⚛️ quantum physics

A Margolus-Levitin speed limit for observables: mean energy bounds expectation-value change quadratically

This paper establishes a new Margolus-Levitin-type quantum speed limit for observables, proving that the time required to change an expectation value is fundamentally bounded by the square of the change divided by the mean energy above the ground state, thereby closing a gap in speed-limit theory where mean energy previously could not constrain observable evolution.

Original authors: Bryan Nasr

Published 2026-08-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Bryan Nasr

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the quantum world, time is not a passive backdrop but a resource that can be spent, saved, or wasted. Just as a car engine requires fuel to move a vehicle, a quantum system requires energy to change its state. For decades, physicists have known there are hard limits on how fast this change can happen. Two famous rules, established long ago, act as the speedometers for this microscopic motion. One rule says that if you know how much the energy of a system fluctuates, you can calculate the minimum time it takes for the system to become completely different from its starting point. The other rule uses the average amount of energy the system holds above its lowest possible state to set a similar limit. These rules have been incredibly useful for understanding how quantum states evolve, but they have a blind spot: they describe the motion of the entire system, not the specific measurements we actually take in a lab.

When scientists run an experiment, they rarely watch the whole quantum system transform in a blur. Instead, they measure a specific property, like the spin of a particle or the position of an atom, and watch how the average value of that property changes over time. Until now, the rules for how fast these specific measurements can change have relied almost entirely on the first rule about energy fluctuations. The second rule, which uses the average energy, had never been successfully applied to the change of a specific measurement. This left a gap in our understanding of the quantum speed limit, specifically for the things we actually observe.

A researcher named Bryan Nasr has now filled that gap, proving that the average energy does indeed control how fast a measurement can change, but in a way that was previously unexpected. The paper demonstrates that while the old rules suggest a straight-line relationship between energy and speed, the new rule reveals a curved, quadratic relationship. This means that for small changes in a measurement, the average energy is a much weaker constraint than previously thought, but as the change gets larger, the energy requirement shoots up dramatically. The study proves that it is impossible to have a simple, straight-line rule connecting average energy to the speed of a measurement change. Instead, the most efficient path requires a specific, curved trade-off that holds true for every type of quantum system, whether it is a single atom or a complex mixture of many particles.

The core of this discovery is a mathematical proof that rules out the possibility of a linear connection. If one were to guess that doubling the average energy would simply double the speed at which a measurement changes, the new work shows this is false. Instead, to double the speed of a change, one must quadruple the average energy. This quadratic law is the best possible limit; no quantum system can break it. The researcher calculated a precise number that defines this limit, a constant that does not depend on the size or complexity of the system. This constant, approximately 0.1725, acts as a universal floor. No matter how cleverly a system is designed, it cannot move a measurement faster than this floor allows given its available energy.

To find this limit, the study looked at the simplest possible scenarios, specifically systems with just two energy levels, and showed that even in these basic cases, the quadratic law holds. The researchers then proved that this rule applies just as strictly to complex, mixed states where the system is not in a single, pure condition but a jumble of possibilities. They also explored what happens when the measurement starts from a special position where it is momentarily still. In those specific cases, the quadratic rule relaxes back into a linear one, but for almost all other situations, the curved, quadratic law is the governing principle. This distinction is crucial because it tells us exactly when the average energy is the most important factor in limiting speed.

The implications of this finding are most clear in the design of autonomous quantum clocks. These are devices that keep time using a fixed, unchanging energy source, much like a pendulum clock uses gravity. The new rule sets a fundamental limit on how precisely such a clock can tick. It establishes that the energy stored in the clock's mechanism sets a minimum time required for the clock's hand to move a measurable distance. If the clock has very little energy, it cannot move its hand quickly, no matter how well it is built. This provides a new way to certify the energy efficiency of quantum devices: by observing how fast a measurement changes, one can calculate the minimum amount of energy the device must be using, without needing to look inside the machine or know its internal state.

The study also clarifies where this rule does not apply. It does not govern systems that are driven by external, changing forces, such as the pulses used to control quantum computers. It is also less relevant for very large systems where the total energy grows with the size of the object, making the limit too weak to be useful for local measurements. However, for small, isolated systems that rely on their own internal energy to evolve, this new law is the definitive guide. It completes the map of quantum speed limits, showing that while energy fluctuations set one kind of speed limit, the average energy sets a different, often stricter one for the things we actually measure. By proving that this limit is quadratic rather than linear, the work provides a clearer, more accurate picture of the fundamental constraints on time and change in the quantum realm.

Technical Summary: A Margolus–Levitin Speed Limit for Observables

Problem Statement
Quantum speed limits (QSLs) traditionally quantify the minimum time required for a quantum state to evolve to an orthogonal state. Two foundational bounds exist: the Mandelstam–Tamm (MT) bound, which relies on energy variance ( $\Delta H$ ), and the Margolus–Levitin (ML) bound, which relies on the mean energy above the ground state ( $\langle H \rangle - E_0$ ). While QSLs for the evolution of observables (the rate of change of an expectation value $\langle A(t) \rangle$ ) have been established, they have exclusively utilized the energy variance (the MT lineage). The mean energy resource, which drives the ML bound for states, has not previously been applied to bound the change $\Delta \langle A \rangle$ of a bounded observable in a fixed state. This paper addresses this gap, investigating whether a mean-energy bound exists for observable evolution and determining its functional form.

