Search arXiv⌕ Search

arXiv subjects

Sareen Sabra

Publications and source records attributed to Sareen Sabra.

1 recordsLinked to original sources

Seeding and Amplification of Chaos in Long-Range Interacting Spin Chains

Truncating weak interactions is a natural simplification of a many-body Hamiltonian, but the validity of this approximation need not be uniform in distance and time. In a chaotic system, a weak long-range coupling can seed a small disturbance far from its source, after which the local dynamics can amplify that disturbance exponentially. We study this mechanism in an infinite-temperature classical Heisenberg spin chain with a dominant nearest-neighbor exchange and a weak interaction tail. Using a classical decorrelator, we compare exponentially and power-law decaying tails over several decades in interaction strength. For exponentially decaying interactions, the long-distance front remains ballistic when the tail is sufficiently shallow, with a velocity controlled by the competition between the spatial decay of the seed and the local chaotic growth. For sufficiently rapid exponential decay, the nearest-neighbor front, which is defined by an intrinsic butterfly velocity $v_b$, is recovered within the accessible space-time window. For power-law interactions, the arrival time is logarithmic in distance for all exponents studied, corresponding to an exponentially expanding spatial front; the power-law exponent controls the coefficient of this logarithm. We interpret both behaviors using a seeding-and-amplification picture in which long-range couplings generate weak disturbances throughout the chain whose growth is governed primarily by the local nearest-neighbor dynamics. In this picture, the overall strength of the tail primarily controls when the long-range front becomes visible, while its spatial decay controls the propagation law. Thus a weak interaction tail can leave conventional spin-dynamical observables essentially unchanged while qualitatively altering the long-distance propagation of decorrelation.

cond-mat.stat-mech↗