B14-65666: Bright Lyman-Break Galaxy System

• B14-65666 is a luminous Lyman-break galaxy system at z≈7.15, noted for its bright UV and far-infrared emissions and detailed multi-phase diagnostics.
• It displays clear merger-induced starburst features with two resolved components and complex kinematics evidenced by high [OIII]/[CII] ratios and large Lyα velocity offsets.
• Advanced ALMA, HST, and JWST observations reveal low metallicity, high ionization, and extreme star formation rates, informing early galaxy evolution in the reionization epoch.

Searching arXiv for recent and foundational papers on B14-65666 and related diagnostics. B14-65666, also known as “Big Three Dragons,” is a luminous Lyman-break galaxy system at $z \simeq 7.15$ in the epoch of reionization. It is one of the brightest unlensed $z>7$ LBGs in the rest-frame ultraviolet continuum and in far-infrared fine-structure lines, and it has become a benchmark target because ALMA, HST, and JWST together resolve its stellar, dust, ionized-gas, and neutral-gas structure. Across the literature it is described as a major merger-induced starburst, an evolved massive merging system, and an advanced merger, with strong [OIII] $88\,\mu{\rm m}$, [CII] $158\,\mu{\rm m}$, and dust-continuum emission, sub-solar metallicity, compact star-forming cores, and an unusually large Ly$\alpha$ velocity offset (Hashimoto et al., 2018, Hashimoto et al., 2022, Sugahara et al., 2024, Jones et al., 2024, Prieto-Jiménez et al., 9 Jul 2025).

1. Identification, redshift, and basic observational status

B14-65666 was first identified in the UltraVISTA survey and subsequently confirmed spectroscopically through ALMA detections of [OIII] $88\,\mu{\rm m}$ and [CII] $158\,\mu{\rm m}$, with Ly$\alpha$ also detected by Subaru/FOCAS (Hashimoto et al., 2018, Hashimoto et al., 2022). Its integrated UV absolute magnitude is reported as $M_{\rm UV}\approx -22.4$ to $-22.5$, making it about $z>7$0–$z>7$1 brighter, or about four times brighter, than the characteristic UV magnitude at $z>7$2 (Hashimoto et al., 2018, Hashimoto et al., 2022, Sugahara et al., 2024).

The systemic redshift from the ALMA FIR lines is reported as $z>7$3 for the S/N-weighted mean of the whole system, with the abstract of the discovery paper quoting $z>7$4 (Hashimoto et al., 2018). For the whole system, the measured line widths are $z>7$5 and $z>7$6 (Hashimoto et al., 2018). Ly$z>7$7 is detected at $z>7$8, corresponding to a velocity offset of $z>7$9, with rest-frame equivalent width $88\,\mu{\rm m}$0 $88\,\mu{\rm m}$1; this is explicitly noted as among the largest Ly$88\,\mu{\rm m}$2 offsets reported at $88\,\mu{\rm m}$3 (Hashimoto et al., 2018).

The source has progressively moved from being a bright FIR-line LBG to a fully resolved multi-phase merger target. Early HST and ALMA work established the two-component structure and the “Big Three” combination of [OIII], [CII], and dust (Hashimoto et al., 2018). Later ALMA Band 7 and Band 3 studies added a $88\,\mu{\rm m}$4 dust-continuum detection, a stringent [NII] $88\,\mu{\rm m}$5 upper limit, and non-detections of CO(6–5), CO(7–6), and CI (Sugahara et al., 2021, Hashimoto et al., 2022). JWST NIRCam, NIRSpec IFU, and MIRI then resolved the rest-optical morphology, optical emission-line structure, and H$88\,\mu{\rm m}$6 kinematics in detail (Sugahara et al., 2024, Jones et al., 2024, Prieto-Jiménez et al., 9 Jul 2025).

