Atmospheric aerosols and their impacts on clouds are considered to be important for climate-induced radiative forcing, but their impacts on ozone are still small.
Globally, about half of cloud condensation nodules originate from atmospheric vapor nucleation.
It is believed that sulfuric acid is a necessary condition to initiate the formation of most particles in the atmosphere 3,4, while the role of ions is relatively small.
However, some laboratory studies reported that the formation of organic particles did not intentionally add sulphuric acid, although pollution could not be ruled out 6,7.
Here, we provide evidence for the formation of aerosol particles from highly oxidized biological vapors in atmospheric conditions without sulfuric acid.
It is produced by the decomposition of alpha ozone. -pinene.
We found that ions from cosmic rays in the Milky Way increase the nucleation rate by one or two orders of magnitude compared with neutral nucleation.
Our experimental results are supported by quantum chemical calculations of binding energies of representative Homs clusters. Ion-
Induced nucleation of pure organic particles is a potential source of aerosol particles in terrestrial environment polluted by low sulfuric acid.
CERN cloud experiment aims to study the effects of cosmic rays on aerosols, cloud droplets and ice particles under precisely controlled laboratory conditions. The 3-m-
Diameter Stainless Steel-
The cleanliness and performance of steel cloud chamber and its gas system meet the highest technical standards.
Cloud chambers are cleaned regularly by ultrasonic cleaning of walls. -
Pure water, then heated to 373 K, then washed with humidified synthetic air and ozone at high speed. (
How many parts per million (by volume).
Pollutant Level of Condensable Steam-p. p. t. v. range.
High Cleanness and Large Volume of Combustor（26. 1u2009m)
The highly stable operating conditions allow the formation of particles to be studied at a nucleation rate of about 0 under atmospheric conditions.
001 cm s and 100 cm s.
The loss rate of condensable vapor and particles on the wall of the chamber is similar to that of the environmental condensation pool in the original boundary layer.
Indoor ion production can be controlled by an internal electric field. (
Or a tunable beam from the CERN proton synchrotron.
The crossbeam is-
Focus on about 1 lateral dimension. 5u2009mu2009×u20091.
5 m through the cloud chamber.
When the electric field is set to zero, the equilibrium ion-
Because of the effect of GCR, the paired concentration in the incubator is about 700 cm.
By using the beam, the value can be increased to any value about 3000cm.
Therefore, the ion concentration corresponding to any height of the troposphere can be generated in the cloud chamber.
The precise control of indoor Micro-Steam and ambient temperature is carried out between 300 K and 203 K.
Uniform Mixing of Magnetically Coupled Stainless Steel-
Steel fans installed at the top and bottom of the combustion chamber.
The characteristic gas mixing time in the chamber is several minutes, depending on the fan speed.
The photochemistry process is caused by ultraviolet optical fiber irradiation. -
Optical systems, providing highly stable gases-
Phase response with precise start time.
Continuous analysis of the contents of the pharmacy is carried out by means of an instrument connected to a sampling probe inserted into the pharmacy.
Sampling analyzers are tailored for each experimental activity, but usually consist of about 30 to 35 instruments, up to 10 of which are mass spectrometers.
For the results reported in this paper, the analytical instruments installed in the test box include chemical ionization mass spectrometer. (CIMS)
For HSO concentration;
Atmospheric pressure interface time-of-flight (APi-TOF;
Aviation Research Corporationand Tofwerk AG)
Molecular Composition Mass Spectrometer for Positive and Negative Charge Ions and Clusters;
Bichemical Ionization Atmospheric Pressure Interface Time-of-flight (CI-APi-TOF;
Aviation Research Corporationand Tofwerk AG)
Molecular Composition and Concentration Mass Spectrometer of Neutral Gases-
Stage HSO and HOMS;
Proton transfer reaction time-of-flight (PTR-TOF;
Ionicon Analytick Ltd.
Organic vapor mass spectrometer;
Neutral Cluster and Air Ion Spectrometer（NAIS; Airel Ltd)
For the concentration of positive ions, negative ions and charged clusters in the range of 1-40 nm; a nano-
Radial Differential Fluidity Analyser（nRDMA)
And nano-scanning mobility particle meter（nano-SMPS)
Particle size spectrum;
And several agglomerated particle counters（CPCs)
The detection efficiency threshold of 50% is between 1 nanometer and 4 nanometer: two Airmodus A09 particle size magnifiers, psm, (one fixed-
Threshold and other scans)
Diethylene glycol CPC, diethylene glycol-
CPC, butanol TSI 3776 CPC and water TSI 3786 CPC（TSI Inc. ).
