2003 Progress Report: Ion-Induced Nucleation of Atmospheric AerosolsEPA Grant Number: R829620
Title: Ion-Induced Nucleation of Atmospheric Aerosols
Investigators: McMurry, Peter H. , Eisele, Fred
Institution: University of Minnesota , Georgia Tech Applied Research Corporation
Current Institution: University of Minnesota
EPA Project Officer: Hahn, Intaek
Project Period: January 1, 2002 through December 31, 2004 (Extended to December 31, 2005)
Project Period Covered by this Report: January 1, 2003 through December 31, 2004
Project Amount: $400,000
RFA: Exploratory Research: Nanotechnology (2001) RFA Text | Recipients Lists
Research Category: Safer Chemicals , Nanotechnology
The objectives of this research project are to study the role of ion-induced nucleation as a mechanism for producing new nano-sized particles in the atmosphere. We hypothesize that: (1) nucleation processes in different locations are driven by different gas-phase species, and can be homogeneous and/or ion-induced depending on time and locale; and (2) ion-induced nucleation events can be due to the growth of either positive or negative ions, and different gas-phase species are responsible for bursts of intermediate and large positive and negative ions. The ultimate objective of this research project is to develop experimentally verified models for the formation of ultrafine atmospheric particles by nucleation.
During Year 2 of this project, we carried out field measurements in Boulder, CO, and have been working on interpreting those data, as well as the data obtained in Atlanta, GA, during Year 1. We also have been working on the design of improved instrumentation for nanoparticle measurements.
Measurements of ion mobility spectra and ion composition were begun at the National Center for Atmospheric Research’s (NCAR) Marshall field test site in March 2004. The Marshall site is in a rural area about 5 miles southeast of Boulder, CO.
Ion Mobility Spectra
An example of ion mobility spectra obtained during this study is shown in Figure
1. This figure includes four contour plots. For each plot, particle size is
shown on the vertical axis, and time of day is on the abscissa. Particle concentration
is represented by color. The top plot shows particle size distributions for
particles between 6 and 100 nm. The next two plots show ion mobility distributions
for positive and negative ions with sizes between 0.4 and 6 nm. The bottom plot
shows the ratio of positive to negative ion concentrations as a function of
size and time. Note that on this day, a nucleation event occurred between 7:00
and 8:00 a.m., and that over the course of several hours, freshly nucleated
particles grew from molecular clusters smaller than 1 nm to particles of 40-50
nm. We are very excited to have data that extend to particles smaller than 1
nm, and that these data are in such good qualitative agreement with size distributions
measured independently. We have been carrying out such measurements on a daily
Figure 1. Particle Size Distributions (6-100 nm; Top Panel), Positive Ion Size Distributions (0.4-6 nm; 2nd Panel), Negative Ion Size Distributions (0.4-6 nm; 3rd Panel), and Ratio of Positive to Negative Ion Concentrations (Bottom). In each case, the vertical axis shows size, the horizontal axis shows time of day, and color is a value of the size distribution function.
Measurements of Ion Composition
Measurements of ion composition were carried out from mid-March to early April 2004. The study began with some brief measurements of negative ions to make sure that the instrument was working correctly and to get some idea of how similar the ions at this site were to those previously measured at the Atlanta site. As can be seen in Figure 2, the negative spectrum contains NO3¯ and HSO4¯ ions similar to the Atlanta spectrum. As observed in Atlanta and again at the Marshall field test site, the negative spectrum often is dominated by two ion cluster families: those centered around the HSO4¯ core ions, which typically are present when gas phase sulfuric acid is photochemically produced and concentrations are relatively high, and those clustered around the NO3¯ core ion, when sulfuric acid is relatively low. The HSO4¯ ion family contains sulfuric acid clusters in large part because of the very low volatility of sulfuric acid and is probably far more likely to be involved in ion-induced nucleation than other previously or presently observed ambient negative ions. Thus, negative ion studies tend to be focused on specific compounds such as the HSO4¯ ion family.
The positive spectrum, as expected, is far more complex than the negative spectrum. In the positive ion spectrum, there is no clear compound that would be expected to dominate the ion-induced process, if such a process even occurs, and this spectrum also tends to contain a far greater number of observed peaks. This much larger number of peaks is expected to correspond to a wide range of fairly alkaline organics. This added complexity makes the process of interpreting the observed positive ion spectrum and identifying a potential charged nucleating agent particularly difficult. Rather than trying to chemically identify specific ion peaks that are already just barely detectable because of the total ion charge being spread between so many different masses, a more general approach was used. Low resolution measurements of mass were made (+/- 5 amu) stepping about 5-10 masses at a time to improve sensitivity and the speed at which meaningful spectra could be obtained. Unfortunately, there was little sunny weather during the time in which positive ion measurements were conducted. There were some small possible nucleation events, but no significant correlation with changes in ion spectra was observed.
