Particle Ambiguity in Curved Spacetime

1. Frequency, Energy, and Spacetime Symmetry

Let's revisit the cornerstone of our previous discussion: the split into positive and negative frequency modes. For the Klein-Gordon equation in Minkowski space, $(\square +m^2)\varphi =0$, we found plane-wave solutions of the form $\varphi \propto e^{-i{k}_{\mu }{x}^{\mu }}=e^{-i(\omega t-\mathbf{p}\cdot \mathbf{x})}$. We designated modes with $\omega >0$ as "positive frequency."

Why was this choice so natural and unambiguous? It's because Minkowski spacetime possesses a high degree of symmetry, described by the Poincaré group. Specifically, it has a global timelike Killing vector field, ${\xi }^{\mu }=(\partial /\partial t{)}^{\mu }$. A Killing vector field generates an isometry, a transformation that leaves the metric unchanged. The condition is that the Lie derivative of the metric with respect to the vector field vanishes:
${\mathcal{L}}_{\xi }{g}_{\mu \nu }={\nabla }_{\mu }{\xi }_{\nu }+{\nabla }_{\nu }{\xi }_{\mu }=0$
A timelike Killing vector, like $\partial /\partial t$, corresponds to a time-translation symmetry. A spacetime that possesses such a symmetry is called stationary.

In a stationary spacetime, we can find solutions to the field equations that are also eigenfunctions of the Lie derivative operator associated with the Killing vector:
${\mathcal{L}}_{\xi }\varphi ={\xi }^{\mu }{\partial }_{\mu }\varphi =-i\omega \varphi$
Here, $\omega$ is a constant associated with each mode. Because ${\xi }^{\mu }$ is a symmetry of the spacetime, this $\omega$ corresponds to a conserved notion of energy for the mode. This allows for a globally consistent and physically meaningful separation of all solutions into three classes: positive frequency ($\omega >0$), negative frequency ($\omega <0$), and zero frequency ($\omega =0$). This gives us a natural, preferred way to define creation and annihilation operators, and thus a unique, preferred vacuum state.

2. When Symmetry Breaks: The Generic Spacetime

The problem is that a generic curved spacetime does not possess a global timelike Killing vector.

• An expanding Friedmann-Robertson-Walker (FRW) universe is changing in time, so it has no time-translation symmetry.
• A spacetime describing a collapsing star that forms a black hole is dynamic; it is stationary long before the collapse and long after the black hole has settled down, but the process of collapse breaks the time-translation symmetry globally.

Without a global timelike Killing vector, there is no universally preferred time coordinate and no conserved notion of energy for field modes across all of spacetime. This means there is no natural, unique way to perform the positive/negative frequency split. Different observers, using different local coordinate systems to define their notion of time, will perform this split differently.

This is the origin of the ambiguity. If different observers disagree on what constitutes a positive-frequency mode, they will disagree on the definition of annihilation operators, and therefore, they will disagree on what constitutes the vacuum state.

3. A Tale of Two Vacua: Bogoliubov Transformations

Let's make this more concrete. The core of the issue is a choice of basis.

To quantize the field $\varphi$, we need to find a complete set of basis functions $\{u_i\}$ that solve the Klein-Gordon equation $(g^{\mu \nu }{\nabla }_{\mu }{\nabla }_{\nu }+m^2)\varphi =0$. We can then promote $\varphi$ to an operator by expanding it in this basis:
$\hat{\varphi }(x)={\sum }_i\left({\hat{a}}_i{u}_i(x)+{\hat{a}}_i^{†}{u}_i^{\ast }(x)\right)$
The coefficients ${\hat{a}}_i$ and ${\hat{a}}_i^{†}$ become the annihilation and creation operators. The vacuum state for this observer, let's call her Alice, is defined as the state annihilated by all her ${\hat{a}}_i$:
${\hat{a}}_i|0_A⟩=0\text{for all }i$

Now, consider another observer, Bob, who is in a different state of motion (e.g., accelerating relative to Alice). He will naturally choose a different complete basis of solutions, $\{v_j\}$, which are better adapted to his coordinates. He expands the same field operator $\hat{\varphi }$ in his own basis:
$\hat{\varphi }(x)={\sum }_j\left({\hat{b}}_j{v}_j(x)+{\hat{b}}_j^{†}{v}_j^{\ast }(x)\right)$
Bob's vacuum state is defined by his annihilation operators:
${\hat{b}}_j|0_B⟩=0\text{for all }j$

Since both $\{u_i\}$ and $\{v_j\}$ are complete bases for the same space of solutions, any function in one basis can be written as a linear combination of functions in the other. Specifically, Bob's basis functions can be expanded in terms of Alice's:
$v_j={\sum }_i\left({\alpha }_{ji}{u}_i+{\beta }_{ji}{u}_i^{\ast }\right)$
The matrices ${\alpha }_{ji}$ and ${\beta }_{ji}$ are called Bogoliubov coefficients. The $\alpha$ coefficients mix positive-frequency modes with positive-frequency modes, while the $\beta$ coefficients mix positive-frequency modes with negative-frequency modes (from Alice's perspective, since they multiply $u_i^{\ast }$).

