CRITICAL REVIEW pubs.acs.org/est
Organic Contamination of Settled House Dust, A Review for Exposure Assessment Purposes Fabien Mercier,†,‡,* Philippe Glorennec,†,‡,§ Olivier Thomas,†,‡ and Barbara Le Bot†,‡ †
EHESP School of Public Health, Avenue du Professeur Leon Bernard, 35043 Rennes Cedex, France Irset, University of Rennes 1, 35000 Rennes, France § Inserm, U625, University of Rennes 1, 35000 Rennes, France ‡
bS Supporting Information ABSTRACT: People spend a considerable amount of time indoors. As a result, exposure to indoor contaminants is of great concern, notably via settled dust ingestion in particular for infants and toddlers. This paper proposes a critical review on the organic contamination of settled house dust and human exposure over the past 10 years and focused on sources, contaminations and measurement methods (sampling, pretreatment, storage and analysis). As many compounds were identified, arises the question of which ones to consider. Sensitive and selective analytical methods for simultaneous determination of targeted substances should be developed and evaluated. Various methods were described for sampling and sample preparation. Harmonization and standardization are needed to enable comparison of results from similar studies. Finally, an integrated multipollutant and multicompartment (settled dust, suspended particles and air) approach appears essential in order to determine the extent of the threat to public health posed by indoor contaminants.
’ INTRODUCTION People spend much of their time in indoor environments such as homes, workplaces, and schools; for instance in the U.S., 21 h/day for adults and 17 19 h/day for children.1 The development of building materials and consumer products over the past fifty years has led to an increase in the number of new chemicals such as plasticizers, pesticides, and flame retardants that are found in indoor environments.2 There are many sources of chemical contamination in the home caused by (i) the degradation of building materials, (ii) the widespread use of cleaning products, cosmetics, biocides, textiles, house furnishings, electronic devices, and other household products, (iii) indoor activities such as cooking, smoking and burning incense and candles, (iv) and the intrusion of material from outdoors. Over the past 10 years, considerable attention has been paid to the exposure of vulnerable groups, such as infants and pregnant woman, to indoor contaminants to assess the impact on health3 and reduce identified human health risks.4 This has led to settled dust being considered as an exposure medium,5 in particular for infants and toddlers, who are at highest risk owing to hand-tomouth activities. It is thought that the ingestion of settled dust may constitute a significant part of the exposure to some phthalates,6 polybromodiphenylethers (PBDEs)7 and pesticides.8 Settled dust could also be a global indicator of residential contamination,9 in particular for two other compartments of the indoor environment, the gas phase and suspended particles.10 This paper provides a critical analysis of scientific literature on the organic contamination of settled house dust over the r 2011 American Chemical Society
past 10 years. It first reviews the sources and occurrence of contaminants and then considers measurement methods (sampling, pretreatment, storage and analysis). These issues are important for human risk assessment (Figure 1) and are discussed with a view to assessing exposure.
’ SOURCES AND OCCURRENCE The literature review showed that a large number of chemicals have been studied in recent years. Table 1 indicates uses and sources for each chemical family (more details are given in the Supporting Information (SI)). The greatest number of references found was to PBDEs and pesticides, followed by phthalates, polycyclic aromatic hydrocarbons (PAHs), polychlorobiphenyls (PCBs) and perfluorinated compounds (PFCs). There were fewer papers on other compounds such as alkaloids, alkylphenols and other phenol derivatives, dioxins and furans, musk fragrances, organophosphate esters (OPs), organotin compounds, parabens, and polychlorobenzenes. Volatile organic compounds (VOCs)11 and organic UV filter compounds12 were also studied in settled house dust. Data on the occurrence of organic contaminants in settled dust (detection frequency, contamination and location) is given Received: March 19, 2011 Accepted: June 13, 2011 Revised: June 6, 2011 Published: June 13, 2011 6716
dx.doi.org/10.1021/es200925h | Environ. Sci. Technol. 2011, 45, 6716–6727
Environmental Science & Technology
CRITICAL REVIEW
Figure 1. Scope of the review in the risk assessment context.
in the SI. Phthalates were the most concentrated, ranging from μg/g to mg/g. PBDEs were the most widely investigated substances, in particular BDE-47, 99, 100, 153, 154, 183 and 209, with mean concentrations ranging from ng/g to μg/g. Bisphenol A, pesticides such as permethrin, and the majority of phthalates, PBDEs, PAHs and OPs were detected in most of the samples.
