Environ. Sci. Technol. 2009, 43, 4715–4721
Connection of Sulfuric Acid to Atmospheric Nucleation in Boreal Forest T. NIEMINEN,† H. E. MANNINEN,† S.-L. SIHTO,† T. YLI-JUUTI,† ¨ JA ¨ ,† R. L. MAULDIN, III,‡ T. PETA I. RIIPINEN,† V.-M. KERMINEN,§ AND M . K U L M A L A * ,† Department of Physics, University of Helsinki, P.O. Box 64, 00014 Helsinki, Finland, Atmospheric Chemistry Division, Earth and Sun Systems Laboratory, National Center for Atmospheric Research, P.O. Box 3000, Boulder, Colorado 80307-5000, and Finnish Meteorological Institute, P.O. Box 503, 00101 Helsinki, Finland
Received November 7, 2008. Revised manuscript received February 16, 2009. Accepted March 9, 2009.
Gas to particle conversion in the boundary layer occurs worldwide. Sulfuric acid is considered to be one of the key components in these new particle formation events. In this study we explore the connection between measured sulfuric acid and observed formation rate of both charged 2 nm as well as neutral clusters in a boreal forest environment. A very short time delay of the order of ten minutes between these two parameters was detected. On average the event days were clearly associated with higher sulfuric acid concentrations and lower condensation sink (CS) values than the nonevent days. Although there was not a clear sharp boundary between the nucleation and no-nucleation days in sulfuric acid-CS plane, at our measurement site a typical threshold concentration of 3 · 105 molecules cm-3 of sulfuric acid was needed to initiate the new particle formation. Two proposed nucleation mechanisms were tested. Our results are somewhat more in favor of activation type nucleation than of kinetic type nucleation, even though our data set is too limited to omit either of these two mechanisms. In line with earlier studies, the atmospheric nucleation seems to start from sizes very close to 2 nm.
1. Introduction Formation of new atmospheric aerosol particles by nucleation is a frequent phenomenon that has been observed to take place almost everywhere in the Earth’s atmosphere (1). After their growth in size, nucleated particles may contribute to regional cloud condensation nuclei (CCN) populations (2-5) and to participate into cloud forming processes (6). Model studies suggest that nucleation may be a globally important source of aerosol particles and CCN in the global atmosphere (7-10). In order to quantify the climatic and other potential effects of aerosols, better understanding on atmospheric nucleation mechanisms and their relation to nucleating vapors are needed. In the continental atmosphere, a close connection between nucleation and sulfuric acid (H2SO4) has frequently * Corresponding author e-mail:
[email protected]. † University of Helsinki. ‡ National Center for Atmospheric Research. § Finnish Meteorological Institute. 10.1021/es803152j CCC: $40.75
Published on Web 04/06/2009
2009 American Chemical Society
been observed (e.g. refs 1 and 11-13). Detailed analyses of measurement data have revealed that nucleation rates tend to be proportional to the power 1-2 of the gaseous H2SO4 concentration (13-16). Such findings argue against traditional thermodynamic nucleation mechanisms, including the binary water-sulfuric acid nucleation and ternary watersulfuric acid-ammonia nucleation (17). A direct proportionality between the nucleation rate and H2SO4 concentration could be explained by heterogeneous nucleation, or activation, of pre-existing molecular clusters by gaseous sulfuric acid (17). If, however, nucleation was driven by kinetic collisions between H2SO4 molecules, the nucleation rate should be proportional to the square of the gaseous H2SO4 concentration. Earlier studies looking at connections between atmospheric nucleation and gaseous sulfuric acid have relied on aerosol measurements that do not extend to sizes smaller than 3 nm of particle diameter (particle sizes are given in this paper as Millikan diameters, also called electrical mobility diameter; see ref 18 for a comparison between different conversions from electrical mobility to particle diameter). This is not an ideal situation, since atmospheric nucleation is expected to start from sizes of about 2 nm or even below (19). Here, we will employ new neutral cluster and air ion spectrometer measurement data on neutral and charged sub-3 nm clusters to get further insight into this issue. More specifically, we aim to address the following questions: i) how the formation rate of 2 nm particles depends on the gaseous H2SO4 concentration, ii) how different-size clusters respond to changes in the H2SO4 concentration, and iii) how the dynamic behavior of the system differs between days with active nucleation and days when no nucleation is observed to take place. The measurements considered here were made at the SMEAR II station in Hyytia¨la¨, Finland, during the spring 2007.
