THE BLACK PEARL: A DUST FINGERPRINT 500 MILLION YEARS AFTER THE BIG BANG

The James Webb Space Telescope keeps finding that galaxies filled up with metals and dust surprisingly fast, within the first billion years of cosmic history. Dust is hard to study that far away: its own heat glow is extremely difficult to detect. But dust also leaves marks on the starlight that passes through it, and one mark is especially telling.

A dip in the ultraviolet

In the Milky Way, dust dims ultraviolet light with a broad “bump” centred near 2,175 ångströms: a dip in the spectrum, the strongest feature of our Galaxy’s extinction curve. It is widely attributed to small carbon-rich grains or molecules, though which ones exactly is still debated — small graphite grains, polycyclic aromatic hydrocarbons (PAHs) and amorphous carbon are among the candidates. Along different lines of sight in the Milky Way, the bump’s width and strength vary, but its central position is remarkably stable.

Webb has pushed the search into the early universe. A landmark detection came in a galaxy at redshift 6.71, about 800 million years after the Big Bang, with a bump peaking at about 2,263 Å — redder than in the Milky Way. Another, at redshift 7.11, peaks at 2,257 Å. On average, the bump seems to weaken towards earlier times.

A galaxy at redshift 9.33

Giulia Rodighiero, of the University of Padua and INAF, and about fifty co-authors in Europe, the Americas and Asia went through Webb’s NIRSpec spectra from two coordinated programmes, ZENITH and EMBER. They looked at sources beyond redshift 8, where the 2,175 Å region is covered, without picking those that already showed a bump. One stood out. They call it the Black Pearl — the paper does not say why.

Hydrogen and oxygen emission lines place it at redshift 9.3335, seen as it was about 500 million years after the Big Bang. It is a fairly small galaxy, with a stellar mass of about 10^8.35 suns, forming roughly 2.4 suns’ worth of stars per year, and one of the most ultraviolet-luminous galaxies confirmed spectroscopically beyond redshift 9.

Two ways to see the same dip

Its spectrum shows a broad depression exactly where a dust bump would sit. The team measured it two independent ways.

First, directly: fit the ultraviolet continuum with a smooth curve and measure the extra dimming. The dip is detected at 17.6 sigma, and adding a bump-shaped profile improves the fit enormously (a χ² of 474 falls to 8.2). The profile peaks at 2,132 ± 17 Å, with a width of 413 ± 54 Å and a peak dimming of 0.21 magnitudes. Checks rule out a nearby carbon emission line and a blend of unresolved lines as the cause.

Second, with a full model of the galaxy — stars, gas, dust and photometry — using a flexible dust law in which the bump’s position and width are free. Both star-formation histories tested recover a bump, at about 2,147 and 2,168 Å, consistent with the direct measurement though less precise.

Could a black hole be faking it? An active nucleus can produce a forest of iron lines that mimics a dip. The emission lines point to a hard radiation field and leave room for some contribution from a nucleus, but not for one strong enough to dominate the ultraviolet. The authors favour dust, while noting that a nucleus, “potentially even substantial”, cannot be firmly excluded.

Shifted the wrong way

The width matches the Milky Way’s. The position does not: 43 Å bluer than our Galaxy’s average, at about 2.5 sigma, and some 127 Å bluer than the two bumps known around redshift 7, which are shifted towards the red. For the authors, that diversity argues against a single universal bump in the early universe: the dust grains behind it probably differed from galaxy to galaxy.

What are they? Calculations for the smallest graphite grains come close to the Black Pearl’s profile, unlike the earlier redshift-6.71 case. Some individual PAH molecules can also reproduce it — but PAHs are scarce in metal-poor galaxies at lower redshift, and the intense radiation inferred here may destroy the smallest ones. Both remain plausible.

Dust in 200 million years

The relevant clock is not the 500 million years since the Big Bang but the time since the galaxy started making stars. If that began around redshift 14, about 200 million years were available. Supernovae, grain growth between the stars and perhaps the most rapidly evolving AGB stars could contribute; how much each one did remains open.

This is one galaxy, its bump shift is significant at only about 2.5 sigma, and attenuation curves do not map uniquely onto grain types. The paper has been submitted to the journal ApJ. Its message is that the shape of the bump, and not only its presence, can now track how the first dust was made.