’ MEASUREMENT OF ORGANIC CONTAMINANTS Details of the measurement methods for organic contaminants in settled house dust are provided in the SI. Sampling, pretreatment, storage, and analytical methods are discussed below. Sampling. This section discusses the sampling methods used considering technical limitations and exposure assessment objectives. Description of Sampling Methods. Both passive and active sampling methods were found. Passive methods were rarely used but they involved collecting settled dust on items such as stationary beakers or nonelectrostatic plates9 or on dry filter paper in open Petri dishes placed on a flat surface for a given time (one week).13,14 Active methods, such as vacuuming, wiping, or brushing, were the main sampling methods used. The most commonly used method consisted in collecting settled dust using a vacuum cleaner. The U.S. Environmental Protection Agency (U.S. EPA) has developed a special vacuum cleaner (HVS3: high volume small surface sampler) to search for pollutants on indoor floors (rugs, carpets and various other types of floor-covering). The method is described in ASTM D 5438 (Standard practice for collection of floor dust for chemical analysis) and has been widely used.15 27 However, a conventional household vacuum cleaner was less expensive and easier to use than the HVS3 and could be used with a simple paper vacuum bag28 35 or with special fittings. In several studies, the Nilfisk Sprint plus 1600 W vacuum cleaner was used with a nylon sample sock (25 μm pore size) fitted to the furniture attachment tube of the vacuum cleaner.36 41 The Eureka Mighty-Mite model was used with a cellulose extraction thimble inserted between the crevice tool and the vacuum tube extender and secured, if necessary, with a rubber O-ring.7,42 47
Other accessories used included cellulose coffee filters attached at the end of the vacuum cleaner tube,48,49 phthalate-free ALK dust sampling devices (ALK dust collector and filter),50,51 and 90 mm pure cellulose membrane filters in holders made of styreneacrylonitrile polymer mounted on a polypropylene sampler connected to a vacuum cleaner.52,53 Some studies used other devices such as special vacuum bags,54 vacuum bags with a known porosity (1 μm),14 particle filters,55 a mouthpiece with a dust filter,11 extraction thimbles in place of conventional vacuum bags,56 glass fiber filters placed in special filter holders in vacuum cleaner tube57 and hand-held vacuum cleaners containing vacuum filter cones.58 These devices were mainly intended to prevent dust coming into contact with the inner parts of the vacuum cleaner, especially plastics that might contain phthalates that would contaminate the samples. Sampling had to be carried out with extreme care to quantify such substances in settled dust as illustrated by the use of a cellulose extraction thimble inserted in a custom PTFE crevice tool.7,45,46 An interesting alternative in terms of cost, implementation and sampling duration was collection using household vacuum cleaner bags.43,47,54,59 82 Another active method that did not use any appliances was “the wipe method”, which consisted in wiping a delimited sampling area on the floor by hand using a medium wetted with a solvent, usually isopropanol or a mixture of methanol and distilled water.13,14 The media used were gauzes,22,44 glass fiber filters,21,83,84 filter papers 13,14 and sponges.17 The wiped areas varied from 0.03113 to 1 m2.83 Another method proposed was collecting settled dust by carefully brushing the areas being studied.85 Another method suggested was the use of a small pair of steel tweezers to collect the dust from air conditioning unit filters or the blades of ceiling fans.86,87 Choice of Sampling Method. Each method had advantages and disadvantages depending on the aims and technical limitations of the study. Passive methods were rarely used probably because of the time necessary to obtain a sufficient quantity of dust. But they allow the collection of airborne samples without 6717
