A Case of Mistaken Identity?

A new company needed a permit for proposed wastewater discharge. The regulatory agency rejected its application for a proposed discharge of 15 parts per thousand of a chemical of concern. The company improved its process with a reduced discharge of 15 ppm. The agency again rejected the discharge limitation as too high. The company went back, and after further refinement, was able to propose a 15 ppb discharge. When the agency again rejected the application, the company advised that it was impossible to achieve lower than 15 ppb with current technology. The agency responded that it was not that pp-stuff that was the concern, it was the 15.

Introduction

With recent media attention to potential exposures to lead, phthalates, and other chemicals in consumer products ' especially children's products ' there have been federal and state efforts to reduce the levels of such chemicals in consumer products. The 2008 federal Consumer Product Safety Improvement Act (HR 4040, “CPSIA”) has set very low standards for paint, and lead and several phthalates in children's products. For example, by Aug. 14, 2009, the lead content in paint will decrease from 600 ppm to 90 ppm, and lead in other materials in children's products will decrease from 600 to 300 ppm and to 100 ppm by Aug. 14, 2011. Under the CPSIA, the level of six specific phthalates in children's products cannot exceed 1000 ppm. While these standards pose little challenge where they are present as contaminants in aqueous media such as wastewater, accurate measurements at these low levels in complex media such as plastic, metal, and other materials used in consumer products pose significant challenges as to accuracy. This is of great concern because of the enormous costs associated with a reported exceedance. Products with reported exceedances are subject to recalls under the CPSIA which involve costs for shipping and disposal in addition to the cost of the goods and possible indemnity claims ' all for naught if the reported result is inaccurate.

While chemicals can be measured in water at the part per billion level (ppb), assays in more complex media such as plastics, paint, and even metal are more problematic. It is difficult to isolate the analyte of interest from other chemicals, and there can be interferences from other components in the material. This article discusses the challenges of accurately measuring chemicals and the procedures for sample preparation.

Analytical Measurements

Although we think of a reported result as the concentration of a chemical in a matrix, analytical measurements are estimates of the concentration. Even with the same instrument on the same day, good laboratories will report a range of measurements on the same sample of about 10%. The variability is greater among different laboratories using the same type of instrument, and greater still among different instruments. Yet often, compliance is based on a single measurement.

There are several reasons for the variability. Although instruments are routinely calibrated, they can “drift” throughout the day. The more frequent the calibration, the lower the variability.

Measurements can also vary with materials in which the chemical of interest is not uniformly dispersed in a solid material, such as metal or plastic. Thus, a sufficient sample size is needed to better assure that the sample to be tested is representative. However, in many cases, measurements are being made on very, very small amounts of material. Different portions of the same materials can have different measurements of a chemical, with one being reported as a pass and the other as a fail.

Analytical Methods

A reliable analytical method is one that has a detailed procedure on sample preparation and has been tested in several laboratories and found to produce accurate, reproducible results. While there are numerous analytical methods for analyzing chemicals in water and soil, few have been developed for materials used in consumer products. Paint, plastics, metal, and other materials in consumer products are complex materials, and some of their components can interfere with analyses of lead, phthalates, and other chemicals of concern. Validated methods are especially important at the low regulatory levels of lead and phthalates in consumer products.

The CPSC has recently issued analytical methods for lead in paint, metal and nonmetal materials as well as phthalates in polyvinyl chloride (PVC). Significantly, the methods for lead in paint and nonmetal and the method for phthalate have a disclaimer that these methods have been developed by staff and have not been reviewed or approved by the CPSC. Indeed, it is not clear that any of these methods have been reviewed and approved or validated by any independent group, such as the AOAC International, (formerly Association of Official Agricultural Chemists), ASTM (originally known as the American Society for Testing and Materials) or even commercial laboratories.

Round-Robin Testing

An important part of developing analytical methods is testing at different laboratories to determine whether the different laboratories will obtain comparable results with various test samples. Generally, labs are sent samples of several well characterized test materials and asked to analyze them and report the results. The composition of the test materials is known to the sender, but not the laboratories. If laboratories have comparable results, it indicates that the method is satisfactory. If the laboratories report results outside of acceptable ranges, it indicates that additional work is needed on the method. This is an important part of the validation process for analytical methods, to assure that comparable results are obtained when the method is implemented in different laboratories.

