Environ. Sci. Technol. 2008, 42, 1324–1331
VOC and Particulate Emissions from Commercial Cigarettes: Analysis of 2,5-DMF as an ETS Tracer SIMONE M. CHARLES,† CHUNRONG JIA, STUART A. BATTERMAN,* AND CHRISTOPHER GODWIN Environmental Health Sciences, University of Michigan, Ann Arbor, Michigan 48109-2029
Received August 17, 2007. Revised manuscript received November 4, 2007. Accepted November 13, 2007.
Emissions of particulate matter (PM) and a broad suite of target volatile organic compounds (VOCs) in total, main-stream (MS) and side-stream (SS) smoke emissions are measured for six types of commercial cigarettes. The suitability of 2,5dimethyl furan (DMF) as a tracer for environmental tobacco smoke (ETS) is investigated using laboratory results and a field study of 47 residences. Over 30 VOCs were characterized in cigarette smoke, including several that have not been reported previously. “regular tar”, “low tar”, menthol, and nonmenthol cigarettes showed only minor differences in PM and VOC emissions. When total emissions are considered, PM emissions averaged 18 ( 2 mg cigarette-1 and VOC emissions averaged 3 644 ( 160 mg cigarette-1. DMF appears to satisfy all requirements for a tracer, namely, uniqueness, detectability, similar emission factors across tobacco products (211 ( 16 µg cigarette-1), consistent proportions to other ETS compounds, and behavior similar to other ETS components in relevant environments. On the basis of field study results, DMF more reliably indicated smoking status than occupant-completed questionnaires, and cigarette smoking was responsible for significant fractions of benzene (50%), styrene (49%), and other VOCs in the smoker’s homes.
Introduction Environmental tobacco smoke (ETS) primarily arises from the burning tip of the tobacco product between puffs, the so-called side-stream (SS) smoke, and the smoke exhaled by the smoker that has been puffed through the cigarette, the main-stream (MS) smoke (1, 2). Despite increasingly restrictive policies against smoking, ETS remains an important indoor air pollutant (3–5). ETS exposure is commonly assessed using questionnaires, measurement of airborne contaminants (including tracers), and measurement of biological markers (2). The use of airborne tracers has several advantages, which include facilitating a true risk estimation and exposure measurement and allowing an assessment of the effectiveness of exposure controls (3). Effective ETS tracers must satisfy five criteria (6): uniqueness, easily detected at low smoking rates, similar emission factors across different tobacco products, consistent proportions to other ETS compounds for different environ* Corresponding author e-mail:
[email protected]; phone: +1 734/763-2417; fax: +1 734 /764-9424. † Current address: Jiann-Ping Hsu College of Public Health, Georgia Southern University, Statesboro, GA. 1324
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 4, 2008
ments and tobacco products, and dynamic behavior in relevant environments similar to other ETS components (2). The traditional ETS tracers meet some of these criteria. For example, gas-phase nicotine is relatively easy and inexpensive to measure and unique to tobacco smoke; however, its decay pattern differs from those of many other ETS constituents (2, 7, 8), and thus, nicotine’s relationship to other ETS markers varies considerably (9). Solanesol, an alcohol in the tobacco leaf, is unique to tobacco smoke, but quantitation at low concentrations may be difficult (2), and concentrations decay under intense UV light (10). Another gas-phase tracer, 3-ethenylpyridine (3-EP), is also unique to tobacco smoke and has been correlated to smoking activity (11), however, it adsorbs to surfaces, although to a lesser extent than nicotine (9, 12). This study investigates the suitability of 2,5-dimethyl furan (DMF) as an ETS tracer. Laboratory tests are used to measure emission factors for DMF, other volatile organic compounds (VOCs), and particulate matter (PM) in total, SS, and MS smoke. We evaluate the variability of emissions across six types of commercial cigarettes, including “regular tar”, “low tar”, menthol, and nonmenthol cigarettes. A field study of 47 homes is used to evaluate the tracer.
