Environ. Sci. Technol. 2006, 40, 590-594
Removal of Lead Contaminated Dusts from Hard Surfaces R O G E R D . L E W I S , * ,† SRIDHAR CONDOOR,‡ JOE BATEK,‡ KEE HEAN ONG,† DENIS BACKER,† DAVID STERLING,† JEFF SIRIA,§ JOHN J. CHEN,| AND PETER ASHLEY⊥ School of Public Health, Saint Louis University, St. Louis, Missouri, 63104, Parks School of Engineering, Aviation, and Technology, Saint Louis University, St. Louis, Missouri, 63103, Global Environmental Laboratories, St. Louis, Missouri, 63123, Department of Preventive Medicine, Stony Brook University, Stony Brook, New York 11794, and Office of Healthy Homes and Lead Control, U.S. Department of Housing and Urban Development, Washington, DC, 20410-3000
Government guidelines have widely recommended trisodium phosphate (TSP) or “lead-specific” cleaning detergents for removal of lead-contaminated dust (LCD) from hard surfaces, such as floors and window areas. The purpose of this study was to determine if low-phosphate, non-lead-specific cleaners could be used to efficiently remove LCD from 3 types of surfaces (vinyl flooring, wood, and wallpaper). Laboratory methods were developed and validated for simulating the doping, embedding, and sponge cleaning of the 3 surface types with 4 categories of cleaners: lead-specific detergents, nonionic cleaners, anionic cleaners, and trisodium phosphate (TSP). Vinyl flooring and wood were worn using artificial means. Materials were ashed, followed by ultrasound extraction, and anodic stripping voltammetry (ASV). One-way analysis of variance approach was used to evaluate the surface and detergent effects. Surface type was found to be a significant factor in removal of lead (p < 0.001). Vinyl flooring cleaned better than wallpaper by over 14% and wood cleaned better than wallpaper by 13%. There was no difference between the cleaning action of vinyl flooring and wood. No evidence was found to support the use of TSP or lead-specific detergents over all-purpose cleaning detergents for removal of lead-contaminated dusts. Nophosphate, non-lead-specific detergents are effective in sponge cleaning of lead-contaminated hard surfaces and childhood lead prevention programs should consider recommending all-purpose household detergents for removal of lead-contaminated dust after appropriate vacuuming.
Introduction The presence of lead-contaminated dust (LCD) on flooring, window sills, and window troughs is highly correlated with * Corresponding author phone: 314-977-8151; fax: 314-977-8150; e-mail:
[email protected]. † School of Public Health, Saint Louis University. ‡ Parks School of Engineering, Aviation, and Technology, Saint Louis University. § Global Environmental Laboratories. | Stony Brook University. ⊥ U.S. Department of Housing and Urban Development. 590
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 2, 2006
blood lead levels in children (1). A combination of appropriate cleaning techniques, awareness, education, and appropriate behavior is necessary for reduction of blood lead levels in children (2). Government guidelines have widely recommended trisodium phosphate (TSP) and to a lesser extent, “lead-specific”, detergents for cleaning of lead-contaminated dust from hard surfaces, such as vinyl flooring or wood floors and window areas, after vacuuming with a high-efficiencyfilter equipped vacuum cleaner (3). However, the U. S. EPA said in one government report that TSP may not be necessary; but this recommendation was based on two studies, one of which did not use TSP for removal of leaded dust and one which did use TSP but not against actual leaded dust (4-6). The literature to date on cleaning lead from hard surfaces consists of two laboratory reports published by the EPA in 1997 and 1998 (5, 7) and a more recent report of a randomized field trial (8). The 1997 EPA study was an ambitious project that explored the efficacy of over 32 cleaners in removing two types of soilsoily and drysfrom several hard surfaces. The authors of the 1997 cleaning study evaluated several markers of cleaning in each detergent including surface tension, pH, and phosphate concentration. They found no significant factors that related lead cleaning efficiency with specific detergents, type of soil, or surface, with the exception of surface tension, which was marginally significant. The 1998 EPA study examined a wide range of surface tension and phosphate concentrations for removal of LCD by adjusting these parameters in one hand-dishwashing detergent. The 1998 EPA study demonstrated no significant effect of surface tension or phosphate strength on removal of LCD from enamel- or latex-paint-coated wood surfaces. The two EPA studies made use of reference soils that were deposited on unused or new surfaces using artificial soil, suspended in liquid (mineral spirits), and applied without embedding the lead-containing soil using forces that would represent walking. Therefore, the system used to test cleaning agent efficacy may not be generalizable to real world conditions of surface soiling. A study by Rich et al. (8) evaluated 127 homes in a randomized trial to determine if low-phosphate, non-TSP detergents and non-HEPA vacuum cleaners could perform just as well as TSP and HEPA vacuum cleaners in removal of LCD in urban homes. The study resulted in inconsistent findings where TSP detergents were marginally superior to non-TSP detergents in lead cleanup for floors and window sills but non-TSP detergents (Spic and Span) were significantly superior over TSP in window trough cleanup (21% difference). Although the study by Rich et al. (8) inventoried different materials such as vinyl flooring, painted wood, and tile on lead removal they did not evaluate the effects of these materials on lead removal. It is quite possible that the different surface textures of the materials may account for differences in cleaning results. The 1997 and 1998 lab studies used an indirect method for determining cleaning by using a wipe to remove residual lead prior to and after cleaning (5, 7). A study that analyzes the actual cleaned surface, a direct evaluation of cleaning, may avoid sampling error where a wipe is used after cleaning. These wipes may have extremely small quantities of lead in them after cleaning, making it difficult to find significant differences in cleaning. Although the previously described cleaning studies had some methodological shortcomings, several key findings were established: recognition that household cleaning agents do not have major chemical differences and do not vary in such a manner that specific factors such as pH or phosphate content could be used to predict LCD cleaning efficacy; surface tension is not an important factor in lead cleaning 10.1021/es050803s CCC: $33.50