Laboratory Analytical Standards

To verify the calibration, laboratory standards of known concentration are used to calibrate analytical instruments, and reference materials should also be analyzed. Calibration is generally performed by preparing solutions containing known amounts of the analyte to be measured and developing a curve based on the instrument response as a function of concentration. Generally, a blank and a minimum of three standards bracketing the expected concentrations of the analyte in the test samples are used to generate a linear calibration function. For example, concentrations of 0, 1 ppm, 2 ppm, 5 ppm, 10 ppm, and 20 ppm are prepared for samples with expected concentrations of 1-20 ppm of the test analyte. Calibration solutions can have additional analytes, provided that they do not interfere with each other. Linear regression statistics are commonly used to prepare calibration functions. Typically, the most accurate readings will be in the middle of the calibration range with greater error at the high and low end of the calibration function.

The calibration function must also be verified with an independent calibration standard, and, if possible, a reference sample with a known concentration of the analyte. Obviously, the calibration function is subject to gross errors if the standards used to prepare it were incorrectly prepared or have changed due to age or contamination during use. Thus, the use of an independent verification standard is the best check for this type of error. For best results, the calibration function and all sample processing steps should be verified using known reference materials in the same type of material (e.g., PVC) as the samples in question. Such known reference materials validate not only the calibration function, but also all sample preparation steps such as digestion, extraction, or dilutions included in the final calculation, and validates the laboratory's ability to make the measurement in the sample matrix. Reference materials may be available from government agencies, including the National Institute of Standards and Technology (NIST) in the U.S. and Canada, and from the Institute for Reference Materials and Measurements (IRMM) in Europe. Reference materials have long been available from commercial vendors for standard environmental matrices (soil, water, wastewater, etc), but are just becoming available for some of the newer regulated materials.

Test samples are then measured and their response is correlated with the concentration on the calibration function. Any samples with reported concentrations exceeding the concentration curve are diluted until they fall within the range. Again, care must be taken both in preparing and calculating the dilution. As with weighing errors, dilution errors are difficult to determine independently.

The test samples should also be accompanied by regular quality control samples (blanks, duplicates, spiked samples or reference samples) to monitor ongoing contamination control, precision, and accuracy associated with the measurements.

Sample Preparation

Proper sample preparation is as critical for accurately measuring chemicals in products as the proper operation of equipment. A major preliminary issue is whether to wash the test sample to remove any surficial contaminants on product surfaces that are not indigenous to the product. Of course, care must be taken to ensure that the soap and water or other cleaning materials do not introduce any contamination, cause interference in the analytical instrument, or remove the analyte of interest from the actual sample matrix. Some plaintiffs object to washing samples due to a concern for altering the sample, but sound science requires analytical results that are free from contamination due to external sources.

Most analytical instruments require that samples be in a liquid form, which means that solid samples such as metals and plastics must be dissolved in a liquid prior to analysis. (This article excludes discussion of analysis by x-ray fluorescence.) Paint and other coatings must be removed from their substrate. Most often, razors or other instruments are used to scrape off the paint or coating, with care taken that the tool does not introduce contamination. The CPSC procedures state that care should be taken to remove as little as possible of the substrate. However, even a small amount of metal with a higher lead content could result in an erroneous higher reported value for lead in a coating, and falsely indicate an exceedance of the paint standard as discussed below.

A weighed amount of paint is then dissolved, in the solvent of choice ' generally a strong acid such as concentrated nitric acid. With metal, an aliquot is cut or ground from the test sample and is then weighed and placed in concentrated nitric acid and heated until dissolved. Plastics do not readily dissolve in acid, and a weighed amount is often cut into small pieces or milled and placed in concentrated nitric acid and heated. In all cases, care must be taken to ensure that tools used to cut or scrape materials and the container and solvent used to dissolve the test specimen do not introduce contaminants.

