Time’s Arrow Signals Quantum Phase Shifts

Author: Denis Avetisyan

New research reveals a surprising link between the direction of time and the critical behavior observed during measurement-induced phase transitions in quantum systems.

The system demonstrates a transition in the adiabaticity of a single qubit governed by a specific Hamiltonian <span class="katex-eq" data-katex-display="false">Eq. (13)</span>, where, at extended time scales, the ratio of adiabaticity to time remains consistent, and the excited-state probability saturates above a critical measurement rate, indicating a fundamental limit to sustained quantum coherence. — The system demonstrates a transition in the adiabaticity of a single qubit governed by a specific Hamiltonian $Eq. (13)$ , where, at extended time scales, the ratio of adiabaticity to time remains consistent, and the excited-state probability saturates above a critical measurement rate, indicating a fundamental limit to sustained quantum coherence.

The arrow of time acts as a thermodynamic indicator of measurement-induced phase transitions, detectable through changes in quantum entanglement and non-equilibrium dynamics.

Diagnosing quantum phase transitions typically relies on entanglement measures, yet a complementary thermodynamic perspective remains largely unexplored. In ‘Arrow of Time as an indicator of Measurement-Induced Phase Transitions’, we introduce the arrow of time-a quantification of irreversibility arising from repeated quantum measurements-as a novel diagnostic for these transitions. Our analysis of random quantum circuits reveals that the arrow of time exhibits non-analytic behavior at the critical point, mirroring the hallmarks of a phase transition, and establishing it as a locally accessible order parameter. Does this thermodynamic lens offer a more intuitive understanding of measurement-induced phenomena and pave the way for characterizing broader non-equilibrium dynamics in quantum systems?

The Fragile Bloom of Irreversibility

Quantum systems possess a unique characteristic – entanglement – where two or more particles become linked, sharing the same fate regardless of the distance separating them. This interconnectedness isn’t found in classical physics, and represents a powerful resource for tasks like quantum computation and communication. However, this delicate quantum state is exceptionally vulnerable; any attempt to observe or measure the system fundamentally disturbs the entanglement. This isn’t a limitation of experimental technique, but an inherent property of quantum mechanics – the very act of gaining information necessitates a disruption of the fragile correlations that define entanglement, potentially collapsing the quantum state and forcing it towards classical behavior. $\psi = \sum_{i} c_{i} |i\rangle$ describes this superposition, easily disturbed by observation.

Measurement-induced phase transitions (MIPTs) denote a fundamental alteration in the way quantum entanglement-a cornerstone of quantum mechanics-is organized within a many-body system. These transitions aren’t driven by changes in external parameters, but rather by the act of measurement itself. As measurements are repeatedly applied, entanglement doesn’t simply degrade gradually; instead, it undergoes a sudden, qualitative shift, restructuring from a widespread, interconnected resource into localized, fragmented pockets. This reorganization signifies a loss of quantum coherence – the ability of a system to exist in multiple states simultaneously – and marks a critical boundary where quantum correlations effectively break down. Consequently, MIPTs demonstrate how classical behavior can emerge from purely quantum systems, driven not by inherent properties, but by the unavoidable interaction with an observing system-a process central to understanding the limits of quantum technologies and the very nature of the quantum-to-classical transition.

The significance of measurement-induced phase transitions (MIPTs) extends beyond fundamental quantum mechanics, directly impacting the viability of quantum technologies. These transitions define a critical boundary where the delicate resource of quantum entanglement-essential for tasks like quantum computation and communication-collapses into classical correlations. Determining the precise conditions under which MIPTs occur is therefore paramount; it establishes the limits of how complex a quantum system can be before its quantum advantages are lost to decoherence and measurement backaction. Consequently, research into MIPTs isn’t merely an academic exercise, but a necessary step in engineering robust quantum devices and ultimately understanding the very border between the quantum and classical worlds.