2006 American Chemical Society Published on Web 12/07/2005
TABLE 1. Chemical and Physical Properties of Detergents in Screening Procedurea
test or analyte surface tension, neat (dynes/cm) surface tension, diluted (dynes/cm) pH, neat PH, dilute water content (%) non-ionic surfactant (%) anionic surfactant (%) cationic surfactant (%) a
one-step wood “environment floor all-purpose all-purpose all-purpose all-purpose lead-specific cleaner friendly” detergent cleaner #1 cleaner # 2 cleaner # 3 cleaner # 4 cleaner 31.7 NM 9.7 NM 99.6 0.5 ND ND
33.0 34.6 9.3 9.3 92.1 4.9 ND ND
32.5 33.0 5.6 7.5 91.3 4.3 ND present
29.7 30.1 10.0 8.4 83.1 5.5 2.0 ND
33.1 34.7 9.0 8.8 89.9 7.0 ND present
30.8 32.0 9.9 9.5 90.4 6.6 2.25 ND
30.9 33.5 9.9 9.3 93.3 4.6 1.5 present
34.5 35.1 5.0 5.7 85.2 2.6 1.5 ND
NM ) Not measured. ND ) not detected. Present ) substance detected but not quantifiable.
efficacy; and a high phosphate content may not be necessary for effective cleaning of LCD. The purpose of this study was to determine if low-phosphate, non-lead-specific cleaners could be used to efficiently remove lead-contaminated dust from commonly used hard surface materials that would be found on walls, floors, window sills, or window troughs. Three hypotheses were posed: (1) low-phosphate, general-purpose/ floor detergents do not differ from trisodium phosphate in cleaning efficiency; (2) low-phosphate, general-purpose/floor detergents do not differ from lead-specific cleaning agents in cleaning efficiency; and (3) surface type will affect LCD removal.
Experimental Section Trisodium phosphate and a lead-specific detergent were chosen at the outset to be tested for removal of LCD. Other detergents were screened in order to arrive at generic categories of cleaners that would be used to remove LCD from hard surfaces. This left a large number of general and somewhat specialized floor cleaners to be screened to arrive at two or three other categories of cleaners. Unlike previous studies that attempted to characterize a broad spectrum of detergents and cleaning products for use in LCD removal, we chose to characterize only well-recognized and highvolume general floor and wall cleaners. Market leaders were determined from a Lexis-Nexis search of business and industry databases for sales of household floor and wall cleaners and counter cleaners for 52 weeks of sales during 1999 (9, 10). All-purpose floor and wall cleaners, a so-called environmentally friendly counter and window detergent, a wood cleaning floor and wall detergent, and a one-step floor cleaner that requires no rinsing step represented categories of cleaners that could be used for floors, window sills, and window trough cleaning. In addition to high-volume consumer use, detergents were categorized by surfactant ionic content, i.e., anionic, cationic, or nonionic. The surface tension, pH, and general compositional analysis of 8 market-leading detergents were evaluated (Table 1). Surface tension and density were performed at 77 °C using standard methods (ASTM D133-89 and ASTM D 1475-98, respectively) (11, 12). The laboratory that performed this work participates in an ASTM program for intra-laboratory exchange of test liquids. Some detergents are not intended for dilution such as the wood cleaner and therefore no analysis was done on a diluted form. Dilutions for detergents were made up according to manufacturer specifications. Compositional analysis was performed on neat (undiluted) solutions. The American Association for Laboratory Accreditation accredits the independent laboratory that performed this work. The organic materials were separated as neutral ethanol solubles. The alcohol solubles were then dried to constant weight at 105 °C to determine the nonvolatile organics. The nonvolatile organics were redissolved in ethanol, diluted to a known volume, and aliquots
were taken for determination of nonionic, anionic, and cationic surfactants. An aliquot of the organics was passed through a mixedbed ion-exchange resin. The effluent was evaporated to a constant weight at 105 °C. The weight of residue represented the nonionic surfactant (ASTM D 4252-89) (13). Confirmation was obtained by FTIR (ASTM D 2357-74) (14). An aliquot of the organics was analyzed for anionic content by cationic titration (ASTM D 1681-92) (15). Results were calculated as C12LAS. An aliquot of the organics was passed through an anion-exchange resin. The effluent containing nonionic and cationic surfactants was examined for cationic surfactant using reverse-phase titration with anionic surfactant (ASTM D 1681-92) (15). A reference lead-contaminated dust (LCD) was generated from sanding an old window frame containing 20 mg/cm2 of lead based on XRF analysis. Approximately 150 g of dust was prepared. A small quantity of fumed silica was added to the reference dust to increase the effectiveness of aerosol dispersion and uniform deposition on the specimens. After the dust was mixed with 1% fumed silica (w/w), samples were separated into 3 50-g samples. The samples were sieved to 10 µm. The composition of this dust by sieve fraction, dust mass, and lead mass is provided as Supporting Information. Dust that was