Another concern is accuracy in weighing of representative samples. There are several concerns for error with small sample sizes for small samples. The sample size for closed vessel microwave dissolution procedures is generally limited to 50 milligrams or 0.05 grams; by contrast, samples sizes of 1 gram can be used with hot plate digestion. A weighing error can have a disproportionate effect with small samples. For example, a 10 mg weighing error for a 1 g sample will have a 1% error on the sample weight, while it will have a 20% error on the sample weight of a 50 mg sample. It is highly unlikely to detect independently whether there was a weighing error.

Small samples are also more subject to errors, due to uneven distribution of the analyte in the sample material. For example, if the concentration of lead varies in a polymer sample, a 50 mg sample may not contain sufficient material to represent the variability adequately. Finely dividing or milling the sample may minimize this effect as long as particle size is sufficiently small to provide adequate selection of different portions for the 50 mg sample. Another approach to minimize this error is to test multiple samples and average the results. This approach also provides information on the variability in the measurement due to sample selection.

Small sample sizes are also more prone to contamination effects. For example, assume that the glassware (or plasticware) in the preparation process contributes 10 :g of lead to the sample. For a 1 g sample, this will result in a positive error of 10 :g/g, or 10 ppm. For a 50 mg sample, this sample 10 :g contamination in glassware will contribute a positive error of 200 :g/g, or 200 ppm. Thus, attention to control of contamination becomes even more critical as sample size is reduced.

Plastic containers are generally used for microwave digestion and glass or plastic tubes for hot plate digestion. Cross contamination is possible with the vessels that are re-used, and generally, disposable digestion containers provide the greatest freedom from contamination, particularly at trace levels. Again, because of the significantly smaller amount of sample, cross contamination is of greater concern with microwave digestion than with techniques using larger sample sizes.

Analytical Instruments

Once the test material is dissolved in a suitable medium, it is generally injected into an analytical instrument with a separator and detector. The detector generates an electronic signal which is then interpreted by the operator. If the signal is too strong, the test sample must be diluted, which can introduce measurement and/or calculation errors as well as cross contamination from the diluent. A calculation error may be discoverable if calculations are provided with the result, but is often undetectable except by reference to additional testing.

Analysis of Metals

Atomic absorption (AA) is a common instrument to determine concentrations of metals. The electronic transitions of metals in the samples absorb light energy, and different metals absorb different wavelengths. The detector determines which wavelengths were absorbed and the resulting signal can be converted into a concentration after calibrating the instrument with standards. Another instrument commonly used for analysis of metals is inductively coupled plasma atomic emission spectroscopy (ICP-AES). In this technique, characteristic atomic emissions of metals in the sample are observed after excitation of the atoms in a plasma. However, there can be interferences with the wavelength signals, and measurements by AA and ICP-AES are generally not as accurate as measurements using inductively coupled plasma-mass spectroscopy (ICP-MS). With ICP-MS, the level of detection is lower by a factor of 10 to 100 and it is less susceptible to inferences than AA, because the detection is based on the mass of the element rather than absorption at a characteristic wavelength. With all instruments, other components can cause interferences, and some samples must be diluted, which again can introduce errors.

Analysis of Organics

Concentrations of organic compounds generally require that the organic compound of interest be isolated from the matrix by solvent extraction followed by a measurement of its concentration. Again, the test samples need to be dissolved prior to injection into an analytical instrument.

A common instrument to separate volatile chemicals is gas chromatography (GC). The CPSC method for phthalates uses GC to measure the amount of phthalates in plastics. In gas chromatography, a sample is injected into a column consisting of a solid material over which carrier gas is flowing. The components are separated from each other based on their relative affinities for the solid phase. As a component leaves the column, its concentration is determined by a signal at the detector. In simple mixtures, components can be cleanly separated from each other into separate signals. Complex mixtures, however, often have overlapping signals due to insufficient separation. In addition, because of other constituents in PVC, it is often difficult to achieve a good separation and accurate quantitative assay of phthalates in PVC.