Additional gas analyzers include dew points-point sensors (EdgeTech)
Thermo Fisher Science Company42i-TLE)and ozone (
Thermal Environment Instrument (TEI 49C).
In some experiments, HONO vapor is supplied to the test chamber and photolysed by ultraviolet light to produce OH without O.
Continuous mixing of HSO with nanoparticles produces gaseous HONO. (ref. )
Specially designed stainless steel-
The steel reactor is then steadily flowed into the reactor chamber.
The HONO analyzer consists of a specially designed probe that passes air samples from the chamber through HSO and sulfonamide solutions and then on-line analyses with a long-path absorption spectrophotometer. (LOPAP).
Nucleation rate（in cmu2009s)
Measurement under neutral point（), ground-level GCR ()and beam ()conditions.
The neutral nucleation rate is measured by a clearance field electrode set at (+30 kv), which forms an electric field of about 20 kV m in the chamber.
This is completely suppressed. -
Induced nucleation, because under these conditions, small ions or molecular clusters are swept out of the chamber within about 1 s.
Since all the nucleation and growth processes considered take place over a considerable time scale, the neutral nucleation rate can be measured at zero ion background. -
For GCR and beam conditions, the electric field is set to zero, resulting in equilibrium ions. -
The paired concentrations were about 700 cm and 3000 cm, respectively.
The nucleation rate only measures the neutral rate and the sum of neutral ions. -
Induced nucleation rate, +.
The nucleation rate reported in this paper is mainly through airmodus scanning PSM at 1. 8-nm threshold (PSM1. 8)
And TSI 3776 CPC（CPC2. 5), nominally 2. 5-
NM threshold, but measured at 3:00. 2-
Nano-threshold and WO particles.
The nucleation rate is 1. 7-
Nanomobility diameter（1. 4-
Nano Mass Diameter)
It is generally believed that the size of particles is higher than the critical size, so the thermodynamic stability is achieved.
The critical size corresponds to the cluster size when the evaporation rate and growth rate are equal.
It varies with temperature, chemical substance, charge and vapor concentration. When the evaporation rate is highly inhibited, it may not even exist, for example, for sulfuric acid-dimethylamine clusters.
Our measurements show that the smallest neutral HOM clusters are relatively unstable. therefore, 1.
7 nm, equivalent to about 5 Hom monomer unit, is a reasonable size for deriving the experimental nucleation rate.
Determine the nucleation rate at 1:00. 7u2009nm, the time-
Dependent particle concentration was measured by PSM1. 8 and CPC2.
5. Using a simplified numerical model（AEROCLOUD)
Particle nucleation and growth are dynamically processed at the molecular level.
The model uses the production of HOM monomer, HOM dimer and HSO, all of which come from CI. -APi-
TOF experimental data.
The measured HOM productivity is scaled at a scale of 1.
8 to match the observed particle appearance time and growth rate.
In the whole experimental range of HOM concentration, this scale makes the model in good agreement with the experimental data.
The scale factor is within the measurement uncertainty of CI system. -APi-
TOF, if nitrate CI-APi-
TOF did not detect all HOMs that contributed to the growth of particles.
In the model, the first-order ions of GCR are generated at a known rate = 1.
Seven pairs of ions per cubic centimeter per second.
A fixed parameter of the model explains the charge symbolic asymmetry caused by the difference of diffusion loss rates of positive and negative ions to the wall: the parameter is determined by the concentration of positive and negative ions in the measured NAIS, and the value is 0. 52.
Molecules and particles collide in motion and aggregate with each other.
The model uses a reduced clustering probability. (
Hereinafter referred to as "probability of adhesion".
Explain small unstable clusters rather than evaporating them once they form.
This greatly improves the computational speed.
If the particle formed by collision exceeds a certain size（
That's about one. 7-
The nano-mobility diameter of pure biological clusters; see below)
Then it is assumed to be effective and stable, and then grow near the dynamic limit.
The growth rate of particles between PSM1. 8 and CPC2.
Therefore, in the model, 5 is implicitly regarded as the kinetic restriction of particle condensation plus HOM and HSO vapor condensation.
Particles grow through small size linear and logarithmic spacing (from about 2 nm to the maximum size of 400 nm). The time-
The steps of the clustering process start at 0.
9 s to 10 s depend on the operating conditions of the experiment under analysis. The time-
For all other processes, step 10（
For example, renewal gas concentration, high-
Field change, fan change and particle loss due to dilution of indoor contents or diffusion to the wall).
