Formation and Growth of Atmospheric Aerosols September 8-9 , 2006 Co-Chaired by Peter McMurry and Markku Kulmala Preface : Workshop on the Formation and Growth of Atmospheric Aerosols
D. R. Hanson,M. Kulmala,Hanna Mannimen,M. Sipilä,T. Petäjä,G. Mordas,A. Mirme,M. Vana,T. Nieminen,J. Allan,R. Burgess,H. Coe,M. Cubison,R. Saunders,J. Plane,A. Saiz‐Lopez,E. Tamm,A. Mirme,H. Tammet,A. Hirsikko,L. Laakso,E. Gorsel,H. Cleugh,R. Leuning,S. Zegelin,D. Hughes,M. Kitchen,Richard J. Hurley,L. Sogacheva,L. Laakso
2006-01-01
Abstract:We are developing a new Nano-SMPS system that detects sub 3 nm particles and samples at high particle count rate. Efforts also have been made to make the system safe, stable, and user friendly. The system consists of an aerosol charger, differential mobility analyzer (DMA), and condensation particle counter (CPC) (Figure 1). The aerosol charger is based on the surface discharge microplasma aerosol charger (SMAC) operating in the negative unipolar charging mode. The major advantage of the SMAC is that it does not use a radioactive source. The original design of Kwon et al, (2006) was modified to simultaneously trap the large primary particles and enhance interactions between the sampled particles and generated ions. The design of the mobility classifier region is based on the radial DMA (Zhang et al., 1995). The effect of wall diffusion losses within the channel between the changer and classifier is minimized by extracting the core portion of the annular flow at the aerosol inlet slit. The classifier region is designed to maximize the sheath flowrate without causing secondary flow formation. The geometry of the aerosol inlet slit at the confluence of the sheath flow and the aerosol inlet flow was redesigned to ensure that two flows merge smoothly. The height of the classifying disks will be adjusted to choose the height that gives the best mobility resolution. We tentatively plan to use a laminar flow CPC (LCPC) because LCPCs have performed stably during our past sampling campaigns. However, the modular design of the new SMPS system will also accommodate turbulent Mixing CPC that has recently been shown to detect particles as small as 1nm at high count rates (Kim et al., 2003; Sgro and Fernandez de la Mora 2004). We will choose a CPC working fluid that gives the lowest minimum detectable size. The overall detection efficiency of the entire system, nonideality in voltage mobility relation and mobility resolution will be experimentally determined by tandem differential mobility analysis (TDMA) experiments. Figure 1: Schematics of the nano-SMPS system REFERENCE: Kim, C. S., K. Okuyama and J. Fernandez de la Mora (2003), Performance evaluation of an improved particle size magnifier (PSM) for single nanoparticle detection, Aerosol Sci. Technol., 37, 791-803. Kwon, S. B., H. Sakurai and T. Seto (2006a), Unipolar charging of nanoparticles by the Surfacedischarge Microplasma Aerosol Charger (SMAC), submitted to J. Aerosol Sci., Sgro, L. A. and J. Fernandez de la Mora (2004), A Simple Turbulent Mixing CNC for Charged Particle Detection Down to 1.2 nm, Aerosol Science and Technology, 38(1), 1, 1-11. Zhang, S.-H., Y. Akutsu, L. M. Russell, R. C. Flagan and J. H. Seinfeld (1995), Radial differential mobility analyzer, Aerosol Sci. Technol., 23, 3, 357-72. Measuring Aerosol Particles below 3 nm by Condensation Particle Counter UF–02proto Mikko Sipilä, Tuukka Petäjä, Hanna Manninen, Genrik Mordas, Ilona Riipinen and Markku Kulmala Department of Physical Sciences, University of Helsinki, PO Box 64, 00014 Helsinki, Finland INTRODUCTION Condensation particle counters (CPC) are widely used in atmospheric aerosol particle measurements. Typically particle size range starting from 3 nm can be covered with commercial instruments. Charged particles or clusters smaller than 3 nm can be measured by ion spectrometers but detection of neutral clusters smaller than 3 nm has turned out to be quite complicated. In this paper we describe the studies of detection efficiency of UF–02proto conductive cooling type CPC. The instrument is described in detail by Mordas et al. (2005). Results from the field measurements with UF–02proto are presented by Riipinen et al. (2006). EXPERIMENT A tube furnace was used to generate polydisperse silver aerosol. At approximately 980oC silver evaporated into 5 l min nitrogen flow. After the heating section of the furnace the flow cooled down and silver vapour started to nucleate. To prevent the growth of the particles to bigger sizes the flow was diluted immediately after the furnace by ca. 15 l min of purified compressed air. As a result a particle distribution with the mode around 2.5 nm (mobility equivalent diameter) was achieved. After passing through the neutralizer (Am) a monodisperse fraction was extracted from the aerosol in a differential mobility analyzer (DMA). An aerosol electrometer was used as a reference. Because the transfer functions of the DMA are reasonably wide at the particle sizes close to 2 nm (Stolzenburg, 1988, Reischl et al., 1997) a simple correction that takes account of the initial distribution was applied. Detection efficiency was investigated at constant condenser temperature (Tcond = +10oC) and at saturator temperatures from +43 oC up to +62 oC. RESULTS Typical detection efficiency curve is presented in Fig. 1. At normal operation conditions (Tcond = +10oC, Tsat = +43oC) a 50%–detection limit of (4.2±0.4) nm was reached. Increasing the saturator temperature finally led to (1.8±0.2) nm detection limit at saturator temperature of +54oC. After that the heating of the saturator did not any more affect the detection limit, but homogeneous nucleation rate started to increase slowly. Measurement with Tsat = 62oC was conducted with background aerosol (ca. 15 nm particles) concentration of 4000 cm to prevent homogeneous nucleation. Figure 1. Detection efficiency as a function of particle size with 44 oC temperature difference between saturator and condenser. Figure 2. 50%–detection limit as a function of saturator temperature when condenser temperature is kept constant (+10oC). In case of saturator temperature of 62 oC a background aerosol (4000 cm, diameter ca. 15 nm) was used to prevent homogeneous nucleation. DISCUSSION It is obvious that aerosol was not completely saturated with butanol vapour in saturator, thus high temperature differences were needed before homogeneous nucleation started to dominate over the heterogeneous nucleation on calibration aerosol particles. It should be mentioned that these calibrations were performed using negatively charged particles. It is known fact that activation of charged particles can take place with lower supersaturations than activation of neutral ones. Thus the 50%–detection limit of (1.8±0.2) nm obtained in this study is very probably slightly different for neutral particles. Also, one should keep in mind that the particle diameter in size range below 3 nm is not a well defined concept. In this study particle diameter is expressed as mobility equivalent diameter (Millikan – Fuchs diameter) whereas application of e.g. Tammet’s mobility-diameter conversion (Tammet, 1995) would lead to very different diameters. As an example 1.8 nm mobility equivalent diameter corresponds to Tammet’s diameter of ca. 1.5 nm. CONCLUSIONS Despite of all small obscurities related to particle sizes, performed detection efficiency studies show that with UF–02proto CPC 50%–detection limits far below 3 nm can be reached.