A search for optical counterparts in quiescent black hole X-ray transients

This paper presents ULTRACAM photometric observations and public survey data to identify optical counterparts for nine quiescent black hole X-ray transients, resulting in the first optical identifications and precise astrometry for four targets, refined coordinates for an active source, and magnitude limits for five others to constrain their companion star properties.

The Gist

Imagine the universe as a giant, dark ocean. Most of the time, the most dangerous creatures in this ocean—black holes—are sleeping. They are invisible, hiding in the dark, waiting for a snack. But occasionally, they wake up, grab a passing star, and start eating. This creates a massive, bright flare of X-rays and light, like a lighthouse beam cutting through the fog. Astronomers call these "X-ray transients."

The problem is that once the black hole finishes its meal, it goes back to sleep. The light fades away, and the black hole becomes invisible again. To truly understand these cosmic monsters (specifically, how heavy they are), astronomers need to find the "sleeping" black hole and study the faint, dim star that is orbiting it.

This paper is like a night-vision search party. The team of astronomers went out with powerful telescopes to find nine specific black holes that had been spotted eating (in outburst) but had never been seen sleeping (in quiescence).

Here is a breakdown of their adventure:

1. The Mission: Finding the Invisible

Out of 73 known black hole candidates in our galaxy, astronomers have only successfully found the "sleeping" partner for 34 of them. The other 39 are like ghosts; we know they are there because of the X-ray flares, but we can't see them when they are quiet.

The team picked nine of these "ghosts" to hunt down. They knew roughly where to look based on where the black hole was when it was eating, but they needed to find the exact spot where the faint, sleeping star would be hiding.

2. The Tools: Super-Cameras and Image Stacking

To find these faint stars, they used a special camera called ULTRACAM mounted on a telescope in Chile. Think of this camera as a high-speed, super-sensitive eye.

• The Challenge: The stars are so faint that a single photo would be too blurry or noisy to see them. It's like trying to hear a whisper in a noisy room.
• The Solution: They took hundreds of short photos and stacked them on top of each other, like layering sheets of tracing paper. This "stacking" technique makes the faint signal (the star) stand out against the background noise. They also had to be careful to only stack the clearest photos, ignoring the ones where the atmosphere was wobbly (bad "seeing").

3. The Results: Four Found, Five Still Hiding

After their hard work, the team had a mixed bag of results:

• The Successes (4 Found): They successfully identified the faint, sleeping stars for four of the targets. It's like finding four lost keys in a dark room. For these, they could measure the star's color and brightness.
  • Why does this matter? By measuring the color, they could guess what "type" of star it is (like guessing a person's age by their skin tone). By measuring how much the star dimmed from its "eating" phase to its "sleeping" phase, they could estimate how long it takes the star to orbit the black hole.
• The Misses (5 Still Hidden): For the other five targets, the stars were either too faint or the weather conditions weren't perfect. They couldn't see the stars, but they could set a "limit." They said, "We know the star is fainter than this specific brightness." It's like saying, "The thief is definitely smaller than 6 feet tall," even if you didn't see them.
• The Surprise: One system, 4U 1755-338, was actually still awake! It was still eating and glowing brightly when they looked at it. This allowed them to refine its location, but they couldn't see the "sleeping" version yet.

4. The Detective Work: Solving the Puzzle

Once they found the stars, they played detective to figure out the black hole's secrets:

• The "Amplitude" Clue: They compared how bright the system was when the black hole was eating versus how dim it was when sleeping. The bigger the difference, the shorter the orbit. It's like a swing: if you push it hard (bright outburst), it swings fast (short period). If the swing is slow, the push isn't as dramatic.
• The "Color" Clue: They looked at the color of the faint star. Redder stars are usually cooler and smaller; bluer stars are hotter and larger. This helped them guess the star's "spectral type" (its family name).

5. Why This Matters

Finding these sleeping black holes is the first step to weighing them. Currently, we only know the exact mass of about 20 black holes in our galaxy. This is a tiny sample size, like trying to understand all humans by studying only 20 people.

By finding more "sleeping" black holes and measuring their orbits, astronomers can finally build a better picture of how black holes are born, how they grow, and how common they really are.

In a nutshell: This paper is a report from a team of cosmic detectives who used super-cameras to find nine sleeping black holes. They found four of them, learned a bit about their partners, and set strict limits on where the other five are hiding, paving the way for future discoveries with even bigger telescopes.

Technical Summary

Here is a detailed technical summary of the paper "A search for optical counterparts in quiescent black hole X-ray transients" by Yanes-Rizo et al.

1. Problem Statement

Galactic stellar-mass black holes (BHs) are primarily identified in X-ray transients (XRTs). While 73 BH candidates in XRTs have been identified, only 20 have been dynamically confirmed (via radial velocity measurements). This disparity stems from the difficulty in detecting the faint optical/near-infrared (NIR) counterparts of these systems during quiescence. These faint magnitudes, combined with high interstellar extinction in the Galactic plane, hinder the acquisition of spectra necessary for precise mass determinations. Consequently, the demographic understanding of BH formation is limited by small sample sizes and selection biases.

Objective: To expand the sample of known quiescent BH XRTs by conducting a systematic photometric search for optical counterparts in nine candidate systems lacking reported quiescent identifications, and to refine astrometry for others.

2. Methodology

Sample Selection

The authors selected nine BH XRT candidates from the BlackCAT catalogue that possessed accurate astrometry (positional uncertainty < 1 arcsec) but lacked reported quiescent optical counterparts.

