Article
Microphysical complexity of black carbon particles
restricts their warming potential
Graphical abstract
Highlights
d
Inadequate BC microphysics understanding hinders its
climate assessment
d
Coating heterogeneity and morphology co-explain the lower
BC absorption by
50%
d
Off-center BC-core position is also an essential factor in
absorption gap
d
Integrating an observation-based framework in this model
reduces global BC DRF by up to 23%
Authors
Xiao-Feng Huang, Yan Peng,
Jing Wei, ..., Hongjun Mao,
John H. Seinfeld, Yuan Wang
Correspondence
pengjianfei@nankai.edu.cn (J.P.),
hely@pku.edu.cn (L.-Y.H.),
yzwang@stanford.edu (Y.W.)
In brief
The warming effect of black carbon (BC)
is uncertain due to a persistent model-
measurement gap in light absorption
enhancement. Here, we reveal multi-level
microphysical complexities in BC-
containing particles, and these explain
the lower observed BC absorption. In
addition to coating heterogeneity and
non-spherical shape, the off-center BC-
core position is also an essential factor in
the absorption gap. When considering
the observation-based framework in the
BC microphysical representation, BC’s
global direct radiative forcing reduces by
up to 23%.
Huang et al., 2024, One Earth
7
, 136–145
January 19, 2024
ª
2023 The Author(s). Published by Elsevier Inc.
https://doi.org/10.1016/j.oneear.2023.12.004
ll
Article
Microphysical complexity of black carbon particles
restricts their warming potential
Xiao-Feng Huang,
1
,
6
Yan Peng,
1
,
6
Jing Wei,
1
,
6
Jianfei Peng,
2
,
7
,
*
Xiao-Yu Lin,
1
Meng-Xue Tang,
1
Yong Cheng,
1
Zhengyu Men,
2
Tiange Fang,
2
Jinsheng Zhang,
2
Ling-Yan He,
1
,
*
Li-Ming Cao,
1
Chao Liu,
3
Chenchong Zhang,
4
Hongjun Mao,
2
John H. Seinfeld,
5
and Yuan Wang
4
,
*
1
Laboratory of Atmospheric Observation Supersite, School of Environment and Energy, Peking University Shenzhen Graduate School,
Shenzhen 518055, China
2
Tianjin Key Laboratory of Urban Transport Emission Research & State Environmental Protection Key Laboratory of Urban Ambient Air
Particulate Matter Pollution Prevention and Control, College of Environmental Science and Engineering, Nankai University, Tianjin
300071, China
3
Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science &
Technology, Nanjing 210044, China
4
Department of Earth System Science, Stanford University, Stanford, CA 94305, USA
5
Divisions of Chemistry and Chemical Engineering and Engineering and Applied Science, California Institute of Technology, Pasadena,
CA, USA
6
These authors contributed equally
7
Lead contact
*Correspondence:
pengjianfei@nankai.edu.cn
(J.P.),
hely@pku.edu.cn
(L.-Y.H.),
yzwang@stanford.edu
(Y.W.)
https://doi.org/10.1016/j.oneear.2023.12.004
SUMMARY
Black carbon (BC) strongly absorbs solar radiation, but its warming effect on climate is poorly quantified. A
key challenge is to accurately assess BC light absorption after BC is mixed with non-BC components. How-
ever, there has consistently been a large observation-modeling gap in BC light absorption estimation, reflect-
ing the insufficient understanding of realistic BC complexity. Here, we conduct comprehensive
in situ
mea-
surements of BC single-particle microphysics, e.g., size, coating amounts, density, and shape, along with
optical closure calculation. Specifically, the observed particle-to-particle heterogeneities in size and coating
and the non-spherical BC shape only explain the lower observed BC absorption by
20% and
30%, respec-
tively. A remaining gap for fully aged spherical BC-containing particles is related to the off-center BC-core
position. The global climate model assessment shows that fully accounting for the observed BC complexity
in the aerosol microphysical representation reduces the global BC direct radiative forcing by up to 23%.
INTRODUCTION
Black carbon (BC) is ubiquitous in the atmosphere and profoundly
affects global and regional climate owing to its efficient radiative
absorption capacity.
1–3
The direct radiative forcing (DRF) of BC
particles represents one of the largest sources of uncertainty in
present and future climate assessments.
4
Estimation of BC DRF
depends explicitly on the particle’s light absorption cross-section,
which is further determined by the mixing state of BC with non-BC
materials.
5–8
A widely adopted assumption of the mixing state in
SCIENCE FOR SOCIETY
Black carbon (BC) strongly absorbs solar radiation, but its warming effect is still
highly uncertain. One crucial uncertainty stems from insufficient understanding of BC microphysics and op-
tics, which is evident as the constant model-measurement discrepancy in BC light absorption enhancement
in previous studies. Therefore, characterizing the complexity of BC microphysics in the realistic atmosphere
is essential for constraining BC light absorption ability. Here, we present direct observations of the micro-
physical properties of individual particles across the entire BC-containing particle population, providing
unique insight into multi-level BC microphysical heterogeneity that largely reconciles the model-measure-
ment gap. Our observation-based framework facilitates in-depth assessment of BC environmental and
climate effects in measurement and model studies.
136
One Earth
7
, 136–145, January 19, 2024
ª
2023 The Author(s). Published by Elsevier Inc.
This is an open access article under the CC BY-NC license (
http://creativecommons.org/licenses/by-nc/4.0/
).
ll
OPEN ACCESS
climatemodelsisa‘‘core-shell’’structurewithBCcorescoatedby
non-BCmaterials.Intheory,thecoatingmaterialsproducea‘‘lens-
ing effect’’ that significantly enhances BC light absorption.
1
,
2
,
8
Such large light absorption enhancement (
E
abs
) has been consis-
tently observed for coated BC particles in previous laboratory
and field chamber studies.
6
,
9–11
In contrast, field measurements
frequently reported much weaker
E
abs
.
7
,
10
,
12–16
Such a disagree-
ment indicates that both the Mie core-shell model and the labora-
tory experiments fail to account for the realistic optics of BC-con-
taining particles in the atmosphere; a better understanding of
atmospheric BC complexity is thus essential.
Atmospheric BC-containing particles exhibit complex varia-
tions in chemical and microphysical properties, such as size,
chemical composition, and morphology. Oversimplified treat-
ment of this complexity in BC microphysics by the Mie core-shell
models inevitably leads to significant uncertainties in estimating
BC
E
abs
. For instance, chamber studies have found that the non-
ideal morphology lowers BC
E
abs
compared to the ideal core-
shell assumption in the initial atmospheric aging stage.
6
,
17
,
18
Current interpretations of this non-ideal shape impact include
(1) the collapse of fresh BC branches to form a more compact
and smaller light absorption cross-section shape
6
and (2) the un-
even coating on BC during initial aging to form a quasi-external
mixing state of BC and non-BC components.
7
Moreover, field
measurements have frequently observed that even for the
aged compact BC-containing particles, the BC core may be
encapsulated within, embedded in, or attached to the non-BC
components, producing complex micromorphological struc-
tures and likely regulating
E
abs
.
