advances.sciencemag.org/cgi/
content/full/4/12/eaau5363/DC1
Supplementary Materials for
Multicomponent new particle forma
tion from sulfuric acid, ammon
ia,
and biogenic vapors
Katrianne Lehtipalo*, Chao Yan, Lubna Dada, Federico Bianchi, M
ao Xiao, Robert Wagner
, Dominik Stolzenburg,
Lauri R. Ahonen, Antonio Amorim
, Andrea Baccarini, Paulus S. Ba
uer, Bernhard Baumgartner, Anton Bergen,
Anne-Kathrin Bernhammer, Martin
Breitenlechner, Sophia Brilke,
Angela Buchholz, Stephany Buenrostro Mazon,
Dexian Chen, Xuemeng Chen, Antonio Dias, Josef Dommen, Danielle
C. Draper, Jonathan Duplissy, Mikael Ehn,
Henning Finkenzeller, Lukas Fis
cher, Carla Frege, Claudia Fuchs
, Olga Garmash, Hamish Gordon, Jani Hakala,
Xucheng He, Liine Heikkinen, Mar
tin Heinritzi, Johanna C. Helm,
Victoria Hofbauer, Christopher R. Hoyle, Tuija Jokinen,
Juha Kangasluoma, Veli-Matti Kerminen, Changhyuk Kim, Jasper Ki
rkby, Jenni Kontkanen, Andreas Kürten,
Michael J. Lawler, Huajun Mai
, Serge Mathot, Roy L. Mauldin II
I, Ugo Molteni, Leonid Nichman, Wei Nie,
Tuomo Nieminen, Andrea Ojdanic, Antti Onnela, Monica Passananti
, Tuukka Petäjä, Felix Piel, Veronika Pospisilova,
Lauriane L. J. Quéléver, Matti P
. Rissanen, Clémence Rose, Nina
Sarnela, Simon Schallhart, Simone Schuchmann,
Kamalika Sengupta, Mario Simon, Mikko Sipilä, Christian Tauber,
António Tomé, Jasmin Tröstl, Olli Väisänen,
Alexander L. Vogel, Rainer Volkam
er, Andrea C. Wagner, Mingyi W
ang, Lena Weitz, Daniel
a Wimmer, Penglin Ye,
Arttu Ylisirniö, Qiaozhi Zha, Ke
nneth S. Carslaw, Joachim Curti
us, Neil M. Donahue, Richard
C. Flagan, Armin Hansel,
Ilona Riipinen, Annele Virtanen,
Paul M. Winkler, Urs Baltenspe
rger, Markku Kulmala*, Douglas R. Worsnop
*Corresponding author. Email: ka
trianne.lehtipalo@helsinki.fi (
K.L.); markku.kulmala@helsinki.fi (M.K.)
Published 12 December 2018,
Sci. Adv.
4
, eaau5363 (2018)
DOI: 10.1126/sciadv.aau5363
This PDF file includes:
Supplementary Materials and Methods
Fig. S1. The effect of different
additional vapors on the NPF r
ates (
J
2.5
).
Fig. S2. The effect of different
additional vapors on the bioge
nic nucleation rate (
J
1.7
) at different
NO
x
concentrations.
Fig. S3. Nucleation rates (
J
1.7
) as a function of the MT to NO
x
ratio (MT/NO
x
).
Fig. S4. Nucleation rates (
J
1.7
) as a function of NH
3
mixing ratio.
Fig. S5. Modeled versus measur
ed nucleation rates.
Fig. S6. Modeled versus measured GRs.
Fig. S7. Positive ions and ion clusters detected during multico
mponent NPF in the CLOUD
chamber.
Fig. S8. Global annual mean concen
trations of vapors involved i
n NPF.
Table S1. Pearson’s correlation coefficient (
R
) between
J
1.7
and the concentration of different
precursors in the chamber.
References (
41
–
56
)
Supplementary Materials and Methods
Main
instrumentation
The
particle
concentration
and
number
size
distribution
in
the
chamber
were
measured
with
several
independent
instruments.
The
particle
size
magnifier
(PSM
(
41
)
,
Airmodus
Ltd.),
together
with
a
condensation
particle
counter
(CPC)
was
used
to
determine
the
number
concentration
s
of
the
smallest
particles.
