Document 5kR1BYVvKgbXparXj4024Xg6e

DownloadSearchable PDFRandom document
i CMm CHEMICAL MANUFACTURERS ASSOCIATION January 3, 1991 TO: Vinylidene Chloride Producers Group RE: Lance Wallace Study JAN - T199! 11 Lance Wallace recently sent me the two enclosed published reports. I was expecting some unpublished data from him; nevertheless, this is what I received. The TEAM Study report is quite large and I did not photocopy the document other than the title page. Wallace himself is unsure of the source of vinylidene chloride in his study; however, he is confident it is not a sampling and/or analytical error. I will be in touch with Zeb Bell about our next step. Sincerely, J Jon Busch Manager, Vinylidene Chloride Panel '""501 M Stic;et, NW, Washington. DC 20037 202-887 1100 Telex 89617 (C.MA WSFI) SL 063785 w * SERA United States Office of Acid Deposition, Environmental EPA/600/6-87/002b Environmental Protection Monitoring and Quality Assurance June 1987 Agency Washington DC 20460 Research and Development_____________________________________________________________ The Total Exposure Assessment Methodology (TEAM) Study: Elizabeth and Bayonne, New Jersey, Devils Lake, North Dakota and Greensboro, North Carolina: Volume II. Part 2 SL 063786 6PA/600/6-87/002b June 1987 TOTAL EXPOSURE ASSESSMENT METHODOLOGY (TEAM) STUDY: ELIZABETH AND BAYONNE, NEW JERSEY, DEVILS LAKE, NORTH DAKOTA AND GREENSBORO, NORTH CAROLINA VOL. II (SECTION 8 AND REFERENCES) FINAL REPORT PART II by E. D. Pelllzzarl, K. Perrltt, T. D. Hartwell, L. C. Michael, C. M. Sparaclno, L. S. Sheldon, R. Whitmore, C. Lenlnger, H. Zelon, R. W. Handy and D. Smith Research Triangle Institute Post Office Box 12194 Research Triangle Park, North Carolina 27709 Project Officer L. Wallace Air, Toxics, and Radiation Monitoring Research Division Office of Monitoring, System and Quality Assurance U.S. ENVIRONMENTAL PROTECTION AGENCY OFFICE OF RESEARCH AND DEVELOPMENT WASHINGTON, DC 20460 SL 063787 DISCLAIMER This report was prepared under contract to an agency of the United States Government. Neither the U.S. Government nor any of Its employees, contractors, subcontractors, or their enployees makes arty warranty, expressed or Implied, or assumes any legal llbablllty or responsibility for any third party's use or the results of such use of any Information, apparatus, product, or process disclosed In this report, or represents that Its use by such third party would not Infringe on privately owned rights. Publications of the data In this document does not signify that the contents necessarily reflect the joint or separate views and policies of each sponsoring agency. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. ii SL 063788 CONTENTS Figures Tables Sectjan 8 Statistical Analysis of Oata References ........................................................... Bias iv xii ill SL 063789 Environment International, V'ol. 8, pp 269-282. 1982 Pnnied m the USA. All rights reserved. 0160-4120/82/0*0269-14103 00/0 Copyright e 1982 Pergamon Press Ltd MONITORING INDIVIDUAL EXPOSURE. MEASUREMENTS OF VOLATILE ORGANIC COMPOUNDS IN BREATHING-ZONE AIR, DRINKING WATER, AND EXHALED BREATH Lance Wallace Office of Monitoring Systems and Quality Assurance, Office of Research and Development, U.S. Environmental Protection Agency, Washington, DC 20460, USA Ruth Zweidmger, Mitch Erickson, S. Cooper, Don Whitaker, and Edo Pellizzari Analytical Sciences Division. Chemistry and Life Sciences Group. Research Triangle Institute, Research Triangle Park. North Carolina 22709, USA Methods for determining individual exposure to volatile organic compounds (VOQ during normal daily activities were Held tested on university student volunteers in Texas and North Carolina. The equipment tested included a personal monitor employing Tenax GC* to collect organic vapors for later analysis by GC-MS, and a specially designed spirometer for collecting samples of expired human breath on duplicate Tenax cartridges. The personal monitor and spirometer proved feasible for collecting abundant quan titative data on most of the 15 target organics. Air exposures to many VOC varied widely, sometimes over three orders of magnitude, among students on the same campus who had been monitored over the same time period and day. A log-linear relationship between breathing-zone air exposures and concentrations in exhaled breath was suggested for three chemicals: tetrachloroethylene, 1,1,1-trichioroethane. and vtnylidene chloride. Air was the main route of exposure for all target compounds except the two trihalomethanes (chloroform and bromodichloromethane), which were transmitted mainly through water. Estimated total daily intake through air and water of the target organics ranged from 0.3 to 12.6 mg, with 1,1,1-irichioroethane at the highest concentrations in both geographic areas. Introduction Few studies (Pellizzari et ai., 1979; Zweidinger et al., 1980) have attempted to measure individual human ex posure to organic substances simultaneously with measurements of body burden. Yet a knowledge of ex posure and body burden is crucial in arriving at deci sions of great economic consequence concerning the regulation of these substances. The present study is a pilot effort to evaluate the methods required to deter mine individual human exposure and body burden for a number of volatile organic compounds (VOC). Study objectives The main objective of the study was to field test the following methods for measuring human exposure to VOC; A personal air-quality monitor to sample breathingzone air; A specially designed spirometer to sample exhaled breath; Analytical protocols for measuring VOC in air, tap water, and breath. A second objective was to compare levels of VOC in breathing-zone air and drinking water with levels of the same compounds in human breath (Table 1). Selection of sites Two areas were selected for study; a petrochemical manufacturing center in Texas and a nonindustrial com munity in North Carolina. Volunteers from local uni versities were sought. Beaumont, TX, was selected to represent the petro chemical area. Lamar University is bordered on the north and south by oil storage tank farms, and on the northwest by the urban area of Beaumont. For 9 months of the year, prevailing winds from the south cross over 063190 ro L. Wallace, R. Zweidinger, M, Erickson, S. Cooper. D. Whitaker, and E Petlizzari Table 1 Target chemicals in Lamar University and University of North Carolina studies. Chemical Air and Breath Drinking Water 1. Benzene -1 Chloroform 3. 