19–21
Notably, recent modeling
studies proposed that the particle-to-particle heterogeneities in
the mixing state exert a dominant impact on
E
abs
.
22–24
Neverthe-
less, the gap in BC
E
abs
has never been fully reconciled through
in situ
field observations due to the challenge of determining
complex BC microphysics at single-particle levels.
Here, two intensive field observations were conducted in
Shenzhen, China: one at an urban site in winter and another at
a regional site in late summer. A suite of state-of-the-art instru-
ments was utilized to simultaneously obtain the magnitude and
evolution of the various microphysics at both bulk-averaged
and individual-particle levels under diverse atmospheric environ-
ments. Further, the influences of different BC microphysics on
E
abs
are quantified separately to explain and reconcile the dis-
crepancies between models and observations. By introducing
a set of new
E
abs
parameterizations in a state-of-the-art global
climate model, we demonstrate that BC DRF can be significantly
decreased by accounting for the real-world microphysics of BC
particles in climate models. Our observation-based framework
facilitates in-depth assessment of BC environmental and climate
effects in measurement and model studies.
RESULTS AND DISCUSSION
Evolution of BC microphysics in the atmosphere
Microphysics of each BC-containing particle are quantified dur-
ing field observations in terms of BC-core mass equivalent diam-
eters (
D
c
), coating-to-BC mass ratios (
M
R
), effective density (
r
eff
),
and dynamic shape factor (DSF) via a comprehensive scheme
for integration, extrapolation, and correction of the measurement
data (see
experimental procedures
). Instruments deployed in the
field observations include a soot particle-aerosol mass spec-
trometer (SP-AMS; Aerodyne Research) and a custom-devel-
oped tandem differential mobility analyzer (DMA; TSI)-single-
particle soot photometer (SP2; DMT) system (
Figure S1
). The
measured bulk-averaged
M
R
over the entire population of BC-
containing particles is utilized to reflect their mean atmospheric
aging stage.
10
,
25
Atmospheric processes lead to evidently size-dependent evo-
lution in BC microphysics. During the observations at the urban
site, the measured
M
R
of single particles exhibits a considerable
variation among different particles, with the calculated average
standard deviation of
log
10
(
M
R
) about 0.5 (
Figure 1
A). This stan-
dard deviation exerts positive correlations with two photochem-
ical parameters, i.e., hydroxyl radical exposure and carbon
oxidation state of BC-containing particles, and negative correla-
tion with NO
x
concentrations (see
Note S1
and
Figures S4
E, and
S4F). Such relationships imply that atmospheric aging pro-
cesses enhance particle-to-particle
M
R
heterogeneity. This
enhanced
M
R
heterogeneity, as shown in
Figure 1
B, is also re-
flected by the evolution of size-resolved
M
R
. The values of
size-resolved
M
R
are in a relatively small range in the initial aging
stage (bulk-averaged
M
R
= 3–3.5) but exhibit large variation over
the fully aged stage (bulk-averaged
M
R
= 6–6.5) (
Figure 1
B).
Apparently, compared to the particles with larger BC cores,
those with smaller BC cores exhibit a more rapid increase (slope
1.24 vs. 0.56) in coating amounts during aging (
Figure 1
B). Par-
ticle density and DSF are key parameters for describing the
shape of BC-containing particles. Here, the bulk-averaged
r
eff
over the entire population of ambient BC-containing particles
is directly measured by combining SP-AMS and DMA-SP2 mea-
surements (see
experimental procedures
). The measured bulk-
averaged
r
eff
monotonically increases with bulk-average
M
R
and is close to material density over the fully aged stage (bulk-
averaged
M
R
= 6–6.5) (
Figures 1
C and
S5
); this suggests the
compact shape for most of the fully aged BC-containing parti-
cles. Size-resolved
r
eff
is also explored for size-dependent anal-
ysis (see
Note S3
). Significantly lower size-resolved
r
eff
with
larger BC cores are observed for relatively fresh BC particles
(bulk-averaged
M
R
= 3–3.5) than those with smaller BC cores
(1.0 vs 1.3 g cm
3
;
Figure 1
C), consistent with previous studies
in which large, fresh BC particles were found to be more
fractal.
6
,
26
,
27
Furthermore, according to the single-particle
r
eff
and DSF information, we classify all BC-containing particles
into two groups, i.e., near-spherical (DSF
%
1.05) and non-
spherical (DSF >1.05) particles (see
experimental procedures
).
The mean fraction of the spherical particles (
F
spherical
) raises
from 26% in the initial aging stage (bulk-averaged
M
R
= 3–3.5)
to 75% over the fully aged stage (bulk-averaged
M
R
= 6–6.5) (
Fig-
ure 1
D). Results also prove a higher possibility for smaller BC-
core particles to form a spherical shape.
27
,
28
The measurements presented here reveal the actual variation
and evolution of BC microphysics in ambient, regulated by atmo-
spheric aging processes. We categorized our data according to
sites, seasons (urban sites in autumn-winter and regional recep-
tor sites in autumn), and emission features (industrial + biomass-
burning emissions and shipping emissions). For different sites
and emission processes, the microphysical characteristics of
BC, including size distribution and mixing state evolution during
atmospheric aging, exhibited strong similarities (
Figures S6
and
ll
OPEN ACCESS
Article
One Earth
7
, 136–145, January 19, 2024
137
S7
). Both results exhibit that the aging of BC particles expands
particle-to-particle microphysical heterogeneity and reshapes
the particles to be more compact and spherical. Furthermore,
small-diameter BC particles tended to have higher coating
amounts and
r
eff
. This observation is crucial, as it suggests
that the evolution of BC particles in the atmosphere follows
similar patterns across diverse locations and emission sce-
narios. Due to the non-linear responses of
E
abs
to coating
amount (
Figure S8
), the oversimplified BC optical calculation us-
ing bulk-averaged
M
R
cannot accurately capture the realistic
E
abs
. Besides, two pieces of evidence, i.e., the higher increase
in
M
R
and mobility diameter for smaller BC-core particles
(
Figures 1
B and 1E) and the increasing BC cores with the aging
degree (
Figure 1
F), imply the importance of coagulation on BC
microphysical evolution. Such coagulation processes might pro-
duce complex morphological features and likely regulate
E
abs
.
Note that the conventional aerosol microphysics models do
not capture such
M
R
heterogeneities and morphological fea-
tures, leading to potentially significant bias in the evaluation of
BC optical properties.
Figure 1. Measured variation and evolution
of BC microphysics during the field obser-
vation at an urban site in China
(A) The BC mass distribution in each BC-core
mass equivalent diameters (
D
c
) and coating-to-BC
mass ratios (
M
R
). The average standard deviation
of
log
10
(
M
R
) is approximately 0.50.
(B) Size-resolved
M
R
as a function of bulk-aver-
aged
M
R
.
(C) Size-resolved effective density (
r
eff
)asa
function of bulk-averaged
M
R
. The gray belt rep-
resents the standard deviation of the material
density (calculated by chemical compositions).