The
PSM
uses
diethylene
glycol
as
working
fluid
and
achieves
supersaturated
conditions
by
turbulently
mixing
heated
satur
ated
air
with
the
sample
flow.
Particle
growth
takes
place
in
a
cooled
growth
tube,
and
in
the
external
CPC,
which
is
also
used
to
count
the
particles.
Since
the
saturation
ratio
can
be
quickly
adjusted
by
altering
the
flow
rate
of
the
saturated
air,
the
c
ut
-
off
diameter
of
the
PSM
can
be
varied.
In
this
study
two
PSMs
were
operated
with
fixed
cut
-
off
sizes
and
one
in
scanning
mode,
which
allows
determining
the
particle
concentration
at
several
different
cut
-
off
sizes
(1.7
nm
used
in
this
study),
as
well
as
the
number
size
distributions
between
about
1
and
3
nm
(
42
)
.
The
instruments
were
calibrated
before
the
campaign
using
size
-
selected
tungsten
oxide
particles.
Additionally,
a
conventional
butanol
ultra
-
fine
CPC
(TSI
3776)
with
a
cut
-
off
size
of
ca.
2.5
nm
was
used.
The
size
distribution
of
particles
between
1.7
and
8
nm
was
measured
with
a
newly
developed
instrument,
the
DMA
-
train
(
43
)
.
It
consists
of
six
differential
mobility
analyzers
(DMAs)
and
six
CPCs
operated
in
parallel
at
fixed
sizes.
This
provides
high
time
-
resolution
and
allows
exploitation
of
the
full
counting
statistics
at
all
six
sizes.
Thereby
high
sensitivity
to
low
particle
concentrations
is
obtained.
For
larger
particles
we
used
a
commercial
nano
-
SMPS
(TSI
3938)
together
with
a
water
-
CPC
(TS
I
3788),
and
a
home
-
built
SMPS,
consisting
of
a
TSI
X
-
ray
source,
a
long
DMA
and
a
CPC
(TSI
3010).
The
full
size
distribution
produced
by
combining
the
different
measurements
from
different
instruments
thus
spanned
a
range
from
about
1
to
500
nm.
The
ion
concentration
and
size
distribution
were
measured
using
a
neutral
cluster
and
air
ion
spectrometer
(NAIS
(
44
)
,
Airel
Ltd.).
It
simultaneously
determines
the
number
size
distribution
of
positive
and
negative
ions
in
the
range
of
0.75
–
45
nm
mobility
diameter
with
two
cylindrical
mobility
spectrometers
in
parallel,
one
for
each
polarity.
Additionally,
a
corona
charger
is
p
eriodically
switched
on
to
charge
the
aerosol
for
the
detection
of
the
total
particle
size
distribution
in
the
size
range
of
2
–
45
nm.
Concentrations of sulfuric acid and highly oxygenated molecules (HOMs) were measured
with a nitrate
-
ion based chemical io
nization atmospheric pressure interface time
-
of
-
flight
mass spectrometer (CI
-
APi
-
TOF
(
34, 45
)
). The nitrate ions are produced by exposing HNO
3
containing sheath flow to an x
-
ray source. After charging in a drift tube, the sample enters
the APi where it gets
focused and the pressure is gradually reduced to ca. 10
-
6
mbar.
Subsequently, the sample is guided to the TOF region, where the molecules are separated
according to their mass
-
to
-
charge ratio and detected by a microchannel plate detector.
Similar instrume
nts without the chemical charging unit (APi
-
TOF
(
46
)
) were used in
negative and/or positive mode to detect and identify negative and positive ions and charged
clusters.