• Prediction: Quiescent $r$-band magnitudes were estimated using typical outburst amplitudes ($\Delta V \approx 6$ mag) or pre-existing NIR magnitudes converted to optical using standard K5 donor colors and extinction maps.
• Target Range: Predicted magnitudes were in the range $r \sim 21 - 24$.
• Additional Targets: Three sources (SWIFT J1539.2-6227, XTE J1817-330, XTE J1818-245) were analyzed using archival outburst data to refine coordinates. XTE J1650-500 was included for astrometric refinement.

Observations

• Deep Imaging: Time-resolved photometry was conducted using the ULTRACAM camera on the 3.5m New Technology Telescope (NTT) in Chile during two campaigns (March 2021 and March 2023).
  • Filters: $u_s, g_s, r_s$ (SDSS super filters).
  • Strategy: Short 20s exposures to build deep images ($r \sim 25-26$) while capturing variability.
  • Image Processing: A "seeing cut-off" filter was applied to select frames with the best atmospheric seeing (FWHM) to maximize Signal-to-Noise Ratio (SNR) in co-added images.
• Archival Data: Analysis of images from DECaPS2, Pan-STARRS, FORS2 (VLT), and LDSS3/IMACS (Magellan) was performed to supplement ULTRACAM data and analyze outburst phases.

Data Analysis Techniques

• Astrometry: Cross-matching with Gaia DR2 using the Gaia image tool to achieve RMS residuals of 0.02–0.08 arcsec.
• Photometry:
  • Aperture Photometry: For isolated stars, calibrated against DECaPS2 or Pan-STARRS.
  • PSF Photometry (DAOPHOT): Used for crowded fields (e.g., XTE J1817-330, XTE J1818-245) to model and subtract light from interloper stars, revealing faint underlying counterparts.
• Physical Constraints:
  • Extinction Correction: Applied Schlafly & Finkbeiner (2011) relations and $N_H$-to-$A_V$ conversions (Güver & Özel 2009) to derive intrinsic colors.
  • Orbital Period ($P_{orb}$): Estimated using the empirical $\Delta V - P_{orb}$ correlation (Shahbaz & Kuulkers 1998).
  • Spectral Type: Constrained by comparing intrinsic colors to standard indices and by calculating mean stellar density ($\bar{\rho}$) derived from $P_{orb}$, assuming Roche-lobe filling main sequence companions.

3. Key Results

Counterpart Detections

The study successfully identified optical counterparts for four targets and provided 3$\sigma$ lower limits for five others.

1. MAXI J1348-630: Detected as a blend with a brighter star. PSF photometry yielded $g=23.6, r=21.18$. Implies $P_{orb} \leq 7.9$ h and spectral type later than K2 V.
2. SWIFT J1539.2-6227: Detected in quiescence ($g=22.40, r=21.19$). Refined coordinates provided. Implies $P_{orb} \leq 6.7$ h (based on color constraints) and spectral type later than K0 V.
3. XTE J1726-476: Detected ($g=24.2, r=23.7$). Implies $P_{orb} \leq 10.6$ h and spectral type later than F8 V.
4. XTE J1817-330: Detected in DECaPS2 ($g=22.33, r=21.24$) after PSF subtraction. Implies $P_{orb} \leq 8.2$ h and spectral type K3 V.

Non-Detections and Limits

For the remaining five targets, no counterparts were detected, providing deep lower limits:

• MAXI J0637-430: $g > 25.8, r > 25.3$. Implies $P_{orb} \leq 4.5$ h (very faint, likely distant or high extinction).
• MAXI J1803-298: $g > 21.2, r > 20.5$. Field confusion prevented detection; deeper observations needed.
• XTE J1818-245: $g > 21.7, r > 20.8$.
• MAXI J1828-249: $g > 22.3, r > 21.7$.
• 4U 1755-338: Found to be active during observations ($g \approx 18.9$). Refined coordinates provided. Quiescent limits suggest it is too faint ($g \sim 28$) for current detection.

Astrometric Refinements

Precise coordinates were established for:

• SWIFT J1539.2-6227: Improved from 0.5" to 0.07" uncertainty.
• XTE J1650-500: Refined from 0.31" to 0.05" uncertainty using archival FORS2 data.
• XTE J1817-330 & XTE J1818-245: Improved radio positions using optical outburst data.

Physical Constraints

Using the empirical $P_{orb} - T_{eff}$ relation (Casares et al. 2025), the authors derived more realistic spectral type constraints (Table 4), generally suggesting late-type main sequence companions (K to M dwarfs) for the detected systems.

4. Significance and Contributions

• Sample Expansion: This work increases the number of BH XRTs with identified quiescent optical counterparts, providing a larger pool of targets for future dynamical mass measurements.
• Methodological Advancement: Demonstrates the efficacy of combining deep, time-resolved photometry (ULTRACAM) with public survey data (DECaPS2, Pan-STARRS) and advanced PSF photometry to resolve counterparts in crowded fields.
• Target Prioritization: The derived constraints on orbital periods and spectral types allow astronomers to prioritize targets for future spectroscopic follow-up. Specifically, the four detected targets with $r < 22$ are prime candidates for time-resolved spectroscopy to confirm BH masses.
• Astrophysical Insights: The results reinforce the correlation between outburst amplitude and orbital period, and suggest that the majority of these systems harbor low-mass main sequence donors, consistent with standard XRT evolution models.
• Future Outlook: The paper highlights that the five non-detected targets will likely require next-generation telescopes (e.g., LSST, ELT) for detection, setting the stage for future observational campaigns.

5. Limitations

The authors note that orbital period constraints derived from the $\Delta V - P_{orb}$ relation are approximate due to the small calibration sample and potential projection effects (inclination). Furthermore, spectral type constraints based on quiescent colors are upper limits, as they do not account for potential contamination from residual accretion disk emission.