi} = -1$$

Raising both sides of the equality to the power of $i$ gives, after simplification:

$$e^{-\pi} = i^{2i}$$

Square rooting both sides gives:

$$e^{-\frac{\pi}{2}} = i^{i} \approx 0.2078 $$

2) The formula for the infinite tetration of some $z$

For the general case of some base $z$ , using the Lambert W function, one can find that, if:

Let $c = z^{z^{z^{...}}} = z^{1/z} $, '$...$' denoting the iterated exponentiation (tetration) of $z$.

Then:

$$c = -\frac{W(-\ln(z))}{\ln(z)}$$


3) Evaluation for $z=i^i$ as shown in 1)

Now let in 2) the value $z = i^{i} = e^{-\frac{\pi}{2}} \approx 0.2078 $ as found in 1), then it can be computed that the infinite tetration of $z=i^{i}$ looking like $$ c = (z)^{(z)^{(z)^\ldots}} = (i^i)^{(i^i)^{(i^i)^\ldots}} $$ is equal to:

$$c=\frac{2W(\frac{\pi}{2})}{\pi} \approx 0.4745409995$$

Is this correct?!

$\endgroup$
  • 2
    $\begingroup$ Exponentiation is not commutative, so we should not expect the infinite tetrations of $i$ and $i^i$ to agree, i.e., that $$i^{i^{i^{i^{\cdots}}}} = (i^i)^{(i^i)^{\cdots}}.$$ Indeed, the number $\alpha := \frac{2}{\pi} W \left(\frac{\pi}{2}\right)$ does not satisfy $i^\alpha = \alpha$ for any choice of branch. One does find a solution to this equation (for a particular choice of branch), however, at $\approx 0.438 + 0.361 i$. $\endgroup$ – Travis Jan 22 '16 at 14:48
  • $\begingroup$ Is your question about finding value of $i^i$ $\endgroup$ – Archis Welankar Jan 22 '16 at 15:26
  • $\begingroup$ you are missing the "i"; $W(-\frac{\pi i}{2})$ should work, but it is difficult to work with the lambert W function in the complex plane. I get $-W(-\frac{\pi i}{2})/\frac{\pi i}{2}\approx 0.4383+0.3606i$ $\endgroup$ – Sheldon L Jan 24 '16 at 18:18
  • $\begingroup$ @ArchisWelankar His question is on $$i^{i^{i^{\dots}}}$$ $\endgroup$ – Simply Beautiful Art Jan 25 '16 at 23:56
  • $\begingroup$ The subject-line is about tetration with base $z=i$, and the last question is about tetration with base $z = i^i \approx 0.2078... $. It's difficult to decide how to help ... $\endgroup$ – Gottfried Helms Mar 5 '17 at 14:22
1
$\begingroup$

You made a small mistake:

$$c=z^{z^{z^{\cdots}}}=i^{i^{i^{\dots}}}$$

Hence:

$$z=i\ne i^i$$

$$i=e^{i\pi/2}$$

So, reevaluating, manipulating your formula with a Lambert W function identity:

$$c=e^{-W(-\ln(e^{i\pi/2}))}=e^{-W(-i\pi/2)}=\frac{W(-i\pi/2)}{-i\pi/2}$$

This should then evaluate to the actual answer.

$\endgroup$

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.