To quantify sulfuric acid and HOMs, we conducted a calibration and applied corrections
similar to our previous work
(
15, 32
)
. Briefly, in separate experiments the OH concentration
in the chamber was determined using 1,3,5
-
trimethylbenzene in CLOUD10 and 1,2,4
-
trimethylbenzene in CLOUD11. Sulfuric acid production rates were calculated based on
the
OH and SO
2
concentration in the chamber. Then, the concentration of sulfuric acid in the
chamber could be determined using the production rate and losses, including wall loss and
condensation loss to aerosol particles. Additional corrections including
the instrument
transmission correction
(
47
)
, as well as corrections for sampling losses for HOMs were
applied. The overall uncertainty in the sulfuric acid and HOMs concentrations is estimated
to be ca. 40%. The raw data were analyzed with the MATLAB tofto
ols package
(
46
)
. The
elemental composition of each peak was identified using high
-
resolution peak fitting, based
on which we further categorized the HOMs into four different groups: non
-
nitrate HOM
monomer (C
4
-
10
H
x
O
y
), non
-
nitrate dimer (C
11
-
20
H
x
O
y
),
organonitrate monomer (C
4
-
10
H
x
O
y
N
1
-
2
), and organonitrate dimer (C
11
-
20
H
x
O
y
N
1
-
2
). The CI
-
APi
-
TOF mainly detects
hi
ghly oxygenated compounds, with
y≥4 for monomers and y≥6 for dimers.
Ammonia
(NH
3
) concentrations were measured with a quadrupole chemical io
nization mass
spectrometer (CIMS) equipped with an APi inlet
(
48
)
. Positively charged water clusters
((H
2
O)
n∙
H
3
O
+
) were used for the detection of ammonia
(
49
)
. The primary ions are formed by
ionizing humidified synthetic air through a corona discharge at amb
ient pressure
(
50
)
.
Neutral ammonia molecules in the sample air interact with the ionized water clusters
forming (H
2
O)
n
NH
4
+
, which are mainly detected as NH
4
+
since most of the water molecules
evaporate in the collision
-
dissociation cell of the CIMS. The in
strument was calibrated
before and after the experiments for the relevant range of NH
3
; the calibration curves
indicated an excellent linearity and a low detection limit of around 20 pptv. The instrumental
background was found to be approximately 100 pptv.
The measurements have an estimated
overall uncertainty of a factor of two because different inlet systems had to be used between
the instrument calibration and the sampling from the CLOUD chamber. For some of the
early experiments in CLOUD10 the CIMS was
not available, therefore different methods
had to be used for deriving the ammonia mixing ratios. Recently, it was reported that
ammonia can also be detected in the negative ion mode using (HNO
3
)
n
NO
3
–
primary
ions
(
51
)
. The observed NH
3
(HNO
3
)
1,2
NO
3
–
clusters were used to quantify the ammonia
concentrations with the CI
-
APi
-
TOF from a cross calibration with the CIMS when both
instruments were measuring in parallel during later experiments. However, when only very
small amounts of ammonia were added to
the CLOUD chamber, the sensitivity of the CI
-
APi
-
TOF method was not high enough. Therefore, for those experiments, the mixing ratio
was estimated based on the flow of ammonia into the chamber and an experimentally
determined wall loss life time
(
21
)
. The ov
erall scale uncertainty of these methods is
bracketed by a factor 2.5 towards lower values and a factor of 4 towards higher values. In
CLOUD11, NH
3
concentrations were measured with a high
-
resolution time
-
of
-
flight mass
spectrometer (H
-
TOF) using protonate
d water clusters, and the values at high NH
3
concentration were cross
-
checked against a commercial PICARRO NH
3
analyzer. The
estimated uncertainty due to the calibration methods is ca. ± 50%.
The
concentrations
of
monoterpenes
and
other
volatile
organic
c
ompounds
were
measured
with
a
newly
developed
version
of
the
proton
transfer
reaction
time
-
of
-
flight
mass
spectrometer
(PTR
-
TOF
-
MS;
model:
PTR3
(
52
)
).
The
PTR3
has
a
new
inlet
using
center
-
sampling
through
a
critical
orifice
reducing
wall
losses
of
low
vola
tility
compounds.
In
addition,
the
new
ionization
chamber
allows
a
30
-
fold
longer
reaction
time
and
a
40
-
fold
pressure
increase
compared
to
standard
PTR
-
TOF
-
MS
instruments.
Coupled
to
the
lates
t
quadrupole
-
interfaced
Long
-
ToF
mass
analyzer
(TOFWERK),
sensi
tivities
of
up
to
20
000
cps/ppbv
at
a
mass
resolution
of
8000
m/Δm
were
achieved.
Gas
monitors
were
used
to
measure
the
concentration
of
sulfur
dioxide
(SO
2
,
Thermo
Fisher
Scientific,
Inc.
42i
-
TLE),
ozone
(O
3
,
Thermo
Environmental
Instruments
TEI
49C)
and
water
(dew
point
mirror
from
EdgeTech).