(D) Size-resolved fraction of the non-spherical
particles (
F
non-spherical
) as a function of bulk-aver-
aged
M
R
.
(E) The particle mobility diameter (
D
p_mob
)asa
function of bulk-averaged
M
R
.
In (B)–(E), the shadows represent the the upper
and lower limits of relevant parameters for parti-
cles with different
D
c
ranges, i.e., blue and red
shadows for 100–120 nm and 280–300 nm,
respectively. The dashed lines represent the
slopes of microphysics changes for particles with
D
c
of 100–120 (blue line) and 280–300 nm (red
line), respectively. The black dotted lines repre-
sent the bulk-average or mean values.
(F) The
D
c
distribution (log-normal fit) under
different bulk-averaged
M
R
. The gray shade rep-
resents the average distribution.
Reconciling BC
E
abs
In this study, BC
E
abs
exhibits distinct
differences between observations and
theoretical calculations. The observed
bulk-averaged
E
abs
over the entire popu-
lation values at the wavelength of 870 nm
are 1.12 ± 0.07 and 1.10 ± 0.07 at the ur-
ban and regional sites, respectively.
Moreover, the observed
E
abs
increases
only slightly with elevated bulk-averaged
M
R
(
Figures 2
A and
S9
). On the other hand, the original core-shell
Mie model, which assumes a homogeneous mixing state and
a core-shell structure for all particles, predicts
E
abs
of up to
1.8, approximately 50% higher than the measured values
(
Figures 2
A,
S10
A, and S10B). Such a model-observation
discrepancy is consistent with previous field studies in California
(USA), Toronto (ON, Canada), Nagoya (Japan), and other places
worldwide.
10
,
13–15
Having measured single-particle micro-
physics for an entire BC-containing particle population, we
derive an observational-based framework to address the
model-measurement discrepancy.
Considering heterogeneities in BC-core diameters and
M
R
jointly explains
20% of the overestimation in BC
E
abs
by the
original core-shell Mie model (
Figures 2
A and
S10
B). As dis-
cussed above, the bias results from the nonlinear responses of
E
abs
to coating amount, particularly for particles with different
BC-core sizes (
Figure S8
). Compared to the particles with large
BC cores, those with small cores exhibit lower size-resolved
E
abs
in the initial aging stage, although their
M
R
is comparable
(
Figure 2
B). During the aging process, the particles with small
ll
OPEN ACCESS
Article
138
One Earth
7
, 136–145, January 19, 2024
cores exhibit a more rapid increase in
M
R
, yet their absorption
does not significantly enhance (
Figure 2
B). Moreover, the
explainable proportion is relatively smaller than the previous
study using a particle-resolved model when considering
M
R
heterogeneities.
22
This may be caused by the smaller standard
deviation of the particle-to-particle
M
R
observed in our observa-
tions compared to their model simulation.
Irregularity in particle shape is one essential microphysical
feature of BC-containing particles that the original core-shell
Mie model does not possess. In the previous studies, the impact
of irregular shape on
E
abs
was evaluated by simply assuming a
critical
M
R
for determining thinly coated BC-containing particles
and a small
E
abs
for them.
24
,
29
Here, leveraging the single-parti-
cle shape information, we evaluate this impact by adopting a
relatively low
E
abs
value of 1.1 for all non-spherical particles.
6
On average, the modified light-absorbing capability for non-
spherical particles explains
30% of the overestimation in BC
E
abs
by the original core-shell Mie model (
Figure 2
A). Such modi-
fication of light absorbing is more significant on particles with
larger BC cores (
Figure 2
C), likely due to the larger proportion
of non-spherical particles (
Figure 1
D). In particular, the modified
E
abs
reasonably reproduces the measured
E
abs
in the initial aging
stage (bulk-averaged
M
R
= 3–3.5) but still exhibits non-negligible
biases over the fully aged stage (bulk-averaged
M
R
= 6–6.5) (
Fig-
ure 2
A). To further confirm this non-negligible gap, we also
employ two existing parameterizations to calculate the shape
impacts on
E
abs
.
7
,
22
Not surprisingly, none of the above three
shape-related modification methods are sufficient to resolve
the gap in bulk
E
abs
estimation (
Figure S11
;
Table S1
).
Since the joint impacts of the
M
R
heterogeneity and non-spher-
ical shape have been evaluated to be insufficient to interpret the
model-measurement gap, the remaining difference in
E
abs
is
necessarily related to the complexity of spherical BC-containing
particles. The core-shell model explicitly assumes that the BC
core resides in the center of the host particle to form a lensing ef-
fect. This assumption is generally reasonable for BC-containing
particles in laboratory and field chamber experiments because
their coatings are evenly distributed on fresh BC particles by
gas-particle phase partitioning.
6
,
7
,
18
In field measurements, as
illustrated in the transmission electron microscope (TEM) images,
BC cores of spherical BC-containing particles are rarely located in
the center of the spherical hosts but tend to be randomly distrib-
uted within the host particles or even on their surface (
Figure S12
).
This off-center BC-core position likely results from atmospheric
processes such as coagulation between BC and non-BC particles
and liquid-liquid phase separation during the aging processes.
32
Figure 2. Measured and modeled
E
abs
of BC-containing particles and the influences of real-world BC microphysics on
E
abs
during the field
observation at an urban site in China
(A) Measured and modeled (consider microphysics at different levels)
E
abs
as a function of bulk-averaged
M
R
at a wavelength of
l
= 870 nm. Four different model
frameworks, i.e., (I) original core-shell model, (II) considering
D
c
-
M
R
heterogeneity, (III) considering
D
c
-
M
R
heterogeneity and shape, and (IV) considering
D
c
-
M
R
heterogeneity, shape, and BC-core position, are employed to calculate the
E
abs
of BC particles. The upper and lower lines of the shaded area show the 85
th
and 15
th
percentiles, respectively. The box and whisker plots show the mean (
-
), median (
), lower and upper quartile (boxes), and 10
th
and 90
th
percentiles (whiskers).
(B) Size-resolved
E
abs
in case II as a function of bulk-averaged
M
R
.
(C) Size-resolved
E
abs
in case III as a function of bulk-averaged
M
R
.
In (B) and (C), the shadows represent the upper and lower limits of relevant parameters for particles with different
D
c
ranges.
(D) The overestimated absorption percentage of the ideal core-shell model prediction compared to off-center core particles in other papers.
20
,
21
,
29
,
30
,
31
ll
OPEN ACCESS
Article
One Earth
7
, 136–145, January 19, 2024
139
Theoretical studies have estimated that, compared to the original
core-shellmodel,BC-containingparticleswithoff-centerBCcores
exhibit a smaller lensing effect by 15%–40% (
Figure 2
D), varying
with particle
M
R
and BC-core position within their host
particles.
