Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

Why is $\sqrt{2\sqrt{2\sqrt{2\sqrt{2\sqrt{2\sqrt{2\cdots}}}}}}$ equal to 2? Does this work for other numbers?

share|cite|improve this question
By any chance did you mean ${\sqrt 2}^{{{\sqrt 2}^{\sqrt 2}}^{\dots}} = 2$? If so, look at Or maybe you meant $\sqrt{2+\sqrt{2+\sqrt{2+\dots}}}$? If so, look at – sdcvvc Jun 28 '13 at 13:11
With the reformulated question you have $x^2=2x$ on the assumption of convergence. – Mark Bennet Jun 28 '13 at 13:46
I cleared up some comments from before the question was edited into present form. That should include the comment of @sdcvvc, but said comment includes some useful links to related, but not identical, questions. – Willie Wong Jun 28 '13 at 13:59
The dots should be on the left side of the expression... in fact one starts the operation from the inner part – Emanuele Paolini Jun 28 '13 at 14:03
I personally have great difficulty with such "infinite expressions" because of the inherent ambiguity. Although I can understand the remark by @EmanuelePaolini that expression must be evaluated from the inside outward, that would not be possible for the variation $\sqrt {1\sqrt {2\sqrt {3\sqrt {4\sqrt{5\sqrt\cdots}}}}}$ which nevertheless might be given some meaning. My real problem with the question is the final "$2$"; as there are infinitely many operations around in, one never gets to the point of inserting it. I would say, if you mean to take a limit, write a limit. – Marc van Leeuwen Jun 28 '13 at 14:20
up vote 35 down vote accepted

The number in question is simply

$$2^{1/2+1/4+1/8+\cdots} = 2^1 = 2$$

Yes, this works for other numbers. More interesting is if the numbers are not equal inside the radicals. For example, say you have a positive sequence element $a_n$ inside the $n$th radical. If we assume the expression converges to some value $P$, then

$$\log{P} = \sum_{k=1}^{\infty} \frac{\log{a_n}}{2^n}$$

share|cite|improve this answer
Perhaps it is more precise to say that the number in question is the limit of the sequence $a_k = 2^{1/2+1/4+\cdots+1/2^k}$ for $k\geq 1$. – Álvaro Lozano-Robledo Jun 28 '13 at 13:57
@ÁlvaroLozano-Robledo: perhaps. But I will sleep well with what I have posted. – Ron Gordon Jun 28 '13 at 14:00

A general way to ascribe a value to this non rigorously defined formula and to many others similar to it, is to consider that they describe the "infinite" iteration of a given function. Thus, one considers a sequence $(x_n)$ such that $x_{n+1}=\sqrt{2x_n}$ for every $n\geqslant0$ and $x_0\geqslant0$. Then, indeed the value of the formula is $2$ in the following sense:

For every $x_0\gt0$, $x_n\to2$.

To see this, note that $x_{n+1}=u(x_n)$, where the function $u:x\mapsto\sqrt{2x}$ is such that $x\lt u(x)\lt2$ for every $x$ in $(0,2)$, $u(2)=2$, and $2\lt u(x)\lt x$ if $x\gt2$. Thus $(x_n)$ is increasing if $0\lt x_0\lt2$ and decreasing if $x_0\gt2$ and converges to $2$ in both cases (and I will let you solve the case $x_0=2$).

Does this work for other numbers?

Indeed, if one means by this the fact that:

$\sqrt{a\sqrt{a\sqrt{a\sqrt{a\sqrt{a\sqrt{\cdots}}}}}}=a$, for every $a\geqslant0$.

To see this, consider $a\gt0$ and $x_{n+1}=u_a(x_n)$, where $u_a$ is the function $u_a:x\mapsto\sqrt{ax}$. Note that $x\lt u_a(x)\lt a$ for every $x$ in $(0,a)$, $u_a(a)=a$, and $a\lt u_a(x)\lt x$ if $x\gt a$. Thus $(x_n)$ is increasing if $0\lt x_0\lt a$ and decreasing if $x_0\gt a$ and converges to $a$ in both cases (and I will let you solve the case $x_0=a$).

share|cite|improve this answer
For completeness, it shouldn't be too hard to answer OP's follow up question ("Does this work for other numbers?") as well. – Lord_Farin Jun 28 '13 at 13:56
@Lord_Farin Done. Thanks. – Did Jun 28 '13 at 14:04
Yes! A downvote... Halelujah! – Did Jun 28 '13 at 14:07

Let $x=\sqrt{2\sqrt{2\sqrt{2\sqrt{2\sqrt{2\sqrt{\cdots2}}}}}}$ then


so $x(x-2)=0$ since $x\neq 0$


share|cite|improve this answer
You should prove the convergence first. – eccstartup Jun 28 '13 at 13:53

In the case of $\sqrt{2+\sqrt{2+\sqrt{2+\cdots}}}$, this is a limit of the short-chord of a polygon $2^n \cdot x$ as $n$ goes large. The short-chord is the chord of a regular polygon that spans two edges. There is a triangle formed by two edges and the short-chord. As the polygon gets more sides, this triangle approaches a straight line, and the short-chord approaches two edges in length.

share|cite|improve this answer

$$k=\sqrt{a\sqrt{a\sqrt{a\cdots}}}$$ $$k^2/a=k$$



share|cite|improve this answer

Your Answer


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.