Figure 3 shows positive ion data from March 27, 2004, a day
in which a small burst of ultrafine particles was observed. There was no clear
change in ion mass distributions observed throughout that day. We hope to be
able to make another series of ambient ion measurements later this summer (2004).
Figure 2. Two Diurnal Variations Are Shown for the Major Negative Ions. Note the anti-correlation between NO3¯-ions and the H2SO4-family of ions, with NO3--high at night and H2SO4 increasing during sunlight hours. Cloud cover increased in the afternoon on both days.
Figure 3. The Sum of Positive Ions Observed in 50 Mass Unit Blocks Are Shown for Five Time Periods on March 27 From 3:00 a.m. Until 7:00 p.m. The five bar graphs for each set of masses from left to right are: 3:00-7:00 a.m., 7:00-9:00 a.m., 9:00 a.m.-1:20 p.m., 1:20-4:00 p.m., and 4:00-7:00 p.m. The ultrafine particle burst overlapped the 7:00-9:00 a.m. and 9:00 a.m.-1:20 p.m. periods.
We are in the early phases of examining our atmospheric data to understand its implications regarding nucleation mechanisms. We have solved the birth-death equations for the charge distributions of particles as they grow, assuming that they begin as: (1) neutral clusters; (2) negative ions; and (3) positive ions. These calculations show that regardless of their initial charge state, particles approach steady-state charge distributions after about 1 hour of growth. At this point, particles have grown to a size of 5-10 nm, depending on the day. We now are using the results of these calculations to see if we are able to infer the likely initial charge state of nucleated clusters. We are doing this by converting measured ion mobility spectra to total size distributions, and comparing these with size distributions measured independently using a scanning mobility particle-sizing system. Results of this work will be discussed in our next report.
The data for positive and negative ion mobility spectra shown in Figure 1 were measured with the Inclined Grid Mobility Analyzer (IGMA) that was built by Professor Hannes Tammet for this project. The IGMA measures concentrations of atmospheric ions (very small particles that carry a positive or negative charge). Also, it is important to know the concentrations of neutral particles, because a knowledge of both charged and uncharged fractions would enable us to infer the relative contributions of neutral nucleation and nucleation on positive and negative ions. Kenjiro Iida is interested in this question, and has begun to design an instrument system that will enable such measurements.
Our approach involves designing a scanning mobility particle sizer (SMPS) optimized for measuring total size distributions as small as 1 nm. Instruments of this type currently are used to measure total size distributions of particles larger than 3 nm, but do not perform well for smaller particles. The 1 nm SMPS would be designed to minimize the diffusional losses of nanoparticles, and would include a unipolar charger to adjust the nanoparticle charge distribution in a controlled and predictable manner prior to measurement.
Mr. Iida has modified currently available instruments so that
they can be used for measurements in the 1 nm range. In one set of measurements,
he used this apparatus to measure the size distributions of fullerenes (C60)
produced by evaporation from bulk samples in a furnace. We used C60
because it provides an excellent monodisperse aerosol standard for instrument
calibration. Sample data from these measurements are shown in Figure 4. The
red and green curves reflect different approaches to converting from mobility,
which is measured, to size. These data show that this modified instrument system
is able to size C60 with reasonable accuracy.
Figure 4. Laboratory Measurements of Fullerene (C60) Size Distributions. The red and green curves reflect results obtained when two approaches are used to convert measured mobility to particle size.
Mr. Iida is now designing a new 1 nm SMPS that integrates the unipolar charger, the mobility classifier, and the condensation particle counter. This instrument will be designed to ensure that diffusional losses of particles are as small as possible and can be quantified. This will be, to our knowledge, the first instrument system of this type, and it will move us much closer to being able to quantify the total (charged plus neutral) size distribution of sub 3 nm particles. These data, combined with our ability to measure total size distributions with the IGMA, will enable us to quantify the relative rates of ion-induced and neutral nucleation for new particle formation in the atmosphere.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
|Other project views:||All 26 publications||2 publications in selected types||All 2 journal articles|
||Kulmala M, Vehkamäki H, Petäjä T, dal Maso M, Lauri A, Kerminen V-M, Birmili W, McMurry PH. Formation and growth rates of ultrafine atmospheric particles: a review of observations. Journal of Aerosol Science 2004;35(2):143-176.||