This relationship between the basis functions implies a relationship between the operators. A calculation shows that Bob's annihilation operators are a mix of Alice's annihilation and creation operators:
${\hat{b}}_j={\sum }_i\left({\alpha }_{ji}^{\ast }{\hat{a}}_i-{\beta }_{ji}^{\ast }{\hat{a}}_i^{†}\right)$

Now for the crucial question: What does Alice's vacuum look like to Bob? Bob would try to measure the number of "B-particles" in Alice's vacuum state, $|0_A⟩$. The number operator for B-particles in mode $j$ is ${\hat{N}}_{b_j}={\hat{b}}_j^{†}{\hat{b}}_j$. Let's calculate its expectation value:
$⟨0_A|{\hat{N}}_{b_j}|0_A⟩=⟨0_A|{\hat{b}}_j^{†}{\hat{b}}_j|0_A⟩$
Substituting the transformation for ${\hat{b}}_j$, we get:
$⟨0_A|\left({\sum }_k({\alpha }_{jk}{\hat{a}}_k^{†}-{\beta }_{jk}{\hat{a}}_k)\right)\left({\sum }_l({\alpha }_{jl}^{\ast }{\hat{a}}_l-{\beta }_{jl}^{\ast }{\hat{a}}_l^{†})\right)|0_A⟩$
Since ${\hat{a}}_l|0_A⟩=0$ and $⟨0_A|{\hat{a}}_k^{†}=0$, the only terms that survive are those with ${\hat{a}}_k{\hat{a}}_l^{†}$. The expression simplifies to:
$⟨0_A|{\sum }_{k,l}{\beta }_{jk}{\beta }_{jl}^{\ast }{\hat{a}}_k{\hat{a}}_l^{†}|0_A⟩={\sum }_{k,l}{\beta }_{jk}{\beta }_{jl}^{\ast }⟨0_A|[{\hat{a}}_k,{\hat{a}}_l^{†}]|0_A⟩$
Using the commutation relation $[{\hat{a}}_k,{\hat{a}}_l^{†}]={\delta }_{kl}$, we arrive at the final result:
$⟨0_A|{\hat{N}}_{b_j}|0_A⟩={\sum }_k|{\beta }_{jk}{|}^2$

This is the central result. If the Bogoliubov coefficient $\beta$ is non-zero, then the number of particles Bob measures in Alice's vacuum is non-zero. What one observer calls a vacuum, another sees as a state containing particles. The particle content of a quantum field is not an absolute property of the state; it is relative to the observer. This only happens if $\beta \ne 0$, which is precisely the case when the observers' notions of positive frequency do not align.

4. A Word from a Founder

This idea that particles are an observer-dependent concept can be counter-intuitive. It's a radical departure from our classical intuition and even from standard quantum mechanics. The modern viewpoint, which you should adopt, is that the field is fundamental, while particles are a derived, observer-dependent manifestation of the field.

Let's hear this directly from one of the pioneers of the field, William Unruh, who discovered that accelerating observers perceive the Minkowski vacuum as a thermal bath of particles (the Unruh effect).

Unruh's statement—"what a particle is depends on your state of motion"—is the essential physical takeaway of this lesson. The mathematical reason is the lack of a globally preferred way to define time, which, in the language of geometry, is the absence of a global timelike Killing vector.

Conclusion

We have now confronted the fundamental ambiguity at the heart of quantum field theory in curved spacetime. The comfortable, solid notion of a 'particle' that we built in flat space has dissolved into an observer-dependent concept.

Key Takeaways:

• In Minkowski spacetime, the existence of a global timelike Killing vector $\partial /\partial t$ provides a natural, observer-independent way to define positive and negative frequency modes. This leads to a unique vacuum state.
• A general curved spacetime (like our expanding universe or a collapsing star) lacks a global timelike Killing vector. Consequently, there is no preferred, unique definition of positive frequency.
• Different observers will choose different bases of modes to define their "vacuum" and "particle" states.
• These different bases are related by Bogoliubov transformations. If these transformations mix positive and negative frequency modes (i.e., the $\beta$ coefficients are non-zero), then one observer's vacuum will appear to contain particles to another observer.
• The fundamental entity is the quantum field. Particles are a derived concept whose definition depends on the observer's state of motion and the spacetime geometry.

Preview of the Next Lesson:

We have seen conceptually why particle definition is ambiguous and have been introduced to the name for the mathematical tool that quantifies this ambiguity. In the next lesson, we will dive deeper into the mathematics of Bogoliubov transformations. We will formalize their properties and see how they relate different sets of creation and annihilation operators, preparing us to calculate the particle spectra seen by different observers in phenomena like the Unruh and Hawking effects.