1.2-Dichloroethane 4 1.1,1-Trichloroethane 5. Trichloroethylene 6. Tetrachloroethylene 7. Bromodichloromet hane 8. Chlorobenzene 9. Vinylidene chloride 10. l.l-dichloropropane 11. 1,2-dichloropropane 12. Dibromochloromethane 13. Ethylene dibromide 14. Dichlorobenzene (m~ or p-isomer) 15. o-dichlorobcnzene X X X X X X X X X X X X X X X X X X X X X X X X major refineries and petrochemical plants before reach ing the university. Chapel Hill, NC, was selected to represent the nonindustrial area. Students at the University of North Carolina (UNC) formed the study population. Selection of subjects At both universities, student volunteers were selected only if ihey were not currently enrolled in a course in volving direct contact with organic chemicals, not employed in occupations involving exposure to organic chemicals, and not engaged in hobbies involving poten tial exposure to organic chemicals. A questionnaire developed by the University of Miami Medical School was administered to each student to determine factors that might be related to exposure, such as residence on or off campus, dietary habits, hobbies, parents' occupations, and so on. The question naire had been approved by the human rights committee at the University of Miami Medical School. In all, 17 students were selected: 11 at Lamar Univer sity (five sampled on March 4, 1980, and six on the following day); and six at UNC (three students sampled on June 10,1980, and three on the following day). Each participant signed a consent form and received a small incentive when sampling was completed. samples were returned and breath samples were col- ( lected. A personal sampler employing Tenax GC* polymer to collect organic compounds was used to collect all air samples in the study. The sampler consisted of an MSA Model C-200 pump and an attached Tenax cartridge. Breath samples were collected on Tenax GC cartridges via a specially designed spirometer, as shown in (Fig. 1). Air flow from the pure-air tank was allowed to fill the 50-L Bag A. When the bag was about half full, a clamp and plug were removed from the mouthpiece, the sub ject's nose was clipped closed, and the subject began to breathe through the mouthpiece from Bag A into Bag B. After a few minutes, the transfer of the exhaled breath to the Tenax cartridge was started using the Nutech sampler pump at a flow of approximately 7 L/tnin. (The flow was adjusted to approximate the individual sub ject's respiration rate.) The subject was asked to breathe until the air passing through Bag B was estimated to be about 73 L. Then the subject was asked to stop. Bag B was clamped closed near the mouthpiece, and the re mainder of the contents of Bag B was sampled. The Tenax cartridges were removed and stored in culture tubes. Sample collection -- University of North Carolina During June, 1980, tap water, breath, and air sam ples were collected at UNC. The criteria and methods of sampling were the same as for Lamar University. Six students participated as subjects. Sampling and Analysis Sample collection-Lamar University Each morning air monitors and water sample vials (3 per person) were distributed to each of the Lamar stu dents to be tested that day. Participants carried the monitors for a 5-9-h period while they attended classes, ate lunch, commuted, or carried out other normal daily activities. They filled a water sample vial each time they drank. At the end of the day, air monitors and tap water Fig. 1. Schematic diagram of the spirometer. SL 063791 Monitoring individual exposure n Table 2 Percent recovery of selected test substances in air and breath control samples (Lamar University study). Air Breath Compound Benzene Chloroform 1,2-Dichloroethane 1,1,1 -Trichloroeihane Trichloroethylene Tetrachloroethylene Bromodichloromethane Chlorobenzene m-Dichlorobenzene Observed (ng/cart.) X x S.D. (C.V.) 570 * 177 (31) 270 * 91 (33) 750 x 285 (38) 505 x 175 (35) 606 x 155 (26) 694 x 142 (21) 286 x 90 (32) 255 * 49 (19) 567 x 174 (31) Actual (ng/cart.) 528 200 581 368 605 536 227 219 490 Percent 108 x 33 135 x 45 129 * 49 137 x 47 100 x 26 129 x 26 126 x 40 116 x 22 116 x 35 Observed (ng/cart.) X x S.D. (C.V.) 643 x 265 (41) 219 x 74 (34) 707 x 243 (34) 501 x 115 (23) 615 x 140(23) 572 x 171 (30) 249 x 68 (27) 218 x 90(41) 499 x 124 (25) Percent 122 x 50 109 x 37 122 x 41 136 x 31 102 x 23 107 x 32 110 x 30 99 x 41 102 x 25 Analysis procedures Air and breath. Analysis of the Tenax cartridges was performed by a thermal desorption GC-MS procedure described fully in previous publications (Pellizzari, 1977; Pellizzari, 1979). Water. Tap water samples were analyzed by a purge and trap method (Bellar and Lichtenberg, 1974). Both a Hall Electrolytic Conductivity Detector and a flame ioniza tion detector were operated simultaneously to detea both the halogenated compounds and benzene. Stan dards, blanks, and controls were interspersed through out the analysis period. The standards were prepared fresh daily and transferred to smaller containers which were stored in the refrigerator until used. Two tap water samples were selected to be purged onto Tenax GC cartridges for broad spearum analysis by GC-MS. The selection was determined by the number and amount of compounds detected by the purge and trap analysis. Quality control The blanks and controls were prepared 1 day prior to the sampling trips. Lab blanks and controls remained refrigerated in the laboratory during the trip; field blanks and controls were transported to and from the field alongside the samples at ambient temperatures. The air controls were spiked with the following seven compounds via a permeation system; chloroform. 