There are six phthalates subject to a 1000 ppm standard under the CPSIA: di-ethyl hexyl phthalate (DEHP), dibutyl phthalate (DBP), benzyl butyl phthalate (BBP) diisononyl phthalate (DINP), diisodecyl phthalate (DIDP), and di-n-octyl phthalate (DnOP). The separation by GC of these phthalates from each other as well as other phthalates is difficult because they have similar chemical properties. For example, DEHP and DnOP have the same chemical formula but slightly different structure, making separation and identification difficult. DINP and DIDP differ only by two carbon atoms from each other and are very similar to DEHP and DnOP. A further complication is that commercial formulations of these phthalates are not pure materials and often include mixtures of different phthalates, phthalate isomers, and non-phthalate plasticizers which further complicates separation.

Common Laboratory Contaminants

Both lead and phthalates are common laboratory contaminants. Lead has been commonly used in pigments and various metal alloys. Lead concentrations in paints, polymers and metals can vary from percent levels to trace ppm levels. Phthalates are volatile and found in a variety of products from vinyl coverings on computer and power cords to plastic tubing on laboratory equipment to vinyl coverings on chairs. Phthalate levels can also vary in plastics from percent levels to low trace levels. Care must be taken to ensure that the results of the test sample have not been contaminated, and that high concentrations in one sample do not contribute contamination to low level samples processed at the same time. Generally, labs will prepare and analyze a sample of a material thought to have low lead in a similar manner to that of the test sample; if the material has no detectable lead or phthalate, the laboratory results on the test sample will likely not have been contaminated. If, however, the material has detectable lead, it may indicate laboratory contamination due to laboratory handling or due to cross-contamination from high level samples.

Summary and Conclusions

Accurate assays for lead and phthalates in consumer products are critically important. Current analytical methods have not been subjected to rigorous peer review and validation to demonstrate that they are reliable and reproducible at the levels required for legal determination of compliance. Additional method development and validation by regulating authorities and by testing laboratories is needed to ensure reliable results at these new regulatory standards in complex media in consumer products.

Eileen Nottoli is an attorney at Allen Matkins Leck Gamble Mallory & Natsis, a California-based real estate, land use, and environmental law firm that has focuses on sustainable development and addressed water recycling.

A new company needed a permit for proposed wastewater discharge. The regulatory agency rejected its application for a proposed discharge of 15 parts per thousand of a chemical of concern. The company improved its process with a reduced discharge of 15 ppm. The agency again rejected the discharge limitation as too high. The company went back, and after further refinement, was able to propose a 15 ppb discharge. When the agency again rejected the application, the company advised that it was impossible to achieve lower than 15 ppb with current technology. The agency responded that it was not that pp-stuff that was the concern, it was the 15.

Introduction

With recent media attention to potential exposures to lead, phthalates, and other chemicals in consumer products ' especially children's products ' there have been federal and state efforts to reduce the levels of such chemicals in consumer products. The 2008 federal Consumer Product Safety Improvement Act (HR 4040, “CPSIA”) has set very low standards for paint, and lead and several phthalates in children's products. For example, by Aug. 14, 2009, the lead content in paint will decrease from 600 ppm to 90 ppm, and lead in other materials in children's products will decrease from 600 to 300 ppm and to 100 ppm by Aug. 14, 2011. Under the CPSIA, the level of six specific phthalates in children's products cannot exceed 1000 ppm. While these standards pose little challenge where they are present as contaminants in aqueous media such as wastewater, accurate measurements at these low levels in complex media such as plastic, metal, and other materials used in consumer products pose significant challenges as to accuracy. This is of great concern because of the enormous costs associated with a reported exceedance. Products with reported exceedances are subject to recalls under the CPSIA which involve costs for shipping and disposal in addition to the cost of the goods and possible indemnity claims ' all for naught if the reported result is inaccurate.

While chemicals can be measured in water at the part per billion level (ppb), assays in more complex media such as plastics, paint, and even metal are more problematic. It is difficult to isolate the analyte of interest from other chemicals, and there can be interferences from other components in the material. This article discusses the challenges of accurately measuring chemicals and the procedures for sample preparation.