Modifying measurements in a random unitary circuit of <span class="katex-eq" data-katex-display="false">LL</span> qudits with local Hilbert space dimension <span class="katex-eq" data-katex-display="false">q</span> allows mapping the system's density matrix to a statistical mechanics model on the honeycomb lattice, which simplifies to a square lattice in the large <span class="katex-eq" data-katex-display="false">q</span> limit and ultimately reveals a connection between the minimum information perturbation test and bond percolation. — Modifying measurements in a random unitary circuit of $LL$ qudits with local Hilbert space dimension $q$ allows mapping the system’s density matrix to a statistical mechanics model on the honeycomb lattice, which simplifies to a square lattice in the large $q$ limit and ultimately reveals a connection between the minimum information perturbation test and bond percolation.

Charting the Quantum Trajectory

The Quantum Trajectory Framework facilitates the simulation of quantum dynamics by representing the evolution of a quantum state as a series of conditional states, each determined by a specific measurement outcome. Rather than propagating a single wavefunction, this approach propagates an ensemble of trajectories, where each trajectory is updated based on the result of a continuous or discrete measurement. This conditioning on measurement outcomes allows for a detailed reconstruction of the system’s evolution, effectively mapping the complex dynamics onto a stochastic process. The framework is particularly useful for systems undergoing frequent measurements, as it naturally incorporates the information gained from each measurement into the subsequent trajectory, providing a computationally efficient method for analyzing non-equilibrium quantum dynamics and implementing feedback control.

Continuous measurements, as employed within the quantum trajectory framework, utilize invertible Kraus operators to update the quantum state based on information gained from ongoing, weak measurements. Unlike projective measurements which cause immediate state collapse, these operators allow for a smooth evolution of the wavefunction conditioned on the measurement record. The invertibility of the Kraus operators is crucial, enabling the reconstruction of the pre-measurement state and the consistent tracking of individual quantum trajectories. This approach avoids the discontinuities inherent in standard measurement schemes, providing a detailed and continuous description of the system’s dynamics as it unfolds, and allowing for the calculation of conditional wavefunctions $\psi(t| \{x_{\tau}\})$ based on the history of weak measurement outcomes $\{x_{\tau}\}$ .

No-click dynamics represent a specific implementation of continuous measurement in quantum mechanics where the measurement process is designed to avoid the instantaneous state collapse typically associated with projective measurements. This is achieved through the use of Kraus operators that evolve the quantum state continuously, providing information about the system’s evolution without abruptly forcing it into a definite eigenstate. The resulting dynamics are described by a stochastic Schrödinger equation, where the stochastic force is determined by the measurement record. This allows for the simulation of quantum trajectories that are smooth and continuous, facilitating the investigation of quantum phenomena without the artificial effects of discrete measurement events, and is particularly useful for modeling weak measurements and post-selection schemes.

Trajectories of a four-state qudit under different dynamics-measurement only, perturbed identity, and Haar-random unitaries-reveal a consistent increase in the average time to teleportation (<span class="katex-eq" data-katex-display="false">\text{AoT}</span>) except in the Haar-random case where the state is driven away from measurement eigenstates, and the expectation values of the measurement operators (<span class="katex-eq" data-katex-display="false">\langle P^j \rangle</span>) do not distribute uniformly. — Trajectories of a four-state qudit under different dynamics-measurement only, perturbed identity, and Haar-random unitaries-reveal a consistent increase in the average time to teleportation ( $\text{AoT}$ ) except in the Haar-random case where the state is driven away from measurement eigenstates, and the expectation values of the measurement operators ( $\langle P^j \rangle$ ) do not distribute uniformly.

Constructing the Seeds of Disorder

Random Quantum Circuits (RQCs) are constructed by applying a sequence of randomly chosen unitary gates to an initial quantum state, typically $|0\rangle^{\otimes n}$ . The universality of RQC-generated states allows them to approximate any other disordered quantum state within a certain fidelity. Each gate in the circuit is selected independently from the Haar measure over the unitary group $U(d)$ , where d is the local Hilbert space dimension. The randomness inherent in gate selection creates complex entanglement and correlations, resulting in a highly disordered initial state suitable for benchmarking Many-Body Localization (MBL) and probing the ergodic-to-localization transition. The depth of the circuit, or the number of gates applied, controls the degree of disorder and entanglement present in the generated state.

