[Paper Review] Optical Coherence Tomography with a nonlinear interferometer in the high parametric gain regime

Gerard J. Machado, Gaetano Frascella+2 more|arXiv (Cornell University)|Jun 3, 2020

Optical Coherence Tomography Applications39 references46 citations

TL;DR

The paper demonstrates optical coherence tomography using an SU(1,1) nonlinear interferometer in the high parametric gain regime, enabling high photon flux, standard detectors, and increased sensitivity to small losses.

ABSTRACT

We demonstrate optical coherence tomography based on an SU(1,1) nonlinear interferometer with high-gain parametric down-conversion. For imaging and sensing applications, this scheme promises to outperform previous experiments working at low parametric gain, since higher photon fluxes provide lower integration times for obtaining high-quality images. In this way one can avoid using single-photon detectors or CCD cameras with very high sensitivities, and standard spectrometers can be used instead. Other advantages are: higher sensitivity to small loss and amplification before detection, so that the detected light power considerably exceeds the probing one.

Motivation & Objective

• Motivate and demonstrate OCT with a nonlinear interferometer operating at high parametric gain to improve image acquisition speed and sensitivity.
• Show that high-gain operation yields higher photon flux, enabling ordinary detectors while maintaining or improving axial resolution.
• Compare high-gain performance to low-gain regimes, and illustrate both time-domain and Fourier-domain OCT in this architecture.

Proposed method

• Use an SU(1,1) interferometer formed by two passes through a nonlinear crystal pumped by a 532 nm Nd:YAG laser to generate correlated signal and idler photons.
• Operate the system at high parametric gain G (measured G = 1.7 ± 0.2 for the first pass) to achieve large photon flux (≈ 13,000 idler photons per pulse probing the sample and ≈ 4×10^5 signal photons detected per pulse).
• Derive and use the multimode interference visibility V as a function of sample and system reflectivities (Eq. 4 in the paper) and show its nonlinear dependence on r_s and r_i.
• Implement both time-domain OCT and Fourier-domain OCT by scanning phase for fringes or analyzing the spectrum of the output signal, with path-length difference constraints (Eq. 8).
• Provide analytic expressions for the spectrum S(Ω) of the signal photons (Eq. 1) and the Bogoliubov transformations governing the first and second pass through the crystal (S1–S4 in Supplementary Material).
• Demonstrate axial resolution and spectral analysis via Fourier transform of the detected spectrum, yielding an OCT depth profile.

Experimental results

Research questions

• RQ1 Does operating an SU(1,1) interferometer in the high parametric gain regime improve OCT performance compared with low-gain regimes?
• RQ2 Can high-gain OCT utilize standard detectors (CCD cameras or spectrometers) while maintaining image quality and sensitivity to small losses?
• RQ3 What is the influence of sample reflectivity on interference visibility in high-gain SU(1,1) OCT, and how does it compare to low-gain behavior?
• RQ4 How can Fourier-domain OCT be realized in this nonlinear interferometer, and what axial resolutions are achievable given the PDC bandwidth?
• RQ5 What are the practical considerations (path-length difference, losses, and spectral resolution) for implementing high-gain OCT in this architecture?

Key findings

• Demonstrated interference visibility up to 90% in high-gain OCT with r_s = 0.6 and varying r_i.
• Found nonlinear dependence of visibility on sample reflectivity, with high gain enhancing sensitivity to small losses.
• In the very high-gain regime, visibility approaches V = 2|r_s|/(|r_s|^2 + |r_i|^2) × |r_i|, and for matched losses (r_s = r_i) visibility equals 1.
• FD-OCT is feasible in this setup, with an axial resolution of approximately 60 μm under the current bandwidth and spectrometer constraints.
• Achieved high photon flux allowing the use of standard detectors (CCD or spectrometer) rather than single-photon detectors, while probing with near-infrared light and detecting in the visible range.
• Provided a practical framework showing that detected power can exceed probing power due to parametric amplification after interaction with the sample.