1,2-dichloroethane, 1,1,1-trichloroethane, carbon tetra chloride, trichloroethylene, bromodichloromethane, and benzene. The breath blanks were run on the spirometer with the mouthpiece plugged and the pure air forced through the valves from Bag A to Bag B and pumped through a blank Tenax cartridge. The controls were collected in the same manner except each cartridge was spiked with about 500 ng of each of the compounds listed above. Water blanks were prepared from deionized water. Controls were spiked to give a concentration of 10 ng/mL of each of the following nine compounds: chlo roform, 1,2-dichloroethane, 1,1,1-trichloroethane, carbon tetrachloride, trichloroethylene, tetrachloroethylene, chlorobenzene, o-dichlorobenzene, and bromodichloromethane. Results Quality control results Air and breath. Percent recoveries for the air and breath analyses generally ranged between 95 % and 140% (Tables 2 and 3). Coefficients of variance de creased considerably for 1,1,1-trichloroethane and tri chloroethylene between the Lamar and UNC visits, but were unchanged for the other chemicals. Water. The results of the quality control sample analyses are presented in Tables 4 and 5. The peaks for trichloro- Table 3. Percent recovery of selected test substances in air and breath control samples (University of North Carolina study). Air Breath Compound Observed (ng) X x S.D. (C.V.) Actual (ng) Percent Observed (ng) X x S.D. (C.V.) Percent Benzene Chloroform 1.2-Dichloroethane 1,1,1 -Trichloroethane Trichloroethylene Tetrachloroethylene Bromodichloromet hane Carbon tetrachloride 450 x 77 (17) 208 x 59 (28) 548 x 236 (43) 600 x 31 (5) 889 x 56 (6) 852 x 276 (32) 292 x 98 (33) 508 x 39 (8) 530 200 567 455 742 556 225 343 85 x 14 104 x 29 97 x 42 132 x 7 119 x 7 153 x 50 130 x 43 148 x 11 501 x 291 (58) 644 x 74(11) 456 x 143 (31) 828 x 41 (5) -x. 484 x 145 (30) 94 x 55 _ 113 x 13 100 x 31 111 x 6 _ 141 x 42 Si 537g2 2': L, Wallace. R. Zweidinger. M Erickson, S Cooper. D Whitaaer, ana E' Peihzz.,-' Table 4. Quanrines of larget compounds recovered in lap water blanks and control for Lamar University studs (ng ml). Sample CF* DE MCF CT BCM TCE PERC CB NX ater blank (7,1 Field blank Lab blank Field control Lab control 1.2 X 2 0 3 x 0.1 30 X 4 94 x 4 (94) c 11 X 4 (109) _b 8.3 X 0,5 (83) V3l (93) 3 9 2 (37) 4.6 X 0.7 (46) _ 3 4 x 0.1 (34) 0.5 7.6 5.J (55) 5.8 x 0.1 (58) - NO NO - 2.4 0.2 (26) 2.6 * 0 1 (25) - 7.5 x 0,2 (76) 7 5 *1 (75) *CF * chloroform. DE 1,2-dtchloroethane; MCF = l.l.l-irichloroethane; CT = carbon chloroeihyJene or 1.1,2-tnchloroethane; PERC = tetrachloroethylene; CB * chlorobenzene, k- indicates not deiecied, `Mean * S.D. (percent recovered). tetrachloride; 8CM bromodichloromethane; ICE iri- ethylene and 1,1,2-trichloroethane overlap, so it was not possible to determine whether one or both compounds were present, nor to quantitate either compound. The percent recoveries given in Table 4 indicate that results for drinking water control samples in the Lamar study varied widely, from 25% for tetrachloroethylene to > 100% for chloroform. Adjustments in the analytical protocol led to the greatly improved percent recoveries observed at Chapel Hill three months later (Table 5). Mean recoveries ranged between 92% and 118% for the seven spiked compounds. The precision (i.e., relative standard deviation, or coefficient of variation) ranged from 4% to 12% for the UNC study. In addition to the quality control samples, three blind quality assurance samples were prepared for water. These samples were encoded prior to submission to the analyst. The results, presented in Table 6, indicate good recoveries for chloroform and chlorobenzene; moderate recoveries for carbon tetrachloride; but only 50% re covery of tetrachloroethylene. Recoveries of 1,1,1-trichloroethane. on the other hand, were consistently 40%-50% greater than the spiked value. Field results Air. Of 15 target compounds, six were found in all 17 samples, and four others in more than one-half of the samples (Tables 7 and 8). Six of these 10 compounds showed high variability, ranging over 2-3 orders of magnitude. Geometric means for one compound (1,1,1-trichloroethane) exceeded 50 jig/m1 in each stu dent group. Geometric means for seven other com pounds generally fell between 1 and 10 for each group. The mean concentrations showed no significant difference between the two student groups for any com pound, as determined by a Mest. Breath. Five compounds were found in all 17 samples, and two others were found in more than 50% of the samples (Tables 9 and 10). Five of these seven com pounds showed high variability. Geometric means for these seven compounds ranged from about 1 to 15 Atg/m1. Water. UNC tap water showed consistently higher mean chloroform values (220 ng/mL) than the Lamar Univer sity sources (150 ng/mL) (Tables 11 and 12). Tetrachloroethylene values were also higher in the UNC water supplies. Bromodichloromethane values were similar in the two water supplies (20 ng/mL at Lamar; 17 ng/mL at UNC). Total trihalomethanes exceeded the recom mended interim drinking water standard of 100 ng/mL in all 38 water samples from the two areas. (A substan tial portion of surface supplies from around the nation exceeded this value in the late 1970's.) All of the tap water samples contained chloroform and bromodichloromethane. Some samples contained small amounts of tetrachloroethylene and chloro benzene. No benzene, carbon tetrachloride, 1,2-dichloroethane, vinylidene chloride, or 1,1,1-trichloro ethane was detected. Summary statistics The frequency of detection, arithmetic mean, stan dard deviation, and range of each target compound in air, breath, and water are displayed in Tables 13-16. Each statistic is computed using the actual or estimated values for all participants except where missing data oc cur. The percent detected indicates the percentage of measurements greater than the minimum limit of detec tion (LOD). The median and the arithmetic mean are provided as measures of central tendency. The mean is calculated using the following conventions: the quan- Table 5. Quantities of target compounds recovered in tap water blanks and controls for UNC study (ng/mL). Sample Field blink Lab bleak Field control Lab conlrol CF* Tb 1.3 x 0.5 11 X 1.4(8)C 9.8 x 0 3 (88) DE MCF BCM TCF 10 x 0.9(100) 9 2 x 0.1 (92) 12 x 1 (118) 10.8 x 0.1 (107) 10 x 1 (102) 11 x 0(106) 9.5 x 0 (95) 10.4 x 0.2(104) PERC CB 12.5 x 0.2(125) 12 x 1 (107) 10 x 0.5 (102) 10 x 0 5 (100) Tabic 4 for codes. * Tract, cMean * S.D (percent recovered). I SL 063793 Monitoring individual exposure Table 6. Results of water blind study.