# Mapping a vector to its correspondence in tensor product space [closed]

Lets assume we have two Hilbert spaces $H_1$ and $H_2$. Can you show me how can we map a vector $v\in H_1$ into a vector $v' \in H_1\otimes H_2$?

## closed as off-topic by sammy gerbil, ZeroTheHero, Jon Custer, M. Enns, Cosmas ZachosAug 30 '17 at 16:36

This question appears to be off-topic. The users who voted to close gave this specific reason:

• "Homework-like questions should ask about a specific physics concept and show some effort to work through the problem. We want our questions to be useful to the broader community, and to future users. See our meta site for more guidance on how to edit your question to make it better" – sammy gerbil, Jon Custer, M. Enns, Cosmas Zachos
If this question can be reworded to fit the rules in the help center, please edit the question.

• I think this is impossible as without sense. At least, may be you could map to a vector $\,v_1\in H_1\,$ the following subspace $\,V_1 \subset H_1\otimes H_2\,$ : $$v_1 \longrightarrow V_1 =\lbrace v_1\otimes v_2 \in H_1\otimes H_2 : v_2 \in H_2\rbrace$$ My feeling is that the subspace $\,V_1 \subset H_1\otimes H_2\,$ is a Hilbert space too , but this must be proved. – Frobenius Aug 29 '17 at 7:18
• Note that something like that you ask for could be done with linear transformations. That is, if $\,\mathrm{A_1}\,$ is an operator on $\,H_1\,$ then we could map to it its extension on $\,H_1\otimes H_2\,$ : $$\mathrm{A_1} \text{ on } H_1 \longrightarrow \left(\mathrm{A_1}\otimes\mathrm{I_2}\right) \text{ on } H_1\otimes H_2$$ where $\,\mathrm{I_2}\,$ the identity operator on $\,H_2\,$. – Frobenius Aug 29 '17 at 7:35
• @Frobenius $V_1$ is a Hilbert space isomorphic to $H_2$. – Noiralef Aug 29 '17 at 14:29
• @Noiralef : You are absolutely right. – Frobenius Aug 29 '17 at 15:21
• I rephrased the question. Would you please check it again? – user650585 Aug 29 '17 at 15:27

What you are asking physically is: "My first system $H_1$ is in the state $v$, I don't know the state of my second system $H_2$. What is the state of the total system $H_1 \otimes H_2$?

Of course, this question does not make sense. So, there is no canonical embedding of $H_1$ in $H_1 \otimes H_2$. However, if we fix a state $w \in H_2$, then there is of course the map $$\iota_w: H_1 \to H_1 \otimes H_2, v \mapsto v' = v \otimes w .$$

• In Quantum Mechanics it is said that if their entangled state $\alpha |f_1 e_1\rangle + \beta|f_1 e_2\rangle + \gamma |f_2 e_1\rangle +\delta |f_2 e_2\rangle$ can be written as $\alpha_1\alpha_2 |f_1 e_1\rangle + \beta_1\beta_2|f_1 e_2\rangle + \gamma_1\gamma_2 |f_2 e_1\rangle +\delta_1\delta_2 |f_2 e_2\rangle$ then they are not entangled. I want to know if in such case $\alpha_1 |f_1 e_1\rangle + \beta_1|f_1 e_2\rangle + \gamma_1 |f_2 e_1\rangle +\delta_1 |f_2 e_2\rangle$ is the mapping of one of the original states from $H_1$ to $H_1\otimes H_2$? – user650585 Aug 29 '17 at 15:53
• @user650585 You might want to ask this as a new question. But what you write in the comment / in the question (edit v3) is all wrong, sorry. Entangling two systems does not change the Hilbert space, it is always $H_1 \otimes H_2$. However, if the systems are not entangled, their state will be in the subspace $H_1 \times H_2 \subset H_1 \otimes H_2$, and there is a canonical map $H_1 \times H_2 \ti H_1 \otimes H_2$. – Noiralef Aug 29 '17 at 18:18
• Writing $H_1\times H_2\subset H_1\otimes H_2$ as in your comment is misleading, since the map from one to the other is not injective, so you cannot identify the image with $H_1\times H_2$ itself. – ACuriousMind Aug 29 '17 at 18:21
• @Noiralef : As ACuriousMind pointed out the relation $H_1\times H_2\subset H_1\otimes H_2$ is misleading. If $H_1,H_2$ are finite complex Hilbert spaces of dimensions $r,s$ respectively then in a sense $H_1\equiv\mathbb{C}^r,H_2\equiv\mathbb{C}^s$ so $H_1\times H_2 \equiv \mathbb{C}^{r+s}$ but $H_1\otimes H_2 \equiv \mathbb{C}^{r\cdot s}$ that is the two spaces not even have the same dimension in general. Here accidentally $r=2=s \Longrightarrow r+s=4=r\cdot s$. – Frobenius Aug 29 '17 at 19:17
• True, I was being imprecise in my comment above -- sorry. – Noiralef Aug 29 '17 at 20:42

