Relayed-loop optical scan amplification

Harishankar Jayakumar · 2025

Read Relayed-loop optical scan amplification by Harishankar Jayakumar (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-18399.txt.

Introduction

Many tools sweep a laser beam across points in sequence, from sensors to fabricators. Harishankar Jayakumar and six co-authors present a way to stretch how far one sweeper can reach. Their 2025 paper tackles a known trade: faster sweepers can address fewer spots. Passive optics could double the reach at best.

Their method, called ReLOAD, loops the bent beam back onto the same deflector inside an optical cavity. Each pass multiplies the sweep range. With an electro-optical deflector, prized for speed but short reach, they reached eight times the normal range. The setup hit frame rates of ten thousand per second and line rates of one million per second. Step times fell to millionths of a second. The reach went well past what such deflectors can usually address.

Catalog record

CREATORHarishankar Jayakumar
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE42 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-18399

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Relayed-loop optical scan amplification Harishankar Jayakumar Abstract Abstract Many modern sensing, processing, and fabrication technologies depend upon the sequential scanning of laser light. Due to inertial, thermal, and electrical limitations, the speed of a typical laser deflector is inversely related to its optical invariant and therefore the number of resolvable spots it can address. Passive optical systems have been developed that can effectively double the optical invariant and maximum throughput of a single deflector. We present a method for increasing the effective optical invariant beyond 2 \times by repeated relaying of the deflected beam onto the deflector in an optical loop created within a cavity. We use this Relayed-Loop Optically Amplified Deflection (ReLOAD) approach to accomplish 8 \times amplification of electro-optical (EO) deflection, a deflection technology with unparalleled speed, but limited optical invariant. Using ReLOAD, we demonstrate 10 kHz frame rate imaging, 1 MHz line scan rate, and \mu s step times across an addressable space well beyond the capabilities of typical electro-optical deflectors. High-speed laser deflectors find a wide range of applications, including imaging, mapping, manufacturing, information storage, and computation [ 1 ] . The utility of a deflector across these domains is strongly influenced by both its operating speed and the number of resolvable spots it can address, as these parameters set limits on the field-of-view (FOV) at a given optical resolution. In certain use cases, both high speed and high number of resolvable spots are critical in order to record [ 2 ] or control [ 3 ] fast processes occurring across many locations. However, inertial, thermal, and electrical constraints typically impose a trade-off between the address rate and the number of resolvable spots. This trade-off is apparent for mechanical deflectors, such as galvanometers and resonant mirrors, as well as non-mechanical deflectors, such as acousto-optical deflectors (AODs) and electro-optical deflectors (EODs). Recently, some optical systems have been developed that circumvent this trade-off, achieving a 2 \times amplification of the effective throughput ( i.e., resolvable spots per second) of reflective deflectors [ 4 , 5 , 6 , 7 ] . Figure 1: Illustrations of transmissive ReLOAD configured for 4 \times amplification. (a) Optical path of a laser beam with no deflection applied at the transmissive deflector. (b) Identical to (a), but with deflection of \theta_{0} applied at the deflector. (c) xz cross-section at the inject/eject mirror (orange solid), yz projection at the deflector (teal dashed) and xz projection at the tilted end mirror (light blue dotted) shows the beam paths traversed by an undeflected (top panels) and deflected (bottom panels) beam. Black dashed lines indicate locations of the inject and eject mirrors and the dotted lines indicate the extent of the bidirectional deflection. The number of arrow heads indicate the loop number. To achieve scan amplifications greater than 2 \times , we have designed a method that repeatedly relays the deflected