Section i Sample Preparation and Sample Pretreatment Handbook of Spectroscopy Volume 1. Edited by Gunter Gauglitz and Tuan Vo-Dinh lSBN3-527-29782-0
Section I Sample Preparation and Sample Pretreatment Handbook of Spectroscopy, Volume 1. Edited by Günter Gauglitz and Tuan Vo-Dinh Copyright 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN 3-527-29782-0
Introduction Douglas A. Lane In order to obtain high-quality analytical data, the primary objective of the analyti- cal scientist must be, ideally, to obtain an artifact-free sample for the analysis. This is seldom a simple matter and presents many challenges to the investigator. It is often the case that many sampling programs frequently select the sampling me- thods based on what equipment is available rather than on what question is to be answered or what problem is to be addressed. It is important that sampling ob- jectives be defined first and then a suitable method be selected. The question,Can the sampling method I select provide me with the answers I am looking for? "must always be answered in the affirmative Sampling methodology differs greatly depending upon whether the sample is the gaseous, liquid or solid phase. If the sample is in the liquid or solid phase, is the sample an aerosol or particle that exists in a particular gaseous phase? In or on what medium shall the sample be collected and retained? How shall the sample be stored and/or trans ported prior to analysis? Must the sample the analysis to concentrate or isolate the analyte(s) of interest from the sample matrix before analysis? If the sample is to be used for legal purposes, a chain of custody(not discussed here)needs to be developed. A Quality Assurance/Quality Control (QA/QC) program will likely have to be developed for the analytica In situations where the analyte is present in trace quantities(as usually occurs in nvironmental samples ), it is vitally important to maximize the recovery of the ana lyte from the sample matrix and to lose as little of the analyte as possible during any subsequent processing or"work-up"stages in the analytical process. Extensive recovery testing is usually required to determine the efficiency of the collection and processing procedures The following two chapters address the question of sampling methods for the three phases in which a sample may occur -gaseous, liquid and solid. Different sampling approaches (active vs. passive) are considered as are the specific ap- proaches for a wide variety of sample types and matrices. Handbook of Spectroscopy Volume 1. Edited by Gunter Gauglitz and Tuan Vo-Dinh lSBN3-527-29782-0
Introduction Douglas A. Lane In order to obtain high-quality analytical data, the primary objective of the analytical scientist must be, ideally, to obtain an artifact-free sample for the analysis. This is seldom a simple matter and presents many challenges to the investigator. It is often the case that many sampling programs frequently select the sampling methods based on what equipment is available rather than on what question is to be answered or what problem is to be addressed. It is important that sampling objectives be defined first and then a suitable method be selected. The question, “Can the sampling method I select provide me with the answers I am looking for?” must always be answered in the affirmative. Sampling methodology differs greatly depending upon whether the sample is in the gaseous, liquid or solid phase. If the sample is in the liquid or solid phase, is the sample an aerosol or particle that exists in a particular gaseous phase? In or on what medium shall the sample be collected and retained? How shall the sample be stored and/or transported prior to analysis? Must the sample be processed before the analysis to concentrate or isolate the analyte(s) of interest from the sample matrix before analysis? If the sample is to be used for legal purposes, a chain of custody (not discussed here) needs to be developed. A Quality Assurance/Quality Control (QA/QC) program will likely have to be developed for the analytical method. In situations where the analyte is present in trace quantities (as usually occurs in environmental samples), it is vitally important to maximize the recovery of the analyte from the sample matrix and to lose as little of the analyte as possible during any subsequent processing or “work-up” stages in the analytical process. Extensive recovery testing is usually required to determine the efficiency of the collection and processing procedures. The following two chapters address the question of sampling methods for the three phases in which a sample may occur gaseous, liquid and solid. Different sampling approaches (active vs. passive) are considered as are the specific approaches for a wide variety of sample types and matrices. Handbook of Spectroscopy, Volume 1. Edited by Günter Gauglitz and Tuan Vo-Dinh Copyright 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN 3-527-29782-0
Collection and Preparation of Gaseous Samples Douglas A. Lane The collection of artifact-free gas phase samples is not a simple process. Unless the gas is a highly filtered and purified gas, it will most likely be a complex mixture of gases and vapors, liquids(aerosols)and solids(particles). For example, one of the most sampled, yet most complex sources is the earth's atmosphere. The atmo- phere is a mixture of gases(both organic and inorganic), liquids(such as rain dro- lets and aerosols) and solids(particles such as windblown dusts, pollens and fly ash from a myriad of combustion processes). The atmosphere is also irradiated with sunlight, which can initiate many photochemical reactions. It can truly be said that the atmosphere is like a giant chemical reactor in which all but the most inert compounds are chemically modified, dispersed and eventually depos ited to the earth [1]. It is not a simple matter to collect atmospheric samples, or other gaseous samples for that matter, without modifying the sample during the collection process. After all, one really wants to know the gaseous composition at the time of collection, not as modified by a particular sampling process Prior to using the sophisticated techniques described in the rest of this book to analyze a sample, one must first collect the sample and then prepare it for the final analysis. Analytical techniques have become incredibly sophisticated and more se. lective and sensitive over the past 20 to 30 years. Sampling methods, unfortunately, have not kept pace with the advances in the analytical technology despite the fact will still yield a poor result. To borrow an expression from the nalytical that a poorly collected sample, no matter how sophisticated the analytical method Garbage in equals garbage out". The ultimate challenge is to collect a sample that reflects the composition of the sample at the time of collection. To achieve this ob- ample collection metho as possible from all artifacts of the sampling procedure and be appropriate for the objective of the pro- gram for which the measurements are made. It is just as important to maintain the integrity of the sample after the sampling has been completed and during any work-up procedure to prepare the sample for analysis. o 20o3 wiLE Y VCH Verlag temb &u ce, KGal, weinheim
1 Collection and Preparation of Gaseous Samples Douglas A. Lane 1.1 Introduction The collection of artifact-free gas phase samples is not a simple process. Unless the gas is a highly filtered and purified gas, it will most likely be a complex mixture of gases and vapors, liquids (aerosols) and solids (particles). For example, one of the most sampled, yet most complex sources is the earth’s atmosphere. The atmosphere is a mixture of gases (both organic and inorganic), liquids (such as rain droplets and aerosols) and solids (particles such as windblown dusts, pollens and fly ash from a myriad of combustion processes). The atmosphere is also irradiated with sunlight, which can initiate many photochemical reactions. It can truly be said that the atmosphere is like a giant chemical reactor in which all but the most inert compounds are chemically modified, dispersed and eventually deposited to the earth [1]. It is not a simple matter to collect atmospheric samples, or other gaseous samples for that matter, without modifying the sample during the collection process. After all, one really wants to know the gaseous composition at the time of collection, not as modified by a particular sampling process. Prior to using the sophisticated techniques described in the rest of this book to analyze a sample, one must first collect the sample and then prepare it for the final analysis. Analytical techniques have become incredibly sophisticated and more selective and sensitive over the past 20 to 30 years. Sampling methods, unfortunately, have not kept pace with the advances in the analytical technology despite the fact that a poorly collected sample, no matter how sophisticated the analytical method, will still yield a poor result. To borrow an expression from the computer industry: “Garbage in equals garbage out”. The ultimate challenge is to collect a sample that reflects the composition of the sample at the time of collection. To achieve this objective, the sample collection method selected must be as free as possible from all artifacts of the sampling procedure and be appropriate for the objective of the program for which the measurements are made. It is just as important to maintain the integrity of the sample after the sampling has been completed and during any work-up procedure to prepare the sample for analysis. Handbook of Spectroscopy, Volume 1. Edited by Günter Gauglitz and Tuan Vo-Dinh Copyright 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN 3-527-29782-0