2. Morphology, component structure, and merger kinematics

Early HST/WFC3 imaging showed that B14-65666 comprises two spatially separated rest-UV clumps, and ALMA later demonstrated that [OIII], [CII], and dust-continuum peaks align with those UV components, with no significant multi-wavelength offsets (Hashimoto et al., 2018). The projected separation is reported as $88\,\mu{\rm m}$7–$88\,\mu{\rm m}$8 kpc in the HST/ALMA analysis, $88\,\mu{\rm m}$9 kpc in the ALMA reconstruction, $158\,\mu{\rm m}$0 kpc in JWST/NIRCam imaging, and $158\,\mu{\rm m}$1 kpc in the MIRI/H$158\,\mu{\rm m}$2 analysis (Hashimoto et al., 2018, Sugahara et al., 2024, Prieto-Jiménez et al., 9 Jul 2025). The literature therefore consistently describes two principal components, although the exact quoted separation depends on the dataset and measurement definition.

Later JWST work names the two main galaxies E and W. Galaxy E contains a bright compact core surrounded by diffuse extended rest-optical emission interpreted as tidal tails, while galaxy W is elongated and clumpy (Sugahara et al., 2024). The E-core is unresolved even in NIRCam F115W, with $158\,\mu{\rm m}$3 ($158\,\mu{\rm m}$4 pc; $158\,\mu{\rm m}$5 limit), and the higher-resolution MIRI/NIRCam structural analysis gives $158\,\mu{\rm m}$6 pc (Sugahara et al., 2024, Prieto-Jiménez et al., 9 Jul 2025). Galaxy W has a length $158\,\mu{\rm m}$7 ($158\,\mu{\rm m}$8 kpc) and a circularized effective radius $158\,\mu{\rm m}$9 pc (Sugahara et al., 2024, Prieto-Jiménez et al., 9 Jul 2025). Statmorph measurements give shape asymmetry $\alpha$0 for E and $\alpha$1 for W, consistent with disturbed merger morphologies (Prieto-Jiménez et al., 9 Jul 2025).

The kinematic evidence also favors a merger rather than a smoothly rotating disk. In the ALMA decomposition, the two components are separated by $\alpha$2, and moment-1 maps show an $\alpha$3 gradient across the system (Hashimoto et al., 2018). The field is explicitly described as not smoothly rotating, favoring a merger interpretation (Hashimoto et al., 2018). JWST/MIRI H$\alpha$4 later measured $\alpha$5, $\alpha$6, and $\alpha$7, in agreement with the ALMA FIR-line separation (Prieto-Jiménez et al., 9 Jul 2025).

Dynamical mass estimates reinforce the picture of a major merger. The ALMA virial estimates give $\alpha$8 and $\alpha$9 for the two main clumps, totaling $88\,\mu{\rm m}$0 (Hashimoto et al., 2018). The resolved NIRCam+ALMA analysis interprets the system as a major merger with a stellar mass ratio of $88\,\mu{\rm m}$1 to $88\,\mu{\rm m}$2 (Sugahara et al., 2024).

3. Far-infrared lines, dust continuum, and the ionized ISM

ALMA spatially resolved [OIII] $88\,\mu{\rm m}$3, [CII] $88\,\mu{\rm m}$4, and the underlying dust continuum in B14-65666, making it exceptional among $88\,\mu{\rm m}$5 galaxies (Hashimoto et al., 2018). For the whole system, the integrated line fluxes are $88\,\mu{\rm m}$6 and $88\,\mu{\rm m}$7, corresponding to $88\,\mu{\rm m}$8 and $88\,\mu{\rm m}$9 (Hashimoto et al., 2018). The standard conversion used is

$158\,\mu{\rm m}$0

The luminosity ratio is $158\,\mu{\rm m}$1, with clump-level values of $158\,\mu{\rm m}$2 and $158\,\mu{\rm m}$3 (Hashimoto et al., 2018).

The FIR-line phenomenology is interpreted as highly ionized ISM conditions. The high $158\,\mu{\rm m}$4 ratio implies that ionized gas dominates the FIR line budget and is described as consistent with highly ionized H II regions, fewer or less luminous PDRs, and possibly low-to-moderate metallicity (Hashimoto et al., 2018). The same study argues that a strong UV radiation field in the starburst raises the ionization parameter and boosts [OIII] relative to CII.