2. Materials and Methods 2.1. Measurements. Measurements for this study were made at the SMEAR II station of the University of Helsinki located in Hyytia¨la¨, Southern Finland. The station is located in a rural environment surrounded by large areas of pine forest. Detailed descriptions of the station and the continuous and comprehensive measurements carried out there can be found in refs 20 and 21. In addition to the continuous measurements at the station, an intensive campaign was carried out during March 6 and June 30, 2007 within the EUCAARI (European integrated project on aerosol, cloud, climate, and air quality interactions) project (22). We will utilize in this study sulfuric acid concentration measurements made during that period. Aerosol particle number size distributions have been measured continuously at the SMEAR II station since 1996 (23) with a DMPS (Differential Mobility Particle Sizer) system. The DMPS measurement system operating in Hyytia¨la¨ consists of two differential mobility analyzers. The DMPS measures total particle size distributions in a diameter range of 3-1000 nm. The Hyytia¨la¨ DMPS system is described in detail by Aalto et al. (24). Starting from 2003 also air ion (charged particle) size distributions have been measured with two air ion spectrometers, namely Balanced Scanning Mobility Analyzer (BSMA) and Air Ion Spectrometer (AIS). Information about ion spectrometers can be found in refs 25 and 26. In 2006 a Neutral cluster and Air Ion Spectrometer (NAIS) was installed at the station (19). The NAIS measures total particle (both charged and electrically neutral) concentrations in diameter range starting from about 2 nm and extending to 40 nm. The lower detection limit of the NAIS VOL. 43, NO. 13, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4715
FIGURE 1. An example of new particle formation day on 15.4.2007 (a), and 22.3.2007, a day when there is no particle formation (b). On the x-axis is time and on the y-axis particle diameter. The figure shows data measured by the NAIS between 0.7 and 40 nm. Below 2 nm the data should be considered to be only semiquantitative because of the ions produced inside the corona chargers in the NAIS.
FIGURE 2. Median diurnal concentrations of negative ion clusters in different size classes (solid lines) and sulfuric acid (dashed line) during new particle formation days (a) and on days when no particle formation is detected (b). is determined by ability to distinguish the charged particles produced in the instrument during sample charging from the atmospheric particles. According to calibrations performed by Asmi et al. (27), the concentration of charger produced ions falls rapidly with particle size after approximately 1.5 nm. Gas-phase sulfuric acid and hydroxyl radical concentrations were measured with a technique utilizing selected chemical ionization and detection with a mass spectrometer (CIMS) (28, 29). The measurement setup used in Hyytia¨la¨ has been described in detail by Peta¨ja¨ et al. (30) and references therein. 2.2. Data Analysis. In this paper we are examining connections between freshly formed atmospheric aerosol particles and ambient concentrations of sulfuric acid. Based on the aerosol spectrometer size distribution measurements, each day is classified to be either an event day (when we see new particle formation, that is particles between 2-3 nm, and growth of these particles into larger sizes), a nonevent day (when no new particles are seen to appear), or an FIGURE 3. Sulfuric acid concentration and condensation sink undefined day (when it cannot be determined reliably value at the start of new particle formation detected with ion whether there is new particle formation or not). The spectrometer BSMA for negative (blue crosses) and positive classification is done based on criteria developed by Dal Maso (red) ions and with NAIS for all particles (green). Black et al. (31). Here the size range representing freshly formed markers show median values between 9-15 on the days when no particle formation is detected. particles was taken to be 2-3 nm in Millikan diameter. 4716
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 13, 2009
Total formation rates of new 2-3 nm particles (J2) were calculated from NAIS measurement data with the following formula J2 )
dN2-3 GR + CoagS2·N2-3 + ·N dt 1 nm 2-3
(1)
Here N2-3 represents the number concentration of 2-3 nm particles, CoagS2 is the coagulation sink for 2 nm particles, and GR is the particle growth rate between 1.5 and 3 nm. Particle growth rates were determined from measured size distributions by first determining the peak concentration in each size channel of the instrument and then fitting a linear equation into these points. Further details of this method can be found in refs 32 and 31. When calculating the formation rate of charged 2-3 nm particles, also ion-ion recombination and charging of neutral particles were taken into account. The formation rate of charged 2 nm particles (J2() was calculated by
J( 2 )
( dN2-3 GR ( ( ( + + R·N2-3 ·N
