QFT vs Wave Function: Understanding Particle States

[MichPod]

My level is not sufficient enough to easily understand QFT yet there is some basic question I need to understand in it - what in QFT corresponds to a wave function in QM, for a single particle case and, say, for a more general case of multiparticle nonseparable state (suppose the particles are identical). I do understand Fock space idea as well as creation-annihilation operators idea, yet I cannot grasp what is that operator field considered in QFT and how it may correspond to the general multiparticle case of QM. If this field defines an operator at each space point, then on which vector space does it act and how it may be related to the multiple particle wave function? Or maybe it is not the operator field itself which corresponds in QFT to the wave function/particles state, but something else?
I tried to look through some discussions here and on other sites but did not succeed to see or grasp what is the solution.

Thank you.

 

[A. Neumaier]

The quantum field $\phi(x)$ corresponds in meaning to the quantum mechanical $q_j$ (specifically, $\phi\to q$ and $x\to j$, but with a continuum of indices). But its action on an $N$-particle state gives (in the nonrelativistic case) a linear combination of $(N-1)$-particle and $(N+1)$-particle states.
See https://en.wikipedia.org/wiki/Second_quantization.

 

[MichPod]

But how the quantum field and particle states are related? I was under an impression that quantum particles are an emergent property of a quantum field, that is particles are some discrete modes of oscillation of that quantum field. So I expected that a state of the quantum field itself may somehow be interpreted as particle state.

From what I can see and understand in second quantization, operator field introduced there cannot contain any state-specific information.

 

[atyy]

The creation operator acting on the vacuum state creates an excited state, which is a state with one particle.

When you read about non-relativistic second quantization, first keep in mind non-relativistic quantum mechanics of many identical particles (ie. with the standard interpretation of the wave function etc.) Second quantization is simply non-relativistic quantum mechanics of many identical particles, but formulated in a different language.

 

[A. Neumaier]

operator field introduced there cannot contain any state-specific information.

Just like in ordinary quantum mechanics, the position operator cannot contain any state-specific information. The latter is always in the state, not in the operators.

 

[Demystifier]

In QFT there are 3 different things which, in some sense, correspond to the wave function $\psi({\bf x},t)$ in QM.
1. Field operator $\hat{\phi}({\bf x},t)$
2. Wave functional $\Psi[\phi({\bf x}), t]$
3. Wave function itself, which in QFT can be calculated as $\psi({\bf x},t)=\langle 0|\hat{\phi}({\bf x},t)|\Psi\rangle$

 

[kith]

From what I can see and understand in second quantization, operator field introduced there cannot contain any state-specific information.

Yes. The field operator $\phi(\vec x)$ is not the state but it acts on the state. One way to picture it is by its action on the vacuum. The vacuum state is the state with zero particles which is denoted by $|0\rangle$. If I act on it with the field operator, I get a particle localized at location $\vec x$ (compare this with Demystifier's point 3 in post #6 above):
$$\phi(\vec x) |0\rangle = |\vec x\rangle$$
In ordinary QM, the position wavefunction can be written as a superposition of momentum eigenstates. Analogously in QFT, the field operator can be written as a sum of operators which create or annihilate a particle with definite momentum. If I act with the field operator on the vacuum, it creates in effect a particle with definite location.

(What I have written here is by no means rigorous and there are quite a few difficulties. For example, we cannot define a position operator with the usual properties for massless particles like photons. So the very concept of localizing particles isn't always well-defined.)

 

[kith]

Your confusion about the "operator field" and the "state of a quantum field" may be related to the fact that the term "field" in these quotes refers to different things. I started a thread about this a couple of years ago.

 

[vanhees71]

Yes. The field operator $\phi(\vec x)$ is not the state but it acts on the state. One way to picture it is by its action on the vacuum. The vacuum state is the state with zero particles which is denoted by $|0\rangle$. If I act on it with the field operator, I get a particle localized at location $\vec x$ (compare this with Demystifier's point 3 in post #6 above):
$$\phi(\vec x) |0\rangle = |\vec x\rangle$$
In ordinary QM, the position wavefunction can be written as a superposition of momentum eigenstates. Analogously in QFT, the field operator can be written as a sum of operators which create or annihilate a particle with definite momentum. If I act with the field operator on the vacuum, it creates in effect a particle with definite location.