beam back to the deflector in an optical loop within a cavity [ 8 ] (Fig. 1 ), which we call Relayed-Loop Optically Amplified Deflection (ReLOAD). Here, we focus on an implementation of ReLOAD for amplifying a transmissive non-mechanical deflector. However, ReLOAD is a versatile N -fold deflection amplification scheme that is also compatible with reflective mechanical deflectors (Fig. S1 , see Visualization 1 ). Conceptually, ReLOAD requires a cavity with a loop dimension ( xz Fig. 1 (a)) that sets the number of interactions of light with the active deflector before ejection from the cavity. This loop dimension is orthogonal to the active deflection dimension ( yz ), allowing the amplification factor to be set independently. In our physical implementations of ReLOAD, an optical cavity is realized with afocal relays and a deflector placed at the afocal plane. For transmissive deflectors, the deflector is placed at the common afocal plane of two sequential afocal relays (Fig. 1 ). Light enters the cavity via a small inject mirror placed at the focal plane of the cavity. A loop control mirror on one end of the cavity is tilted at an angle \phi_{0} (Fig. 1 (c)). The applied angle (2 \phi_{0} ) causes the beam to follow a converging figure-eight trajectory, with each subsequent loop spanning a smaller angle in the loop dimension. In this way, the beam interacts with the deflector multiple times – accumulating additional angle ( \theta_{0} ) along the deflection axis with each pass (Fig. 1 (b)), so that the deflection angle after N passes is N \theta_{0} . The looping beam exits the cavity by reflecting off an eject mirror placed at the focal plane (Fig. 1 (c)). The number of loops is determined by the tilt angle ( \phi_{0} ) of the loop control mirror, with the maximum number of loops limited by the width of the eject mirror, as well as other practical constraints ( e.g., acceptance angle of the system, cumulative aberrations, and cumulative power loss). To demonstrate the utility of ReLOAD, we applied it to electro-optical deflection (Fig. 2 ), a deflection technology with unparalleled speed, but low optical invariant and number of resolvable spots. We constructed an EOD based on the perovskite crystal, potassium tantalate niobate (KTN, KTa 1-x Nb x O 3 ). In its paralectric phase, the large quadratic electro-optic coefficient and a high dielectric constant of KTN [ 9 ] yields an unusually large deflection angle per unit of electric field compared to other EOD technologies [ 10 , 11 ] . KTN-based EODs can operate at up to MHz rates [ 12 , 13 ] . However, EODs designed for high-speed ( > 50 kHz line rate) operation can address a very limited number of resolvable spots due to a combination of issues, including clear aperture of the crystal, EOD capacitance, and electroelastic damage [ 14 , 13 ] . Commercial KTN-based deflectors can address a maximum of \approx 32 resolvable spots at a 200 kHz line rate and only \approx 4 resolvable spots at 1 MHz line rate [ 15 ] . Our custom-built KTN-based EOD (Fig. S2 ) contains a 4 mm \times 3.15 mm \times 1.2 mm KTN crystal plated with layered titanium/platinum/gold electrodes (NTT Advanced Technology Corp.). A broadband magnesium fluoride anti-reflective coating was applied to transmissive faces via electron beam evaporation. Copper interfaces provided electrical contact with the crystal and thermal contact with thermoelectric modules that maintained the KTN at 5 °C above its Curie temperature (T c = 23 °C), ensuring operation in the paraelectric phase (see Supporting Information ). The deflector was operated in a space-charge mode [ 16 ] by applying a temporary 400 VDC bias to fill electron traps in the crystal. According to Gauss’s law, this charge results in a 1-dimensional parabolic gradient refractive index profile (cylindrical convex GRIN lens) orthogonal to the electrodes [ 17 ] , with a back focal length of \approx 32-40 mm (Fig. 2 (a)). Under our operating conditions, charges remained trapped in the crystal for hours [ 18 ] . Figure 2: Illustrations and schematics of EO-ReLOAD with a KTN EOD deflector. (a) A cylindrical convex gradient index (GRIN) formed within the charged KTN (b) Drive voltage shifts the optical axis of the GRIN, deflecting the beam. Red line: pivot plane of the deflected beam (c) Increasing

Read the complete text