Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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


  1. $x,y,z\geq0$

  2. $xy+yz+zx=1$

Prove that $\displaystyle \frac 1 {x+y}+\frac 1 {y+z}+\frac 1 {z+x}\geq \frac 5 2$.

I tried using Cauchy's inequality LHS $\geq\frac 9 {2a+2b+2c}$, but failed. Please give me some ideas. Thank you.

share|cite|improve this question
Did you try to use Lagrange multipliers ? Some context may be usefull – Belgi Mar 3 '13 at 13:17
I haven't heard of Lagrange multipliers... – ᴊ ᴀ s ᴏ ɴ Mar 3 '13 at 13:18
Did you try expanding, assuming, WLOG x<y<z and using rearrangement inequality? It should work. – user45099 Mar 3 '13 at 13:22
I tried, but nothing happens. – ᴊ ᴀ s ᴏ ɴ Mar 3 '13 at 13:30
up vote 1 down vote accepted

assume $z$ is the smallest. if we fix the sum $x+y$,

$\dfrac{1}{x+y} + \dfrac{1}{z+x}+\dfrac{1}{z+y} = \dfrac{1}{x+y} +\dfrac{x+y+2z}{1+z^2}$, take it as a function of $z$, you may see, it is increasing.[you may differentiate it or just plugin $z=0$ and compare]

Thus $z=0$, it will be smallest.

The rest is easy.


After we get $z=0$, the restriction on $x,y,z$ turns out to be $xy=1$, and the goal is to minimize


Since now $x=\dfrac{1}{y}$, thus

$\dfrac{1}{x+y} +x+y = \dfrac{1}{x+\dfrac{1}{x}}+x+\dfrac{1}{x}$. Take $t = x+\dfrac{1}{x}$, we know that $t\ge 2$. then the objective function is to minimize


for $t\ge 2$.

You can take a derivative to see that $t+1/t$ is increasing when $t\ge 1$. So minimum will be obtained at $t = 2$. which is $x=y=1$.

share|cite|improve this answer
When you fix the sum $x+y$, $z$ can only take one value based on the restriction $xy+yz+zx$=1; is it meaningful to take the expression as a function of $z$? – Vincent Tjeng Mar 3 '13 at 15:59
when I fix $x+y$, $z$ still can change. I only fix the sum, the product can change. – Yimin Mar 3 '13 at 16:00
@VincentTjeng even you don't fix the sum, you can also see when $z=0$,by taking partial derivative w.r.t $z$, the function will obtain the minimum. – Yimin Mar 3 '13 at 16:02
@Yimin: What are the rest? Minimum of $x+y+1/(x+y)$ is 2, not 5/2 ... – ᴊ ᴀ s ᴏ ɴ Mar 5 '13 at 14:25
@jasoncube, you have $z=0$, then there is a restriction on $xy=1$. – Yimin Mar 5 '13 at 14:32

Well,I think you can use the famouse Iran 96 inequality as a result to solve this problem. the Iran 96 inequality

Let $x,y,z\geq 0$ we have

$$ \frac{1}{(x+y)^{2}}+\frac{1}{(y+z)^{2}}+\frac{1}{(x+z)^{2}}\geq \frac{9}{4(xy+yz+zx)} $$ square both side,we can rewrite the inequality into $$ \sum_{cyc}{\frac{1}{(x+y)^{2}}}+2\sum_{cyc}{\frac{1}{(x+y)(x+z)}}\geq \frac{25}{4}$$ Now,Using this inequality as a known result,it's suffice to prove $$ \sum_{cyc}{\frac{1}{(x+y)(x+z)}}\geq 2 $$ after simple homogenous,it's $$ (xy+yz+xz)(x+y+z)\geq (x+y)(y+z)(z+x) $$ Or $$ xyz\geq 0 $$ Which is obviously true,Equality occurs if and only if $x=y=1, z=0 $ or it's permutation. The proof is complete

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.