Galilean Invariance in the Characterization of Light Drag in Moving Atomic Vapor

TL;DR

This work experimentally verifies Galilean invariance for transverse light drag in a highly dispersive slow-light medium by comparing two inertial frames: a moving medium with a stationary probe and a moving probe with a stationary medium, using non-degenerate Zeeman EIT in rubidium. The authors harness slow light to produce large group delays $\tau$ and measure the lateral displacement $\Delta x$, confirming the relation $\Delta x = v\tau$ across both configurations. Observed delays in the tens of microseconds and linear $\Delta x$ versus velocity demonstrate frame-invariant transverse drag, enabling precise velocimetry and potential stand-off sensing in challenging environments. The setup offers a robust platform for exploring light-matter interactions in moving media and could be extended with optical storage and gravitational-field tests to enhance sensitivity and broaden applications in quantum information and metrology.

Abstract

Light experiences drag effects when it propagates through a moving medium. The study of light drag has provided foundational insights into light-matter interactions. While longitudinal drag has been extensively characterized, transverse drag, where the medium moves perpendicular to the light's propagation, is subtler and requires advanced techniques for detection. In this work, we experimentally investigate transverse drag in a highly dispersive slow-light medium using non-degenerate Zeeman electromagnetically induced transparency (EIT) in rubidium vapor. By systematically comparing configurations where the light beam and the medium serve as the moving frame, we leverage Galilean invariance to analyze transverse light-drag in this optical context. Thus, we provide a platform for future tests of fundamental principles on strong experimental grounds, which offers promising applications in precision velocimetry, accelerometry, quantum information, and light storage technologies.

Figures (5)

• Figure 1: (a) Schematic representation of the transverse light-drag experiment. The beam experiences a lateral displacement as it propagates through the moving rubidium cell. (b) Non-degenerate Zeeman EIT atomic configuration
• Figure 2: Experimental setup to generate and measure slow light from non-degenerate Zeeman EIT.
• Figure 3: Probe and reference pulses shown on the oscilloscope. The red and blue curves are Gaussian fits of the signals for two different extinction ratios between the control and probe beams, achieved by adjusting the probe beam power.
• Figure 4: : Left: (a) Transverse drag enhanced by slow light in a medium moving at $v=350 \text{ }\mathrm{mm/s}$; the green arrow indicates the direction of motion. Right: Experimental configurations to investigate transverse light-drag in moving media. Both depicted setups display only the part of the setup used to measure transverse light drag using slow-light induced by non-degenerate Zeeman EIT. (b) Setup for a moving medium with a stationary probe beam. The trigger mechanism consists of an extension barrier attached to the moving medium, and an iris placed such that the reference beam is temporally aligned with the probe beam exactly when the probe passes through the center of the Rb cell. (c) Setup for a moving probe beam with a stationary medium; we used a PBS to extract part of the moving probe beam to trigger the camera, exactly when the probe passes through the center of the Rb cell.
• Figure 5: Transverse drag measurement versus velocity, for the cases when the motion is in the left direction (negative speeds) and in the right direction (positive speeds). (a) Moving medium frame: The linear fits are in good agreement with expected values measured of time delays of $36$, $40$ and $60 \mu s$, within less than 10%. (b) Moving probe frame: The linear fits are also in good agreement with that expected from the measured time delays of $24$, $30$ and $40 \mu s$, within less than 10%. The data points are the means of 50 measurements, and the small error bars are their standard deviations.