1 Collection and Preparation of Gaseous Samples An artifact is something not nanu rally present in the sample but is introduced during the sampling or work-up procedure. Artifacts include the oxidation of the collected sample during the sampling process, adsorption of gas phase compounds by a particle collection filter, volatilization of particle associated compounds that subsequently are trapped by an adsorbent and assessed as gas phase material, irr versible adsorption or reaction of the gases or vapors with the sampling substrate, ondensation of water on the sample, loss of adsorbed sample during sampling, transport or sample work-up and chemical alteration of the sample during sample extraction and/ or preparation. There are many other potential artifacts, all of which should be minimized. Part of this process is the selection of the proper sampling method for the problem at hand Since the focus of this book is the analysis of collected samples, only integrated samples will be considered here. Furthermore, this chapter will look only at the col lection and preparation of gaseous samples. The collection of particle and liquid samples will be considered in the following chapter. Because of their complexity, ill center on the collection and of atmospheric samples, however, the principles involved relate to any gaseous The most important aspect of sampling is to know what problem is to be solved or addressed and then to select the appropriate method. Proper selection of sam pling method is critical to the solution of a problem. Sampling considerations In the atmosphere, gases and vapors co-exist as gas phase material. Each chemical has its own vapor pressure and saturated vapor concentration. Chemical com pounds that have a subcooled liquid phase vapor pressure greater than approxi- mately 10-2 Pa will exist entirely in the gas phase. These compounds include gases such as ozone, oxides of nitrogen, carbon monoxide, carbon diox fur dioxide and vapors from volatile organic compounds(vOcs), such as methy lene chloride, acetone, and isoprene, low molecular weight aliphatic compounds and aromatic compounds such as benzene, toluene and xylene. Contaminants ith subcooled liquid phase vapor pressures less than 10 Pa will exist almost en tirely in the particle phase, while contaminants with vapor pressures between 10-2 and 10 Pa will partition themselves between the gaseous and particulate phases These are the so-called semivolatile organic compounds(SvOCs), and include many of the polychlorinated biphenyls(PCBs), some of the polycyclic aromatic compounds, many of the dioxins and many pesticides. Semivolatile compounds exist in the atmosphere at or near equilibrium between the gas phases. In Fig. 1. 1, the gas phase fraction of SvOC components is plotted versus the log of the subcooled liquid phase vapor pressure(log Pi)of the SVOCs. This shows graphically that compounds with a log P between approximately 10 and 10- Pa will partition between the gaseous and particulate phases in the atmo-
An artifact is something not naturally present in the sample but is introduced during the sampling or work-up procedure. Artifacts include the oxidation of the collected sample during the sampling process, adsorption of gas phase compounds by a particle collection filter, volatilization of particle associated compounds that subsequently are trapped by an adsorbent and assessed as gas phase material, irreversible adsorption or reaction of the gases or vapors with the sampling substrate, condensation of water on the sample, loss of adsorbed sample during sampling, transport or sample work-up and chemical alteration of the sample during sample extraction and/or preparation. There are many other potential artifacts, all of which should be minimized. Part of this process is the selection of the proper sampling method for the problem at hand. Since the focus of this book is the analysis of collected samples, only integrated samples will be considered here. Furthermore, this chapter will look only at the collection and preparation of gaseous samples. The collection of particle and liquid samples will be considered in the following chapter. Because of their complexity, most of the discussions in this chapter will center on the collection and analysis of atmospheric samples, however, the principles involved relate to any gaseous sample. The most important aspect of sampling is to know what problem is to be solved or addressed and then to select the appropriate method. Proper selection of sampling method is critical to the solution of a problem. 