20
,
21
,
29
,
30
,
31
By averaging existing theoretical estimates,
we introduce a correction factor of 30% for the near-spherical
BC-containing particles in the original core-shell Mie model to
represent the overall effects of BC-core position. Accounting for
the off-center BC-core position of spherical BC-containing parti-
cles explains the rest
40% of the overestimation in BC
E
abs
by
the original core-shell Mie model. More importantly, the modified
E
abs
well reproduces the observed
E
abs
over the entire bulk-aver-
aged
M
R
range (
Figures 2
Aand
S10
B).
The reproduced
E
abs
suggest that light absorption of BC-
containing particles depends not only on bulk-averaged
M
R
but also on
M
R
heterogeneity, shape, and BC-core position.
Accounting for these realistic BC microphysics bridges the
model-measurement discrepancy of
E
abs
for observations at
both the urban and regional sites in the present study
(
Figures 2
Aand
S9
). By fitting with the observational data in
Fig-
ure 2
, we develop a set of BC absorption parameterizations
(see
experimental procedures
) that explicitly address the real-
istic variation and evolution of BC microphysics in the aerosol
optical module. By applying those parameterizations to the
observed aerosol properties at different sites in previous
studies, the reconstructed
E
abs
better match the observed
E
abs
than the original core-shell model and reduce the model-
observation biases to less than 20%
33
(
Figure 3
;
Table S2
).
Climate implications
The present study introduces an observational-based frame-
work and elucidates the underlying causes of BC absorption
overestimation in numerical models. We highlight the complexity
in BC aging processes as well as the applicability of the current
aerosol optical assumption used in global BC DRF estimation.
The widely observed model-measurement gap has been attrib-
uted to (1) shortcomings of the ideal internal mixing model for
fresh and thinly coated particles
6
,
7
and (2) challenges of chamber
experiments in representing realistic particle-to-particle hetero-
geneity of
M
R
during atmospheric BC aging.
22
The present mea-
surements described here confirm the significant role of particle
shape and heterogeneity in
M
R
suggested previously but, more-
over, reveal that they are quantitively insufficient in bridging the
model-measurement gap. We further highlight that the spatial
arrangement of BC and non-BC components within individual
spherical BC-containing particles is also an essential factor in
the model-measurement gap (
Figure 4
).
The present results are important for estimating the global
radiative forcing of BC. A set of observation-based
E
abs
-related
parameterizations have been implemented in the NCAR Com-
munity Earth System Model v.1.2 (CESM1) to evaluate the
possible impacts of a wide range of absorption enhancements
Figure 3. Simulated magnitude and global distribution of BC DRF in the updated NCAR CESM1 with the observation-constrained BC absorp-
tion parameterizations from this study
Top: simulated global mean BC DRF and
E
abs
near the surface from four types of treatment of BC absorption in NCAR CESM1 according to
Figure 2
; middle:
simulated global distribution of BC DRF changes induced by the observation-derived BC absorption parameterization in case IV; and bottom: the avera
ge
absorption enhancement calculated by the parametric method of case IV, homogeneous core-shell model, and measured values at different sites attach
ed in
Table S2
.
10
,
15
,
25
The
E
abs
of BC-containing particles in cases II–IV are parameterized as an exponential square function of
D
p
(a total of 16 size bins were divided)
(see
experimental procedures
).
ll
OPEN ACCESS
Article
140
One Earth
7
, 136–145, January 19, 2024
on the global radiative forcing assessment. Model setup and
experiment descriptions can be found in the
experimental pro-
cedures
. By comparing the DRF and
E
abs
using different param-
eterizations that consider different levels of BC microphysics
(
Figure 3
, top), we find that taking the heterogeneities of BC-
core diameters and
M
R
(from case I to case II) into account con-
tributes most to the
E
abs
and DRF reductions. The global mean
DRF can be reduced by 13%. Considering the BC-core position
induces an additional 7% reduction in DRF (from case III to case
IV). In contrast, the effect of shape is relatively small (from case II
to case III), likely due to the large
M
R
predicted in this model.
Globally, with all observational constraints on
E
abs
, we find the
global BC DRF is reduced from 0.30 ± 0.01 to 0.23 ± 0.01 W
m
2
, corresponding to about a 23% fractional change (
Figure 3
).
Spatially, larger reductions in BC DRF occur near the BC source
regions, such as East Asia, Central Africa, and the Southern
Amazon. Higher primary BC emissions generally come with a
larger overestimation of DRF.
Reducing BC emissions is desirable for both air pollution con-
trol and climate mitigation. Due to the homology of CO
2
and BC
emissions, the reduction of fossil fuel combustion driven by the
‘‘carbon neutral’’ strategy will inevitably lead to a reduction of
BC emissions. Observationally constrained BC-forcing assess-
ments will help assess the environmental and climatic benefits
of future BC reduction scenarios. A smaller BC positive forcing
also may have profound implications for the Earth’s climate
sensitivity and future warming degree.
Conclusions
This study employs
in situ
measurements of BC microphysical
properties in the atmosphere to address the influence of BC
complexity within aerosols on its light-absorption capability. The
research reveals the real-world variation and evolution of BC
microphysics, regulated by atmospheric aging processes. Aging
of BC particles expands
M
R
heterogeneity and reshapes the parti-
clesasmorecompact and spherical,aspects not fully capturedby
theoretical assumptions from laboratory and modeling studies.
Specifically, the observed heterogeneity in size and coating, along
withthenon-sphericalshapeofBCparticles,accountsforapprox-
imately 20% and 30% of the observed lower BC absorption ca-
pacity, respectively. However, the combined impacts of the
M
R
heterogeneityandnon-sphericalshapearequantitivelyinsufficient
in bridging the model-measurement gap. We further find that a re-
mainingabsorptiongapforfullyagedsphericalBC-containingpar-
ticles is necessarily related to the spatial arrangement of BC and
non-BC components within individual BC-containing particles.
By considering comprehensively all observed BC microphysical
attributes, the estimated global BC DRF in climate models is
reduced from 0.30 ± 0.01 to 0.23 ± 0.01 W m
2
, corresponding
to about a 23% fractional change.
This study highlights the critical importance of accounting for
the real-world complexity of BC microphysics in measurements
and models when accurately assessing BC environmental and
climate effects. Future research and models should strive to
more accurately depict the diversity of BC within atmospheric
aerosols, thereby enhancing our understanding and estimation
of its role in global climate change.
EXPERIMENTAL PROCEDURES
Resource availability
Lead contact
Further information and requests for resources and reagents should be directed
to and will be fulfilled by the lead contact, Jianfei Peng (
pengjianfei@nankai.
edu.cn
).
Materials availability
This study did not generate new unique materials.
Data and code availability
Data for the figures in the main text have been deposited at Zenodo under the
https://doi.org/10.5281/zenodo.10208500
and are publicly available as of the
date of publication. The original code of the CESM model used in this study is
available at
https://www2.cesm.ucar.edu/models/cesm1.2/
. Custom code for
this work is available from the
lead contact
upon reasonable request. Any addi-
tional information required to reanalyze the data reported in this paper is avail-
able from the
lead contact
upon request.
