Black carbon or brown carbon? The nature of light-absorbing carbonaceous aerosols

For years, we talked about the dark stuff in the air as if it were all the same: soot. It’s black, simple, and powerful at soaking up sunlight. But turn the dial toward ultraviolet light, or UV, and a different voice comes through — brown carbon. This isn’t soot; it’s organic, aromatic, and messy. Once you hear it, you can't unhear it. It forces a rewrite of the story of light-absorbing carbon in the atmosphere. Here's the lay of the land. Soot is the black material produced by combustion. When we say soot carbon, we mean carbon particles with that classic soot morphology — tangled aggregates of graphene-like layers with a little oxygen and other heteroatoms on the surface. Brown carbon is everything that absorbs light but isn't soot: humic-like organics, tarry smoke products, bioaerosols, and secondary material that forms in the air. Put both together and you get light-absorbing carbonaceous matter, the big umbrella. Now, the terms we use to measure them are trickier. Elemental carbon is a thermal definition — the refractory carbon left after you drive off the more volatile stuff. Apparent elemental carbon is the piece that burns off above a particular temperature in a protocol. Black carbon is an optical proxy for soot. Equivalent black carbon — often abbreviated as BCe — is a pragmatic translation: how much soot-like carbon would produce the same optical signal I just measured? The catch, as Andreae and colleagues emphasized, is that none of these are chemical species. They're operational. And when brown carbon shows up, it can mess with every one of them. What convinced researchers that brown carbon deserved a seat at the table? Several threads came together. If you measure the absorption spectrum of aerosols near certain sources — think smoldering biomass or tobacco smoke — the curve leans steeply upward into the UV, unlike pure soot. If you take continental aerosols and extract them with water, the solution behaves like humic and fulvic substances from soils — dark, chromophoric, and strongly absorbing at shorter wavelengths. Laboratory chemists have also watched brown carbon form in real time: aromatics reacting in acidic droplets, lignin pyrolysis producing brown oligomers, and dienes in sulfuric acid building chromophores. And the chemistry looks familiar. Humic-like substances, or HULIS, long known in natural waters, have clear analogs in the air, as Krivácsy, Decesari, and Kiss each showed in different contexts. The upshot is simple but profound: there isn't a single "brown carbon molecule." There's a continuum from black to brown, and in places dominated by low-temperature combustion or biogenic processes, brown can win. Optically, that continuum shows up as a spectral fingerprint. Soot's absorption with wavelength is fairly gentle — if you plot absorption versus wavelength on a logarithmic scale, the slope is near one. Brown carbon's slope is steeper. You'll hear people summarize this with a power law: the mass-specific absorption, call it alpha, at a given wavelength, scales like a constant times the wavelength raised to a negative exponent. That exponent, known as the Ångström exponent for absorption, tells you how fast the curve climbs into the UV. Soot-dominated aerosols usually sit near one. Biomass smoke often sits at two or three. Tobacco smoke can push toward three and a half. And water-soluble HULIS can be extreme — six to seven — as Kirchstetter and Hoffer reported for some smoldering systems. That steepness is the giveaway: a spectrum that rockets up in the near UV signals brown carbon. Brown carbon also brings something soot mostly doesn't: water solubility. Pull the organics into water, and many of the chromophores keep their color. That opens doors in clouds and fog. A droplet with brown carbon at just two micromolar — that’s two millionths of a mole per liter — can absorb about six times more light at blue wavelengths than pure water. Now think about millions of those droplets in a cloud deck. Small numbers, big effect. Chemically, soot and brown carbon part ways again. Soot's hydrogen-to-carbon ratio is tiny, about 0.15 with a wiggle of plus or minus 0.05 either way. Atmospheric HULIS sit way higher, around 1.4 to 1.6. Lignin-derived char can land anywhere from 0.3 up to about 1.3 depending on how it's processed. Analysts use those contrasts to read thermograms and mass spectra, but there's a wrinkle: brown carbon chars. Heat a humic-rich sample and up to thirty-seven percent of it can convert into a more refractory, blacker form right on the filter. That char looks like elemental carbon to a thermal instrument, even though it wasn't soot when it came in. And that takes us straight into the measurement maze. In a thermal analysis, you heat the sample in a gas stream and count the carbon off as carbon dioxide. Total carbon is everything; organic carbon is what evolves below a temperature threshold; the remainder gets labeled elemental carbon. That "elemental" tag depends on temperature cut points and optical corrections. In biomass smoke or aged aerosols, the material that looks like elemental carbon by this definition often comes off at four hundred fifty to five hundred fifty degrees Celsius. Urban soot tends to hang on until five hundred fifty to seven hundred degrees Celsius. If you draw the line at a fixed temperature, Novakov and Corrigan — and later Mayol-Bracero in the tropics — showed you can overestimate elemental carbon by factors of two to ten. That's not a rounding error. It's a category mistake driven by brown carbon's broad volatility and its talent for charring. Thermal-optical methods try to fix this by shining light through the filter while it heats. As pyrolyzed carbon darkens the filter, the transmittance drops; when that darkness burns back off, the method declares the split between organic and elemental carbon. It's clever, but it assumes the char's optical properties match those of what just burned away. Huffman pointed out in the IMPROVE network that this assumption can fail badly, biasing elemental carbon low in some cases and high in others. Direct comparisons between thermal-optical reflectance and transmittance reported differences of up to a factor of four for the apparent elemental fraction. Look across labs and protocols and total carbon usually agrees; the elemental slice doesn't. Twofold differences are common, sixfold shows up in brown carbon-rich samples, and even the same reference material — RM 8785 — has yielded two different elemental carbon values under two protocols, as Klouda's team found. Maybe optical instruments can save us? They help, but they're not immune. Only a handful measure absorption of airborne particles directly; the photoacoustic