dx.doi.org/10.1021/es200925h |Environ. Sci. Technol. 2011, 45, 6716–6727
Environmental Science & Technology
CRITICAL REVIEW
Table 1. Investigated Compounds in Settled House Dust: Uses and Sources investigated compounds
uses/applications
sources
references
N
alkaloids
tobacco
cigarette smoke
58,109
2
alkylphenol
nonionic surfactants (substitute for anionic
detergents and cleaning products
15,26,27,45
4
ethoxylates
surfactants)
(APEs) and alkylphenols dioxins and furans
combustion byproducts, contaminants of
outdoor air (industrial process
73,110
2
chlorinated phenols
the production process of PCB mixtures herbicides, molluscicides, dye
contaminations), smoking... cellulose, paper, cardboard, timber,
94
1
34,62,106
3
31 33,55,111
5
outdoor air, PVC pipes
63
1
personal care products, canned food, beverages,
29,30,45,112
4
68,71,75,77,113,114
6
intermediates, bactericides, and
and impregnating textiles and leather
preservatives musks (synthetic aromatic nitromusks
fragrances (substitute for natural musk),
personal care products, cosmetics,
scenting agents
perfumes, cleaning products
and polycyclic musks) organophosphate esters (OPs)
polymers, lubricants and hydraulic fluids, textiles, polyurethane foams, computer,
plasticizers, antifoaming agents, stabilizers, flame retardants, hydraulic fluids, floor finishes, wax, lacquer, paint,
television-sets, electrical cable sheathing...
glue, cosmetic products, fungicides, industrial processes organotin compounds parabens and triclosan
PVC stabilizers, antifouling agents, industrial biocides bactericides, fungicides, antimicrobial agents, biocides, preservatives
textiles... cosmetics, clothes, toys, detergents, plastics for processing or packaging of foodstuffs
perfluorinated compounds (PFCs)
stain, water, oil, soil and grease repellents,
carpets, rugs, drapery, stone, tile, upholstery,
fire-fighting foams, alkaline cleaners,
fabrics, textiles, leather, packaging and
floor polish, sizing agents, leveling agents pesticides
paper products, coatings
agricultural pesticides, termiticides...Indoor
outdoor and/or indoor air, tracked-in dust,
use pesticides phenols and miscellaneous
plasticizers, stabilizers, antioxidants,
phthalic and adipic
plasticizers, solvents, adhesives,
13,14,17 28,34,35,44, 30
outdoor to indoor air transport
45,55 57,60,67,69,
outdoor air, polycarbonates and epoxy resins...
86,103,110,115 117 26,27,45,78,112 5
personal care products and cosmetics,
11,15,26,45,50 55,59, 14
intermediates in organic synthesis acid esters
flexible PVC, PVC flooring, wall covering,
perfume fragrances
62,67,118
electrical cable insulation (PVC insulation polybrominated diphenyl ethers (PBDEs) and other halogenated
flexible), casings and circuit boards foam cushions and mattresses, plastics, textiles,
flame-retardants, intermediate for the production of other commercial flame retardants (2,4,6-TBP)
7,36 43,45,47 49,
electronic devices and other household products, polystyrene, styrenic polymers
43
54,65 67,72,74, 76,79 82,87,101,
flame retardants
102,110,112, 119 132
polychlorinated biphenyls (PCBs)
heat transfer fluids, stabilizers for PVC
oil-filled transformers, capacitors, fluorescent lamp 26,45,46,55,67,86,
wire-insulation, flame-retardants,
ballasts, foam cushions and mattresses, electronic
pesticide extenders, additives
devices and other household products,
9
104,110,133
sealants, adhesives, paints, floor finishes, polychlorobenzenes
carbonless copy paper indoor air, dielectric fluids, PVC
intermediates in the synthesis of pesticides
34
1