Overview of field observations
One field observation was conducted on the campus at Peking University
Shenzhen Graduate School (22.60
N, 113.97
E) in Shenzhen (PKUSZ),
China, during autumn (October 24–November 8, 2020) and winter (January
15–January 31, 2021). This site represents a typical urban environment, with
Figure 4. The evolution of microphysics and
E
abs
of BC-containing particles in numerical
models, chambers, and real atmosphere dur-
ing aging processes
Model schemes (top) consider BC-containing par-
ticles as homogeneous core-shell structures, pro-
ducing a lensing effect that significantly enhances
absorption. Laboratory chamber experiments (mid-
dle) can further address the shape of particles and
determine low
E
abs
for non-spherical particles with
small
M
R
. However, chamber experiments fail to
represent the particle-to-particle heterogeneity of
M
R
during atmospheric BC aging. For realistic BC
aging processes in the atmosphere (bottom), parti-
cles exhibit remarkable
M
R
heterogeneities, non-
spherical shape, and off-center BC-core position.
Taking these complex microphysics into account,
the BC will retain weak absorption even over the fully
aged stage and reduces the BC direct radiative
forcing by
23%. The
E
abs_global
is the simulated
global mean
E
abs
.
ll
OPEN ACCESS
Article
One Earth
7
, 136–145, January 19, 2024
141
vehicle emissions being the most important source of BC, followed by biomass
burning.
34
A series of state-of-the-art instruments were utilized to measure the
chemical and physical properties of ambient BC-containing particles (
Fig-
ure S1
). Since the detected size ranges of DMA-SP2 and SP-AMS are less
than 1,000 nm, a PM
1
cyclone was used for the optical instruments to ensure
consistency with other instruments on particle cutoff size. Furthermore, a
chemiluminescence instrument (Thermo Scientific, Model 42i) equipped with
a molybdenum reactor was used to measure the NOx, and a 6000X2 PTR-
ToF-MS instrument (Ionicon Analytik, Innsbruck, Austria) was used to measure
the volatile organic compounds (VOCs). All sampling air was dried to ensure a
low relative humidity (<35%). Another field observation was conducted on
Yangmeikeng (22.55
N, 114.60
E) in Shenzhen (YMK), a regional site in Shenz-
hen, China, during late summer (October 8–October 30, 2019). YMK is situated
in a less populated area with minimal local influences, predominantly experi-
encing regional transport of pollutants (
Figure S2
D). PM
2.5
cyclones and dryers
were also used during sampling at this site. The two sites allowed us to capture
both urban and regional air pollution characteristics.
During our measurements, the sampled air masses were transported from
different directions (e.g., Southeast Coastal, Northern China, Intra-Guangdong
Province,andSouthwesternCoastal),representingvariouspollutioncharacteris-
tics and emission sources (
Figures S2
A–S2C). At the PKUSZ site, local vehicular
and industrial emissions contributed significantly to atmospheric pollutants.
34
,
35
In contrast, the YMK sitewas primarily influenced by regionaltransport for indus-
trial emissions, biomass burning, and/or ship emissions (
Figure S3
).
36
Measurement of the BC chemical composition
SP-AMS was used to measure the concentrations and composition of BC
-
containing particles
10
,
25
,
37
The SP-AMS is equipped with an intracavity
infrared laser vaporizer (same as the SP2 laser)
37
with the tungsten thermal
vaporizer removed so that absorbing particles are heated to their boiling
point (>1,000
C), incandesce, and vaporize, enabling the detection of BC
and internally mixed non-refractory components.
38
,
39
The SP-AMS mass
spectrometer is operated only in ‘‘V’’ mode (average m/
D
m
3,000) for
higher sensitivity for ions with mass-to-charge ratios up to
m/z
= 450. The
collection efficiency was calculated based on BC measured by the SP2 as
described by Collier et al.,
40
with an average of 0.48 in winter. The calibration
of SP-AMS was carried out before and after the measurement.
14
,
37
,
41
The
corrected BC core and mass concentration were in good agreement with
that measured by DMA-SP2 (
Figure S16
), which indicates the accuracy of
SP-AMS quantification. In addition, the relative ionization efficiencies (RIEs)
of organic, nitrate, sulfate, ammonium, and chloride were 1.4, 1.05, 0.82,
4, and 1.1, respectively, in accordance with most previous studies.
40
Bulk-
averaged
M
R
is determined as the mass concentration ratio of total coatings
and BC core.
Measurement of BC size and mixing state
We combined the DMA and SP2 to measure directly the
D
p_mob
and
D
c
of each
BC-containing particle in PM
1
. The DMA selected dry particles in the
D
p_mob
range of 69.8–582.9 nm, and the SP2 measured single BC-containing particles
with a
D
c
range of 70–580 nm.
42
,
43
Size-selected fullerene soot was used for
the SP2 mass calibration.
44
The data measured by DMA-SP2 were corrected
via several steps, including (1) correction of collection efficiency, (2) correction
of delay time, (3) correction of multi-charged particles, and (4) extrapolation of
the size range (for details, see
Note S2
and
Figures S13–S16
).
Measurement of BC density and DSF
The
r
material_bulk
of coating and particles were computed as the volume-
weighted average of the material densities of components measured by SP-
AMS, where the assumed
r
material
for BC, ammonium nitrate, ammonium
sulfate, and ammonium chloride were 1.8, 1.72, 1.77, and 1.53 g cm
3
, respec-
tively. In addition, the
r
eff_bulk
of BC-containing particles was calculated ac-
cording to
Equation 1
,
r
eff
bulk
=
6
M
p
bulk
p
D
p
mob
3
(Equation 1)
where
M
p_bulk
is the bulk-averaged mass of BC-containing particles measured
by SP-AMS and (
D
p_mob
)3 is the bulk-averaged (
D
p_mob
)3 of BC-containing
particles measured by DMA-SP2. At the same mass, a lower
r
eff
indicates a
more fractal particle shape.
Further, the size-resolved
r
eff
was inverted based on the bulk-averaged
r
eff
,
the bulk-averaged
r
material,
and the size distribution of BC-containing particles.
The main steps included (1) determining
r
eff
for fully fresh and aged particles
under each
D
c
bin, (2) fitting the
r
eff
curve under each
D
c
bin, and (3) selecting
the optimal
r
eff
curve (for details, see
Note S3
).
Therefore, the
M
R
of single BC-containing particles can be calculated by
Equation 2
. The
M
R
in single particles was converted to
M
R
in bulk (
M
R_bulk
)
by summation of the total non-BC and BC single-particle masses over a given
period (
Equation 3
):
M
R
=
r
eff
3
D
3
p
mob
1
:
8
3
D
3
c
1
(Equation 2)
M
R
bulk
=
P
i
M
R
;
i
3
P
i
M
BC
;
i
P
i
M
BC
;
i
(Equation 3)
where
i
denotes the
i
th single particle.