The Haar Random State is a quantum state obtained by averaging over the uniform Haar measure on the space of all pure quantum states. This results in a maximally disordered state, meaning it possesses the highest possible entropy and lacks any preferential direction in Hilbert space. Consequently, the Haar Random State is frequently employed as a stringent benchmark for evaluating the capabilities of Measurement-induced Phase Transitions (MIPTs). Specifically, the ability of a system to exhibit a phase transition when initialized in a Haar Random State indicates a high degree of universality, suggesting the transition is not sensitive to specific initial conditions and relies on fundamental properties of the underlying dynamics.

The statistical properties of disordered quantum states generated by random quantum circuits are characterized by the distribution of local expectation values. These expectation values are effectively modeled using the Beta distribution, providing a quantifiable measure of disorder. Specifically, the variance of these local expectation values exhibits a scaling relationship with the local dimension, $q$ , and the number of qudits, $L^{\prime}$ , following the equation variance ∝ 1/ $q$ ^(L’+1). This inverse relationship indicates that as either the local dimension or the number of qudits increases, the variance of the local expectation values decreases, suggesting a trend towards more ordered, less disordered states.

Distributions of local expectation values <span class="katex-eq" data-katex-display="false">\langle P^j \rangle</span> for Haar-random states of <span class="katex-eq" data-katex-display="false">L^{\\prime}</span>-qudits with dimension <span class="katex-eq" data-katex-display="false">q</span> reveal that qubit distributions narrow with increasing system size, higher local dimensions shift the distribution's peak towards <span class="katex-eq" data-katex-display="false">1/q</span>, and the distribution dramatically changes between volume-law (<span class="katex-eq" data-katex-display="false">L^{\\prime}=\\mathcal{O}(L)</span>) and area-law (<span class="katex-eq" data-katex-display="false">L^{\\prime}=\\mathcal{O}(1)</span>) phases, as modeled by Beta distributions with variance scaling as <span class="katex-eq" data-katex-display="false">1/q^{L^{\\prime}+1}</span>. — Distributions of local expectation values $\langle P^j \rangle$ for Haar-random states of $L^{\\prime}$ -qudits with dimension $q$ reveal that qubit distributions narrow with increasing system size, higher local dimensions shift the distribution’s peak towards $1/q$ , and the distribution dramatically changes between volume-law ( $L^{\\prime}=\\mathcal{O}(L)$ ) and area-law ( $L^{\\prime}=\\mathcal{O}(1)$ ) phases, as modeled by Beta distributions with variance scaling as $1/q^{L^{\\prime}+1}$ .

Echoes of Universality and the Flow of Time

The ReplicaTrick provides a sophisticated analytical approach to unraveling the behavior of complex quantum systems, specifically measurement-induced phase transitions (MIPTs). This technique sidesteps the intractable problem of directly averaging over the vast landscape of possible quantum circuits by employing a mathematical ‘replica’ – effectively creating multiple identical copies of the system and calculating their correlations. While seemingly abstract, this allows researchers to translate the calculation of average properties, like entanglement or the spread of information, into a more manageable problem involving these replicas. The resulting equations, though often complex, reveal crucial insights into how quantum information behaves under repeated measurements, and ultimately define the characteristics of the MIPT. This powerful tool has become central to understanding the emergence of many-body localization and the breakdown of thermalization in these driven quantum systems.

