1

If a quantum state can be represented as

$$|\psi\rangle=\alpha|0\rangle+\beta|1\rangle$$

Then

Because of $|\alpha|^2+|\beta|^2=1$, we may rewrite Equation (1.1) as $$|\psi\rangle=e^{i\gamma}\left(\cos\frac{\theta}2|0\rangle+e^{i\varphi}\sin\frac{\theta}2|1\rangle\right)$$ where $\theta$, $\varphi$ and $\gamma$ are real numbers.

My question is why can we rewrite it like this? I understand that the new coefficients obey the normalisation requirement of $$|\alpha|^2 + |\beta|^2 = 1$$ but why can we be confident that this rewritten version covers all the possible values of and $\alpha$ and $\beta$?

Adam Zalcman
  • 4,884

1 Answers1

0

The key observation is that cosine and sine are bijections, i.e. one-to-one functions, between the set $[0,1]$, where $|\alpha|$ and $|\beta|$ reside, and the set $[0,\frac{\pi}{2}]$, where $\frac{\theta}{2}$ resides.

Bijection argument

First, write $\alpha$ and $\beta$ in polar form

$$ \alpha=re^{ia}\quad\beta=se^{ib}\tag1 $$

with $r,s\in[0,1]$ and $a,b\in[0,2\pi)$ and note that $(1.3)$ imposes no constraints on the phase angles $a$ and $b$. In fact, we can just set $\gamma:=a$ and $\varphi:=b-a$. However, equation $(1.3)$ does constrain the absolute values $|\alpha|=r$ and $|\beta|=s$ by prescribing that

$$ r=\cos\frac{\theta}{2}\quad s=\sin\frac{\theta}{2}\tag2 $$

for some $\theta\in[0,\pi]$. Now, cosine is a bijection between $[0,\frac{\pi}{2}]$ and $[0,1]$, so for any $r\in[0,1]$ there exists a unique $\theta\in[0,\pi]$ such that $r=\cos\frac{\theta}{2}$. Then $s\in[0,1]$ is uniquely determined from $s=\sqrt{1-r^2}$. Moreover,

$$ s=\sqrt{1-r^2}=\sqrt{1-\cos^2\frac{\theta}{2}}=\sin\frac{\theta}{2}\tag3 $$

as expected.

Geometric argument

The above argument can be given geometric interpretation by treating the pair $(r, s)$ as coordinates of a point in the first quadrant of a Cartesian coordinate plane. The points that satisfy the requirement $$r^2+s^2=|\alpha|^2+|\beta|^2=1\tag4$$ lie on the unit circle with the center at the origin. Instead of identifying a point using $(r,s)$ coordinates, we can identify it by specifying the polar angle $\phi\in[0,\frac{\pi}{2}]$ which is related to $(r,s)$ by

$$ r=\cos\phi\quad s=\sin\phi.\tag5 $$

By setting $\theta:=2\phi$ we recover $(2)$ .

Adam Zalcman
  • 4,884