The density of pure Hom clusters is fixed at 1.
3 G cm, 1.
For pure HSO clusters, 85 g cm.
For neutral-neutral collisions, the number of particles in bunkers 1 and 2 that coagulate at intervals of time to produce mass particles is neutral-neutral collision nucleus, particle number concentration and van der Waals enhancement factor. (see below).
The neutral-neutral adhesion probability of pure biological particles is: free parameter.
This parameter effectively defines the threshold quality of the stable cluster, because the adhesion probability is equal to time, and the parameter controls the sharpness of the threshold.
When at least one particle is mainly sulfuric acid, the collision adhesion probability is also defined as: the neutralization is a free parameter.
Neutral-Neutral Collision Nuclei in Equations（1)
It is the Fuchs form of Brown condensation coefficient.
Van der Waals Enhancement Factor is a revision of Fuchs theory, such as references.
For Knudsen number in dynamics（free molecular)regime.
The enhancement factor is: the reduced Hammark constant is: where is the radius of the particle, = 6. 4u2009×u200910u2009J (
Hammark constant of sulfuric acid), u2009=u20090. 0151, u2009=u2009−0. 186, u2009=u2009−0.
It's the Boltzmann constant, it's the temperature.
The same Hamaker constant is used for sulphuric acid and HOMS because it does not significantly change the model predictions.
Ions and charged particle clusters collide according to similar formulas（1)
Where is the enhancement factor for charged collision nuclei? (Describe as follows. .
The probability of adhesion between neutral particles and charged particles is: free parameters and = or biological or acid particles, respectively.
Ion-ion recombination produces a neutral particle that may evaporate at a smaller size.
The model allows partial evaporation of the composite particles.
In this case, the cluster is divided into monomers and the mass is conserved.
The probability of cluster survival after ion-ion recombination is a free parameter.
The Power of Unity（=u20091)
Because the data can't constrain this parameter very well.
In order to obtain a charged collision nucleus, the neutral collision nucleus is multiplied by its size. -
Dependent Enhancement Factor: Collision nuclei and subscripts refer to the charges of colliding particles.
Charged collision nuclei in the equation（2)
Obtained from references.
Sulfuric acid granules.
Since biological particles may have different neutral charged collision nuclei, their enhancement factors are free in fitting: where are the free parameters?
Ions, monomers, clusters and larger particles are constantly lost by diffusing to the wall and diluting the indoor contents with fresh gas mixtures.
Dilution life approaching 3 h（10u2009s)
Depending on the total sampling rate of all instruments connected to the test box.
The wall loss rate is 1.
The HSO monomer is 8 10 S, which decreases to 1/ with the increase of cluster or molecular diameter.
The wall loss rate of HOMS is obtained by using the same proportional law.
In other words, it is assumed that the HOM and particles colliding with the wall are irreversibly lost.
At the beginning of operation, there are a large number of particles in the combustor due to the incomplete cleaning of the combustor. By inserting the initial particle size distribution into the particle size box of the model, the loss of the condensation pool can be calculated.
In order to determine the nucleation rate, the five free parameters of the model（, , , and )
It matches the experimental particle concentration in PSM1. 8 and CPC2. 5 versus time.
For example, there is only one free parameter for the operation of a neutral pure biological source. （)
Participate in fitness.
Values range from 12 to 14, 4 to 6, and 0. 1 to 1.
0, close to 4, close to 10,000.
These parameters are determined by all operations in the global fitting data set and then fixed to these values.
The applicable threshold qualities are 1300 and 700, respectively.
The parameter is set to 1.
0 and set to 0. 52.
The model well reproduces the time evolution of particle number concentration in two counters in all nucleation events in the data set. (
The example is shown below. .
After fitting the data with the model, the nucleation rate is determined as the number of particles with a mobility diameter of 1.
7 nm or larger-
Step by step, divided by time increment.
In each nucleation process under fixed conditions, the maximum time is determined.
Then the running value is calculated as the average measurement value in the interval. (±u2009300u2009s).
There are three main advantages of using data-
The nucleation rate is determined by the driving dynamics model, rather than directly measured by PSM1. 8 or CPC2. 5 data.
Firstly, it avoids the need for time derivatives of data, which are prone to large errors at low counting rates.
Secondly, the growth rate of particles is determined by dynamics, and the growth due to collisions with monomers and other particles is explained appropriately.
Therefore, the model processing of the data avoids the exponential sensitivity of other methods to the experimental growth rate.