Analytical Measurements

Although we think of a reported result as the concentration of a chemical in a matrix, analytical measurements are estimates of the concentration. Even with the same instrument on the same day, good laboratories will report a range of measurements on the same sample of about 10%. The variability is greater among different laboratories using the same type of instrument, and greater still among different instruments. Yet often, compliance is based on a single measurement.

There are several reasons for the variability. Although instruments are routinely calibrated, they can “drift” throughout the day. The more frequent the calibration, the lower the variability.

Measurements can also vary with materials in which the chemical of interest is not uniformly dispersed in a solid material, such as metal or plastic. Thus, a sufficient sample size is needed to better assure that the sample to be tested is representative. However, in many cases, measurements are being made on very, very small amounts of material. Different portions of the same materials can have different measurements of a chemical, with one being reported as a pass and the other as a fail.

Analytical Methods

A reliable analytical method is one that has a detailed procedure on sample preparation and has been tested in several laboratories and found to produce accurate, reproducible results. While there are numerous analytical methods for analyzing chemicals in water and soil, few have been developed for materials used in consumer products. Paint, plastics, metal, and other materials in consumer products are complex materials, and some of their components can interfere with analyses of lead, phthalates, and other chemicals of concern. Validated methods are especially important at the low regulatory levels of lead and phthalates in consumer products.

The CPSC has recently issued analytical methods for lead in paint, metal and nonmetal materials as well as phthalates in polyvinyl chloride (PVC). Significantly, the methods for lead in paint and nonmetal and the method for phthalate have a disclaimer that these methods have been developed by staff and have not been reviewed or approved by the CPSC. Indeed, it is not clear that any of these methods have been reviewed and approved or validated by any independent group, such as the AOAC International, (formerly Association of Official Agricultural Chemists), ASTM (originally known as the American Society for Testing and Materials) or even commercial laboratories.

Round-Robin Testing

An important part of developing analytical methods is testing at different laboratories to determine whether the different laboratories will obtain comparable results with various test samples. Generally, labs are sent samples of several well characterized test materials and asked to analyze them and report the results. The composition of the test materials is known to the sender, but not the laboratories. If laboratories have comparable results, it indicates that the method is satisfactory. If the laboratories report results outside of acceptable ranges, it indicates that additional work is needed on the method. This is an important part of the validation process for analytical methods, to assure that comparable results are obtained when the method is implemented in different laboratories.

Laboratory Analytical Standards

To verify the calibration, laboratory standards of known concentration are used to calibrate analytical instruments, and reference materials should also be analyzed. Calibration is generally performed by preparing solutions containing known amounts of the analyte to be measured and developing a curve based on the instrument response as a function of concentration. Generally, a blank and a minimum of three standards bracketing the expected concentrations of the analyte in the test samples are used to generate a linear calibration function. For example, concentrations of 0, 1 ppm, 2 ppm, 5 ppm, 10 ppm, and 20 ppm are prepared for samples with expected concentrations of 1-20 ppm of the test analyte. Calibration solutions can have additional analytes, provided that they do not interfere with each other. Linear regression statistics are commonly used to prepare calibration functions. Typically, the most accurate readings will be in the middle of the calibration range with greater error at the high and low end of the calibration function.

The calibration function must also be verified with an independent calibration standard, and, if possible, a reference sample with a known concentration of the analyte. Obviously, the calibration function is subject to gross errors if the standards used to prepare it were incorrectly prepared or have changed due to age or contamination during use. Thus, the use of an independent verification standard is the best check for this type of error. For best results, the calibration function and all sample processing steps should be verified using known reference materials in the same type of material (e.g., PVC) as the samples in question. Such known reference materials validate not only the calibration function, but also all sample preparation steps such as digestion, extraction, or dilutions included in the final calculation, and validates the laboratory's ability to make the measurement in the sample matrix. Reference materials may be available from government agencies, including the National Institute of Standards and Technology (NIST) in the U.S. and Canada, and from the Institute for Reference Materials and Measurements (IRMM) in Europe. Reference materials have long been available from commercial vendors for standard environmental matrices (soil, water, wastewater, etc), but are just becoming available for some of the newer regulated materials.