Dust emission has been measured at rest-frame $158\,\mu{\rm m}$5, $158\,\mu{\rm m}$6, and $158\,\mu{\rm m}$7. The reported whole-system flux densities are $158\,\mu{\rm m}$8 at $158\,\mu{\rm m}$9, $\alpha$0 at $\alpha$1, and $\alpha$2 at $\alpha$3 (Hashimoto et al., 2018, Sugahara et al., 2021). Modified-blackbody modeling with CMB corrections gives different characteristic temperatures depending on the adopted emissivity index. Using the two-band $\alpha$4 fit, the whole-system temperature is $\alpha$5 K for $\alpha$6, $\alpha$7 K for $\alpha$8, and $\alpha$9 K for $M_{\rm UV}\approx -22.4$0, each with $M_{\rm UV}\approx -22.4$1 K uncertainty (Hashimoto et al., 2018). Using the three-band $M_{\rm UV}\approx -22.4$2 fit, the best-fit values are $M_{\rm UV}\approx -22.4$3 K for $M_{\rm UV}\approx -22.4$4, $M_{\rm UV}\approx -22.4$5 K for $M_{\rm UV}\approx -22.4$6, and $M_{\rm UV}\approx -22.4$7 K for $M_{\rm UV}\approx -22.4$8 (Sugahara et al., 2021).

The inferred infrared luminosity is correspondingly large. The two-band ALMA analysis gives $M_{\rm UV}\approx -22.4$9, $-22.5$0, and $-22.5$1 for $-22.5$2, $-22.5$3, and $-22.5$4, respectively (Hashimoto et al., 2018). The three-band ALMA Band 7 study reports $-22.5$5 for $-22.5$6 and $-22.5$7 for $-22.5$8, with acceptable fits across the explored $-22.5$9 range (Sugahara et al., 2021). These values place B14-65666 among the most IR-luminous reionization-era galaxies with multi-band dust detections (Hashimoto et al., 2018).

The [NII] $z>7$00 line remains undetected. The $z>7$01 upper limit is $z>7$02, or $z>7$03 (Sugahara et al., 2021). Cloudy modeling using [NII], [OIII], and [CII] finds that if $z>7$04, then $z>7$05 and $z>7$06–$z>7$07, with sub-solar N/O; at $z>7$08, the allowed $z>7$09 rises sharply as metallicity decreases and the N/O constraint weakens (Sugahara et al., 2021).

4. Rest-optical spectroscopy, metallicity, and ionization structure

JWST/NIRCam added a resolved rest-optical view of the merger. F115W, F150W, and F200W trace the rest-UV continuum, while F277W and F356W sample the stellar continuum around the Balmer break and F444W contains H$z>7$10 and [OIII] $z>7$11 (Sugahara et al., 2024). A strong F356W–F444W excess therefore traces nebular line emission. The color excess peaks at the E-core, with $z>7$12 mag, and the inferred rest-frame equivalent widths of $z>7$13 are $z>7$14 $z>7$15 for the total system, $z>7$16 $z>7$17 for E, $z>7$18 $z>7$19 for W, and $z>7$20 $z>7$21 for the E-core (Sugahara et al., 2024).

JWST/NIRSpec IFU observations from the GA-NIFS survey then detected [OII] $z>7$22, [NeIII] $z>7$23, Balmer lines, [OIII] $z>7$24, and weak [OIII] $z>7$25 across six apertures: Full, Core-E, Core-W, Clump-N, Clump-W, and Arc-S (Jones et al., 2024). The spectra are modeled with narrow and broad Gaussian components plus a power-law continuum, and the ISM solution is obtained self-consistently with PyNeb, assuming case B recombination, fixed $z>7$26, fixed $z>7$27, and free $z>7$28 where allowed by the auroral ratio (Jones et al., 2024). The fitted temperatures in the main cores are $z>7$29 K and $z>7$30 K for Core-W, and $z>7$31 K and $z>7$32 K for Core-E (Jones et al., 2024).

Strong-line diagnostics place B14-65666 at relatively modest ionization for a reionization-era IFU target. Using Curti et al. (2020) calibrations, the total gas-phase metallicity is $z>7$33, with $z>7$34 in Core-E and $z>7$35 in Core-W (Jones et al., 2024). The total ratios are $z>7$36, $z>7$37, $z>7$38, $z>7$39, and $z>7$40, giving $z>7$41 for the full system (Jones et al., 2024). The study explicitly notes that the source lies near the intersection of local and high-redshift galaxies in common line-ratio diagrams, indicating lower ionization and higher metallicity than many lower-mass $z>7$42 IFU targets (Jones et al., 2024).