Based on
r
eff
and
r
material
, we determined the DSF by
DSF
=
D
p
mob
C
c
D
p
ve
D
p
ve
C
c
D
p
mob
=
ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r
material
r
eff
3
r
3
C
c
D
p
ve
C
c
D
p
mob
(Equation 4)
where
C
c
is the slip correction factor at a given diameter
45
and
D
p_ve
is the vol-
ume-equivalent particle diameter. If the DSF approaches 1, the particles tend
to be spherical. Here, we take 1.05 as the criterion to classify the BC-contain-
ing particles into two groups, i.e., near-spherical and non-spherical particles.
Experimental results demonstrated that when DSF <1.05, the measured ab-
sorption enhancement of BC was consistent with calculations from the Mie
core-shell model.
6
Measurement of aerosol optical properties
A particle absorption extinction (PAX) meter (Droplet Measurement Technolo-
gies) was used to measure the absorption coefficients (
abs
:Mm
1
)at
870 nm.
7
,
15
,
46
Before sampling, the absorption coefficient of PAX at a wave-
length of 870 nm was calibrated using glassy carbon spheres (distributed by
Alpha Aesar). The scatterplot of the calculated [extinction
scattering] vs.
the measured
abs
is performed in
Figure S15
B, and the slope is close to 1. In-
strument zeros were determined using filtered ambient air for 10 s of every
3 min, and the detection line is less than 1 Mm
1
. Besides, absorption at
880 nm (measured by AE-31, Magee Scientific, Aethalometer
31) was
used to confirm the accuracy and stability of PAX (
Figure S19
A).
Measured and modeled
E
abs
The
E
abs
of BC-containing particles is defined as the ratio of the coated and
uncoated mass absorption cross-section (MAC) (
Equation 5
),
E
abs
meas
=
MAC
BC
MAC
ref
(Equation 5)
where
E
abs_meas
is the measured light absorption enhancement and MAC
ref
is
the mass absorption coefficient for uncoated BC particles,
47
which was deter-
mined by extrapolating MAC
BC
to the limit of bulk-averaged
M
R
= 0 when
calculating
E
abs_meas
(
Figure S19
C).
15
In this study, the MAC
ref
was determined
to be 4.81 at
l
= 870 nm, close to the value recommended by Bond and
Bergstrom.
47
Optical models widely used at present include core-shell Mie theory,
T-matrix, and discrete dipole approximation (DDA). Core-shell Mie theory
48
is used to calculate the
E
abs
based on the BC-core size (
D
c
) and coating thick-
ness. The T-matrix and DDA can fully consider the influence of particle
morphology on optics by inputting detailed three-dimensional information
such as particle fractal dimension and monomer number.
21
,
49
,
50
Considering
the necessary input parameters for optical models and calculation efficiency,
we adopt the core-shell Mie theory for optical calculation in this study. The size
distribution of the BC core and BC-containing particles is measured by the
DMA-SP2 tandem system. All coating mass concentrations are corrected by
size-resolved
r
eff
. For the original core-shell assumption (case I), the
M
R
of
ll
OPEN ACCESS
Article
142
One Earth
7
, 136–145, January 19, 2024
every
D
c
was equal to bulk-averaged
M
R
, as described by Cappa et al.
10
and
Liu et al.
7
The
D
c
-resolved coating thickness was incorporated into the core-
shell model for case II. Based on the identification of spherical and non-spher-
ical particles, only the
D
c
-resolved coating thickness of spherical particles was
characterized in the core-shell model for case III, while the
E
abs
of non-spher-
ical particles was assumed to be 1.1. For case IV, on the basis of case III, only
the
E
abs
of spherical particles was applied by a calibration factor of 30%. After
obtaining the size-resolved MAC, we can calculate the MAC of BC particle en-
sembles as
MAC
BC
=
X
n
i
=
0
MAC
i
3
½
BC
i
½
BC
(Equation 6)
where MAC
i
and [BC]
i
are the size-resolved MAC
BC
and mass concentration of
the BC core, [BC] is the mass concentration of BC particle ensembles, and n is
the number of size bins. Then, the
E
abs
of BC particle ensembles are calculated
according to
Equation 6
. Note that the MAC
ref
here is calculated by the core-
shell Mie model when
D
p
/
D
c
=1.
The refractive indices (
RI
s) of BC and its coatings are assumed to be n =
1.85 + 0.71i and n = 1.5 + 0i.
7
,
10
,
51
Previous studies have pointed out that
the real part of the
RI
in the environment was between 1.4 and 1.6.
52
Here,
we use these values as the upper and lower limits to estimate the influence
of the real part of the
RI
on BC
E
abs
in different model frameworks (
Figure S20
).
Results show that the calculated
E
abs
of each model framework (except model
IV) is significantly larger than the measured value in the entire
RI
range. There-
fore, we believe that the
RI
is not the main reason for the overestimation of the
measured
E
abs
by the original core-shell model.
Parameterization and global model simulations
The Community Atmospheric Model v.5 (CAM5) in NCAR CESM1 is used to
assess the BC DRF at the top of atmosphere (TOA) under different cases as
shown in
Figure 2
. A ‘‘double-call’’ method is used to calculate the differences
in radiative fluxes at TOA.
53
To resolve size-dependent aerosol optical proper-
ties, we implemented a core-shell Mie code with 16 size bins. The
E
abs
of case I
was calculated by the core-shell Mie model, while in case II to case IV,
E
abs
was
parameterized as the exponential square function of
D
p
and bulk-averaged
M
R
(
Table 1
). The calculated values of parameterization in different cases are
shown in
Figure S21
.
CAM5 model is configured with 1
horizontal resolution and 30 vertical
levels. The climatological run is performed for 4 years, in which the first year
is used for spin up and the averages of the remaining years are used for optical
and forcing analyses. The atmospheric component CAM5 by default employs
a three-mode Modal Aerosol Module (MAM3) and parameterizes the aerosol
optical properties using fitting coefficients for each mode.
54
In MAM3, BC is
evenly distributed within the accumulation mode and internally mixed with
other species in that mode. Such a treatment artificially eliminates the separa-
tion of BC-contacting and non-BC particles. We compile the available
in situ
observations worldwide (
Figure S9
C) for which the mean ratio of BC-contain-
ing particles is about 29%. Then, the absorption coefficient is calculated by the
above parametric method and other aerosol optical properties, such as scat-
tering coefficient and asymmetric factors, from the online Mie calculation im-
plemented into the model. Note that CESM1/MAM3 suffers from aerosol
composition biases, so the absolute absorption is of uncertainty compared
to observations.
55
Note that our BC optical parameterizations do not explicitly
treat the loss of coating material that occurs at a longer timescale in clean con-
ditions, such as the upper part of the free troposphere. However, it should not
significantly impact our forcing variation assessment, as BC abundance in
those regions is rather small compared to the near-BC-source regions.
Some extreme cases such as BC injection into the lower stratosphere by
mega wildfire plumes
55
are not considered in the model, which mainly con-
cerns the climatological mean states.
SUPPLEMENTAL INFORMATION
Supplemental information can be found online at
https://doi.org/10.1016/j.
oneear.2023.12.004
.