Methodology and Theoretical Framework
The author employs a geometric and analytic approach to derive the bound, working in units where $\hbar=1$ and shifting the ground energy $E_0=0$ such that $H \succeq 0$ . The derivation proceeds through several key steps:

Geometric-Energetic Composition: The author decomposes the problem into two parts. First, the change in an observable $\Delta = |\langle A(T) \rangle - \langle A(0) \rangle|$ is bounded geometrically using Hölder's inequality and the trace distance, relating it to the state fidelity $F = |\langle \psi(0)|\psi(T) \rangle|$ . Specifically, $\Delta \leq 2\sigma_A \sqrt{1-F^2}$ , where $\sigma_A$ is the spectral spread of the observable. Second, the ML bound is applied to the state fidelity, yielding $1-F \leq K T \langle H \rangle$ , where $K$ is a constant derived from a tangent-line lemma.
The Tangent-Line Lemma: A critical mathematical component is the determination of the constant $K = \sup_{x>0} (1-\cos x)/x$ . This supremum is attained at $x_\star \approx 2.3311$ , the smallest positive root of $\tan(x/2) = x$ , giving $K = \sin x_\star \approx 0.7246$ .
No-Go Theorem: The author first proves that no linear mean-energy bound of the form $T(\langle H \rangle - E_0) \geq \kappa \Delta / \sigma_A$ can exist. They demonstrate that for near-ground two-level systems, the ratio of the required energy-time product to the linear change vanishes as the change $\Delta \to 0$ , forcing the exponent of $\Delta$ to be exactly 2.
Extension to Mixed States: The bound is extended to mixed states using the joint convexity of the trace distance, avoiding a factor-of-two loss that would occur via Uhlmann fidelity.

Key Results

The Sharp Quadratic Bound: The paper establishes the following inequality for any time-independent Hamiltonian $H$ , bounded Hermitian observable $A$ , and state (pure or mixed):
$T (\langle H \rangle - E_0) \geq C_\star \frac{\Delta^2}{\sigma_A^2}$
where $\Delta = |\langle A(T) \rangle - \langle A(0) \rangle|$ , $\sigma_A = (\lambda_{\max} - \lambda_{\min})/2$ , and the dimension-independent constant is:
$C_\star = \frac{1}{8 \sin x_\star} \approx 0.172506267461\dots$
This bound is tight; it is approached (though not strictly attained) by a near-ground two-level family in the limit of small excitation.
Exact Trade-Off Curve: Beyond the asymptotic constant, the author derives the exact, tight trade-off curve $P_\star(\delta)$ relating the energy-time budget $P = T(\langle H \rangle - E_0)$ to the normalized swing $\delta = \Delta / 2\sigma_A$ . This curve interpolates between the quadratic behavior at small swings and a limit of $\pi/2$ as $\delta \to 1$ .
Refinements and Extensions:
- Bandwidth-Limited Generators: If the Hamiltonian spectrum is restricted to a bandwidth $B$ , the constant sharpens to $C_\star(BT)$ , becoming significantly stronger for narrow-band, short-time operations.
- Multiple Observables: The bound generalizes to simultaneous constraints on multiple observables, scaling with the Chebyshev radius of the operator combination.
- Linear Recovery: The quadratic law degrades to a linear law if the initial state is an eigenvector of the observable $A$ . In this specific case, the bound recovers the linear ML state bound form.
Numerical Verification: The author performed extensive numerical verification across dimensions $d=2$ to $10$ and over $1.5 \times 10^6$ sampled configurations, confirming that no configuration violates the bound and that the infimum is indeed $C_\star$ .

Significance and Scope
The paper claims to complete the "speed-limit dichotomy" (Table I), filling the missing corner where the resource is mean energy and the target is an observable. The primary significance is structural: it proves that the mean energy controls observable change quadratically, not linearly. Consequently, this bound is most constraining in regimes where linear variance-based bounds are weakest—specifically for large fractional changes in near-ground states.

The author identifies the "cleanest physical home" for this bound in autonomous quantum clocks, where it provides a coherent mean-energy resolution floor complementary to existing entropy and rate bounds. It is also discussed in the context of autonomous quantum batteries as a mean-energy floor for charging, though the author notes this is loose for typical fast gates.

The paper explicitly delimits the scope of the result:

It applies only to time-independent Hamiltonians; it does not constrain standard time-dependent quantum gates.
For extensive many-body systems, the bound is weak for local observables due to the scaling of total mean energy.
For unbounded observables (e.g., position), the bound is vacuous.

In summary, this work establishes a fundamental, quadratic mean-energy speed limit for observables, providing a model-independent certificate for mean energy based on observable swings and completing the theoretical landscape of quantum speed limits.

Technical Summary: A Margolus–Levitin Speed Limit for Observables

More like this