Infrared Time Lags for the Periodic Quasar PG 1302-102

TL;DR

This study confirms infrared time lags in the quasar PG 1302-102, supporting the hypothesis that optical variability originates from the accretion disk and is echoed in the surrounding dusty torus, consistent with a binary black hole system.

Contribution

First measurement of infrared time lags in PG 1302-102, linking optical periodicity to dust echo and supporting binary black hole interpretation.

Findings

Infrared time lags of ~2200-2400 days at 3.4 and 4.6 μm wavelengths.

Optical variability is echoed in infrared, indicating a dusty torus at ~1.1-1.5 pc.

Supports the binary black hole model with a dusty torus reprocessing the optical signal.

Abstract

The optical light curve of the quasar PG 1302-102 at $z = 0.278$ shows a strong, smooth 5.2 yr periodic signal, detectable over a period of $\sim 20$ yr. Although the interpretation of this phenomenon is still uncertain, the most plausible mechanisms involve a binary system of two supermassive black holes with a subparsec separation. At this close separation, the nuclear black holes in PG 1302-102 will likely merge within $\sim 10^{5}$ yr due to gravitational wave emission alone. Here we report the rest-frame near-infrared time lags for PG 1302-102. Compiling data from {\it WISE} and {\it Akari}, we confirm that the periodic behavior reported in the optical light curve from Graham et al. (2015) is reproduced at infrared wavelengths, with best-fit observed-frame 3.4 and $4.6 \mu$m time lags of $(2219 \pm 153, 2408 \pm 148)$ days for a near face-on orientation of the torus, or $(4103\pm 153, 4292 \pm 148)$ days for an inclined system with relativistic Doppler boosting in effect. The periodicity in the infrared light curves and the light-travel time of the accretion disk photons to reach the dust glowing regions support that a source within the accretion disk is responsible for the optical variability of PG 1302-102, echoed at the further out dusty regions. The implied distance of this dusty, assumed toroidal region is $\sim$ 1.5 pc for a near face-on geometry, or $\sim$1.1 pc for the relativistic Doppler boosted case.