Nitric
oxide
(NO)
concentrations
were
determined
from
a
commercially
available
NO
monitor
(ECO
PHYSICS,
model:
CLD
780
TR)
with
a
chemi
-
luminescence
detector.
The
detec
tion
limit
was
ca.
3
pptv
with
an
integrating
time
of
60
s.
During
CLOUD10,
the
amount
of
nitrogen
dioxide
(NO
2
)
was
measured
with
a
cavity
attenuated
phase
shift
nitrogen
dioxide
monitor
(CAPS
NO
2
,
Aerodyne
Research
Inc.)
at
the
bottom
of
the
chamber
(clo
se
to
the
gas
inlet
ports).
During
CLOUD11,
additionally
a
cavity
enhanced
differential
optical
absorption
spectroscopy
(CE
-
DOAS)
instrument
was
deployed
at
the
level
of
the
sampling
ports.
The
concentrations
measured
at
these
two
locations
generally
agree
d
within
20%
for
different
NO
2
injection
rates
and
UV
-
light
settings,
which
can
be
interpreted
as
an
upper
limit
for
the
chamber
inhomogeneity.
The
baseline
of
the
instruments
was
monitored
periodically
by
flushing
them
with
synthetic
air.
Determining
nucleation
and
growth
rates
The
nucleation
rates
(
J
)
were
calculated
from
the
time
derivative
of
the
total
particle
concentration
and
corrected
for
the
particle
losses
in
the
chamber
u
sing
the
full
size
distribution
퐽
=
푑푁
푑푡
+
푆
푑푖푙
+
푆
푤푎푙푙
+
푆
푐표
푎푔
(cm
-
3
s
-
1
) (1)
where
N
is
the
particle
number
concentration
above
a
certain
cut
-
off
size
(
d
p
)
to
which
the
nucleation
rate
is
calculated.
The
dilution
correction
S
dil
arises
from
the
fact
that
the
chamber
is
constantly
flushed
with
synthetic
air
to
account
fo
r
the
instruments’
sample
flows
푆
푑푖푙
=
푁
·
푘
푑푖푙
(cm
-
3
s
-
1
),
(2)
where
k
dil
= 1.437
·
10
-
4
s
-
1
for CLOUD 10 and 1.58
·
10
-
4
s
-
1
for CLOUD11.
Diffusional
losses to the chamber walls (
S
wall
) were determined empirically by observing the
decay of the sulfuric acid monomer concentration in the chamber. The wall loss rate is
inversely
proportional
to the particle size
푘
푤푎푙푙
(
푑
푝
′
,
푇
)
=
2
.
116
·
10
−
3
·
(
푇
푇
푟푒푓
)
0
.
875
·
(
푑
푝
,
푟푒푓
푑
푝
′
)
(s
-
1
)
(3)
where
d
p
’
is the mobility diameter of the particle,
d
p,ref
is the mobility diameter of the sulfuric
acid monomer (= 0.82 nm),
T
ref
= 278 K, and
T
is the actual chamber temperature. Thus the
total wall loss for particles larger than
d
p
is
푆
푤푎푙푙
(
푑
푝
,
푇
)
=
∑
푁
(
푑
푝
′
)
·
푘
푤푎푙푙
(
푑
푝
′
,
푇
)
푑
푝
,
푚푎푥
′
푑
푝
′
=
푑
푝
(cm
-
3
s
-
1
) (4)
Coagulation losses to the surface of larger aerosol particles (
S
coag
) were calculated from the
measured number size distribution of particles present in the chamber
푆
푐표푎푔
(
푑
푝
,
푘
)
=
∑
∑
훿
푖
,
푗
∙
퐾
(
푑
푝
,
푖
,
푑
푝
,
푗
)
∙
푁
푖
∙
푁
푗
푑
푝
,
푚푎푥
푑
푝
,
푗
=
푑
푝
,
푖
푑
푝
,
푚푎푥
푑
푝
,
푖
=
푑
푝
,
푘
(cm
-
3
s
-
1
)
(5)
where
K(d
p,i
,d
p,j
)
is the coagulation coefficient for particles of size
d
p,i
and
d
p,j
,
N
i
and
N
j
are
the number densities of particles in a size bins i and j, and
δ
i,j
= 0.5, if i = j and
δ
i,j
= 1, if i ≠
j.