OP post the following comment complementary to his/her original question :

In Quantum Mechanics it is said that if their entangled state $\alpha |f_1 e_1\rangle + \beta|f_1 e_2\rangle + \gamma |f_2 e_1\rangle +\delta |f_2 e_2\rangle$ can be written as $\alpha_1\alpha_2 |f_1 e_1\rangle + \beta_1\beta_2|f_1 e_2\rangle + \gamma_1\gamma_2 |f_2 e_1\rangle +\delta_1\delta_2 |f_2 e_2\rangle$then they are not entangled. I want to know if in such case $\alpha_1 |f_1 e_1\rangle + \beta_1|f_1 e_2\rangle + \gamma_1 |f_2 e_1\rangle +\delta_1 |f_2 e_2\rangle$ is the mapping of one of the original states from $H_1$ to $H_1\otimes H_2$?

You are confused. A state in $\,H_1\otimes H_2 \,$ $$w=\alpha |f_1 e_1\rangle + \beta|f_1 e_2\rangle + \gamma |f_2 e_1\rangle +\delta |f_2 e_2\rangle \in H_1\otimes H_2 \tag{01}$$ is not a product state (also called a separable state) if it could be expressed as $$w=\alpha_1\alpha_2 |f_1 e_1\rangle + \beta_1\beta_2|f_1 e_2\rangle + \gamma_1\gamma_2 |f_2 e_1\rangle +\delta_1\delta_2 |f_2 e_2\rangle \tag{02}$$ If so any such state would be a product state and moreover separable in many ways since there exists an infinite number of 8 coefficients

$$\left(\alpha_1,\alpha_2,\alpha_3,\alpha_4,\beta_1,\beta_2,\beta_3,\beta_4\right) \tag{03}$$ satisfying the system of equations $$\alpha_1\alpha_2=\alpha \, ,\quad \beta_1\beta_2=\beta \, ,\quad \gamma_1\gamma_2=\gamma\, ,\quad \delta_1\delta_2=\delta \tag{04}$$ The state $\,w\,$ in equation (01) is a product or separable state (that is not entagled) if there exist at least a solution $\,\left(\phi_1,\phi_2,\epsilon_1,\epsilon_2\right)\,$ of the system of equations $$\phi_1\epsilon_1=\alpha \, ,\quad \phi_1\epsilon_2=\beta \, ,\quad \phi_2\epsilon_1=\gamma\, ,\quad \phi_2\epsilon_2=\delta \tag{05}$$ Then the state $\,w\,$ is a (separable) product state $$w=v_1\otimes v_2 =\left(\phi_1f_1+\phi_2f_2\right)\otimes\left(\epsilon_1e_1+\epsilon_2e_2\right) \tag{06}$$ But then why do you want to call the vector $\,w \in H_1\otimes H_2 \,$ the correspodence of $\,v_1 \in H_1$. It has no sense and doesn't help anywhere.

• Thanks for your answer. Regarding your last question what I want to know is if there is a space in which both $v_1$ and $v_1\otimes v_2$ can be represented at the same time? – user650585 Aug 30 '17 at 2:30