* Blank Spike - 1 Spike - 2 Compound Spike (ng/mL) Found (ng/mL) Spike (ng/mL) Found (ng/mL) Recovery (*) Spike (ng/mL) Found (ng/mL) Benzene Chloroform 1,2-Dichloroethane 1.1,1-Trichloroethane Carbon tetrachloride Bromodichloromethane Trichloroethylene and/or 1,1,2-Trichloroethane Tetrachloroethvlene Chlorobenzene 0 0 0 0 0 0 0 0 0 0 NDb ND ND ND ND ND ND ND ND ND 0 13.5 6 14.1 16.7 0 0 0 17.0 11.6 ND 15.6 ND 20 13.4 ND ND 9.2 12.4 0 ND 100 1.5 1.7 0 ND 142 134 202 80 ND ND 0 ND 0 0.02 0 ND 54 0 ND 107 0 ND 'Samples prepared with known levels of the compounds as indicated and encoded prior to submission for analysis. "Not detected. '3 Recovery W 116 151 Table 7 Estimated levels of selected vapor-phase organics in ambient air associated with human participants (Lamar University student study) (gg/m1). Participant No. Compound 30001 30002 30003 30004 30005 30011 30012 30013 30014 30015 30016 LOD* Benzene Chloroform Vinvlidene chloride 1,1-Dichloroethane 1,2-Dichloroethane 1,1,1-Trichloroethane Trichloroethylene 1,2-Dichloropropane Tetrachloroethvlene Bromodichloromethane Dibromochloromethane Ethylene dibromide Chlorobenzene Dichlorobenzene isomer o- Dichlorobenzene 9.3 1.4 416 1.8 11 592 19 * 174 1.6 8.4 - 2.5 1.5 1.4 0.49 22 7.5 161 1.5 73 . 11 3.8 76 0.93 1.069 26 7.2 -- 8.0 2.9 8.3 1,0 -- 0.32 8.5 1.6 5.4 3.2 ~ 6.9 2.4 3.6 5.2 1.1 -- 0.52 8.3 0.90 - 5.6 1.0 0.47 2.5 0.38 4.8 5.8 4.0 4.8 * 7.0 --- 10 . 0.94 31 62 3.8 2.0 -- 5.0 30 -- 1.1 ---- ---- 6.4 3.0 - 0.20 8.3 6.0 5.7 -- 0.95 72 63 -- 718 3.7 -- -- * 23 - 3.2 3.2 2.1 - 0.72 12 0.99 -- 4.5 0.84 -- -- -- 1.8 - 5.3 1.9 4.6 0.71 67 2.4 -- 172 -- -- -- -- 4.3 - 'Limit of detection (LOD) was defined as S/N * 4 for m/z ion selected for quantification, all values in (ig/m\ ^Quantifiable limit (QL) was defined as 5 x LOD or S/N 20. all values in ng/m\ 386 4.8 - -- 13 40 3.7 9.3 -- -- -- 2.1 33 - 0.08 0.08 0.12 0.12 0.12 0.16 0.16 0.20 0.24 0.24 0.24 0.28 0.16 0.20 0.20 QLb 0.40 0.40 0.60 0.60 0.60 0.80 0.80 1.00 1.20 1.20 1.20 1.40 0.80 1.00 1.00 Table 8. Estimated levels of vapor-phase organics in ambient air for several human subjects (University of North Carolina at Chapel Hill study) (ng/m*). Compound Benzene Chloroform Vinvlidene chloride 1.1-Dichloroethane 1,2-Dichloroethane 1,1.1-Trichloroethane Trichloroethylene 1.2-Dichloropropane Tetrachloroethylene Bromodichloromethane Dibromochloromethane Ethylene dibromide Chlorobenzene Dichlorobenzene isomer o-Dichlorobenzene *-- indicates not detected. 40001 14 7.8 14 -- 165 9.7 - 2.5 -- * - 35 1.5 40002 7.5 5.1 27 -- - 194 2.2 - 2.6 -- - - 0.58 0.32 Participant Number 40003 40011 3.8 3.7 9.8 -- 1.1 93 10.8 - 2.1 -- - - - 0.46 - 3.2 3.2 3.5 0.42 14 4.6 - 1.2 * -- 0.17 0.29 - 40012 4.2 17 5.7 _ 0.45 70 2.2 _ 4,3 4.3 _ _ 0.18 15 0.27 40013 3.0 2.2 7.0 0.63 57 183 127 -- ,, __ _ 0.63 0,14 06379^ SL 274 L. Wallace, R. ZweidingeT, M. Erickson, S. Cooper. D Whitaker, and E Pellizzari Table 9 Estimated levels of selected vapor phase organics in breath (Lamar University student study) (<ig/mJ). Participant No. Compound 30001 30002 30003 30004 30005 30011 30012 30013 30014 30015 30016 Benzene Chloroform Vinylidene chloride 1,1-Dichloroethane 1,2-Dichloroethane 1,1,1-Trichloroethane Trichloroethylene 1,2-Dichloropropane Tetrachloroethylene Bromodichloromeihane Dibromochloromethane Ethylene dibromide Chlorobenzene Dichlorobenzene isomer o Dichlorobenzene 2.9 * 1.4 * 1.1 0.2 T* T 15 0.08 2.8 --- 161 * 16 -- 0.66* 0.16 1.11* 0.04 -- 69 96 5 4 0.20 -- --- *- T 31 * 2.5 -- 0.7* 0.0 T 26 8.0 -- 93 * 21 T - 98 * 1.0 T - 1.71 * 0.28 T 0.08 0.99 0.33 T 2.9 1.8* 0.13 T 0.08 -- ---- TT T T-T - 13 * 1.5 T 13 0.71 - T -- T T - 2.2 2.48 0.08 2.5 T 24 1.3 - 1.7* 1.1 * 0.15 0.15 TT 0.5 5.8* T 1.6 -T-- T 1.7* 0.46 1.45* T 0.10 -161 * 1.0 31 ---- 20 * 6.1 6.2 -- 1.9* 0.20 T 3.8* 1.2 T 6.5* 0,75 1.07* 0.04 176 * 0.91 23 * 3.0 - 1 3* 0.52 T 0.08 " T T T T - LOD 0.11 0.11 0.16 0.16 0.16 0.22 0.2 0.27 0.33 0.33 0.33 0.38 0.22 0.2? 0.27 *T = trace amount. * -- indicates not detected. QL 0.55 0.55 0.82 0.82 0.82 1.10 1.10 1.37 1.65 1.65 1-65 1.92 1.10 1.37 1.37 tifiabie limit (QL) is defined as 5 x LOD; values of compounds not detected are estimated as 0.5 LOD; and trace levels are estimated as 0.5 (QL + LOD). In gen eral, the sample distributions for water show much less variability and skewness than for air and breath. Discussion The concentrations of some chemicals reached levels of 100-1000 fig/m1 in air and 100-200 jig/m* in human breath. Although the air levels are far below the work place standards of the Occupational Safety and Health Administration, their chronic effects are unknown and could be of significant public health concern. The great variability exhibited by seven of the 10 most prevalent compounds indicates that it would have been erroneous to characterize either student group as a geographical cohort with uniform exposures. If vali dated by future studies, this conclusion would have im portant implications for epidemiological studies, which have often traditionally assigned similar exposure histo ries to residents of a given region. Table 10. Estimated levels of .apor phase organics in breath of human subjects (University of North Carolina at Chapel Hill study) (ng/m'). Participant Number Compound 40001 40002 40003 40011 40012 40013 Benzene Chloroform Vinylidene chloride 1.1-Dichloroethane 1,2-Dichloroethane 1.1,1-Trichloroeihane Trichloroethylene 1,2-Dichloropropane Tetrachloroethylene Bromodichloromeihane Dibromochloromethane Ethylene dibromide Chlorobenzene Dichlorobenzene Isomer o-Dichlorobenzene 1.0 * 0.05 2.8 * 1.5 4.5 * 0.39 - 23 * 3.6 1.1 * 0.12 * 3.4 * 0.44 - - - - 0.54 * ' 03 *NC = missing values. t'- indicates noi detected. 'Given value is below the limit of detection. 1.4 * 0.23 3.0 * 0.16 14 * 1.3 - - 48 Sfe 11 0.55 0.19 3.3 A 0.19 - -- * -- 4. J * 0 - 1.4 * 0.18 5.1 * 2.8 5.5 * 1.3 - 19 * J.7 1.2 * 0.36 4.3 * 0.76 - - -- 0.27 2.2 * 0.56 - 1.1 * 0 1.8 * 0.28 3.9 * 0.09 - - 6.1 * 0.04 0.65 * .05 - 8.8 * 1.1 -- -- -- _ 0.92 * 0 - NC* NC 7.7 * 0.05 * 0.37 8.5 * 1.8 0.49 * 0.11 -- 7.5 * 0.68 0.1 lc 5.3 * 0.71 - 1.5 * 0.41 1.7 0.41 7.9 * 0.65 -- 0.48 13 * 0.34 32 * 0.70 -- 48 * 5.5 _ -- _ _ 1.1 * 0.36 - SL 063795 Monitoring individual exposure 27? Table 11. Quantities of target compounds found in tap water (Lamar University) (ng/mL), Number Chloroform l ,2-Dichloroethane 1,1,1 -Trichloroethane Carbon Tetra chloride and/ or Bromodi- methane Trichloroethylene and/or 1,1,2-Tri- chloromethane Tetraehloroethylene Chlorobenzene 1-30001 1-30002 2-30002 1-30003 1-30004 2-30004 1-30005 2-30005 1-30011 2-30011 3-30011 1-30012 2-30012 3-30012 1-30013 2-30013 3-30013 1-30014 2-30014 3-30014 1-30015 2-30015 1-30016 2-30016 3-30016 LOD 110 260 130 550 160 99 140 120 120 120 110 120 170 no 140 160 130 160 130 110 110 140 120 120 120 1.0 _* - - " - - - - - * -- -- -- - - -- - 0.6 Not detected. bNC = missing data. T = trace amount. 'Given value is below limit of detection. _ 16 NCb -* - 18 NC - * -- 23 NC - -- - 44 NC T * -- 25 NC - -- -- 18 NC - _ -- 25 NC - -- -- 22 NC - -- 20 NC -- * -- 18 NC 0.2 O^ - 23 NC -- -- -- 22 NC -- _ -- 26 NC -- _ -- 17 NC -- _ 18 NC -- _ -- 22 NC 0.1d - 18 NC -- -- 20 NC _ 18 NC _ _ _ 13 NC _ -- 7.4 NC 0.2* 0.2" -- 18 NC -h. _ 13 NC mt. -- 13 NC _ -- 14 NC 0.1d 0.2 0.4 1.3 1.1 0.6 Table 12. Quantities of target compounds found in tap water (Chapel Hill) (ng/mL). Sample 1,2-Dichloro 1,1,1-Trichloro- Bromodichloro- Chloroform ethane cthane methane Trichloro ethylene Tetrachloroethylene 40001-1 40001-2 40002-1 40002-3 40003-1 40003-2 40011-1 40011-2 40011-3 40012-1 40012-2 40013-1 40013-2 Meanh SD CV(7,) Median LOD' 260 260 220 250 200 230 200 210 220 210 210 180 200 220 23 10 220 0.05 ND ND ND ND ND ND ND ND ND ND ND ND ND - - - 0.06 ND* ND ND ND ND ND ND ND ND ND ND ND ND - - - 0.1 20 19 17 18 18 18 17 16 17 17 16 15 15 17 2 12 17 0.1 2.8 3.0 ND ND ND ND ND ND ND ND ND ND 1.3 0.6 1 170 ND 0.05 3.8 3.8 1.8 1.8 1.8 1.8 1.7 1.8 1.8 1.8 1.8 1.3 1.3 2.0 0.8 40 1.8 0.05 *Not detected. bMean of all values (ND * '/i LOD). 'Limit of detection (S/N = 3). Chloro benzene ND ND ND ND 1.4 ND ND ND 1.5 ND ND 1.5 ND 0.4 0.6 150 ND ND St 63?96 116 L. Wallace, R. Zweidmger, M Enckson, S Cooper, D, W/httaker, and E Pellizzar Table 13 Summary statistics for estimated levels of selected vapor-phase organics (Lamar University). Delected* fo at Trace Meanb Std. Dev. Median Range Benzene Chloroform Vmylidene chloride 1,1-Dichloroethane 1,2-Dichloroethane 1,1,1-Trichloroethane Trichloroethylene 1,2-Dtchloropropane Tetrachloroethvlene Bromodichloromethane Dibromochloromethane Ethvlene Dibromide Chlorobenzene Dichlorobenzene Isomer o-Dtchlorobenzene Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath too too 100 100 82 35 18 0 91 18 100 100 100 82 0 0 100 100 64 0 0 0 0 0 18 0 100 100 27 0 0 40.30 118.87 5.16 2.46-386.56 0 1.61 0.62 1.70 0.72-2.95 0 4.00 2.13 4.04 1.39-8.33 91 0.42 0.68 0.22 0.22-2.48 0 46.84 124.43 2.14 0.06-416.07 9 4.88 8.19 0.33 0.08-25.17 0 0.30 0.57 0.06 0.06-1.83 0 0.08 0.00 C.08 0.08-0.08 9 2.72 4.66 0.72 0.06-12.80 18 0.12 0.10 0.08 0.08-0,33 0 180.35 339.47 40.00 8.27-1069.04 45 24.37 53,14 0.66 0.44-161.50 0 11.88 18.90 3.67 0.90-63.39 54 0.59 0.43 0.44 0.11-1.45 0 0.10 0.00 0.10 0.10-0.10 0 0.13 0.00 0.13 0.13-0.13 0 117.63 212.38 9.26 4.54-718.20 18 51.36 65.61 13.25 0.66-176.32 0 1.23 1.24 1.00 0.12-3.71 0 0.16 0.00 0.16 0.16-0.16 0 0.12 0.00 0.12 0.12-0.12 0 0.16 0.00 0.16 0.16-0.16 0 0.14 0.00 0.14 0.14-0.14 0 0.19 0.00 0.19 0.19-0.19 0 0.30 0.08 0.61 0.08-2.12 0 0.11 0.00 0.11 0.11-0.11 0 13.56 21.52 6.95 1.83-73.47 54 7,73 11.31 0.56 0.55-30.67 0 0.34 0.67 0.10 0.10-2.40 0 0,13 0.00 0.13 0.13-0.13 an = 11, rAll compounds measured in ng'm1. 1 z 10 dg/m1 18 0 0 0 18 18 0 0 18 0 82 18 27 0 0 0 45 64 0 0 0 0 0 0 0 0 27 27 0 0 t > Relative contributions of individual VOC to exposure of volunteers Table 17 shows the relative contribution of each of 12 volatile organics to the air exposure of each of the 17 subjects. These "profiles" can be used to compare dif ferent study groups or areas with respect to their domi nant pollutants. For both student groups studied, methyl chloroform is the main contributor, supplying over onehalf the total intake at UNO, and more than one-third at Lamar. The relative importance of benzene, chloroform, and vinvlidene chloride was also very similar for each group, ranging between 4% and 8%. However, tetrachloroethylene was far more important for the Lamar group than the UNC group (35%-7%), while trichloro ethylene was relatively more important at UNC (13%3.3%). Breath-air relationships For certain chemicals, a relationship between en vironmental levels and levels in body fluids on tissues has been established. [As an example, the Coburn equa tion relates carbon monoxide levels in air to carboxyhemoglobin levels in blood (Coburn et al., 1965).] Such a relationship is useful because it allows body burden or dose to be estimated by measuring the environment rather than the person. For other chemicals, however, such a relationship has not been established, at least at environmental levels. Therefore a preliminary attempt was made in this study to determine whether levels in breath correlated with previous exposures in air. It should be noted that the method of measuring breath levels involves inhaling pure air. Thus, whatever trace chemicals are exhaled are being transferred to the breath from other body compartments, particularly the blood. The simplest assumption is that all compartments are in equilibrium with a steady-state environmental con centration. In this case, assuming a constant fraction (1 - f) of the chemical is removed by metabolic processes on each "pass" through the body, the breath concentra tion should be directly related to the concentration in air: C. = f CA. This theoretical relationship was tested by calculating Spearman correlation coefficients between air and breath samples for each student group (Table 18). Since no significant differences were observed between the two student groups with respect to air concentrations, they were combined into a single group (N = 17) and SL 063797 Monitoring individual exposure s 77 Table 14. Summary statistics for estimated levels of selected vapor phase organics (University of North Carolina at Chapel Hill study). % Detected* *h at Trace Meanb Std. Dev. Median Range 1A 10 4g/mJ Benzene Chloroform Vjnylidene chloride 1,1-Dichloroethane 1,2-Dichloroethane 1,1,1 - Trichloroethane Trichloroethylene 1,2-Dichloropropane Tetrachloroethv lene Bromodichloromet hane Dibromochloromethane Ethvlene Dibromide Chlorobenzene Dichlorobenzene Isomer o- Dichlorobenzene Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath Air Breath 100 100 100 10 (/t = 5) 100 100 0 0 67 33 100 100 100 100 0 0 100 100 17 17 0 0 0 0 33 17 100 100 50 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5.89 1.29 6.59 2.86 11.21 7.28 0.06 0.08 0.45 0.19 99.07 19.47 35.33 5.94 0.10 0.13 23.34 12.52 0.83 0.29 0.12 0.16 0.14 0.19 0.11 0.14 8.66 2.42 0.40 0.13 4.14 0.22 5.66 1.36 8.66 3.73 0.00 0.00 0.39 0.18 68.06 15.14 72.16 12.60 0.00 0.00 50.94 17.40 1.73 0.30 0.00 0.00 0.00 0.00 0.05 0.0 14.14 2.01 0.55 0.00 4.00 1.38 4.41 2.84 8.40 6.59 0.06 0.08 0.43 0.08 81.81 15.97 7.13 0.86 0.10 0.13 2.56 5.92 0.12 0.16 0.12 0.16 0.14 0.19 0.08 0.11 0.61 1.53 0.18 0.13 2.95-13.65 1.00-1.51 2.25-17.46 1.70-5.06 3.53-27.29 3.94-14.12 0.06-0.06 0.08-0.08 0.06-1.09 0.08-0.48 14.45-193.77 6.10-47,63 2.17-182.43 0.49-31.66 0.10-0.10 0.13-0.13 1.22-127.30 3.30-47.74 0.12-4.36 0.16-0.91 0.12-0.12 0.16-0.16 0.14-0.14 0.19-0.19 0.08-0.18 0.11-0.27 0.29-34.96 0.54-5.30 0.10-1.52 0.13-0.13 17 0 17 0 34 17 0 0 0 0 100 67 34 17 0 0 17 17 0 0 0 0 0 0 0 0 34 0 0 0 n = 6 unless otherwise noted bAll compounds are measured in iig/m1. studied further for correlations between chemicals in air and breath (Table 19). Four of the five chemicals prevalent in the breath samples showed significant correlations with their concentrations in air; only benzene showed little correlation between breathing-level exposure and breath clearance. Regression analyses were performed relating breath levels to air levels for those compounds with significant air-breath correlations. Logarithms were employed because of the wide range of concentrations. Nondetect- able (ND) samples were assigned a value of one-half the LOD; and trace (T) samples a value halfway between the LOD and the quantifiable limit (QL). For three chemicals, 50Vo or more of the variance in breath levels was explained by the preceding air exposures. The F values are all above the 99Ve level. A simple log-linear model appears capable of predicting breath levels to within a factor of 3 or 4, given the air exposures for the preceding 8 h. This approach yields the following regression parameters: Table IS. Summary statistics for estimated levels of selected vapor phase organics in tap water (Lamar University student study). Chloroform 1,2-Dichloroethane 1,1.1-Trichloroethane Bromodichloromethane and/or Carbon Tetrachloride Trichloroethylene and/ or 1,1,2-Trichloroethane Tetrachloroethylene Chlorobenzene Detected 100 0 0 100 - 9 0 *!* at Trace 0 0 0 0 - 9 0 Mean* 172.38 0.30 0.10 21.21 - 0.95 0.30 Standard Deviation 126.56 0 0 8.38 -- 1.33 0 Median 130.00 0.30 0.10 20.33 Range 117.00-550.00 0.30-0.30 0.10-0.10 13.33-44.00 0.55 0.30 0.55-4.95 0.30-0.30 All compounds are measured in ng/mL." SL 063798 278 L. Wallace, R. Zweidinger, M. Erickson, S. Cooper, D. Whitaker, and E. Pellizzari Table 16. Summary statistics for estimated levels of selected organics in tap water (Chapel Hill). % Detected to at Trace Mean*b Standard Deviation Median Chloroform 1,2-Dlchloroethane 1,1,1-Trichloroethane Bromodichloromethane Trichloroethylene Tetrachloroethylene Chlorobenzene 100 0 0 100 23 100 23 0 220 23 0- 0 -- 0 17 2 0 0.6 1 0 2.0 0.8 0 0.4 0.6 220 17 ND 1.8 ND `All compounds are measured in ng/ml. bMean of all values (ND = 2.00). Range 180-260 ND ND 15-20 ND-3.0 1.3-3.8 ND-1.5 tetrachloroethylene: 1,1,1-trichloroethane: vinylidene chloride: log C, = 0.23 0.44 + (0.72 0.13) log C^, log Cm = -0.98 0.54 + (0.91 0.23) log C,, log C,, -0.24 * 0.67 + (0.71 0.17) log Ca, where C, = concentration in breath, CA = concentra tion in air, in /ig/mJ. Figures 2-4 display this possible log-linear relation- ship between air exposures and breath concentrations of several compounds. If this preliminary observation is confirmed by future studies, an exposure-body-burden relationship could be established for some compounds. This would allow recent exposures to be estimated from a single noninvasive test lasting just 15 min; conversely, body burden could be estimated from a single personal monitoring sample. Two possible problems exist in attempting to relate expired air concentrations to a previous exposure aver- Table 17. Percent of individual air exposure supplied by selected vapor-phase orfanics (Both groups). Participant Lamar University 30001 30002 30003 30004 30005 30011 30012 30013 30014 30015 30016 Mean Standard Deviation 44 m 1 M iS I s e aS o xU > 3 8 Iu 8 s i i cIt & 1 i u 5 5 E6 0.76 0.86 0.86 7.1 8.8 8.6 4.7 0.93 11.0 2,0 0.11 0.51 0.30 20 13 7.1 3.9 0.67 11.0 0.7 6.1 34 0.48 6.0 2.4 27 * 5.7 0.67 7.2 1.7 - 0.14 - 0.07 - - - - -- - 0.88 0.17 -- 0.73 1.2 1.8 0.7 0.11 2.4 0.26 16 48 8.0 85 21 20 55 51 8.0 41 25 51 1.5 14 0.13 - 2.6 59 0.52 - 2.1 0.J7 - - 3.9 13 7.8 - 2.2 14 2.4 1.2 6.8 87 -- 1.6 25 0.90 - 7.0 80 3.4 16 0.40 2.8 -- -- 0.9 67 -- -- 4.7 12 - 2.7 0.67 27 0.63 17 6.0 11 2.5 2.5 6.2 1.6 42 * - - 5.9 1.0 0.17 - - 4,7 5.8 7.8 -- 2.2 38 3.8 6.2 11.0 4.4 22 3.5 35 2.0 30 1.4 2.3 _ 11 12 0.6 - UNC 40001 40002 40003 40011 40012 40013 Mean Standard Deviation 5.6 3.1 5.6 - - 66 3.9 1 -- -- 14 0.6 3.1 2.1 11 - - 81 0.9 1.1 -- -- 0.2 0.1 3.0 3.0 7.8 - 0.88 74 8.8 1.7 -- -- 0.4 0.1 11 11 12 - 2.4 9.6 3.2 -- 1.3 47 0.23 40 IS 4.0 -- 0.7 1.2 2.4 2.5 0.1 1.0 0.3 8.5 0.2 0.8 0.6 1.8 - 0.16 15 48 34 - - 0.16 0.03 4.3 4.9 6. _ 0.43 54 13 7.4 0.4 0.1 4.0 0.2 *3.3 3.9 3.8 - 0.49 23 16 12 - - 5.3 0.2 Outlier-not included in calculations. SL 063799 Monitoring individual exposure 279 Table 18. Spearman correlation coefficients between air and breath for estimated levels of selected vapor phase organics (both groups). n UNC Lamar **Air* breath UNC Lamar Benzene 5 11 0.70 0.04 Chloroform J 11 0.60 0.20 Vinyiidene chloride 6 11 0.48 0.67* 1,1-Dichloroeihane 6 11 0.00 0.00 1,2-Dichloroethane 6 11 0.44 0.07 1,1,1-Triehloroethane 6 11 0.94 0.63* Trichloroethylene 6 11 0.94b 0.41 1,2-Dichloropropane 6 11 0.00 0.00 Tetrachloroeth ylene 6 11 0.20 0.80b Bromodichloromethane 6 11 0.20 0.00 Dibromochloromethane 6 11 0.00 0.00 Ethylene dibromide 6 11 0.00 0.00 Chlorobenzene 6 11 0.31 0.00 Dichlorobenzene isomer 6 11 0.03 0.08 o-Dichlorobenzene 6 11 0.00 0.00 p < 0.01. V < 0.05. AilIMM a*- * m TM Hi-- II fee*** Fig. 2. Correlation between air exposures and breath concentrations for tetrachloroethylene. R` " 0.68. aged over 8 h. First, since most of these volatile chem icals have two or more distinct halflives associated with different body compartments (the shorter of which is often measured in minutes) (Andersen et at., 1980) the breath level immediately following exposure is likely to reflect the most recent few minutes of the exposure. Stewart et at. (1970) found, for example, that not until the third hour postexposure to trichloroethylene was the concentration in expired breath able to be directly related to the time-weighted average exposure. Thus a sharp increase or decrease in exposure toward the end of the exposure period is likely to yield a breath level well above or below the value expected on the basis of the mean exposure level over 8 h. Second, it is possible that an unusually high ex posure previous to the measured exposure period would continue to show some effects on expired air concentra tions measured 5-9 h later. This is unlikely to be a prob lem for those chemicals with short halflives, since the ef fect would be "washed out" during the exposure period. (For example, the halflife of 1,1,1-trichloroethane associated with the vessel-rich group is only 0.8 h) (Humbert and Fernandez, 1977). Even for chemicals Fig. 3. Correlation between air exposures and breath concentrations for vinyiidene chloride. R* * 0.33. with longer halflives, such as tetrachloroethylene. Hake et at. (1976) have stated that breath samples taken 8 h after exposure were indicative of the amount adsorbed. Also, only in the case where the previous exposure was higher than the measured exposure would there be a noticeable effect on the expired air concentration, yet for all of these subjects, their previous exposure was during the sleep period, an unlikely time for severe ex posures. Table 19. Significant Spearman correlation coefficients for volatile organic compounds observed in breathing-zone air and in exhaled breath of 17 students at Lamar University and University of North Carolina. Breath Air Benzene Methyl Chloroform Tetrachloroethylene Trichloroethylene Vinyiidene Chloride Benzene Methyl chloroform Tetrachloroethylene Trichloroethylene Vinyiidene chloride Ethylene dichlonde 0.54 0.54 0.74 0.88 0.73 0.33 0.62 0.77 Si 63So0 280 L. Wallace, R. Zweidinser, M, Erickson, S. Cooper, D, WhnakcT, and , Ptlliiian 1 t r>lm ini *> i>MH fcw* t^WH Maa> timi-n cm< CmWW" Fig. 4. Correlation between air exposures and breath concentrations for 1,1,1-tnchloroethane. R1 m 0.68. Other possible confounding variables include biolog ical variation in metabolic rates and short-term altera tion of metabolic processes by medication. It should be noted that if these hypothesized variations in exposures before the beginning or toward the end of the measured exposure period were in fact occurring, the result would reduce or eliminate any correlation between exposures and breath levels; however, four significant correlations were observed. Breath-air ratios Table 20 compares the breath-to-air ratios of the con centrations of seven VOC for each of the 17 student volunteers. Breath-air ratios among the two student groups are similar for benzene (307o), vinylidene chlo ride (80Vb), and trichloroethylene (20V0), For three other compounds, however, (1,1,1-trichloroethane, tetrachloroethylene, and dichlorobenzene isomer), the UNC mean breath-air ratios are several times higher than those of Lamar. Moreover, the UNC breath-air ratio is higher than 100Vo for two of these compounds (tetrachloroethylene and dichlorobenzene isomer). This anomaly is most likely due to the high variance associ ated with the sampling and analytical protocol, as well as the small number of subjects. In the case of chloro form, the considerable difference between UNC and Lamar breath levels is likely to be the result of the higher chloroform levels in UNC drinking water. Estimated total daily intake Drinking water was an important contributor to total intake (air and water) for only two of the seven com pounds measured in tap water samples (Table 21). Assuming daily intakes of 10 m1 of air and 1 L of drink ing water, the water accounted for 79V# of the chloro form intake and 76Vo of the bromodichloromethane in- Table 20. Breath/air ratios' for selected volatile organics for student volunteers from two geographical areas. Participant Number Benzene Chloroform Vinylidene Chloride 1,1,1-Trichloroethane Trichloro ethylene Tctrachloroethylene Lamar University 1 2 3 4 5 6 7 8 9 10 11 0.3 0.6 0.07 0.6 0.3 0.4 0.4 0.2 0.3 0,4 - 0.2 0.2b 0.1b 0.04b 0.06b 0.08b 0.J 0.0Sb 0.lb 0.2b 0.07b 0.04 0.06c 0.3 0.08c 2.5 - 0.01c o.oyh 2.7 0.8 - 0.3 0.03 0.09 0,07b 0.07b 0,02b 0.04 0.01b 0,1 0.1 0.02b 0.01c 0.2 0.03" 0.4b 0.1' 0.2b 0.3b 0.02 0.7b 0.4 0.2b 0.4 0.6 1.4 2.4 2.4 0.2b 0.8 0.2 0.2 1.0 0.1b m-Dichloro benzene 0.1b 0.4 0.1b 0.1b 0,3b 0.1b 0.4 0.8 3.3 5.4 0.03b UNC 1 2 3 4 5 6 0.07 0.2 0.4 0.3 0.5 0.4 0.6 1.4 0.6 - 0.8 0.3 0.5 0.6 1.1 1.4 1.1 0.1 0.1 1.4 0.02b 0.2 0.2 1.2 7.8 0.2 0.1 2.1 4.7 0.4 0.1 7.2 3.1b 0.1 0.2 1.7 3.5 0.2 0.2 0.4 1.7b Arithmetic Mean ( Standard Deviation) Lamar University UNC Combined 0.31 0.18 0.30 0.17 0.31 0.17 0.15 0.14 0.72 0.38 0.33 0.35 0.74 1.1 0.83 * 0.42 0.78 0.86 Calculated using T = (LOD QL); ND = LOD. ''Breath value = trace or below quantifiable limit. 