Recent investigations into measurement-induced phase transitions (MIPTs) have unveiled a remarkable connection to the well-established field of classical critical phenomena, specifically mirroring the behavior of bond percolation. This isn’t merely a superficial resemblance; researchers have demonstrated that MIPTs exhibit a critical exponent of -2/3 – a value precisely matching that observed in bond percolation. This shared exponent signifies that the way entanglement spreads and ultimately collapses under repeated measurements belongs to the same “universality class” as the formation of connected clusters in a randomly filled lattice. Consequently, techniques and insights developed for understanding percolation – a cornerstone of statistical physics – can now be applied to unravel the complexities of these quantum many-body systems, offering a powerful new lens through which to view the emergence of quantum chaos and the breakdown of thermalization.

The directionality of time, often termed the ‘Arrow of Time’, finds a surprising connection to the behavior of quantum systems undergoing measurement. Researchers have demonstrated that this temporal asymmetry can be quantified using $ShannonEntropy$ , effectively measuring the degree of irreversibility within the evolving quantum state. This metric isn’t static; it reveals a distinct transition point that precisely coincides with the measurement-induced phase transition-a critical moment where the system’s behavior fundamentally shifts. This suggests that the emergence of temporal directionality isn’t an inherent property of quantum mechanics, but rather an emergent phenomenon linked to the specific process of measurement and the resulting entanglement structure, offering a novel perspective on how irreversibility arises in quantum systems.

Analysis of the adiabaticity of time (AoT) in a random circuit of LLqudits reveals a transition between continuous and projective measurements parameterized by α, which is accurately modeled using bond percolation and exhibits a finite-size scaling behavior of <span class="katex-eq" data-katex-display="false"> |p-p_{c}|^{2/3} </span> at large system sizes. — Analysis of the adiabaticity of time (AoT) in a random circuit of LLqudits reveals a transition between continuous and projective measurements parameterized by α, which is accurately modeled using bond percolation and exhibits a finite-size scaling behavior of $|p-p_{c}|^{2/3}$ at large system sizes.

The study illuminates how emergent order arises not from imposed control, but from the collective behavior of quantum measurements. This resonates with da Vinci’s observation: “Nature never breaks her laws; she only bends them.” The research demonstrates that the arrow of time, functioning as a thermodynamic indicator, signals critical behavior at measurement-induced phase transitions. Rather than attempting to control the system’s evolution, the researchers observe how local measurement choices lead to global changes in entanglement and irreversibility. This mirrors the principle that influence, revealed through the system’s response to measurement, is far more potent than direct control, allowing for the emergence of complex, ordered behavior from inherently random quantum circuits.

Where Do the Currents Flow?

The observation that the arrow of time-that stubborn insistence on unidirectional change-mirrors the critical behavior at measurement-induced phase transitions feels less like a discovery and more like a recognition. The forest evolves without a forester, yet follows rules of light and water; similarly, irreversibility isn’t imposed on these quantum systems, but emerges from the local interactions shaped by measurement. The challenge now isn’t to engineer a direction for time, but to map the landscapes where these transitions-and their attendant arrows-naturally arise.

Current work, while illuminating, remains largely confined to the study of random quantum circuits. A natural progression lies in extending these insights to more physically realistic systems – those with inherent structure and limited connectivity. Can one identify analogous transitions-and corresponding thermodynamic signatures-in condensed matter systems, or even in biological processes where ‘measurement’ takes the form of environmental decoherence? The replica symmetry breaking, while a powerful tool, hints at a deeper, underlying complexity that deserves further investigation.

Ultimately, the pursuit isn’t about ‘controlling’ the arrow of time, but about understanding the conditions that give rise to its perceived direction. Order is the result of local interactions, not directives. The true utility of this work may lie not in its predictive power, but in its capacity to shift the question: not how to impose order, but where to look for it, already self-organized within the quantum substrate.

Original article: https://arxiv.org/pdf/2604.20828.pdf

Contact the author: https://www.linkedin.com/in/avetisyan/

The Fragile Bloom of Irreversibility

Charting the Quantum Trajectory

Constructing the Seeds of Disorder

Echoes of Universality and the Flow of Time

Where Do the Currents Flow?

See also: