Georgia Institute of Technology School of Earth and Atmospheric Sciences EAS 4641 Spring 2008 Lab 4 Major Anions In Atmospheric Aerosol Particles Purpose of Lab 4: This experiment will involve determining the concentrations of the major ionic chemical components comprising PM2.5 ambient aerosol particles in urban Atlanta. This will involve the use of ion chromotrgraphy and the calibrations from labs 2 and 3. Sources of Anions in Ambient Particles: One way to identify the source of particles is to measure their chemical composition. Salts such as NaCl, CaCO 3, Ca 2 NO 3, (NH 4 ) 2 SO 4, and NH 4 NO 3 comprise a large fraction of aerosol particles. Because these salts will adsorb water at moderate RH, the particles take up water and the various salts are often found as solutions in particles, where the salts dissociate and form cations (positive ions) and anions (negative ions). Certain salts dominate in different locations due to their differing sources. High concentrations of sodium chloride, NaCl (or Na + and Cl - ), are found in marine regions due to aerosol formed from sea spray. Calcium carbonate (CaCO 3 ) and calcium nitrate (CaNO 3 ), and similar mineral compounds, are found in arid regions from suspension of soils. This is also commonly referred to as mineral dust. Ammonium sulfate ((NH 4 ) 2 SO 4 ), and ammonium nitrate (NH 4 NO 3 ) are commonly found in air masses influenced by anthropogenic emissions. In the western regions of the U.S. a major anthropogenic aerosol source is emissions from cars/trucks (mobile sources), which lead to high NH 4 NO 3 concentrations. NO x (NO plus NO 2 ) formed in high temperature combustion reacts with OH to produce HNO 3 (nitric acid). NO 2 + OH + M HNO 3 + M (OH is formed photochemically involving H 2 O) Depending on ambient conditions (T, RH) and the composition of the partice, HNO 3 may condense onto existing particles. This gas-to-particle conversion is even more likely if NH 3 is present, forming ammonium nitrate aerosol (NH 4 NO 3 ). In the eastern US where significant power production involves coal combustion, sulfur associated with the coal is released to form sulfur dioxide (SO 2 ). This SO 2 can form sulfate aerosol (SO 2-4 ) through gas phase reactions, SO 2 + OH + M HSO 3 + M HSO 3 + O 2 SO 3 + HO 2 (fast) SO 3 + H 2 O + M H 2 SO 4 + M (fast) 1
The more common pathway, however, is through oxidation in liquid droplets involving hydrogen peroxide (H 2 O 2 ) or other oxidants. SO 2 (g) SO 2 H 2 O (SO 2 dissolves in the liquid drop and is hydrated) SO 2 H 2 O HSO 3 - + H + (hydrated SO 2 dissociates) H 2 O 2 H 2 O 2 (aq) (H 2 O 2 is produced from self reaction of peroxy-radicals, HO 2 ) HSO 3 - + H 2 O 2 (aq) + H + SO 4 2- + 2H + +H 2 O (dissocated SO 2 is oxidized) Annual Average PM 2.5 in Urban Areas, 2002 The predominance of mobile source emissions (cars etc) in the west and coal combustion emissions in the east results in a dominance of NH 4 NO 3 aerosol particles in the west and (NH 4 ) 2 SO 4 aerosol particles in the eastern U.S. Atlanta is ringed by large coal-fired power plants, and is also impacted from more distant power plants in TOTAL MASS = 30.4 ug/m3 other 17% Tennessee (TVA) and power plants along the Ohio River. The result is that the 'soot' sulfate PM2.5 mass in Atlanta is typically about 3% 35% 1/3 to 1/2 ammonium sulfate. The rest is carbonacous material, with smaller amount of minerls, ammonium nitrate, and various metals. organic C 31% nitrate 2% ammonium 12% Average PM 2.5 Observed Composition 2
Collection of Particles on Filters for Chemical Analysis. In this lab we will collect particles on filters, extract the filters for compounds soluble in water, and then analyze the extract for anions. As a method to capture particles for analysis, particle filtration is much more complex than expected. The mechanism by which particles are captured is not straight forward and potential interferences during the partilce collection is large. Filters are constructed of a variety of materials that include borosilicate glass, quartz, teflon, cellulos, polyvinyl chloride, polyester, polycarbonate, nylon, silver, etc. Filter structures also vary. The most common are fibers, membranes, and pore filters. Fiber and membrane filters capture particles by providing a convoluted flow path and large surface area. The figure shows an example of membrane filter. A teflon membrane filter will be used in this experiment. Pore filters are very different and are used for collecting particles on a flat surface for single particle analysis (e.g., electron microscopy). The type of filter structure and filter material must be carefully considered and will depend on the type of compounds to be analyzed and by what method. Pertinent issues include, filter pressure drop, interferences due to absorption of gases onto filter material, and possible interferences of the filter material with the measurement method. Artifacts can be a large problem when collecting particles on filters for analysis. These artifacts lead to either an over or under-measurement of the compound or parameter of interest, and are respectively referred to as positive or negative artifacts. Many compounds found in particles are semi-volatile (e.g., have fairly high vapor pressures). These compounds exist in the particles due to an equilibrium between the vapor and particle phase. Take for example nitric acid (HNO3) and nitrate aerosol (NO3-). Nitric acid will condense in a liquid particle to form particulate nitrate. Equilibrium shifts to the particle phase at lower temperatures, whereas at higher temperatures the vapor phase is favored. Higher ambient humidities (RH) lead to more condensed water, which favors the condensed phase. (Note, the situation is actually more complicated since droplet ph is also a significant factor; ph depends on the availability of neutralizing cations, such as ammonia, NH3). The problem with filter measurements is that sampling can occur over long periods (hours to days). Particles that are collected on the filter, say at night when it is cooler and higher RH, will be exposed to higher temperature and dryer air 3
during sampling the next day. This can lead to a negative artifact due to evaporation loss of HNO 3 that had been collected at night or in the early morning For compounds that are not volatile, such as sulfate (the corresponding gas phase compound is sulfuric acid, which has a very low vapor pressure), negative filter sampling artifacts are minimual. Other artifact problems include adsorption of gases onto particles collected on the filter and onto the filter media itself. These so-called positive artifiacts can be a major issue when sampling for organic aerosols in urban environments. 4
Experiment No. 4: Determining Concentrations of Chloride, Nitrate, and Sulfate in the Atlanta PM2.5 Aerosol In this experiment, you will collect particles onto a teflon filter, extract the filters and use ion chromatography to determine the concentration of chloride, nitrate, and sulfate in the Atlanta ambient particles. Particle Collection Procedure The system shown below has been set up be the TA in the penthouse lab. Flow direction outside inside Filter Holder Lab 3 Critical Orifice Vacuum Pump PM2.5 Size Selective Inlet This setup will be used for collecting particles on a teflon filter for approximately 24 hours. The particle size selector is a PM2.5 cut cyclone; only particles smaller than 2.5 µm diameter can pass through the cyclone and be collected on the filter. To control the flow rate in this experiment the critical orifice calibrated in Lab 3 is used. 1. Load a clean teflon filter into the filter holder. NEVER touch the filter with anything other than a clean tweezers. 2. Start the vacuum pump and record the start time. 3. Record the vacuum pressure on the pump downstream of the critical orifice to make sure the flow is critical. 4. Check that the met station is running and recording data. 5. Let the system run for roughly 24 hours. 6. Obtain two clean plastic bottles, rinse each three times with ultrapure water. 7. Return to the penthouse lab, stop the pump, record the sample ending time, and remove the filter from the filter holder with clean tweezers. Put your filter in one of the cleaned bottles and put the lid on. 8. Download the met data for the period you were sampling. This data may help in your data analysis below. Do the following steps to determine the concentrations of ions in the ambient aerosol: 5
1. Extract the ions from the filters: Use a clean (rinse 3 times with ultrapure water) graduated cylider or pipet (pipet is better) to measure and transfer 30 ml of ultrapure water into each bottle. The bottle without the filter will be a measurement blank. MAKE SURE THE BOTTLES ARE TIGHTLY SEALED. Sonicate or shake the bottles for about 15 minutes. Determine the uncertainty associated with the 30 ml measurement and record in your lab writeup. For each bottle transfer and filter a protion of the liquid into separate disposable cups. Using a disposable syringe, draw a sample into the syringe, attach a disposable filter and push the liquid through the filter into a disposable cup. Now analayze this liquid as done in Lab 2. Draw sample from the cup into the IC sample loop using a syringe. Pull roughly 5 ml of sample through the sample loop. Run the IC and determine the conductivity (area under curve) for chloride, nitrate, and sulfate for both bottles (filter and blank). Record the data. Data Analysis and Interpretation 1. Calculate the concentration of chloride, nitrate, and sulfate in your liquid sample and the blank using the IC calibration curves from Lab 2. 2. Consider any correction to the data, such as contamination or other interferences. For example, for each analyte, if the blank is above the limit of detection subtract it from the filter liquid concentrations. 3. Now calculate the mass of chloride, nitrate and sulfate in the ambient particles using the known liquid concentration determined in 1) and 2), the amount of liquid used to extract the filter, the sample air flow rate through the filter (use Lab 3 calibration data and the recorded vacuum pressure at the pump was the flow critical), and the length of time the filter was run. 4. Provide an estimate of the measurement uncertainty for each analyte reported and how it was determined. For example, this should include uncertainty in liquid added to extract filter, uncertainty in IC calibration (Lab 2), uncertinty in sampling air flow rate (Lab 3), and uncertainty in the sample collection time. Combine using quadrature sum of squares of relative errors. Thus, each PM2.5 concentration reported should be of the form X ± ΔX (for Cl -, NO 3 -, SO 4 = ). 5. Compare your data to other data collected by the Georgia EPD, http://www.air.dnr.state.ga.us/amp/. For example, find the fraction of the three measured anions to the PM2.5 total mass. You may have to average the GA EPD data to your filter integration time so that you can make a valid comparison. Are your results consistent with expectations of what people believe are the components of the Atlanta aerosol. 6. List possible sources for Cl -, NO 3 -, SO 4 2-. Use met data collected on the roof to help interpret your results. 6