In the Standard Model, electroweak unification is based on the symmetry breaking of $ \text{SU(2)}\times \text{U(1)}_Y \rightarrow \text{U(1)}_{EM}$ by the VEV of a complex doublet $H$ with hypercharge $\frac{1}{2}$.

The Lagrangian is

$$\mathcal{L}=-\frac{1}{4}(W^a_{\mu\nu})^2-\frac{1}{4}B^2_{\mu\nu}+(D_\mu H)^\dagger(D_\mu H)+m^2H^\dagger H-\lambda(H^\dagger H)^2,$$

where $W^a_\mu$ are the $\text{SU(2)}$ gauge bosons and $W^a_{\mu\nu}$ are given by $W^a_{\mu\nu}=\partial_\mu W^a_\nu-\partial_\nu W^a_\mu+gf^{abc}W^b_\mu W^c_\nu$, and $B_\mu$ is the $\text{U(1)}$ gauge bosons and $B_{\mu \nu}=\partial_\mu B_\nu-\partial_\nu B_\mu$.

The covariant derivative is $$D_\mu H=\partial_{\mu}H-igW^a_\mu \tau^aH-\frac{1}{2}ig'B_\mu H.$$

After $H$ gets its VEV $$H_0=\begin{pmatrix} 0 \\ \frac{v}{\sqrt{2}} \end{pmatrix},$$ the mass term of the gauge bosons comes from $$(D_\mu H_0)^\dagger(D_\mu H_0)=\frac{v^2}{8}\left[ g^2(W^1_\mu)^2+g^2(W^2_\mu)^2+g'^2(g'B_\mu-gW^3_\mu)^2 \right].$$

My questions are why we define W bosons as $W^\pm_\mu=\frac{1}{\sqrt{2}}(W^1_\mu\mp iW^2_\mu)$ rather than just $W^1_\mu$ and $W^2_\mu$ respectively, and how to see the charges of $W^\pm$ and $Z$ bosons?

  • 3
    $\begingroup$ Look not at the Higgs sector but on the gauge field self-interaction. It will give you interaction between photon and $W^\pm$ $\endgroup$
    – OON
    Dec 1, 2016 at 14:22
  • $\begingroup$ @ OON The gauge field self-interaction comes from the term $-\frac{1}{4}(W^a_{\mu\nu})^2$, but I expand this term and have the interaction of three W bosons looking like $gWW\partial W$ and that of four W bosons looking like $g^2 WWWW$. I don't know how to get the interaction between photon and $W^\pm$ $\endgroup$ Dec 2, 2016 at 1:42
  • $\begingroup$ Electromagnetic field is not an original $B$-field but superposition of $B_\mu$ and $W_\mu^3$. Here it will appear from $W_\mu^3=\cos\theta_W Z_\mu + \sin\theta_W A_\mu$ where $\tan\theta_W=g_2/g_1$. $\endgroup$
    – OON
    Dec 2, 2016 at 6:04
  • $\begingroup$ @OON So should I transform $W^a_\mu , B_\mu$ basis to $W^\pm_\mu , A_\mu , Z_\mu$ basis and then expand the term $−\frac{1}{4}(W^a_{\mu \nu})^2−\frac{1}{4}(B_{\mu \nu})^2$ and find the interaction term looking like $AW^\pm$ ? But there is still no two interaction. If not, which term I should looking for? $\endgroup$ Dec 2, 2016 at 9:51
  • 1
    $\begingroup$ Ok, write thoroughly the $\frac{1}{4}(W_{\mu\mu}^a)^2$ through components $W_\mu^\pm$, $Z_\mu$ and $A_\mu$. Then omit for simplicity all terms involving $Z$ and $W^\pm$ self-interaction. Then compare it with the following - if we want $W^\pm$ to be electrically charged its kinetic term should take the form $\vert D_\mu W_\nu^\pm-D_\nu W_\mu^\pm\vert^2$ where $D_\mu=i\partial_\mu\mp eA_\mu$. $\endgroup$
    – OON
    Dec 2, 2016 at 10:45

1 Answer 1


The only thing You need is the explicit form of the electric charge generator $Q$, the only one "unbroken" by the Higgs VEV: $$ Q = t_{3} + \frac{Y}{2}, $$ where $t_{3}$ is the weak isospin and $Y$ is the weak hypercharge.

For $SU_{\text{L}}(2)$ bosons, namely, $W_{1},W_{2},W_{3}$, the weak hypercharge $Y$ is zero by the definition (note also that the hypercharge of $U_{Y}(1)$ gauge field, which is $B \sim g_{1}Z - g_{2}A$).

In order to define the isospin $t_{3}$, we have to calculate the commutator $$ \tag 1 [T_{3},T_{i}] = t_{3}T_{i}, $$ where $T_{3} \equiv \sigma_{3}$, with $\sigma_{3}$ being the Pauli matrix, is the isospin generator, while $T_{i}$ is the $SU_{\text{weak}}(2)$ generator associated with given boson. $(1)$ is the standard charge algebra.

From $W_{1},W_{2}, W_{3}$ bosons you can construct three states with given eigenvalues of isospin: $t_{3} = +1, t_{3} = -1, 0$. These states corresponds to generators $$ \sigma_{+} = \frac{1}{2}\left(\sigma_{x} + i\sigma_{y}\right), \quad \sigma_{-} = \frac{1}{2}\left(\sigma_{x} - i\sigma_{y}\right), \quad \sigma_{0} = \sigma_{3} $$ We have $$ [T_{3},\sigma_{\pm}] = \pm \sigma_{\pm}, \quad [T_{3},\sigma_{3}] = 0 $$ Therefore, the combination $W_{1} - iW_{2}$ has the charge $Q = +1$, $W_{1}+iW_{2}$ has the charge $Q = -1$, and $W_{3} \sim g_{1}Z + g_{2}A$ has the charge $Q = 0$ (as well as $Z, A$ separately).

  • $\begingroup$ Why we need the commutator to define the isospin $t_3$ ? $\endgroup$ Dec 2, 2016 at 1:44
  • $\begingroup$ @WilliamHuang Because vector bosons transform as adjoint representation of $SU(2)$ - not as 2-component complex vectors but as traceless hermitian 2x2 matrices. $\endgroup$
    – OON
    Dec 2, 2016 at 6:12
  • $\begingroup$ @OON Do you mean $W^a_{\mu \nu} \rightarrow W^a_{\mu \nu}-f^{abc}\alpha^b W^c_{\mu \nu}$ under $SU(2)$ transformation? But I still cannot realize the connection between the adjoint representation and the commutator and $T_3$. $\endgroup$ Dec 2, 2016 at 9:40
  • $\begingroup$ @WilliamHuang $[T_a,T_b]=f^{abc}T_c$. $\endgroup$
    – OON
    Dec 2, 2016 at 10:38
  • $\begingroup$ @OON But now why do we need $[T_3, T_i]=t_3T_i$ rather than $[T_3, T_i]=t_3T_j$ and why does $f^{abc}$ relate to the isospin? I have a lot of confusion about isospin. Very thanks for more explanation. $\endgroup$ Dec 3, 2016 at 16:14

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.