Optical-to-FIR [OIII] ratios provide an additional metallicity and density lever arm. The attenuation-corrected $z>7$43 ratio is $z>7$44 for the integrated system, $z>7$45 at the E-core, and $z>7$46 at the W-core (Sugahara et al., 2024). Photoionization modeling using this ratio, together with the [OIII] $z>7$47 density diagnostic as an anchor and assuming $z>7$48–$z>7$49, yields $z>7$50–$z>7$51 for the total system and a gradient in which the E-core is lower metallicity than the W-core (Sugahara et al., 2024). The later GA-NIFS analysis reports a global $z>7$52, which it interprets as evidence for elevated electron densities, $z>7$53, in at least part of the [OIII]-emitting gas (Jones et al., 2024).

No unambiguous AGN signature has been reported. The NIRCam study states that there is no evidence for an AGN, and the NIRSpec IFU analysis finds that the weakness of [OIII] $z>7$54 precludes a firm AGN identification, with the broad optical component more plausibly associated with tidal interaction or outflows (Sugahara et al., 2024, Jones et al., 2024).

5. Star formation, stellar populations, and molecular-gas constraints

Published stellar-population parameters depend strongly on the available data and modeling assumptions. Early SED fitting with GALAXEV, nebular lines, Calzetti attenuation, and a Chabrier IMF gave $z>7$55, age $z>7$56 Myr, $z>7$57 mag, $z>7$58, and $z>7$59 (Hashimoto et al., 2018). Resolved NIRCam+ALMA modeling with Bagpipes later yielded $z>7$60, $z>7$61 Myr, $z>7$62, and $z>7$63 for the integrated system (Sugahara et al., 2024). Component-wise CIGALE modeling with JWST/MIRI gives $z>7$64, $z>7$65, and a total of $z>7$66 (Prieto-Jiménez et al., 9 Jul 2025). The literature therefore reports a substantial range in stellar mass, reflecting different datasets and model parameterizations.

The star-formation rate is likewise high in every study, but the quoted instantaneous value depends on tracer and calibration. The early ALMA-assisted SED fit gave $z>7$67 and $z>7$68, explicitly described as well above the star-forming main sequence at $z>7$69–7 (Hashimoto et al., 2018). The Bagpipes analysis finds $z>7$70 and $z>7$71, about $z>7$72 dex above the $z>7$73 main sequence (Sugahara et al., 2024). The GA-NIFS H$z>7$74-based IFU analysis gives a total $z>7$75 (Jones et al., 2024). By contrast, the resolved MIRI H$z>7$76 study measures $z>7$77 and $z>7$78 for a reference solar-metallicity calibration, or $z>7$79 and $z>7$80 when adopting $z>7$81 (Prieto-Jiménez et al., 9 Jul 2025). These published values use different recombination lines, dust corrections, and metallicity-dependent conversion factors.

The resolved starburst structure is extreme. In the RIOJA analysis, E has $z>7$82 mag and W has $z>7$83 mag from NIRCam-only fits; with ALMA included, E remains much dustier than W (Sugahara et al., 2024). The E-core combines $z>7$84 pc with $z>7$85, implying $z>7$86, comparable to local ULIRGs (Sugahara et al., 2024). The MIRI study reports very large H$z>7$87 equivalent widths, $z>7$88 and $z>7$89, with burst ages $z>7$90 Myr and $z>7$91 Myr and ionizing efficiencies $z>7$92 and $z>7$93 (Prieto-Jiménez et al., 9 Jul 2025).

ALMA Band 3 observations probe the molecular-gas reservoir indirectly through non-detections. CO(6–5), CO(7–6), and CI are all undetected, with $z>7$94 limits $z>7$95, $z>7$96, and $z>7$97, corresponding to $z>7$98, $z>7$99, and $88\,\mu{\rm m}$00, respectively (Hashimoto et al., 2022). These imply $88\,\mu{\rm m}$01 and $88\,\mu{\rm m}$02, larger than typical ratios in dusty star-forming galaxies and quasar hosts at similar redshift (Hashimoto et al., 2022). PDR Toolbox modeling yields $88\,\mu{\rm m}$03–5 and $88\,\mu{\rm m}$04–$88\,\mu{\rm m}$05, while XDR-like heating is disfavored (Hashimoto et al., 2022).