ACKNOWLEDGMENTS
This work was supported by the National Key Research and Development Pro-
gram of China (2022YFC3701000 to X.-F.H.), the Science and Technology Plan
of Shenzhen Municipality (JCYJ20200109120401943 to X.-F.H.), and the Na-
tional Natural Science Foundation of China (42175123 to J.P.).
AUTHOR CONTRIBUTIONS
Conceptualization, J.P., L.-Y.H., and Y.W.; methodology, X.-F.H., Y.P., J.W.,
Y.W., J.P., and L.-Y.H.; data analysis, X.-F.H., Y.P., J.W., J.P., Y.W., X.-Y.L.,
M.-X.T., Y.C., Z.M., T.F., J.Z., L.-M.C., and C.L.; visualization, Y.P., J.W.,
and Y.W.; writing—original draft, X.-F.H., Y.P., J.W., J.P., and Y.W.;
writing—review and editing, J.P., Y.W., L.-Y.H., X.-F.H., and J.H.S.
DECLARATION OF INTERESTS
The authors declare no competing interests.
INCLUSION AND DIVERSITY
We support inclusive, diverse, and equitable conduct of research.
Received: August 15, 2023
Revised: October 20, 2023
Accepted: December 4, 2023
Published: December 27, 2023
REFERENCES
1.
Jacobson, M.Z. (2001). Strong radiative heating due to the mixing state of
black carbon in atmospheric aerosols. Nature
409
, 695–697
.
2.
Ramanathan, V., and Carmichael, G. (2008). Global and regional climate
changes due to black carbon. Nat. Geosci.
1
, 221–227
.
3.
Bond, T.C., Doherty, S.J., Fahey, D.W., Forster, P.M., Berntsen, T.,
DeAngelo, B.J., Flanner, M.G., Ghan, S., K
€
archer, B., Koch, D., et al.
(2013). Bounding the role of black carbon in the climate system: A scien-
tific assessment. JGR. Atmospheres
118
, 5380–5552
.
4.
Stocker, T. (2014). Climate Change 2013: The Physical Science Basis :
Working Group I Contribution to the Fifth Assessment Report of the
IntergovernmentalPanel onClimate Change (CambridgeUniversityPress)
.
Table 1. The parameterizations for different scenarios
Case
Case II
Case III
Case IV
Mathematical
form
E
abs
= a1 + a2*exp(
a3*
D
p
)
Parameter 1
a1 = 1.9684 –
0.6634*exp(
0.75333*
M
R_bulk
)
a1 = 1.8836 –
4.3785*exp(
0.7655*
M
R_bulk
)
a1 = 1.4894 –
3.2014*exp(
0.91984*
M
R_bulk
)
Parameter 2
a2 = 1.7099 +
0.98182*exp(
0.29812*
M
R_bulk
)
a2 =
1.6299 +
4.931*exp(
0.63489*
M
R_bulk
)
a2 =
0.94323 +
1.5952*exp(
0.62832*
M
R_bulk
)
Parameter 3
a3 = 0.00751 – 0.00387*
exp(
0.20256*
M
R_bulk
)
a3 = 0.00796 +
0.91038*exp(
1.8842*
M
R_bulk
)
a3 = 0.006248 +
5.8054*exp(
2.5054*
M
R_bulk
)
ll
OPEN ACCESS
Article
One Earth
7
, 136–145, January 19, 2024
143
5.
Shiraiwa, M., Kondo, Y., Moteki, N., Takegawa, N., Sahu, L.K., Takami, A.,
Hatakeyama, S., Yonemura, S., and Blake, D.R. (2008). Radiative impact
of mixing state of black carbon aerosol in Asian outflow. J. Geophys.
Res.
113
, D24210
.
6.
Peng, J., Hu, M., Guo, S., Du, Z., Zheng, J., Shang, D., Levy Zamora, M.,
Zeng, L., Shao, M., Wu, Y.-S., et al. (2016). Markedly enhanced absorption
and direct radiative forcing of black carbon under polluted urban environ-
ments. Proc. Natl. Acad. Sci. USA
113
, 4266–4271
.
7.
Liu, D., Whitehead, J., Alfarra, M., Reyes-Villegas, E., Spracklen, D.,
Reddington, C., Kong, S., Williams, P., Ting, Y.-C., Haslett, S., et al.
(2017). Black-carbon absorption enhancement in the atmosphere deter-
mined by particle mixing state. Nat. Geosci.
10
, 184–188
.
8.
Bond, T.C., Habib, G., and Bergstrom, R.W. (2006). Limitations in the
enhancement of visible light absorption due to mixing state. J. Geophys.
Res.
111
, D20211
.
9.
Shiraiwa, M., Kondo, Y., Iwamoto, T., and Kita, K. (2010). Amplification of
Light Absorption of Black Carbon by Organic Coating. Aerosol Sci.
Technol.
44
, 46–54
.
10.
Cappa, C.D., Onasch, T.B., Massoli, P., Worsnop, D.R., Bates, T.S.,
Cross, E.S., Davidovits, P., Hakala, J., Hayden, K.L., PetA
̈
JA
̈
, T., et al.
(2012). Radiative Absorption Enhancements Due to the Mixing State of
Atmospheric Black Carbon. Science
337
, 1078–1081
.
11.
Helgestad, T.M. (2016). Characterizing the Optical Properties of Coated
Black Carbon Particles (ProQuest Dissertations Publishing)
.
12.
Lan, Z.-J., Huang, X.-F., Yu, K.-Y., Sun, T.-L., Zeng, L.-W., and Hu, M.
(2013). Light absorption of black carbon aerosol and its enhancement by
mixing state in an urban atmosphere in South China. Atmos. Environ. X.
69
, 118–123
.
13.
Nakayama, T., Ikeda, Y., Sawada, Y., Setoguchi, Y., Ogawa, S., Kawana,
K., Mochida, M., Ikemori, F., Matsumoto, K., and Matsumi, Y. (2014).
Properties of light-absorbing aerosols in the Nagoya urban area, Japan,
in August 2011 and January 2012: Contributions of brown carbon and
lensing effect. JGR. Atmospheres
119
, 721–812
.
14.
Healy, R.M., Wang, J.M., Jeong, C.H., Lee, A.K.Y., Willis, M.D., Jaroudi,
E., Zimmerman, N., Hilker, N., Murphy, M., Eckhardt, S., et al. (2015).
Light-absorbing properties of ambient black carbon and brown carbon
from fossil fuel and biomass burning sources. JGR. Atmospheres
120
,
6619–6633
.
15.
Cappa, C.D., Zhang, X., Russell, L.M., Collier, S., Lee, A.K.Y., Chen, C.L.,
Betha, R., Chen, S., Liu, J., Price, D.J., et al. (2019). Light Absorption by
Ambient Black and Brown Carbon and its Dependence on Black Carbon
Coating State for Two California, USA, Cities in Winter and Summer.
JGR. Atmospheres
124
, 1550–1577
.
16.
Zhai, J., Yang, X., Li, L., Bai, B., Liu, P., Huang, Y., Fu, T.-M., Zhu, L., Zeng,
Z., Tao, S., et al. (2022). Absorption Enhancement of Black Carbon
Aerosols Constrained by Mixing-State Heterogeneity. Environ. Sci.