This has to be taken with very great care. E.g., in the case of photons, this "state" (it's a generalized state in the sense of a distribution in any case), this cannot be interpreted as a "generalized position eigenstate" since a photon doesn't permit the definition of a position operator in the strict sense. A photon has not a position as an observable at all (within standard QED)!

 

[bhobba]

But how the quantum field and particle states are related?

There are a number of equivalent formulations of QM:
http://www.colorado.edu/physics/phys5260/phys5260_sp16/lectureNotes/NineFormulations.pdf

QFT corresponds to formulation F - the second quantization formulation based on creation and annihilation operators. As operators they operate on states and that's how states come into it. The interesting thing about QFT is its based on operators with states being of secondary importance. They are there of course as they must be - but most of the time they take a back seat.

Thanks
Bill

 

[MichPod]

In QFT there 3 different things which, in some sense, correspond to the wave function $\psi({\bf x},t)$ in QM.
1. Field operator $\hat{\phi}({\bf x},t)$
2. Wave functional $\Psi[\phi({\bf x}), t]$
3. Wave function itself, which in QFT can be calculated as $\psi({\bf x},t)=\langle 0|\hat{\phi}({\bf x},t)|\Psi\rangle$

It'll take me some time to work on it. Some immediate questions to clarify:

Are the $|0\rangle$ and $|\Psi\rangle$ appearing in the item 3 the wave functionals of the same kind as brought in item 2?

Is the wave function brought in item 3 eqivalent to the single particle QM wave function or something else?

Most importantly, how still can we get, say, a nonseparable wave function of two particles from the wave functional brought in item 2?

 

[bhobba]

Some immediate questions to clarify:

Read the link I gave on the second quantization formulation of QM.

Thanks
Bill

 

[DarMM]

Ignoring distribution theoretic issues, the wave functions in QFT are:

$\Psi[\phi(\vec{x})]$ in the Heisenberg picture. That is a function over the space of field configurations in $\mathbb{R}^{3}$. Roughly speaking it is the amplitude to observe the field in that configuration.

If the field is a free field then it turns out that this space of field wave functionals has a basis where it takes the form of a Fock space over a single particle space Hilbert space. Hence one can view the Hilbert space in either a particle or field manner.

Technical details for others (I'll fill them in when I'm less tired!):
If the space of classical fields is $\mathcal{S}$ and $d\nu$ is a measure on this space obeying the axioms of QFT and the specific identities of a free field, then one can prove that:

$L^{2}(\mathcal{S},d\nu) = \mathcal{F}(\mathcal{S}^{*})$

(The single particle quantum Hilbert space is the dual of the classical field space)

 

[MichPod]

Thanks to everybody!
I think I've got some directions and stuff to work on.

 

[gva]

The creation operator acting on the vacuum state creates an excited state, which is a state with one particle.

When you read about non-relativistic second quantization, first keep in mind non-relativistic quantum mechanics of many identical particles (ie. with the standard interpretation of the wave function etc.) Second quantization is simply non-relativistic quantum mechanics of many identical particles, but formulated in a different language.

What is the relativistic quantum mechanics of many identical particles? Or spacetime background independence theory of many particles.. must we come to this latter for completeness?

 

[bhobba]

What is the relativistic quantum mechanics of many identical particles?

Its called the Quantum Field Theory of Bosons or Femions which are all identical. We either have Bosons or Fermions in QFT - this is the famous Spin Statistics Theorem:
http://www.worldscientific.com/worldscibooks/10.1142/3457

Or spacetime background independence theory of many particles.. must we come to this latter for completeness?

Cant see how background independence has anything to do with this.

Thanks
Bill

 

[gva]

Its called the Quantum Field Theory of Bosons or Femions which are all identical. We either have Bosons or Fermions in QFT - this is the famous Spin Statistics Theorem:
http://www.worldscientific.com/worldscibooks/10.1142/3457
Cant see how background independence has anything to do with this.

Thanks
Bill

Simple. General relativity is not yet united with quantum field theory. When you have general relativistic quantum field theory.. the particles will no longer occurred against a backdrop of fixed spacetime.. but dynamical where it is more than the particles curving spacetime... but particles and spacetime becoming united into a symmetry just like space and time where they become mere shadows. Is this not the final goal of QFT?

 

[bhobba]

Simple. General relativity is not yet united with quantum field theory.

Thats a very common misconception, but totally false:
http://arxiv.org/abs/1209.3511

We have a perfectly valid theory up to about the Plank scale. QED is in the same boat - its only valid to a cutoff before the electroweak theory takes over and none would say QED is not EM united with QFT.