Experimental Section VOC and PM emissions in total, side-stream, and mainstream smoke were measured for six types of commercial cigarettes. Four brands were marketed as “regular” tar/ nicotine content, two as “low” tar/nicotine, and half were mentholated. They are designated as follows: B1-RT-M ) Kool/regular tar/mentholated; B2-RT ) Marlboro/regular tar; B3-RT-M ) Newport/regular tar/mentholated; B4-RT ) Camel/regular tar; B1-LT-M ) Kool/low tar/mentholated; B2-LT ) Marlboro/low tar. The within-cigarette precision was established, after which emissions were measured for each type in duplicate experimental runs (except for B1RT-M, where n ) 3). Total emissions were measured for each cigarette type. SS and MS emissions were measured separately for one mentholated (B1-RT-M) and one nonmentholated (B2-RT) type, both regular tar/nicotine. Results were compared to measurements of research cigarettes that used the same measurement system (13). Measurement Procedures. Laboratory protocols have been detailed elsewhere (13). Cigarettes were purchased locally and smoked according to the Federal Trade Commission method, which follows the ISO standard (14). Integrated PM samples were collected over the entire cigarette burn period on glass fiber filters, while VOCs used 10 mL samples collected over a 2-min period near the middle of the ∼10-min burn period. VOCs were collected on thermal sampling/desorption tubes packed with Tenax GR (Scientific Instrument Services, Inc., Ringoes, NJ) and Carbosieve SIII (Supelco, Bellefonte, PA), and the samples were analyzed by GC/MS for 99 target VOCs, as well as for nicotine and 3-EP. Conditioning, storage, transportation, analysis, and quality assurance/quality control procedures have been detailed elsewhere (15–17). Field Study. VOC concentrations were monitored inside and outside of 47 residences in Dearborn, MI, in fall 2004. These residences were distributed over the city, and most were attached single family homes without attached garages. Eligible participants, adults living in the residences during the study period who promised cooperation with all aspects of the study, represent a subset of a larger study selected using random digit dialing and snowball recruitment meth10.1021/es072062w CCC: $40.75
2008 American Chemical Society
Published on Web 01/15/2008
VOL. 42, NO. 4, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1325
3.90 1.65 3.16 0.94 1.18
8.04 2.68
1.02
10.72 8.04
40–134
NA NA
alkanes n-heptane n-octane n-nonane n-decane n-undecane
alkenes 1-octene 1-nonene
terpenes limonene
carbonyls benzaldehyde acetophenone
phenols phenolb
sum of target VOCs PMc
3880.6 21.5
50.1
91.3 71.1
509.5
56.7 43.9
30.2 19.6 21.1 12.4 7.2
437.1 813.8 117.2 365.1 74.8 7.6 15.5 31.2 63.7 21.9 13.6 49.7 43.9 5.9 138.0 14.6
(1.6) (0.0)
(2.6) (0.2) (3.7) (1.6) (1.7)
48.7
86.5 61.1 (2.3)
(1.0) (1.5)
537.8 (51.6)
48.5 34.4
23.9 17.6 20.1 11.6 6.6
406.9 (0.8) 775.5 (0.8) 107.7 (4.9) 370.3 (18.8) 69.7 (2.1) 6.8 (0.3) 14.0 (1.0) 31.7 (3.4) 62.1 (5.8) 19.8 (1.0) 13.2 (1.4) 47.3 (1.6) 44.8 (5.4) 4.7 (0.1) 127.8 (4.8) 12.8 (2.5)
204.1 (25.6) 141.2 (58.2) 197.4 (93.9)
35.6
87.5 65.0
441.5
58.4 41.9
25.5 20.4 23.2 12.7 5.6
(3.4)
(1.8) (0.1)
(1.5)
(5.0) (3.1)
(0.3) (3.4) (4.1) (0.2) (0.2)
459.4 (1.8) 877.6 (19.4) 129.6 (0.1) 385.4 (3.8) 78.9 (2.2) 7.6 (0.3) 16.2 (0.8) 28.5 (1.4) 66.5 (1.8) 22.9 (0.8) 13.3 (0.9) 51.9 (0.9) 45.1 (0.3) 5.7 (0.4) 154.0 (1.0) 12.6 (0.4)
206.0 (14.7) 211.5 (0.3) 176.8 (42.6)
34.6
86.2 65.6
513.5
56.8 41.9
27.2 20.7 22.9 12.3 6.1
446.1 833.4 117.7 374.6 75.0 7.0 15.6 32.5 64.4 22.2 13.1 48.2 44.2 6.2 139.6 12.7
b
57.4
94.7 65.8
431.7
57.6 42.0
29.7 18.8 19.6 10.8 5.9
(5.0)
(6.2) (1.8)
(7.8)
(5.3) (2.4)
(1.4) (1.8) (1.6) (0.9) (0.2)
413.0 (11.0) 750.5 (13.4) 114.9 (4.9) 378.4 (12.2) 78.5 (6.4) 7.4 (0.0) 15.0 (1.0) 29.2 (0.3) 66.3 (4.0) 22.9 (1.9) 12.7 (1.0) 49.3 (2.8) 43.4 (1.7) 6.0 (1.2) 135.9 (5.2) 14.1 (0.6)
196.6 (14.2) 182.6 (16.9) 217.9 (3.4)
55.4
86.4 62.5
419.8
45.6 31.9
27.6 17.2 19.0 11.1 5.7
370.4 706.8 97.1 320.6 63.4 6.3 12.6 27.7 54.7 17.6 11.5 41.8 38.5 5.5 111.0 12.3
(0.3)
(1.8) (0.9)
(32.1)
(0.9) (1.7)
(2.9) (2.6) (1.8) (1.7) (0.2)
(17.0) (26.3) (0.5) (13.9) (1.0) (0.6) (0.0) (2.2) (3.5) (1.0) (0.8) (0.5) (1.1) (1.1) (8.0) (1.9)
199.3 (3.1) 184.0 (37.6) 371.3 (253.9)
47.0
88.8 65.2
475.6
53.9 39.3
27.4 19.1 21.0 11.8 6.2
422.2 792.9 114.0 365.7 73.4 7.1 14.8 30.1 63.0 21.2 12.9 48.0 43.3 5.7 134.4 13.2
211.3 174.4 241.0
(9.7)
(3.5) (3.4)
(50.3)
(5.4) (4.9)
(2.4) (1.4) (1.7) (0.8) (0.6)
(32.2) (61.3) (10.9) (23.2) (5.9) (0.5) (1.3) (1.9) (4.4) (2.1) (0.7) (3.4) (2.4) (0.5) (14.3) (0.9)
(15.5) (33.6) (84.9)
av emissions (µg cigarette-1)
Emissions may be semiquantitative, see text. c PM emissions in mg cigarette-1 (n ) 2).
