# Homework Help: To prove that vectors (a + b) . (a - b) = a^2 + b^2 iff

1. Jan 10, 2013

### nishantve1

1. The problem statement, all variables and given/known data

The question says :
Prove that
$(\vec{a} + \vec{b}).(\vec{a} - \vec{b}) = \left | \vec{a} \right |^{2} + \left | \vec{b} \right |^{2}$
if and only if $\vec{a} \perp \vec{b}$

2. Relevant equations

This is known that
$\left | \vec{a} + \vec{b} \right | = \left | \vec{a} - \vec{b} \right |$
if $\vec{a} \perp \vec{b}$

3. The attempt at a solution

I tried substituting that and then using Cauchyâ€“Schwarz inequality but somehow I can't open up the absolute brackets .

2. Jan 10, 2013

### Saitama

You can simply use the distributive property.

EDIT: Are you sure that you have copied down the question correctly? I guess the RHS should be $$\left | \vec{a} \right |^{2} - \left | \vec{b} \right |^{2}$$

Last edited: Jan 10, 2013
3. Jan 10, 2013

### nishantve1

@Pranav The question is correct

4. Jan 10, 2013

### Saitama

Substitute $\vec{a}=\hat{i}$ and $\vec{b}=\hat{j}$. The relation doesn't satisfy this.

5. Jan 10, 2013

### jfgobin

That doesn't sound right: take the following example:

$\vec{a}=\left ( 1 , 0\right )$
$\vec{b}=\left ( 0 , -0.1\right )$

$\left ( \vec{a} + \vec{b} \right ) \cdot \left ( \vec{a} - \vec{b} \right ) = 0.99$

but

$\left| \vec{a} \right|^2 + \left| \vec{b} \right|^2 = 1.01$

J.

6. Jan 10, 2013

### HallsofIvy

As several people have told you now, the correct relation is
$$(\vec{a}+ \vec{b})\cdot(\vec{a}- \vec{b})= |\vec{a}|^2- |\vec{b}|^2$$
not what you wrote. You should have realised that from the usual numerical "difference of squares" formula.

As to how to prove it, what "tools" do have to? You write as if you had never seen vectors before but that can't be true if you are expected to do a probolem like this!
How are you working with vectors? In terms of components? Are you allowed to assume two or three dimensions? Or in terms of "length and direction"?

How has the dot product been defined? As "$<a_1, a_2, \cdot\cdot\cdot, a_n>\cdot<b_1, b_2, \cdot\cdot\cdot, b_n>= a_1b_1+ a_2b_2+ \cdot\cdot\cdot+ a_nb_n$? Or as $|\vec{a}||\vec{b}|sin(\theta)$ where $\theta$ is the angle between $\vec{a}|$ and $\vec{b}$?

Either can be used but the proof depends on which you are using.

7. Jan 10, 2013

### D H

Staff Emeritus
This is obviously false. Double check the question. It is probably asking you to prove that $(\vec a + \vec b) \cdot (\vec a + \vec b) = ||\vec a||^2 + ||\vec b||^2$ iff $\vec{a} \perp \vec{b}$.

8. Jan 10, 2013

### Staff: Mentor

But this is true for all a,b.

My guess: Show that
$$(\vec{a}+ \vec{b})\cdot(\vec{a}+ \vec{b})= |\vec{a}|^2+ |\vec{b}|^2$$
is true if and only if $\vec{a} \perp \vec{b}$.

Edit: Same minute :(