Thanks
Bill

 

[atyy]

What is the relativistic quantum mechanics of many identical particles? Or spacetime background independence theory of many particles.. must we come to this latter for completeness?

At the non-rigourous level, the relativistic quantum mechanics of many identical particles is conceptually the same as the non-relativistic quantum mechanics of many identical particles, except that particles can be created and destroyed. However, if you use second quantization in non-relativistic quantum mechanics, there you will already see that there are "particles" that can be created and destroyed, such as phonons in non-relativistic quantum mechanics, eg. http://yclept.ucdavis.edu/course/242/2Q_Fradkin.pdf.

As DarMM points out above, the "particle" picture and Fock space are not rigourously correct when the field theory is interacting. So all the derivations in textbooks are only "physically" correct, although they are mathematically wrong.

Background independence is not required for quantum field theory (or any theory with known physical relevance, including quantum gravity at low energies).

 

[Demystifier]

Are the $|0\rangle$ and $|\Psi\rangle$ appearing in the item 3 the wave functionals of the same kind as brought in item 2?

Yes.

Is the wave function brought in item 3 eqivalent to the single particle QM wave function ...?

Yes.

Most importantly, how still can we get, say, a nonseparable wave function of two particles from the wave functional brought in item 2?

Let me explain it for the simplest case of hermitian field operator. It can be written as
$$\hat{\phi}(x)=\hat{\psi}(x)+\hat{\psi}^{\dagger}(x)$$
where $\hat{\psi}^{\dagger}(x)$ contains the creation operators and $\hat{\psi}(x)$ the destruction operators. Then the non-separable 2-particle wave function is
$$\psi(x_1,x_2)=\langle 0|\hat{\psi}(x_1)\hat{\psi}(x_2)|\Psi\rangle$$

Even though $\hat{\psi}(x_1)$ and $\hat{\psi}(x_2)$ are local operators, the product $\hat{\psi}(x_1)\hat{\psi}(x_2)$ is not a local operator. This is why "local" QFT can have non-local (or non-separable) effects.

 

[MichPod]

the non-separable 2-particle wave function is
$$\psi(x_1,x_2)=\langle 0|\hat{\psi}(x_1)\hat{\psi}(x_2)|\Psi\rangle$$

Thank you! Yet what I basically do not understand in this is how we are able to get both single-particle wave function and 2-particle wave function from the same wave functional? I'd expect that for a given arbitrary field wave functional we can (in the best case) just determine the number of particles (if it is definite) and then determine a QM wave function for this number of particles only. I fail to see how can we extract a QM wave function for whatever number of particles we want from the same field wave functional. Or maybe I fail to see what is the meaning of such a wave function.

 

[A. Neumaier]

what is the meaning of such a wave function.

It is a superposition of wave functions with 0,1,2,... particles. The particle number is not definite.

 

[MichPod]

It is a superposition of wave functions with 0,1,2,... particles. The particle number is not definite.

Yes. But suppose some very specific case - that we have a state of the field where the number of particles is definite and it equals two (can we have a definite number of particles at least for non-relativistic QFT?), then we try to calculate a 3-particle wave function. What are we supposed to get for such a case?

 

[Nugatory]

Yet what I basically do not understand in this is how we are able to get both single-particle wave function and 2-particle wave function from the same wave functional?

Even in ordinary QM, a multi-particle quantum system is a single quantum system with a single wave function. You can see that clearly when you look at a pair of particles in the singlet state - there's a single wave function and the two spins are two different observables on it.

So if you're already OK with the idea that a multi-particle system has a single wave function... The new thing is that you're being asked to do is think of the number of particles itself as an observable with an associated Hermitian operator. The wave function may be eigenvector of that observable or it may he a superposition of eigenvectors with different eigenvalues; in the latter case the particle number will not be fixed.

 

[MichPod]

The wave function may be eigenvector of that observable or it may he a superposition of eigenvectors with different eigenvalues; in the latter case the particle number will not be fixed.

And I meant the case where the wave functon(al) is the eigenvector of the number operator (with eigenvalue 2).
Then we apply $$\psi(x_1,x_2,x_3)=\langle 0|\hat{\psi}(x_1)\hat{\psi}(x_2)\hat{\psi}(x_3)|\Psi\rangle$$ transformation and my question was what can we get? I would be happy if the answer is zero, but I am not sure.