spectrometer does, turning absorbed light into sound. Most field instruments pull air through a filter and report attenuation — how much dimmer a beam gets. That measurement is not pure absorption. Filters scatter light. They shadow particles as the filter loads up. Without careful corrections, you can be off by about a factor of two, as Weingartner, Arnott, and Schnaiter showed in a series of lab and field studies. Side by side with a photoacoustic instrument, differences around a factor of two have shown up too, especially as humidity changes the particles, as Moosmüller and Sheridan reported. Converting those optical signals to a mass of soot takes another step, and another set of assumptions. Bond and Bergstrom reviewed the literature and landed on a mass absorption efficiency for fresh, uncoated soot of about seven point five square meters per gram, with an uncertainty of around one point two. Age those particles in the atmosphere and that number moves. Wrap them in sulfate or organic coatings and the absorption can climb — a lot in extreme cases. Modeling and lab work by Martins and Fuller found boosts that could reach more than twentyfold for very thick coatings on large particles, but in the real atmosphere, a doubling is much more typical. Humidity can add another thirty percent at about ninety-five percent relative humidity, as Redemann and colleagues showed. So the "right" number for soot's absorption per mass isn't a number at all. It's a spectrum of possibilities that depends on mixing state, size, and coatings. Because of that variability, people sometimes work with an attenuation cross section instead — how much attenuation per gram on a particular instrument. In practice, that value is often treated as about ten square meters per gram. But scan the literature and you'll find numbers from one up to thirty, as Liousse, Schnaiter, and Moosmüller each reported in different settings. Even with the same aethalometer, the uncertainty grows when the air is poor in light-absorbing carbon, and the shorter-wavelength channels tend to swing more wildly than the near-infrared one, as Weingartner noted. Now bring brown carbon back in. If you only look at one wavelength, especially in the red or near-infrared, you can miss most of brown carbon's voice because its absorption punches hardest in the UV. Its spectral slope can be as steep as wavelength to the power of minus two, minus three, even out toward minus six in water-soluble biomass smoke extracts. Soot, by contrast, is closer to a minus-one slope in the visible. Andreae coined "equivalent black carbon" to make this honest: report the amount of soot-like carbon that would give you the signal you measured. It's a translation, not a truth. You can't back out a unique soot mass from a single absorption number when brown carbon is present. Kirchstetter showed how to do better: use multiple wavelengths and deconvolve the total absorption into a soot-like component with a relatively flat spectrum and a brown component with a much steeper one. Why does that spectral nuance matter beyond tidy bookkeeping? Because UV light runs the chemistry of the lower atmosphere. Organic aerosol absorption accounts for roughly a quarter of the downward UV attenuation, with non-nitrated aromatics contributing on the order of ten to thirteen percent and nitrated aromatics about nine to fifteen percent. Fold that into a city's chemistry and you move real needles. Model results tied to this spectral behavior suggest a five to eight percent drop in ozone mixing ratios in Los Angeles simply from the UV light absorbed by aerosols. And yet, when radiative-transfer models try to simulate observed UV at the surface, they still overpredict by about eight percent at urban sites and by roughly twenty-five percent at rural ones. That gap tells us that the absorbing pieces — especially the brown, water-soluble ones — are still not fully captured by our measurements and models. Put all of this together and a few themes emerge. First, there is no single conversion from "how dark is it?" to "how much soot is there?" when brown carbon is around. Second, thermal definitions of elemental carbon are context-dependent, and brown carbon's charring and broad volatilization span can shift apparently "elemental" signals around by factors of two, four, even ten. Third, coatings and humidity change soot's absorption enough that a one-size-fits-all efficiency — seven point five square meters per gram — is a starting point, not an answer. And fourth, the spectrum matters. A single red channel can lull you into thinking you've got the absorbing carbon nailed, while the UV tells a very different story. So where do we go from here? The call from the community is clear. Standardize and widen the lens. On the thermal side, that means protocols that explicitly reduce brown-carbon charring — water extraction helps by removing inorganic catalysts and some of the brown organics — and intercomparisons that make the assumptions visible. On the optical side, it means routine multi-wavelength measurements so we can separate soot-like and brown contributions, and more frequent use of instruments that measure absorption directly in the air. It also means being honest in our language — using BCe when we report an optical signal translated into soot-like terms, and saying when brown carbon is likely present. Reference materials help anchor all of this. RM 8785 has already taught us that even a standard can yield two "elemental carbon" answers under two protocols. That's a feature, not a bug — it reveals where methods differ. Broader adoption of such references, plus complementary techniques like fluorescence lidar for remote sensing of chromophores, spectral absorption methods in the lab, and ultra-high-resolution mass spectrometry such as Fourier transform ion cyclotron resonance to map the brown-carbon soup, can close the loop between what we see, what we count, and what we model. If you like a simple punchline, here it is. Black carbon isn't one thing. Elemental carbon isn't one thing. And the atmosphere doesn't care about our categories. Soot and brown carbon share the sky, and they share the job of bending sunlight and steering chemistry. When we measure them with single temperatures, single wavelengths, or single conversion factors, we force a complicated reality through a narrow keyhole. Step back, spread across wavelengths, and let the spectrum tell you who's in the room. Then the numbers — seven point five square meters per gram for fresh soot, a doubling with common coatings, a thirty percent bump at high humidity, and Ångström exponents of one for soot and up to seven for soluble brown — start to fit together. And with that, the story of atmospheric darkness gets a lot clearer and a lot more interesting.