I want to find the wave functions for the three first energy states of 2 identical particles in an infinite square well for two cases of spin 1) both have zero spin 2) both have 1/2 spin I know that $$\psi_n =\psi_{\mathrm{space}} \psi_{\mathrm{spin}}$$ but I have a problem finding both $\psi_{\mathrm{space}}$ and $\psi_{\mathrm{spin}}$ For the wave function of space, I don't know what does it mean for 2 identical particles to be in a specific energy state, what should $n_1$, $n_2$ be. For 1 particle it is $n=1$ for the ground state $n=2$ for the first excited state and so on, but what about 2 particles? And for the wave function of spin, I can't find a formula or method to find it, can you explain the process I must follow?


The first step is to realize that the spatial wavefunctions for two non-interacting particles are of the form $\psi_{n_1}(x_1)\psi_{n_2}(x_2)$ and so have energy $E=E_{n_1}+E_{n_2}\sim n_1^2+n_2^2$. Thus the lowest energy states have the lowest sums $n_1^2+n_2^2$.

For the spin-1/2 case, the total wavefunction must be fully antisymmetric since spin-1/2 particles are fermions. There are four possible types of spin wavefunctions for two spin-1/2 particles: \begin{align} \chi^1_1&= \chi^{1/2}_+(1)\chi^{1/2}_+(2)\, ,\\ \chi^1_{-1}&= \chi^{1/2}_- (1)\chi^{1/2}_-(2)\, ,\\ \chi^1_0&= \chi^{1/2}_+(1)\chi_-^{1/2}(2)+\chi^{1/2}_-(1)\chi^{1/2}_+(2)\, ,\\ \chi^0_0&=\chi^{1/2}_+(1)\chi_-^{1/2}(2)-\chi^{1/2}_-(1)\chi^{1/2}_+(2) \end{align} The first three are symmetric, while the last is antisymmetric under interchange of the particle label $1\leftrightarrow 2$. Thus, the first three must be combined with a spatially antisymmetric wavefunction, and the last to a spatially symmetric wavefunction, to produce a full antisymmetric total wavefunction.

There are three possible types of spatial wavefunctions. The first two are symmetric wavefunctions of the form $$ \psi^+(x_1,x_2)=\left\{\begin{array}{lol} \psi_{n_1}(x_1)\psi_{n_2}(x_2)+\psi_{n_2}(x_1)\psi_{n_1}(x_2)& \hbox{for } n_1\ne n_2\\ \psi_{n}(x_1)\psi_{n}(x_2)&\hbox{for } n_1=n_2=n \end{array} \right. $$ There is also an antisymmetric combination for $n_1\ne n_2$: $$ \psi^-(x_1,x_2)=\psi_{n_1}(x_1)\psi_{n_2}(x_2)-\psi_{n_1}(x_2)\psi_{n_2}(x_1)\, . $$ which can occur when $n_1\ne n_2$. The permutation symmetry here is on the spatial position $x_1\leftrightarrow x_2$, which results from the interchange particle labels $1\leftrightarrow 2$.

The lowest energy wave function is $\psi_{1}(x_1)\psi_1(x_2)$ and is space-symmetric. Thus, it must be combined with $\chi^0_0$ to produce the fully antisymmetric state with lowest energy.

The next lowest are $\psi_1(x_1)\psi_2(x_2)\pm \psi_1(x_2)\psi_2(x_1)$. The symmetric combination must be multiplied by $\chi^0_0$ to make it fully antisymmetric. The antisymmetric combination can be multiplied by any one of the $\chi^1_m$ to make it fully antisymmetric. This should be enough to produce what you need for the spin-1/2 case.

For the spin-0 case there is only one single particle wavefunction and thus there is only one product spin wavefunction: $\chi^0_0(1)\chi^0_0(2)$; it is fully symmetric under the exchange of particle labels $1\leftrightarrow 2$. It must therefore always be multiplied by a spatially symmetric wavefunction since spin-0 particles are bosons. You can work out the details on this using the three types of spatial wavefunctions given above.

  • $\begingroup$ Why is it that the total wave function can be separated into a product of the spatial wave function and the spin wave function? $\endgroup$ – user148792 May 22 '18 at 12:24
  • 1
    $\begingroup$ @delickcrow123 By the method of separation of variables, wavefunctions for different, non-interacting degrees of freedom will be products. Just like the product wavefunctions for the spatial part. $\endgroup$ – ZeroTheHero May 22 '18 at 12:26
  • $\begingroup$ Can't the different degrees of freedom be entangled and, therefore, cannot be written simply as a product? $\endgroup$ – user148792 May 22 '18 at 12:36
  • 1
    $\begingroup$ @delickcrow123 that would be through linear combinations of product states, v.g. Bell states $\sim \vert +\rangle \vert -\rangle \pm \vert -\rangle \vert +\rangle $ $\endgroup$ – ZeroTheHero May 22 '18 at 13:03
  • 1
    $\begingroup$ @delickcrow123 I can search states however I want, but I know that the product states will be a basis and that therefore any state (entangled or not) will be a linear combination of product states. $\endgroup$ – ZeroTheHero May 22 '18 at 15:51

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.