AIRCRAFT STUDY OF THE PHYSICAL AND CHEMICAL EVOLUTION OF AEROSOLS IN SMELTER AND POWER PLANT PLUMES

 

C.M. Banic, W.R. Leaitch, K. Strawbridge, S. Daggupaty, N. Shantz, J.Y. Lu, N. Bhyat, R. Tanabe, Meteorological Service of Canada, 4905 Dufferin Street, Downsview, ON Canada, M3H 5T4;  J.I. MacPherson, National Research Council of Canada;  H.K.T. Wong, National Water Research Institute; Z. Nejedly, J.L. Campbell,  University of Guelph;  C. Gariepy, A. Simonetti, GEOTOP, University of Quebec (Montreal); J. Skeaff, CANMET-Natural Resources Canada; A. Chatt, Dalhousie University; M. Lamoureux, Saint Mary’s University

Cathy.Banic@ec.gc.ca

 

ABSTRACT

 

In January and February of 2000 the National Research Council of Canada Twin Otter aircraft was flown in the plumes of a coal-fired power plant and a copper smelter to investigate the sizes and metals content of the particles emitted.  A scanning lidar system was deployed at a fixed site at the ground to monitor the behaviour of the plume over longer time scales than the aircraft flights. In addition, measurements were made at  ground-based sites to characterize the regional aerosol.  The data for this study are not yet fully analyzed.  Here we give an overview of the aircraft experiment and an indication of the forthcoming results.

 

 INTRODUCTION

 

National and international concern about the health effects and continued use of Pb, Cd, As and Hg as well as other metals has defined a need for improved estimates of the long term risks to ecosystems and human health from metals released from mining, metallurgical, and energy production activities. The Twin Otter aircraft operated by the National Research Council of Canada was used to determine the microphysical and chemical properties of airborne particulate metal emissions from the Nanticoke coal-fired power generating station located on the north shore of Lake Erie, Ontario and the Horne copper smelter at Rouyn, Quebec.  These properties are critical to the determination of the deposition rates of metals emitted, and hence the potential for these species to have impacts on local or distant ecosystems. The measured properties will be used in existing atmospheric models to predict the concentrations and deposition patterns of current emissions of metal in particulate from both individual point sources and the regional-scale background and to estimate the proportion subject to long range transport. The impact of proposed emission-reduction strategies on the local to regional metals burden will be predicted. 

 

 

 

 

METHODS

 


Thirteen flights into the Nanticoke plume were made between Jan. 17 and 28, 2000.  Eleven flights were flown into the plume from the Horne between Feb. 14 and 24, 2000.  Each flight was approximately 2 hours in duration.  Both plumes were sampled from a few to 40 km downwind of the stack.  A variety of conditions was encountered:  sunny and cloudy daytime, night, light to strong winds, well-mixed to stable boundary layers.   The flight track for Jan. 28 is shown in Figure 1.  Winds were from the north west at 350o.  The plume was tracked out to 30 km and then studied with vertical profiles and horizontal passes at ranges of 20 and 10 km followed by a run up the plume to within 2 km of the stack.  These distances correspond to plume ageing times in the environment of 45, 30, 15 and 3 minutes, respectively.  The ambient environment was characterised by measurements made past the edges of the plume in the vertical and horizontal at the 20- and 10-km ranges as well as by a vertical profile taken over the ground site at Simcoe.

 

 


Figure 1:  A representative flight path showing  the investigation of the plume to the southeast of  Nanticoke and the vertical profile upwind. The details of each individual flight vary, but the overall emphasis and pattern are well represented by this example.

 

Particle size distributions were determined using a TSI Scanning Mobility Particle analysis System (SMPS) to measure particles from 0.005 mm to 0.2 mm diameter (uses the principal of mobility of charged particles moving in an electric field), and  PCASP and FSSP-100 mounted under the wing of the aircraft (use light scattering by particles) to measure particles from 0.15 to 3 mm and 2 to 40 mm diameter, respectively.

 

A sampling technique allowing a ½ cubic metre volume in the aircraft to be flushed or filled with air from outside the aircraft in 2 seconds was used to obtain grab samples of the plume.  The grab sample allows sampling of one section of the plume to a filter or other instrument for a longer period of time than the aircraft speed (60 m/s) will permit us to linger.  The volume is contained in a collapsible bag made of a conductive plastic.  The fast fill is achieved by ram air flowing through a large forward-facing inlet (approximately 8 cm by 8 cm)  with electonically controlled valves at the inlet and outlet of the bag.  The SMPS system could sample ambient air directly or the air collected in the bag.

 

Metal content as a function of particle size is determined by the collection of particles on filters and subsequent analysis by a number of techniques including Proton-Induced X-ray Emission Spectroscopy (PIXE), ICP-MS, Pb isotope and single particle analysis.   Multiple filters are collected in parallel from the grab samples or through an isokinetic inlet followed by a short stainless steel line of 4.5 cm diameter with 1 gentle bend.  The PIXE analysis will be applied to impactor samples (9 stages with cut-off diameters between 16 and 0.06 mm) collected from the grab samples.

 

In addition, SO2 was measured continuously using a TECO analyzer and in batch mode with treated filters inserted as a second stage to particulate filters.   Methods and selected results for measurements of speciated mercury are given in these proceedings in a paper by Lu et al.

 

RESULTS AND DISCUSSION

 

Figure 2 shows the particle number concentration and size distribution as determined by the PCASP and FSSP-100 for one pass through the Nanticoke plume compared with the ambient environment into which the plume was released.  For this example the plume has 1 order of magnitude greater number of particles throughout the size range than is found in the ambient environment.  In general, it was found that the number of new particles formed by nucleation in the plume was dependent on the particle loading in the ambient atmosphere with more evidence of nucleation of new particles in relatively clean air.  Analysis of the filter samples continues in order to place the metals into the appropriate size range in the distribution.  This experiment will be repeated in the summer of 2000 so that the temperature, sunlight and air chemistry can be further varied to investigate the behaviour of the metals in the plumes under real atmospheric conditions.

 


 


Figure 2:  Illustration of particle  size distribution in  the plume relative to ambient air.

 

 

 

AcknowledgmentS

Funding for this project was provided by the Toxic Substances Research Initiative, the Mining Association of Canada and Ontario Power Generation Inc. We thank the staff of Noranda Inc.-Horne Smelter and Ontario Power Generation for their assistance.   This work was made possible through the expert support of the pilots and staff of NRC-IAR and Steve Bacic and John Deary of MSC.