1.2 Sampling considerations In the atmosphere, gases and vapors co-exist as gas phase material. Each chemical has its own vapor pressure and saturated vapor concentration. Chemical compounds that have a subcooled liquid phase vapor pressure greater than approximately 102 Pa will exist entirely in the gas phase. These compounds include gases such as ozone, oxides of nitrogen, carbon monoxide, carbon dioxide and sulfur dioxide and vapors from volatile organic compounds (VOCs), such as methylene chloride, acetone, and isoprene, low molecular weight aliphatic compounds and aromatic compounds such as benzene, toluene and xylene. Contaminants with subcooled liquid phase vapor pressures less than 105 Pa will exist almost entirely in the particle phase, while contaminants with vapor pressures between 102 and 105 Pa will partition themselves between the gaseous and particulate phases. These are the so-called semivolatile organic compounds (SVOCs), and include many of the polychlorinated biphenyls (PCBs), some of the polycyclic aromatic compounds, many of the dioxins and many pesticides. Semivolatile compounds exist in the atmosphere at or near equilibrium between the gaseous and particulate phases. In Fig. 1.1, the gas phase fraction of SVOC components is plotted versus the log of the subcooled liquid phase vapor pressure (log PL) of the SVOCs. This shows graphically that compounds with a log PL between approximately 102 and 105 Pa will partition between the gaseous and particulate phases in the atmo- 1 Collection and Preparation of Gaseous Samples 5
Fig. 1.1 Gas Phase fraction as of the log of the subcool vapor pressure(Pa) of SvOcs ,2 LogIPi phere. First described by Junge [2], the theory has been greatly developed by Pankow [3, 4] and Pankow and Bidleman [5] The sampling method selected must be sensitive to the vapor pressure of the ompound, the temperature at which sampling is to take place, the stability of the compound during sampling and the anticipated concentration of the com- pound in the air. The act of drawing an air sample through a sampler requires a the o ure drop across the sampler and this will disturb the equilibrium between the gas and particle phases. Sampling methods must endeavor to minimize this disruption of the equilibrium if gas particle partition measurements are being Since gaseous samples are about 800 times less dense than liquid or solid sam- les and, since the vast majority of these gases and vapors exist in the in extremely low concentrations(often at nanogram to sub-nanogram per cubic meter concentrations), it is necessary to collect large volumes of air in order to col- lect sufficient material to permit both qualitative and quantitative analyses. For ex ample, in gas chromatographic/mass spectrometric analyses, 1 uL of a 1 mL sam- ole will typically be injected onto the column of the gas chromatograph. If the in- strument has sufficient sensitivity to permit quantitation on 50 pg of analyte, the there must be 50,000 pg or 50 ng of the analyte in the 1 mL sample. This, in turn, requires that 50 ng of sample be collected from the air (assuming no losses during the work-up procedure). If the sample exists in the air at a concentration of 1 ng m-3 then it follows that one would need to collect a minimum of 50 m of for the analysis. Likewise, if the compound exists in the air at a concentration 100 ng m, then one need only collect a 0.5 m' sample. It is, thus, important to have some knowledge of the anticipated concentration of the analyte in the gas mixture prior to selecting the sampling method and sampling time
sphere. First described by Junge [2], the theory has been greatly developed by Pankow [3, 4] and Pankow and Bidleman [5]. The sampling method selected must be sensitive to the vapor pressure of the compound, the temperature at which sampling is to take place, the stability of the compound during sampling and the anticipated concentration of the compound in the air. The act of drawing an air sample through a sampler requires a pressure drop across the sampler and this will disturb the equilibrium between the gas and particle phases. Sampling methods must endeavor to minimize this disruption of the equilibrium if gas particle partition measurements are being made. Since gaseous samples are about 800 times less dense than liquid or solid samples and, since the vast majority of these gases and vapors exist in the atmosphere in extremely low concentrations (often at nanogram to sub-nanogram per cubic meter concentrations), it is necessary to collect large volumes of air in order to collect sufficient material to permit both qualitative and quantitative analyses. For example, in gas chromatographic/mass spectrometric analyses, 1 L of a 1 mL sample will typically be injected onto the column of the gas chromatograph. If the instrument has sufficient sensitivity to permit quantitation on 50 pg of analyte, then there must be 50,000 pg or 50 ng of the analyte in the 1 mL sample. This, in turn, requires that 50 ng of sample be collected from the air (assuming no losses during the work-up procedure). If the sample exists in the air at a concentration of 1 ng m3 , then it follows that one would need to collect a minimum of 50 m3 of air for the analysis. Likewise, if the compound exists in the air at a concentration of 100 ng m3 , then one need only collect a 0.5 m3 sample. It is, thus, important to have some knowledge of the anticipated concentration of the analyte in the gas mixture prior to selecting the sampling method and sampling time. 6 1.2 Sampling considerations Fig. 1.1 Gas Phase fraction as a function of the log of the subcooled liquid vapor pressure (Pa) of SVOCs