 

[Demystifier]

Yes. But suppose some very specific case - that we have a state of the field where the number of particles is definite and it equals two (can we have a definite number of particles at least for non-relativistic QFT?), then we try to calculate a 3-particle wave function. What are we supposed to get for such a case?

We get $\psi(x_1,x_2,x_3)=0$

 

[bob012345]

The creation operator acting on the vacuum state creates an excited state, which is a state with one particle.

When you read about non-relativistic second quantization, first keep in mind non-relativistic quantum mechanics of many identical particles (ie. with the standard interpretation of the wave function etc.) Second quantization is simply non-relativistic quantum mechanics of many identical particles, but formulated in a different language.

And what exactly and I mean physically, is a creation operator? Don't give me math, I want physics. Thanks.

 

[A. Neumaier]

what exactly and I mean physically, is a creation operator?

It is the negative frequency part of the Hermitian field operator, and the annihilation operator (its adjoint) is the positive frequency part. For classical fields (and in signal processing) one would call the latter the analytic signal of a real field (or signal).

 

[Traruh Synred]

Interesting thread.

One thing there are no particles in QFT. The electrons are not a particles, but excitations of the electron-field. Particle-like appearance is when the excitations form a packet (many wave lengths) that is localized.

In this sense the wave picture is more correct than the particle in QFT.

 

[Nugatory]

An attempted thread hijack has been successfully prevented, but unfortunately a number of informative and even humorous posts by some of the SAs were lost in the cleanup.

The thread remains open.

 

[bhobba]

And what exactly and I mean physically, is a creation operator? Don't give me math, I want physics.

I am afraid in QFT the difference between math and physics is very blurred - what you ask simply can't be done. The attempt to do it leads to a LOT of misconceptions such as virtual particles that are very very hard to dislodge.

At a basic level see the harmonic occilator
https://en.wikipedia.org/wiki/Quantum_harmonic_oscillator

In a certain sense QFT is an infinite set of harmonic occilators:
http://physics.stackexchange.com/qu...field-an-infinite-set-of-harmonic-oscillators

Again notice:
'This is all nonsense.This might sound strong, but it has been the source of many annoying misunderstandings in the publicization of quantum theories to laypeople. Just because something (ϕϕ) fulfills a wave/oscillator equation and has a mode expansion (as the above is called), it does not mean that anything oscillates. It's just the same type of equation you encounter in oscillator, not the same physical situation. It's a nice pretty picture to tell ourselves that we understand the quantum field, but ultimately, there is nothing there that would justify the oscillator interpretation. Nothing physical is vibrating or oscillating here.'

In QFT the physics is the math.

Thanks
Bill

 

[vanhees71]

Well, although I agree that the only concise picture of nature is given by mathematical models, it is always good advice, to keep an eye on what's measured that's described by the mathematical model. That's most important for QFT.

QFT has a very broad range of applicability. It's the most comprehensive mathematical model we have about nature. The only part that's not satisfactorily covered is gravity, but to discover the quantum theory of gravity is hindered by the fact that we have no observables concerning gravity that would need a quantum description. The classical limit of a (hopefully existing) quantum theory of gravitation is General Relativity, which is a classical field theory.

Relativistic QFT has been discovered by looking at high-energy particle-scattering experiments, and the physical quantities that are observable are corresponding cross sections. In the QFT formalism these are encoded in S-matrix elements which we are able to calculate in renormalized perturbation theory, which is well organized in terms of Feynman diagrams (sometimes with the necessity to resum infinitely many diagrams to get finite results; or using renormalization-group equations to improve the applicability of perturbation theory etc.). Thus, the physical meaning of the formalism is in the S-matrix elements, and to keep this in mind is very important, if you want to understand why relativistic QFT is formulated as it is. In the relativistic realm you need QFT, because particles are usually created and destroyed in scattering processes, which makes the formalism with creation and annihilation operators a very natural way to describe what's going on. The interpretability in terms of "particles" is very limited, since this notion makes only sense in terms of S-matrix elements, i.e., for asymptotic free states.

Another application of QFT in both relativistic and non-relativistic physics is many-body theory, i.e., the theory of very many particles building "matter" in the literal sense. There you can use QFT to derive macroscopic (semi-)classical equations to describe effectively the relevant (macroscopic) degrees of freedom encoded in material parameters like particle densities, energy-momentum densities (energy-momentum-stress tensor), various transport coefficients, etc. All these quantities are given in terms of correlation functions of field operators, often evaluated in thermal equilibrium (usually in the grand-canonical ensemble).