The experimental growth rate is determined by the time of particle reverse rise, and there is great uncertainty between 1 and 3. -nm size range.
Finally, the model needs consistency between PSM1. 8 and CPC2.
Therefore, formation velocity is experimentally confined to the vicinity of 1. 7-
The Nano-threshold is about 3 nanometers.
We had a wide cross-section. -
By calculating the nucleation rate separately in two additional ways, the nucleation rate obtained by the model is checked. （1)
Direct measurement 1.
Using scan PSM and（2)CPC2.
5-step measurement-corrected to 1. 7-
NM threshold size.
Within the range of experimental uncertainty, the nucleation rates obtained by the two methods are in good agreement with the values obtained by the gas cloud dynamics model. The stepwise-
The revision method is described in detail in the references.
But here is a brief summary.
The nucleation rate is derived from the change rate d/d of the formation rate, in which the particle number concentration is measured by CPC2.
It is higher than the detection threshold of 5.
For the loss, dilution and solidification of particles on the wall, the formation rate is corrected in two consecutive steps:（1)
Particle loss is higher than sum（2)
Particle loss during growth from the beginning. 7 nm to .
The dilution rate and wall loss rate are the same as those in the dynamic model.
In order to calculate the coagulation rate, the particles are divided into granular silos, and then the loss rate of each silo is calculated by summing the granularity. -
Particle rates in silos and in all other silos.
The total coagulation loss rate is the sum of the particle loss rate in each silo.
Fixed the loss of particles in the growth process starting from 1. 7u2009nm to (item (2)above)
Particle growth rates need to be understood.
This is determined experimentally by several instruments. For example, by scanning the appearance time measured in the psm, the scanning PSM can detect particles in the threshold diameter range from 1 nm to 2. 5u2009nm.
In addition, several other tools have been used to measure growth rates in different scales, including-
Threshold psm, 2 degrees-
CPCS, TSI 3776 CPC, API-
TOF, AN NAIS, NRDMA and nano-SMPS.
The experimental growth rate is parameterized, because in all events, they cannot be measured accurately and adequately at each time point.
To determine the nucleation rate of 1.
The dimension interval is divided into logarithms at 7 nm away from the corrected formation velocity. -
In this section, the difference between the two instruments is shown by adding uhel. -
CI Characteristics in UFRA Backbrackets-CI. The CI-APi-
TOF has been described previously.
Sample air from the cloud chamber is inhaled through a 1/2"stainless steel tube at a flow rate of 9_l min. (10u2009lu2009min).
An electrostatic filter is installed in front of each instrument to remove ions and charged clusters formed in the ionization chamber.
The geometric structure of both ion sources follows the design of references.
But a corona charger（X-ray generator)
Used for ion formation.
Dry air containing nitric acid vapor is washed on the ionizer to produce ions.
Ions enter the sample stream through an electric field, where they react with sulfuric acid and HOMS.
The reaction time is about 50 ms. (200u2009ms)
Ions pass through a pinhole of 350 um diameter before entering API segment（300u2009μm).
The API consists of three continuous differential pumping chambers, in which the pressure decreases gradually, and the ions are focused by two groups of quadrupoles and an ion lens system. The mass-to-
The charge ratio of ions through these chambers, /, is measured over a period of time. -of-flight (TOF)
Mass Spectrometer（Tofwerk AG).
Voltage Settings in API-
TOF Section Affects Quality-
It depends on the transmission efficiency.
The transmission curves are measured in a series of calibration measurements in which different/various perfluoric acid vapors enter the instrument in sufficient quantities to saturate all primary ions.
In this way, a constant ion signal will be generated at each/place, so the transmission efficiency can be determined relative to the mass range of primary ions. The UFRA-
During the whole data acquisition period, CI works under the same voltage setting. the UHEL-
CI operates in two switching modes for low/high/optimized voltage settings, respectively.
The original data are analyzed by using MATLAB software package.
The calibration accuracy of quality scale is higher than 10 p. p. m. using a two-parameter fit.
The concentration of sulphuric acid is calculated according to the ratio of bisulfate ion counting rate. (in s)
Compared with primary ions, the coefficient corrects the loss in the cloud chamber sampling line.
Calibration coefficient, determined by connecting CI-APi-TOF to a well-
Characteristic HSO generator.
The value depends on the voltage settings in the API-
The TOF section was determined to be 6. 5u2009×u200910u2009cm (1. 2 x 10 cm and 2.
The high/low/setting is 8 10 cm, respectively.