The
nucleation
rates
at
1.7
nm
(
J
1.7
)
were
calculated
from
the
scanning
PSM
and
verified
against
the
values
calculated
from
the
two
other
PSMs
and
the
butanol
CPC
at
fixed
cut
-
off
sizes.
It
should
be
noted
that
there
is
an
uncertainty
of
about
0.5
nm
in
the
cut
-
off
size
of
the
particle
count
ers
due
to
the
effect
of
composition
and
charge
on
the
detection
efficiency
(
53
)
.
To
account
for
this,
we
verified
the
cut
-
off
size
of
the
PSM
for
each
chemical
system
in
the
chamber
by
comparing
the
concentration
and
rising
time
of
the
PSM
at
different
sat
urator
flow
rates
against
the
different
size
bins
of
the
NAIS,
which
has
been
shown
to
be
very
accurate
in
determining
the
ion
mobility
(
54
)
.
The
J
value
given
for
each
experiment
is
the
median
value
after
reaching
stable
conditions.
The
uncertainty
in
the
nucleation
rates
(given
as
error
bars
in
the
fig
ures)
was
calculated
with
error
propagation
method,
taking
into
account
both
the
systematic
and
statistical
errors
and
run
-
to
-
run
repeatability.
The
systematic
errors
include
errors
on
concentration
measureme
nt
(10%),
dilution
(10%),
and
wall
loss
(20%).
The
statistical
errors
include
uncertainty
on
dN/dt
and
coagulation
sink,
which
varied
from
run
to
run
depending
on
the
stability
of
the
measurement
conditions.
The
run
-
to
-
run
repeatability
of
J
in
CLOUD
under
nominally
identical
conditions
is
ca.
30%.
The
growth
rates
were
calculated
using
the
appearance
time
method
(
31,
42
)
from
the
scanning
PSM
(1
-
3
nm),
the
DMA
-
train
(2
-
3
nm
and
3
-
8
nm)
and
the
nano
-
SMPS
(7
-
25
nm).
The
error
in
GR
was
estimated
from
the
95
%
confidence
intervals
for
the
(
d
p
,
time)
–
fits,
which
were
used
to
determine
the
GRs.
The
appearance
times
from
the
different
instruments
were
also
checked
for
consistency
in
the
overlapping
size
regions.
While
the
appearance
time
method
is
simple
to
appl
y
for
different
instruments,
and
provides
a
useful
estimation
of
particle
growth,
it
should
be
kept
in
mind
that
the
growth
rates
especially
in
the
smallest
size
ranges
are
difficult
to
define,
and
different
methods
might
differ
from
each
other
depending
n
on
-
linearly
on
the
environmental
conditions
(
55,
56
)
.
Supplementary
fig
ures
Fig
.
S1.
The
effect
of
different
additional
vapors
on
the
NPF
rates
(
J
2.5
).
All
points
have
similar
monoterpene
(
53
0
-
590
pptv)
and
ozone
(40
ppbv)
mixing
ratios
.
The
leftmost
points
were
measured
with
only
monoterpenes
added
to
the
chamber,
and
each
step
to
the
right
represents
addition
of
on
e
more
component
to
the
system.
Solid
arrows
describe
the
addition
of
ca.
1
ppbv
SO
2
(resulting
in
an
H
2
SO
4
concentration
o
f
1
-
2
·10
7
cm
-
3
)
,
dashed
arrows
the
addition
of
ca.
0.7
ppbv
NO
x
and
dotted
arrows
the
addition
of
ca.
180
pptv
NH
3
.
Circles
are
experiments
at
neutral
(N)
and
diamonds
at
GCR
conditions.
Colors
of
the
symbols
indicate
the
meas
ured
monoterpene
mixing
ratio
.
See
Fig
.
1
for
the
formation
rate
of
1.7
nm
particles.
Fig
.
S2.
The
effect
of
different
additional
vapors
on
the
biogenic
nucleation
rate
(
J
1.7
)
at
different
NO
x
concentrations.
All
points
have
a
sim
ilar
monoterpene
mixing
ratio
(50
0
-
590
pptv).
The
lef
tmost
points
were
measured
with
only
monoterpenes
added
to
the
chamber,
and
each
step
to
the
right
represents
the
addition
of
one
more
component
to
the
system.