Technol.
56
, 1586–1593
.
17.
Zhang, R., Khalizov, A.F., Pagels, J., Zhang, D., Xue, H., and McMurry,
P.H. (2008). Variability in Morphology, Hygroscopicity, and Optical
Properties of Soot Aerosols during Atmospheric Processing. Proc. Natl.
Acad. Sci. USA
105
, 10291–10296
.
18.
Guo, S., Hu, M., Lin, Y., Gomez-Hernandez, M., Zamora, M.L., Peng, J.,
Collins, D.R., and Zhang, R. (2016). OH-Initiated Oxidation of m-Xylene
on Black Carbon Aging. Environ. Sci. Technol.
50
, 8605–8612
.
19.
China, S., Mazzoleni, C., Gorkowski, K., Aiken, A.C., and Dubey, M.K.;
Michigan Technological Univ, H.M.I. (2013). Morphology and mixing state
of individual freshly emitted wildfire carbonaceous particles. Nat.
Commun.
4
, 2122
.
20.
Wang, Y., Li, W., Huang, J., Liu, L., Pang, Y., He, C., Liu, F., Liu, D., Bi, L.,
Zhang, X., and Shi, Z. (2021). Nonlinear Enhancement of Radiative
Absorption by Black Carbon in Response to Particle Mixing Structure.
Geophys. Res. Lett.
48
, e2021GL096437
.
21.
Adachi, K., Chung, S.H., and Buseck, P.R. (2010). Shapes of soot aerosol
particles and implications for their effects on climate. J. Geophys.
Res.
115
.
22.
Fierce, L., Onasch, T.B., Cappa, C.D., Mazzoleni, C., China, S., Bhandari,
J., Davidovits, P., Fischer, D.A., Helgestad, T., Lambe, A.T., et al. (2020).
Radiative absorption enhancements by black carbon controlled by parti-
cle-to-particle heterogeneity in composition. Proc. Natl. Acad. Sci. USA
117
, 5196–5203
.
23.
Wang, J., Wang, J., Cai, R., Liu, C., Jiang, J., Nie, W., Wang, J., Moteki, N.,
Zaveri, R.A., Huang, X., et al. (2023). Unified theoretical framework for
black carbon mixing state allows greater accuracy of climate effect esti-
mation. Nat. Commun.
14
, 2703
.
24.
Wu, Y., Cheng, T., and Zheng, L. (2020). Light absorption of black carbon
aerosols strongly influenced by particle morphology distribution. Environ.
Res. Lett.
15
, 094051
.
25.
Liu, S., Aiken, A.C., Gorkowski, K., Dubey, M.K., Cappa, C.D., Williams,
L.R., Herndon, S.C., Massoli, P., Fortner, E.C., Chhabra, P.S., et al.
(2015). Enhanced light absorption by mixed source black and brown car-
bon particles in UK winter. Nat. Commun.
6
, 8435
.
26.
Olfert, J., and Rogak, S. (2019). Universal relations between soot effective
density and primary particle size for common combustion sources.
Aerosol Sci. Technol.
53
, 485–492
.
27.
Hu, K., Liu, D., Tian, P., Wu, Y., Deng, Z., Wu, Y., Zhao, D., Li, R., Sheng, J.,
Huang, M., et al. (2021). Measurements of the Diversity of Shape and
Mixing State for Ambient Black Carbon Particles. Geophys. Res. Lett.
48
, e2021GL094522
.
28.
Hu, K., Liu, D., Tian, P., Wu, Y., Li, S., Zhao, D., Li, R., Sheng, J., Huang, M.,
Ding, D., et al. (2022). Identifying the Fraction of Core–Shell Black Carbon
Particles in a Complex Mixture to Constrain the Absorption Enhancement
by Coatings. Environ. Sci. Technol. Lett.
9
, 272–279
.
29.
Wu, Y., Cheng, T., Liu, D., Allan, J.D., Zheng, L., and Chen, H. (2018). Light
Absorption Enhancement of Black Carbon Aerosol Constrained by
Particle Morphology. Environ. Sci. Technol.
52
, 6912–6919
.
30.
Liu, C., Li, J., Yin, Y., Zhu, B., and Feng, Q. (2017). Optical properties of
black carbon aggregates with non-absorptive coating. J. Quant.
Spectrosc. Radiat. Transf.
187
, 443–452
.
31.
Zhang, X., Mao, M., Yin, Y., and Wang, B. (2017). Absorption enhancement
of aged black carbon aerosols affected by their microphysics: A numerical
investigation. J. Quant. Spectrosc. Radiat. Transf.
202
, 90–97
.
32.
Zhang, J., Wang, Y., Teng, X., Liu, L., Xu, Y., Ren, L., Shi, Z., Zhang, Y.,
Jiang, J., Liu, D., et al. (2022). Liquid-liquid phase separation reduces radi-
ative absorption by aged black carbon aerosols. Commun. Earth Environ.
3
, 128–129
.
33.
Cao, L.-M., Huang, X.-F., Li, Y.-Y., Hu, M., and He, L.-Y. (2018). Volatility
measurement of atmospheric submicron aerosols in an urban atmosphere
in southern China. Atmos. Chem. Phys.
18
, 1729–1743
.
34.
Huang, X.F., Zou, B.B., He, L.Y., Hu, M., Pre
́
vo
ˆ
t, A.S.H., and Zhang, Y.H.
(2018). Exploration of PM2.5 sources on the regional scale in the Pearl
River Delta based on ME-2 modeling. Atmos. Chem. Phys.
18
,
11563–11580
.
35.
Zhu, B., Huang, X.-F., Xia, S.-Y., Lin, L.-L., Cheng, Y., and He, L.-Y. (2021).
Biomass-burning emissions could significantly enhance the atmospheric
oxidizing capacity in continental air pollution. Environ. Pollut.
285
, 117523
.
36.
Onasch, T.B., Trimborn, A., Fortner, E.C., Jayne, J.T., Kok, G.L., Williams,
L.R., Davidovits, P., and Worsnop, D.R. (2012). Soot Particle Aerosol Mass
Spectrometer: Development, Validation, and Initial Application. Aerosol
Sci. Technol.
46
, 804–817
.
37.
Lee, A.K.Y., Chen, C.-L., Liu, J., Price, D.J., Betha, R., Russell, L.M.,
Zhang, X., and Cappa, C.D. (2017). Formation of secondary organic aero-
sol coating on black carbon particles near vehicular emissions. Atmos.
Chem. Phys.
17
, 15055–15067
.
38.
Massoli, P., Onasch, T.B., Cappa, C.D., Nuamaan, I., Hakala, J., Hayden,
K., Li, S.M., Sueper, D.T., Bates, T.S., Quinn, P.K., et al. (2015).
Characterization of black carbon
-
containing particles from soot particle
ll
OPEN ACCESS
Article
144
One Earth
7
, 136–145, January 19, 2024