The resulting molecular-gas mass remains broad but bounded. Combining [CII]-based, dust-based, and dynamical arguments gives $88\,\mu{\rm m}$06–$88\,\mu{\rm m}$07, $88\,\mu{\rm m}$08–140, and $88\,\mu{\rm m}$09–550 Myr (Hashimoto et al., 2022). The authors state that B14-65666 could be an ancestor of a passive galaxy at $88\,\mu{\rm m}$10 if no gas is fueled from outside the galaxy (Hashimoto et al., 2022).

6. Reionization-era significance and observational outlook

B14-65666 occupies an unusual position among known $88\,\mu{\rm m}$11–9 FIR-line emitters. Its [OIII] and [CII] luminosities are among the highest for normal star-forming galaxies in that redshift range, and its $88\,\mu{\rm m}$12 ratio sits between extreme LAEs such as SXDF-NB1006-2, with $88\,\mu{\rm m}$13, and dusty SMGs, with $88\,\mu{\rm m}$14–1.3 (Hashimoto et al., 2018). Its very large Ly$88\,\mu{\rm m}$15 offset, together with observed trends that brighter UV and higher-[CII]-luminosity galaxies have larger Ly$88\,\mu{\rm m}$16 offsets, places it at the luminous end of the reionization-era population and is presented as consistent with enhanced Ly$88\,\mu{\rm m}$17 visibility in bright systems despite a neutral IGM (Hashimoto et al., 2018).

At the same time, later optical spectroscopy depicts the system as chemically evolved for its epoch. The GA-NIFS analysis places B14-65666 on the high-mass extension of the $88\,\mu{\rm m}$18 mass-metallicity relation, with $88\,\mu{\rm m}$19 and $88\,\mu{\rm m}$20 for the integrated system (Jones et al., 2024). Within the merger, Core-E and Core-W show contrasting combinations of stellar mass, SFR, metallicity, and [CII] prominence, which the authors interpret as distinct evolutionary pathways inside one halo (Jones et al., 2024).

B14-65666 has also served as a forecasting target for facilities and line diagnostics. FIRSTLIGHT-based MIRI/MRS simulations predicted that a deep $88\,\mu{\rm m}$21 ks spectrum would detect H$88\,\mu{\rm m}$22 at very high significance and would strongly detect He I $88\,\mu{\rm m}$23, [SIII] $88\,\mu{\rm m}$24 $88\,\mu{\rm m}$25, and several Paschen lines while placing useful constraints on [NII] and SII. Subsequent MIRI observations indeed detected spatially resolved H$88\,\mu{\rm m}$26 in both principal components (Prieto-Jiménez et al., 9 Jul 2025). Separately, ALMA Band 9 forecasts for [OIII] $88\,\mu{\rm m}$27 show that B14-65666 is among the most promising $88\,\mu{\rm m}$28 targets: with $88\,\mu{\rm m}$29, 1.56, and 0.56, the expected $88\,\mu{\rm m}$30 on-source times are $88\,\mu{\rm m}$31, $88\,\mu{\rm m}$32, and $88\,\mu{\rm m}$33 hours, and even a $88\,\mu{\rm m}$34 h upper limit with $88\,\mu{\rm m}$35 would tighten $88\,\mu{\rm m}$36 by $88\,\mu{\rm m}$37–3 dex and $88\,\mu{\rm m}$38 by up to $88\,\mu{\rm m}$39 dex (Yang et al., 2021).

The system is therefore significant not only because it is bright, but because its brightness has enabled a rare joint reconstruction of merger dynamics, dust geometry, nebular excitation, chemical enrichment, and multi-phase gas structure at $88\,\mu{\rm m}$40. Deeper NIRSpec G395H or MIRI spectroscopy, higher-sensitivity ALMA mapping, and direct [OIII] $88\,\mu{\rm m}$41 observations are specifically identified in the literature as the next steps for refining $88\,\mu{\rm m}$42, $88\,\mu{\rm m}$43, electron density, metallicity gradients, LyC-leakage diagnostics, and the relation between the broad optical component and the colder FIR phases (Yang et al., 2021, Jones et al., 2024).