and other chemicals, and in the production of pyrotechnic materials, components of dielectric fluids, deodorants for restrooms, moth control, insecticides, fungicides, plasticizers agent for PVC 6718
dx.doi.org/10.1021/es200925h |Environ. Sci. Technol. 2011, 45, 6716–6727
Environmental Science & Technology
CRITICAL REVIEW
Table 1. Continued investigated compounds polycyclic aromatic Hydrocarbons (PAHs) UV filters
uses/applications combustion byproducts UV filters
sources outdoor air, cooking, smoking, unvented
references 16,26,45,55,61,
combustion appliances sunscreens and other personal care products,
N 12
64,69,70,83 85,105 12
1
11
1
goods, plastics, varnishes, clothes, furniture, upholstery, paints volatile organic compounds (VOCs)
coarse material; 99% of the collected particles were less than 50 μm in diameter.9 Using an HVS3 vacuum cleaner was a standardized method of collecting settled dust but, as it was expensive and cumbersome, it seemed to be more appropriate for small scale studies. The use of a commercial vacuum cleaner appeared to be a satisfactory alternative with special accessories to prevent sample contamination. The third vacuum cleaner method based on the collection of vacuum cleaner bags from households required voluntary participation or compensation. This technique could be used to obtain a large quantity of dust quickly but had two major disadvantages: the dust was probably contaminated by the inner parts of the vacuum cleaner and was not representative for target population exposure studies. The wipe method was quick and easy to implement and no contamination was expected. It was designed to collect dust likely to adhere to the hands and then be ingested. However, the quantity of sampled dust was relatively small (about 30 mg for 0.1 m2 88), which might alter detection limits. It should be noted that this method was rarely used for organic contaminants. Comparisons between sampling methods were rarely reported for organic contaminants. A significant correlation between a passive method (dry filter paper placed in an open Petri dish for one week) and an active method (wipe method using wet filter paper) was observed for pesticides in farmer’s homes.13 Allen43 found significantly lower (and not strongly correlated) PBDEs concentrations in home vacuum cleaner bag dust than in researcher-collected dust (Eureka Mighty-Mite vacuum cleaner with a cellulose extraction thimble inserted between the crevice tool and the vacuum tube extender), probably because of the choice by the researcher of the sampled rooms according to the presence of potential sources of PBDEs. By its side, Colt89 concluded that the home vacuum cleaner bag method seemed to be a suitable alternative to the HVS3 for pesticides and other compounds including PAHs and PCBs in carpet dust. A few studies compared dust sampling methods for exposure assessment, mainly with respect to lead contamination. A study90 on the relative effectiveness of four sampling methods (wipes and three vacuum cleaners, HVS3, GS 80, and MVM) concluded that the HVS3 vacuum cleaner provided the best correlation with blood lead levels (R2 = 0.33), followed by wipes (R2 = 0.25). Wipes were recommended for the assessment of exposure in public health, even on carpets and rugs, as they were easier to use and cheaper. Lanphear91 also recommended the use of wipes for large-scale studies as they were easier to use, cheaper and more suitable for standardized quality control procedures for laboratory analyses. To our knowledge, no information exists for organic compounds. Regarding organic compounds, Allen et al.43 noticed that PBDE dust concentration and dust loading were strongly correlated. The choice of exposure units was a key issue when choosing a sampling method as they must be related to the aims of the study.