The uncertainty is + 50%/-33%.
The detection limit of HSO is 5 10 cm or slightly lower.
The concentration of Hom at /= is calculated as follows: Here is the background. -
Subtract the HOM count rate.
In the absence of any alpha, background levels are measured by sampling air from clean cloud chambers. -pinene present.
The factor is quality. -
It depends on the transmission efficiency.
The calibration coefficients are the same as those of sulfuric acid, because the molecular collision rates of HOMS and sulfuric acid with nitrate ions are similar.
In addition, this study also found that the binding capacity of HOMS with high oxidation is very strong, so clustering should be carried out near the kinetic limit, just as in sulfuric acid.
The coefficient corrects the loss in the cloud chamber sampling line.
The diffusion coefficients of E and E were measured by experiments.
HOM monomer e is the background. -
Subtract the sum of peak values in/band 235-424;
Hom dimer e is the corresponding sum of 425-625 power.
The instrument contamination peaks are excluded from the band sum because these peaks are assigned to Ro. free radicals. (
CHO, corresponding to /= 293 th, 325 th, 357 th and 389 th).
Total Homs is defined as the sum of RO * e + E.
Ozone decomposition or OH chemical HOM production was calculated by assuming that the yield and loss rate were equal during the stabilization process. -
Status: where yield is a fraction of alpha-pinene (AP)
The oxidation reaction results in the formation of HOM, and "Ox" denotes O or OH.
The Value of Rate Constant（
Alpha oxidation at 278 K-
Pinene = 5.
84 x 10, from the International Federation of Purification and Applied Chemistry（IUPAC)(the α-
The Pinene+o rate constant is updated on the IUPAC website at.
The loss rate of HOM wall is determined to be 1.
1 x 10 s, assuming they are irreversible loss.
Another loss is caused by the dilution of the combustion chamber volume by the supply gas. (0. 1u2009×u200910u2009s).
The total loss rate of HOMS is = 1. 2u2009×u200910u2009s.
In the experiment of pure hydride and nitrite（HONO)
The concentration range is 0. 5u2009p. p. b. v. to 3u2009p. p. b. v.
Ultraviolet radiation of optical fiber system is used to decompose into OH.
This leads to a small amount of NO pollution, which may affect HOM production.
The concentration of OH in the cloud chamber was estimated by ptr. -
TOF measurement of alpha difference-
Pinene concentration without OH（Ultraviolet shutdown)
Oh, now. （
Ultraviolet radiation of different intensity).
Decrease in alpha-
In these experiments, OH reaction is the only cause because there is no O in the room.
It is estimated to be (+30%). 1)
Uncertainty including alpha-
The measurement of pinene and the reaction rate constant lead to the system scale uncertainty of HOM production rate. [AP][OH·], of ±40% (1). However, run-to-
Operational uncertainty greatly affects the overall uncertainty, as shown in the error bar.
SO Chemical Ionization Mass Spectrometer（SO-CIMS)
SO is converted into SO by first-order ions and then measured in a quadrupole mass spectrometer with API interface. (Georgia Tech).
The overall design of the ion source is shown in the references.
But primary ions are produced by corona discharge.
The corona needle holder was improved to supply CO, O and AR directly through corona discharge.
In this way, the direct contact between N sheath flow and discharge needle is avoided, thus reducing the pollution of the pair and maximizing the ratio of pairs.
The ionization reaction scheme of SO-to is referred to in the literature. .
Sulfur dioxide was detected at /= 112 of the mass spectrometer by evaporating water molecules from ions using a dry n buffer stream in front of the pinhole of the mass spectrometer. ().
SO concentration（in p. p. t. v. )
According to the ion counting rate, the calculation is as follows: which corresponds to the background-
The corrected ion counting rate is the primary ion counting rate.
Calibration factors are obtained by periodically calibrating instruments using SO gas standards. (Carbagas AG)
During the campaign.
In the calibration process, the standard gas was diluted by super clean humidified air with relative humidity of 38%. (
The same as provided to the cloud chamber)
The range of SO mixing ratio between 12p is different. p. t. v. and 11u2009p. p. b. v.
The calibration coefficient is 1. 3u2009×u200910u2009p. p. t. v.
The estimated uncertainty is (+11%).
The errors include the uncertainty of flow rate and gas standard concentration in the calibration process, as well as the statistical uncertainty.
However, we also observed that the temperature changes in the laboratory hall caused the background signal to drift when SO was not applied to CIMS.
This effect leads to the overall uncertainty, which mainly affects the measurement at low SO level. (