3

When deriving the Lagrangian for Spin $\frac{1}{2}$ particles we are naturally led to using $\Psi$ and $\bar{\Psi}$. The Euler-Lagrange equations lead us to two wave equations: \begin{equation} (i\gamma_\mu \partial^\mu - m ) \Psi =0 \end{equation}

\begin{equation} (i \gamma_\mu \partial^\mu + m )\bar{\Psi} =0 \end{equation} which differ by a sign in front of the mass term. The same thing happens if we look at the electromagetic coupling of these $\frac{1}{2}$ fields. Again their coupling is different by a sign. This is interpreted as particle and anti-particle having opposite charge. Nevertheless it is unconventional to speak of the anti-particle having negative mass. Why is this the case?

John Rennie
  • 355,118
jak
  • 9,937
  • 4
  • 35
  • 106
  • 2
    More on negative mass: http://physics.stackexchange.com/q/44934/ and http://physics.stackexchange.com/q/34115/ – jinawee Jan 27 '14 at 18:06

1 Answers1

4

The second equation is actually incorrect. It should be written as follows:

$$ i\partial^{\mu}\overline{\Psi}\gamma_{\mu}+m\overline{\Psi}=0. $$ Here, $\overline{\Psi}$ is understood as a 4-component row vector (not in the sense of the vector rep. of the Lorentz group).

At any rate, $\overline{\Psi}$(or $\Psi^{\dagger}$) is not what you obtain from $\Psi$ by exchanging the roles of particles and antiparticles. The result of such operation is $\Psi^{C}\equiv-i\gamma^{2}\Psi^{\ast}$, and it satisfies the same Dirac equation as $\Psi$. Among other things, it means that antiparticles have the same mass as particles.

higgsss
  • 2,832
  • Thanks for you answer, you are of course right, that $\bar \Psi = \Psi^{\dagger} \gamma_0 $ has to stand on the left-hand side of $\gamma_\mu$. Do you have of any tip about interpretation of the second equation, or $\bar \Psi$ or where i can read about it? – jak Jan 28 '14 at 08:15
  • I read about charge conjugation and as far as i understand $\Psi^{C}\equiv-i\gamma^{2}\Psi^{\ast}$ is introduced because it does the job, i.e. changing the sign in front of the coupling to the EM-Field and conventionally this is interpreted as wave-function of the anti-particle. Nevertheless this doesn't seem like something natural to do to me, nor does it prevent us from interpreting $\overline{\Psi}$ as wave function for the anti-particle instead, or am i missing something? $\bar{\Psi}$ has the EM-coupling sign changed, too and additional the mass sign. – jak Jan 28 '14 at 08:39
  • @JakobH Consider the mode expansion $\Psi(x) = \int\frac{d^{3}p}{(2\pi)^{3}\sqrt{2E_{\textbf{p}}}} \sum_{\sigma}\left(a_{\textbf{p}\sigma}u_{\sigma}(p)e^{-ipx}+b_{\textbf{p}\sigma}^{\dagger}v_{\sigma}(p)e^{ipx}\right)$. Then, the natural "antiparticle wave function" is obtained by exchanging particles and antiparticles (i.e., $a$ and $b$) in $\Psi$. It turns out that $\Psi^{C} = -i\gamma^{2}\Psi^{\ast}$ does the job, as you pointed out. – higgsss Jan 28 '14 at 09:00
  • i looked it up, and I find that $\bar \Psi(x) = \int\frac{d^{3}p}{(2\pi)^{3}\sqrt{2E_{\textbf{p}}}} \sum_{\sigma}\left(a_{\textbf{p}\sigma}^{\dagger}u_{\sigma}(p)e^{ipx}+b_{\textbf{p}\sigma‌​}v_{\sigma}(p)e^{-ipx}\right)$ which would mean, that in field theory $\bar \Psi$ creates anti particles, destroying particles as opposite to $\Psi$ – jak Jan 29 '14 at 07:20
  • @JakobH $\overline{\Psi}$ as you defined certainly creates a particle and destroys an antiparticle at the same time, but it isn't something that transforms like a Dirac field under a Lorentz transformation. To make it an object that transforms like a Dirac field, one should also exchange $u$ and $v$ from $\overline{\Psi}$. – higgsss Jan 29 '14 at 21:52
  • If you want something that satisfies the Dirac equation, but with $m$ replaced by $-m$, $\gamma^{5} \Psi$ is what you are looking for. Yet once you do the mode expansion, it will lead to the same $a$ and $b$ ($u$ and $v$ will be multiplied by $\gamma^{5}$) and hence the same mass between particles and antiparticles. – higgsss Jan 29 '14 at 22:04
  • Actually the mass of a particle/antiparticle is what you read off from the Hamiltonian, $H = \int \frac{d^{3}\textbf{p}}{(2\pi)^{3}} \sum_{\sigma} \sqrt{\textbf{p}^{2}+m^{2}}(a_{\textbf{p}\sigma}^{\dagger}a_{\textbf{p}\sigma}+b_{\textbf{p}\sigma}^{\dagger}b_{\textbf{p}\sigma})$. $a_{\textbf{p}\sigma}$ and $b_{\textbf{p}\sigma}$ destroys quanta with the same energy $E_{\textbf{p}}=\sqrt{\textbf{p}^{2}+m^{2}}$, and hence the same mass $E_{\textbf{p}=0}=m$. – higgsss Jan 29 '14 at 22:09