'Breath value below limit of detection. 0.08 0.07 0.23 0.)l 0.13 0.11 0.23 0.21 0.17 0.06 0.21 0.17 0.88 0.83 2.4 2.5 1.4 1.7 1.0 1.7 3.5 2.6 1.9 2.4 SL 063801 * i Monitoring individual exposure 281 Table 21. Estimated daily intake* of selected compounds from water compared to air (g/day). Chloroform Bromodichloromethane Tetrachloroethylene Subject Air Water Percent Intake from Air Air Water Percent Intake from Air Percent Intake Air Water from Air 1 13.9 185 2 15.2 130 3 28 3 550 4 83.3 130 5 52.2 130 6 40.4 117 7 48.4 133 8 60.2 143 9 31.7 133 10 18.6 125 11 48.2 120 12 78.1 255 13 50.9 235 14 37.4 215 15 31.9 210 16 175 210 17 22.5 190 7 16.3 17 10 15.2 23 5 NDb 44 39 32.1 22 29 10.0 24 26 ND 20 27 11.4 22 29 37.1 19 19 8.4 17 13 ND 13 29 ND 13 23 ND 20 18 ND 18 15 ND 18 13 ND 17 45 43.6 16 11 ND 15 49 1750 ND* 100 40 1620 ND 100 3 72,1 T 95 59 54.1 ND 98 29 56.5 ND 98 6 49.8 0.07 100 34 301 ND 100 66 7180 0.03 J00 33 45.4 ND 98 9 1720 0.01 100 9 92.6 0.03 100 6 24.8 3.8 86 7 26.5 1.8 94 7 20.6 1.8 92 7 12.2 1.8 87 73 43.2 1.8 96 8 1270 1.3 100 'Assuming 10 mVday respiration rate and 1 L/day ingestion rate. bND - <1.2 ag/lOm1. 'ND * <1,1 fig/in take. By contrast, drinking water contributed only 7% of the combined (air + water) intake of tetrachloroethylene for the UNC students, and even less for the Lamar students. Each student's estimated daily intake of all the target compounds from air and drinking water is listed in Table 22. The air values range from 0.3 to 12.4 mg/day, with a geometric mean of 1.6 mg/day and a geometric standard deviation of 3.5. The corresponding geometric mean for the water intake was 0.2 mg/day. Summary and Conclusions This report documents the first field effort of a con tinuing exposure monitoring program at the United States Environmental Protection Agency. Sampling equipment and analytical protocols were tested on 17 subjects at two universities. The sampling equipment (personal monitors and a specially designed spirometer) and the analytical protocols worked well for the air, breath, and tap water samples. These results indicate that the concept of making simultaneous direct measurements of individual human exposure to a significant number of volatile organic compounds is feasible. This first effort has resulted in several interesting findings, particularly the wide vari ability of exposures among a homogenous group of sub jects, and the apparent relationship between inhaled and exhaled concentrations of several compounds. These findings could not have been made using standard ap proaches of ambient monitoring. Additional studies (Pellizzari et at., 1980; Pellizzari et Table 22. Estimated daily intake* of 10 volatile organic compounds through air and water for 17 subjects. Subject Air pg/day Water Total Percent from Air 30001 30002 30003 30004 30005 30011 30012 30013 30014 30015 30016 40001 40002 40003 40011 40012 40013 12,400 2,700 12,000 400 300 550 1,250 9,000 300 2,600 1,140 2.470 2.390 1,250 300 1,240 3,800 200 12,600 150 2,850 600 12,600 150 550 150 450 140 690 160 1,410 160 9,160 150 450 140 2,740 130 1,270 280 2,750 260 2.650 240 1,490 240 540 230 1,470 210 4,010 98 95 95 73 67 80 89 98 67 95 90 90 90 84 56 84 95 'Assuming 10 m'/day and 1 L/day intake rates for air and water. o/., 1981) have been undertaken to extend the personal monitoring approach discussed here to a statistically valid sample of several hundred people in an industrial community. References Andersen, M. E., Cargas, M. L., Jones, R. A., and Jenkins. L. J., (1980) Determination of the kinetic constants for metabolism of SL 063802 28: L Wallace, R Zweidinger, M Erickson, S Cooper, D Whitaker. and E Pellizzari inhaled toxicant' in mvo using gas uptake measurements, Toxicol. Appl Pharmacol. 54. 100-116. Bellar. T ^ and .I, Lichtenberg (1924) Determining volatile organics at microgram-per-litre levels bv gas chromatography, J. Am Hater Works 4ssoc 66, "'39-"44 Coburn. R. F.. Forster, R. E , and Kane, J (1965) Consideration of the ph'Siologv and variables that determine the blood carboxyhemoelobin concentration m man, J Chn. Invest. 44. 1899-1910. Hake. C. L.. Stewart. R D.. Wu. A., and Graff, S. A. (1976) Experi mental human exposure to perchloroethylene, Toxicol. Appl. Pharmacol 37, P5, Humbert, B E and Fernandez, J. G., (1977) Exposure to 1,1,1-tn- chloroethane contribution to the study of absorption, excretion, and metabolism in human subjects, Arch. Mai. Prof. 3*. 415-425. Pellizzari. E, D. (1977) The Measurement of Carcinogenic Vapors in Ambient Atmospheres. L.S. Environmental Protection Agency, Research Triangle Park, NC. Pellizzari. E. D. (1979) Analysis of Organic Air Pollutants by Gas Chromatography and Mass Spectroscopy. U.S. Environmental Protection Agencv, Research Triangle Park, NC. Pellizzari, E D,, Erickson, M D., and Zweidinger R. (1979) For mulation of a Preliminary Assessment of Hatogenated Organic Compounds in Man and Environmental Media L .S. Env ironmental Protection Agency, Washington, DC Pellizzari, E. D., Erickson, M. D., Sparacino, C M,. Hartwell. T D , Zelon, H., Rosenzweig. M., and Leimnger. C. (1981) Total Ex posure Assessment Methodology /TEAM) Phase II Work Plan. Vols. Mil. U.S. Environmental Protection Agencv, Washington, DC. Pellizzari, E. D., Hartwell, T., Zelon, H., Leimnger, C., Erickson, M,, and Sparacino, C. (1980) Total Exposure Assessment Methodology (TEAM): Phase I Work Plan, Vols. I. II. U.S. Environmental Protection Agency, Washington, DC. Stewart, R. D., Dodd. H. C,, Gay, H. H., and Erly. D. S., (1970) Experimental human exposure to trichloroethylene, Arch Env Health 20, 64-71, Zweidinger, R. A. el al. (1980) Measurement ofBenzene Body Burden of Potentially Environmentally Exposed Individuals. U.S. En vironmental Protection Agency, Washington DC. 06383