Using the chiral representation of the gamma matrices, Peskin and Schroeder arrive in some expressions for the 4-component spinors $u(p)$ and $v(p)$ in terms of a square root of the Pauli matrices marices, pages 45-48. Doing some research online, I found out that we can express these square roots as

$$ \sqrt{p.\sigma} \equiv \frac{E_p+m-{\bf \sigma}.\bf{p}}{\sqrt{2(E_p+m)}} $$ and $$ \sqrt{p.\bar{\sigma}} \equiv \frac{E_p+m+{\bf \sigma}.\bf{p}}{\sqrt{2(E_p+m)}}. $$

But I could not prove this fact.

  • $\begingroup$ Is the $p$ on the left different from the $\mathbf{p}$ on the right? If so, how is the $0$th sigma matrix defined? $\endgroup$ – G. Smith Jun 8 at 21:14
  • $\begingroup$ On the left the $p$ is a four-vector, on the right it is the three momentum. The 0th sigma is the 2by2 identity matrix. $\endgroup$ – Slayer147 Jun 8 at 21:41

Here's the proof of the first identity; the second is similar.

To show that

$$\sqrt{p \cdot \sigma} = \frac{E + m - {\vec{p} \cdot \vec{\sigma}}}{\sqrt{2(E + m)}},$$

square both sides and multiply through by the denominator on the right side. This gives

$$2(E + m)(p \cdot \sigma) = (E + m - {\vec{p} \cdot \vec{\sigma}})^2.$$

Now use

$$p = (E, \vec{p})$$


$$\sigma = (I_2, \vec{\sigma})$$

to get

$$p \cdot \sigma = E - \vec{p} \cdot \vec{\sigma},$$

where, as in the equation we're trying to prove, we don't bother to write the $I_2$.

Substituting this gives

$$2(E + m)(E - \vec{p} \cdot \vec{\sigma}) = (E + m - {\vec{p} \cdot \vec{\sigma}})^2.$$

Expand this to get

$$2(E + m)E - 2(E + m)\vec{p} \cdot \vec{\sigma} = (E + m)^2 - 2(E + m)\vec{p} \cdot \vec{\sigma} + (\vec{p} \cdot \vec{\sigma})^2.$$

The middle terms on each side cancel, giving

$$2(E + m)E = (E + m)^2 + (\vec{p} \cdot \vec{\sigma})^2.$$

This simplifies to

$$E^2 = m^2 + (\vec{p} \cdot \vec{\sigma})^2.$$


$$(\vec{p} \cdot \vec{\sigma})^2 = p_i p_j \sigma_i \sigma_j = \frac{1}{2} p_i p_j \{\sigma_i, \sigma_j\} =\frac{1}{2} p_i p_j (2\delta_{ij}) = \vec{p}^2$$

so the original equation is equivalent to

$$E^2 = m^2 + \vec{p}^2,$$

which is simply the relationship between energy and momentum.


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.