indoor air
Concentrations of target compounds in settled dust were expressed either as surface loading (μg/m2) or dust concentration (μg/g) depending on the sampling method. The dust concentration was calculated from the mass of analyzed dust. This information was not available in the case of a dust wipe sample as the sample was generally extracted in its entirety, including the wipe. The weight of the wipe could not be determined by weighing before and after sampling because of the abrasion of the wipe during sampling. Surface loading calculations were based on the area of the sampled zone, which was not the case when dust was collected in household vacuum cleaner bags. The sampling method chosen, therefore, conditioned the units of the results and consequently their application. Conversely, the aim of the study implied a sampling method. Surface loading measurements were usually more appropriate for human exposure assessment43 while dust concentration was usually more suitable for source characterization (lower detection limits) and health risk assessment (calculating the ingested dose from the quantity of ingested dust in mg/d was based on the mass concentration of a compound in dust). Another key point was the sampled locations when using vacuum cleaner method. If the researchers are primarily interested in source characterization, then multiple surfaces may be sampled including rugs, upholstery, wood floors, windowsills, ceiling fans, and furniture, as done by Rudel.45,46 However, if the researchers are primarily interested in human exposure and predictors of body burden, then the main living area of the floor should be preferred. The spatiotemporal variability of dust concentration must be taken into account in order to define the sampling strategy as the quantity and composition of house dust depends on numerous seasonal and environmental factors:9 surroundings, exchange with outside air and ground, age of the house, building materials and their condition, density of furniture and carpets and their state of preservation, ventilation and heating systems, cleaning habits, and time activity pattern. Furthermore, the concentration of an indoor pollutant is governed by its behavior in the indoor environment (emission, phase equilibrium, dilution, and degradation). It depends not only on its indoor emission rate but also on the rate at which it is transported from outdoors to indoors, and the rate at which it is collected by indoor surfaces, consumed by indoor chemical processes or removed by ventilation or filtration.2 Allen et al.43 examined critical factors influencing exposure estimates and comparisons between studies for PBDEs. In particular, temporal and spatial variability of concentrations in the same home, and correlation between air and settled dust concentrations were discussed. The authors observed that (i) PBDE concentrations did not vary in time but spatial variability within a home suggested that specific sources played a significant role in the distribution of PBDEs in household dust and (ii) air 6719
dx.doi.org/10.1021/es200925h |Environ. Sci. Technol. 2011, 45, 6716–6727
Environmental Science & Technology
CRITICAL REVIEW
Figure 2. Critical issues to be considered when sampling indoor settled dust for exposure assessment purposes.
(vapor and particulate phase) and dust concentrations were correlated for pentaBDE but not for decaBDE. The confirmation and generalization of these results required additional similar works, notably on other SVOCs. However, it was important to collect dust from the rooms most frequented by the target population. On the basis of this information, Figure 2 attempts to display the critical factors in sampling indoor settled dust for exposure assessment purposes with regard to the different exposure pathways: non dietary ingestion of dust (especially by infants and toddlers), ingestion of particles adhering to food, surfaces in the home (toys, etc.) and skin, inhalation of suspended and resuspended particles (because of typical human activities such as walking, cleaning, or vacuuming) and dermal contact (absorption through the skin). It appeared that the three compartments of the indoor environment (gas phase, suspended particles and settled dust) must be studied for human exposure assessment purposes. One study92 was carried out on the distribution of phthalate esters in indoor environments between these three compartments. It was reported that measurements of phthalate ester concentrations in settled dust could provide an estimate of their concentration in suspended particles and in the gas phase, provided that concentrations of suspended particles were available. The contribution of each compartment to indoor exposure could, therefore, be estimated from the concentrations in settled dust and concentrations of suspended particles. The authors indicated that this
could be applied to most SVOCs since the same sorption processes were likely to be implicated in the association between the compounds and both suspended particles and settled dust. Pretreatment. Before analysis, two pretreatments were suggested for stabilizing dust and selecting the part to be analyzed: sterilization and sieving. Sterilization. Dust samples can be sterilized by exposure to ionizing radiation (X, β, or γ-rays) to remove the microbial activity and extend the conservation time of the dust sample. So far as we are aware, only one study77 used irradiation to eliminate microbiological activity. The authors commented on the lack of published data concerning the storage stability of the target compounds (perfluorinated compounds) in dust or the potential impact of irradiation. The National Institute of Standards & Technology (NIST) also sterilized vacuum cleaner bags by γ radiation for the preparation of Standard Reference Material 2585 (SRM 2585, Organic Contaminants in House Dust).93 Sieving. Coarse material can be removed from the raw dust and particles separated from fibers by sieving. This step is essential but an analysis of the scientific literature produced no consensus in the choice of the dust fraction selected for analysis, beyond the fact that it was necessary to obtain a more homogeneous sample. The following mesh sizes were reported: < 2 mm,55,60,65,74,76,79 < 1 mm,94 between 75 μm and 1 mm,73 < 500 μm,36 39,41 43,66 < 150 μm,16,18,19,25,26,45,46,58,68,70,75,77,80 82 < 125 μm, 7 < 100 μm, 67 < 75 μm, 29 < 63 μm, 59,78 and