Test samples are then measured and their response is correlated with the concentration on the calibration function. Any samples with reported concentrations exceeding the concentration curve are diluted until they fall within the range. Again, care must be taken both in preparing and calculating the dilution. As with weighing errors, dilution errors are difficult to determine independently.

The test samples should also be accompanied by regular quality control samples (blanks, duplicates, spiked samples or reference samples) to monitor ongoing contamination control, precision, and accuracy associated with the measurements.

Sample Preparation

Proper sample preparation is as critical for accurately measuring chemicals in products as the proper operation of equipment. A major preliminary issue is whether to wash the test sample to remove any surficial contaminants on product surfaces that are not indigenous to the product. Of course, care must be taken to ensure that the soap and water or other cleaning materials do not introduce any contamination, cause interference in the analytical instrument, or remove the analyte of interest from the actual sample matrix. Some plaintiffs object to washing samples due to a concern for altering the sample, but sound science requires analytical results that are free from contamination due to external sources.

Most analytical instruments require that samples be in a liquid form, which means that solid samples such as metals and plastics must be dissolved in a liquid prior to analysis. (This article excludes discussion of analysis by x-ray fluorescence.) Paint and other coatings must be removed from their substrate. Most often, razors or other instruments are used to scrape off the paint or coating, with care taken that the tool does not introduce contamination. The CPSC procedures state that care should be taken to remove as little as possible of the substrate. However, even a small amount of metal with a higher lead content could result in an erroneous higher reported value for lead in a coating, and falsely indicate an exceedance of the paint standard as discussed below.

A weighed amount of paint is then dissolved, in the solvent of choice ' generally a strong acid such as concentrated nitric acid. With metal, an aliquot is cut or ground from the test sample and is then weighed and placed in concentrated nitric acid and heated until dissolved. Plastics do not readily dissolve in acid, and a weighed amount is often cut into small pieces or milled and placed in concentrated nitric acid and heated. In all cases, care must be taken to ensure that tools used to cut or scrape materials and the container and solvent used to dissolve the test specimen do not introduce contaminants.

Another concern is accuracy in weighing of representative samples. There are several concerns for error with small sample sizes for small samples. The sample size for closed vessel microwave dissolution procedures is generally limited to 50 milligrams or 0.05 grams; by contrast, samples sizes of 1 gram can be used with hot plate digestion. A weighing error can have a disproportionate effect with small samples. For example, a 10 mg weighing error for a 1 g sample will have a 1% error on the sample weight, while it will have a 20% error on the sample weight of a 50 mg sample. It is highly unlikely to detect independently whether there was a weighing error.

Small samples are also more subject to errors, due to uneven distribution of the analyte in the sample material. For example, if the concentration of lead varies in a polymer sample, a 50 mg sample may not contain sufficient material to represent the variability adequately. Finely dividing or milling the sample may minimize this effect as long as particle size is sufficiently small to provide adequate selection of different portions for the 50 mg sample. Another approach to minimize this error is to test multiple samples and average the results. This approach also provides information on the variability in the measurement due to sample selection.

Small sample sizes are also more prone to contamination effects. For example, assume that the glassware (or plasticware) in the preparation process contributes 10 :g of lead to the sample. For a 1 g sample, this will result in a positive error of 10 :g/g, or 10 ppm. For a 50 mg sample, this sample 10 :g contamination in glassware will contribute a positive error of 200 :g/g, or 200 ppm. Thus, attention to control of contamination becomes even more critical as sample size is reduced.

Plastic containers are generally used for microwave digestion and glass or plastic tubes for hot plate digestion. Cross contamination is possible with the vessels that are re-used, and generally, disposable digestion containers provide the greatest freedom from contamination, particularly at trace levels. Again, because of the significantly smaller amount of sample, cross contamination is of greater concern with microwave digestion than with techniques using larger sample sizes.

Analytical Instruments

Once the test material is dissolved in a suitable medium, it is generally injected into an analytical instrument with a separator and detector. The detector generates an electronic signal which is then interpreted by the operator. If the signal is too strong, the test sample must be diluted, which can introduce measurement and/or calculation errors as well as cross contamination from the diluent. A calculation error may be discoverable if calculations are provided with the result, but is often undetectable except by reference to additional testing.

Analysis of Metals

Atomic absorption (AA) is a common instrument to determine concentrations of metals. The electronic transitions of metals in the samples absorb light energy, and different metals absorb different wavelengths. The detector determines which wavelengths were absorbed and the resulting signal can be converted into a concentration after calibrating the instrument with standards. Another instrument commonly used for analysis of metals is inductively coupled plasma atomic emission spectroscopy (ICP-AES). In this technique, characteristic atomic emissions of metals in the sample are observed after excitation of the atoms in a plasma. However, there can be interferences with the wavelength signals, and measurements by AA and ICP-AES are generally not as accurate as measurements using inductively coupled plasma-mass spectroscopy (ICP-MS). With ICP-MS, the level of detection is lower by a factor of 10 to 100 and it is less susceptible to inferences than AA, because the detection is based on the mass of the element rather than absorption at a characteristic wavelength. With all instruments, other components can cause interferences, and some samples must be diluted, which again can introduce errors.

Analysis of Organics

Concentrations of organic compounds generally require that the organic compound of interest be isolated from the matrix by solvent extraction followed by a measurement of its concentration. Again, the test samples need to be dissolved prior to injection into an analytical instrument.

A common instrument to separate volatile chemicals is gas chromatography (GC). The CPSC method for phthalates uses GC to measure the amount of phthalates in plastics. In gas chromatography, a sample is injected into a column consisting of a solid material over which carrier gas is flowing. The components are separated from each other based on their relative affinities for the solid phase. As a component leaves the column, its concentration is determined by a signal at the detector. In simple mixtures, components can be cleanly separated from each other into separate signals. Complex mixtures, however, often have overlapping signals due to insufficient separation. In addition, because of other constituents in PVC, it is often difficult to achieve a good separation and accurate quantitative assay of phthalates in PVC.

There are six phthalates subject to a 1000 ppm standard under the CPSIA: di-ethyl hexyl phthalate (DEHP), dibutyl phthalate (DBP), benzyl butyl phthalate (BBP) diisononyl phthalate (DINP), diisodecyl phthalate (DIDP), and di-n-octyl phthalate (DnOP). The separation by GC of these phthalates from each other as well as other phthalates is difficult because they have similar chemical properties. For example, DEHP and DnOP have the same chemical formula but slightly different structure, making separation and identification difficult. DINP and DIDP differ only by two carbon atoms from each other and are very similar to DEHP and DnOP. A further complication is that commercial formulations of these phthalates are not pure materials and often include mixtures of different phthalates, phthalate isomers, and non-phthalate plasticizers which further complicates separation.

Common Laboratory Contaminants

Both lead and phthalates are common laboratory contaminants. Lead has been commonly used in pigments and various metal alloys. Lead concentrations in paints, polymers and metals can vary from percent levels to trace ppm levels. Phthalates are volatile and found in a variety of products from vinyl coverings on computer and power cords to plastic tubing on laboratory equipment to vinyl coverings on chairs. Phthalate levels can also vary in plastics from percent levels to low trace levels. Care must be taken to ensure that the results of the test sample have not been contaminated, and that high concentrations in one sample do not contribute contamination to low level samples processed at the same time. Generally, labs will prepare and analyze a sample of a material thought to have low lead in a similar manner to that of the test sample; if the material has no detectable lead or phthalate, the laboratory results on the test sample will likely not have been contaminated. If, however, the material has detectable lead, it may indicate laboratory contamination due to laboratory handling or due to cross-contamination from high level samples.

Summary and Conclusions

Accurate assays for lead and phthalates in consumer products are critically important. Current analytical methods have not been subjected to rigorous peer review and validation to demonstrate that they are reliable and reproducible at the levels required for legal determination of compliance. Additional method development and validation by regulating authorities and by testing laboratories is needed to ensure reliable results at these new regulatory standards in complex media in consumer products.

Eileen Nottoli is an attorney at Allen Matkins Leck Gamble Mallory & Natsis, a California-based real estate, land use, and environmental law firm that has focuses on sustainable development and addressed water recycling.