The
solid
arrows
describe
the
addition
of
NO
x
(~
0.7,
2
or
5
ppbv
),
the
dashed
arrows
the
addition
of
~1
ppbv
SO
2
(resulting
in
an
H
2
SO
4
concentration
of
1
-
2∙10
7
cm
-
3
)
and
dotted
arrows
the
addition
of
~180
pptv
NH
3
.
Circles
are
experiments
at
neutral
and
diamonds
at
GCR
conditions.
The
c
olor
of
the
symbol
indicates
the
measured
NO
x
concentration.
Fig
.
S3.
Nucleation
rates
(
J
1.7
)
as
a
function
of
the
MT
to
NO
x
ratio
(MT/NO
x
).
All
experiments
were
performed
without
added
NH
3
at
a
constant
NO/NO
2
ratio
of
ca.
0.6%
.
The
c
olor
indicates
the
sulfuric
acid
concentration.
The
blue
points
(lowest
H
2
SO
4
)
were
measured
without
SO
2
added
to
the
chamber
(pure
biogenic
nucleation).
The
d
ashed
line
gives
the
maximum
rate
from
ion
-
induced
nucleation,
based
on
the
ion
pair
production
rate
in
CLOUD
under
GCR
conditions
14
.
Fig
.
S4
.
Nucleation
rates
(
J
1.7
)
as
a
function
of
NH
3
mixing
ratio.
Open
circles
refer
to
neutral
experiments,
closed
diamonds
to
GCR
experiments,
and
the
color
refers
to
the
H
2
SO
4
concentration.
All
points
were
mea
sured
at
278
K
and
38%
RH
with
constant
MT
(ca.
250
pptv)
and
NO
x
(ca.
2
ppbv)
mixing
ratio
s
.
Due
to
the
unavailability
of
the
CIMS
to
measure
NH
3
in
this
set
of
experiments,
the
lowest
NH
3
values
(<200
pptv)
were
estimated
from
the
NH
3
flow
to
the
chamber,
while
the
values
larger
than
200
pptv
were
derived
from
the
CI
-
APi
-
TOF
(see
Materials
and
Methods).
Fig
.
S5.
Modeled
versus
measured
nucleation
rates.
(
A
,
D
,
G
)
modelled
nucleation
rates
using
equation
(1),
(
B
,
E
,
H
)
modelled
nucleation
rates
using
equation
(1)
with
[MT/NO
x
]
in
pl
ace
of
[HOM
di
]
and
(
C
,
F
,
I
)
modelled
nucleation
rates
using
an
earlier
CLOUD
parametrization
(
11
)
without
NH
3
dependency,
assuming
[BioOxOrg]=[HOM].
The
data
were
colored
either
by
the
NH
3
(
A
-
C)
,
H
2
SO
4
(
D
-
F
)
or
non
-
nitrate
HOM
dimer
(G
-
I
)
concentration.
Only
neutral
experiments
are
presented
for
clarity.
R
is
Pearson’s
correlation
coefficient
between
log
10
(
J
meas
ured
)
and
log
10
(
J
modelled
).
Fig
.
S6.
Modeled
versus
measured
GRs.
The
modelled
growth
rates
w
ere
calculated
using
equation
(4
)
separately
for
4
different
size
ranges
(
A
-
D
).
The
data
points
are
colored
by
the
non
-
nitrate
HOM
dimer
concentration.
It
can
be
seen
from
the
values
of
the
free
parameters
k
1
-
k
3
that
the
importance
of
sulfuric
acid
decreases
and
the
importance
of
organic
s
increases
towards
larger
sizes.
For
particles
larger
than
7
nm
a
different
set
of
organics
should
probably
be
considered,
as
the
correlation
coefficient
R
between
measured
and
modelled
GRs
starts
to
decrease.
GR
in
the
size
range
1.5
-
2.5
nm
was
determine
d
from
the
PSM,
1.9
-
3.5
nm
and
3.5
-
7
nm
from
the
DMA
-
train
and
>7
nm
from
the
nano
-
SMPS
(see
Materials
and
Methods).
R
is
Pearson’s
correlation
coefficient
between
log
10
(GR
meas
ured
)
and
log
10
(GR
modelled
).
Fig
.