(82.8) 3568.6 (13.5) 3434.7 (327.7) 3643.9 (160.4) (1.3) 15.8 (1.1) 16.9 (0.1) 17.9 (2.0)
(13.9)
(4.7) (2.6)
(22.6)
(1.2) (1.6)
(0.9) (0.5) (0.6) (0.6) (0.3)
(13.5) (27.9) (6.8) (20.0) (3.5) (0.2) (0.8) (1.1) (3.2) (0.9) (1.5) (1.3) (2.3) (0.5) (8.5) (1.2)
229.1 (12.6) 126.4 (30.2) 162.3 (134.0)
(71.7) 3554.6 (98.8) 3766.7 (67.3) 3658.0 (0.3) 17.1 (0.2) 17.0 (0.9) 18.9
(11.2)
(10.9) (9.5)
(41.6)
(3.3) (2.3)
(1.7) (2.3) (5.0) (1.8) (1.2)
(14.6) (42.4) (2.4) (5.9) (1.5) (0.1) (0.6) (1.5) (1.8) (1.3) (0.4) (2.3) (1.8) (0.4) (3.7) (1.3)
232.5 (7.8) 200.8 (10.1) 320.4 (125.9)
Standard deviation in parentheses. NA ) not available.
0.71 0.87 0.51 0.81 0.57 0.48 0.95 0.94 1.25 1.33 0.82 0.75 0.92 1.37 0.57 1.40
aromatics benzene toluene ethylbenzene p-, m-xylene o-xylene isopropylbenzene n-propylbenzene p-isopropyltoluene 4-ethyl toluene 2-ethyl toluene 1,3,5-trimethylbenzene 1,2,4-trimethylbenzene 1,2,3-trimethylbenzene n-butylbenzene styrene naphthaleneb
a
2.08 10.85 14.61
ETS tracers 2,5-DMF 3-EPb nicotineb
VOC/PM
MDL B1-RT B2-RT B3-RT B4-RT B1-LT-M B2-RT (µg cigarette-1) (µg cigarette-1) (µg cigarette-1) (µg cigarette-1) (µg cigarette-1) (µg cigarette-1) (µg cigarette-1)
total emissions by cigarette brand
NA 11.5–16
49–394
7.7–17.1 NA
269–480
NA NA
NA NA NA NA NA
238–610 311–1100 69–165 85–470 40–98 4–5.9 8.5–17.3 NA NA NA 6.2–19 22–74 32.70 5.2–6.1 107–203 17–54
127–266 82–890 323–3740
literature reported (µg cigarette-1)
19
3, 3, 3, 3, 3,
18, 18, 18, 18, 18,
19, 19, 19, 19, 19,
23, 23, 23, 23, 23,
32 32 32 32 32
19, 32
3, 18
19
1, 19, 23, 32
1, 19 18, 19 1 19 1, 3, 18, 19, 23, 32 18
1, 1, 1, 1, 1, 1 1,
19 1, 3, 18, 19, 23, 32 3, 18, 19, 32
TABLE 1. Method Detection Limits (MDLs, Based on a 10 mL Sample), Emission Factors for Six Types of Commercial Cigarettes (n = 2), 6-Brand Average, and Range of Emission Factors in the Literaturea
TABLE 2. Main- (MS) and Side-Stream (SS) Emissions for P1-RT-M and P2-RT Cigarettes and Ratio of SS to MS Emissions (SS/ MS)a mainstream (MS) and sidestream (SS) emissions by cigarette brand B1-RT-M VOC
MS (µg cigarette-1)
SS (µg cigarette-1)
B2-RT SS/MS
mass balance
MS (µg cigarette-1)
SS (µg cigarette-1)
SS/MS
mass balance
ETS tracers 2,5-DMF 3-EPb nicotineb
58.3