S7.
Positive
ions
and
ion
clusters
detected
during
multicomponent
NPF
in
the
CLOUD
chamber.
The
mass
defect
shows
the
difference
between
nominal
and
exact
mass
of
the
ions
detected
with
the
positive
APi
-
TOF
during
experiments
with
ca.
600/1200
pptv
MT,
1
-
2∙10
7
cm
-
3
of
H
2
SO
4
,
1
ppbv
of
NO
x
(ca.
70
pptv
of
NO)
and
200
pptv
of
NH
3
in
the
chamber.
Colored
symbols
are
identified
ions
(see
legend)
and
open
symbols
are
unidentified
ions.
The
symbol
size
corresponds
to
the
relative
signal
intensity
on
a
logarithmi
c
scale.
Fig
.
S8.
Global
annual
mean
concentrations
of
vapors
involved
in
NPF.
The
colors
indicate
the
mixing
ratios
in
pptv,
at
approximately
500
m
above
the
surface
(cloud
base
level),
from
the
TOMCAT
chemical
transport
model
32
with
the
embedded
GLOMAP
aerosol
model
33
.
(
A
)
shows
sulfuric
acid,
(
B
)
ammonia
and
(
C
)
the
gas
`SecOrg’,
which
is
a
proxy
for
low
volatility
organic
vapo
rs
produced
from
monoterpenes.
The
spatial
distribution
of
SecOrg
matches
approximately
the
distributi
on
of
biogenic
HOMs
in
the
atmosphere.
Panel
(
D
)
shows
the
overlap
of
the
vapor
concentrations,
following
the
functional
form
of
eq.
(1).
Table
S1.
Pearson’s
correlation
coefficient
(
R
)
between
J
1.7
and
the
concentration
of
different
precursors
in
the
chamber.
The
data
set
from
CLOUD10
was
divided
into
experiments
with
and
without
NH
3
,
and
into
neutral
(N)
and
GCR
conditions.
R
is
the
correlation
coefficient,
and
p
its
statistical
significance
level
(0.05
mean
s
a
5%
chance
of
getting
the
given
correlation
by
chance).
Coefficients
higher
than
0.75
are
colored
red.
The
NaN
values
are
due
to
missing
ammonia
measurement
s
during
the
first
part
of
the
experiment,
when
no
NH
3
was
added.
Correlation coeffici
ent with
J
1.7
No NH
3
GCR
No NH
3
N
NH
3
GCR
NH
3
N
Nr of data points
41
28
30
40
R
p
R
p
R
p
R
p
H
2
SO
4
-
0.06
0.69
0.04
0.85
0.22
0.24
0.14
0.38
MT
0.60
<0.01
0.79
<0.01
0.02
0.90
0.03
0.84
NO
x
-
0.50
<0.01
-
0.07
0.74
0.03
0.88
0.05
0.76
MT/NO
x
0.81
<0.01
0.46
0.02
0.00
0.98
0.00
0.98
HOM monomers
0.08
0.61
0.61
<0.01
-
0.10
0.60
0.02
0.89
HOM dimers
0.86
<0.01
0.82
<0.01
-
0.10
0.61
-
0.02
0.89
Non
-
nitrate HOM monomers
0.63
<0.01
0.82
<0.01
-
0.09
0.65
0.03
0.86
Nitrate HOM monomers
-
0.42
0.01
0.01
0.94
-
0.07
0.73
0.01
0.97
Non
-
nitrate HOM dimers
0.97
<0.01
0.83
<0.01
-
0.04
0.84
0.00
1.00
Nitrate HOM dimers
-
0.34
0.03
-
0.10
0.63
-
0.28
0.13
-
0.08
0.63
Total HOMs
0.17
0.28
0.65
<0.01
-
0.11
0.58
0.02
0.92
Total non
-
nitrate HOMs
0.72
<0.01
0.85
<0.01
-
0.08
0.68
0.02
0.88
Total nitrate HOMs
-
0.42
0.01
0.00
0.99
-
0.09
0.65
0.00
0.99
H
2
SO
4
* HOM
di
0.45
<0.01
0.44
0.02
0.12
0.54
0.12
0.45
H
2
SO
4
* HOM
di
* NH
3
NaN
NaN
NaN
NaN
0.82
<0.01
0.89
<0.01