Environ. Sci. Technol. 2006, 40, 1133-1138
Determination of 14 Polycyclic Aromatic Hydrocarbons in Mainstream Smoke from U.S. Brand and Non-U.S. Brand Cigarettes YAN S. DING, XIZHENG J. YAN, RAM B. JAIN, EUGENE LOPP, AMEER TAVAKOLI, GREGORY M. POLZIN, STEPHEN B. STANFILL, DAVID L. ASHLEY, AND CLIFFORD H. WATSON* Emergency Response and Air Toxicants Branch, Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, 4770 Buford Highway NE, Mailstop F-47, Atlanta, Georgia 30341
Tobacco smoke contains thousands of chemical compounds, including many carcinogenic polycyclic aromatic hydrocarbons (PAHs). To determine the concentration ranges of PAHs in tobacco smoke and to understand what factors alter their levels, we quantitatively measured 14 PAHs in mainstream smoke from a transnational U.S. brand (Marlboro) and from locally popular brand cigarettes from 14 countries. We used standardized machine smoking conditions (35mL puff volume, 60-s puff interval, and 2-s puff duration), extraction of total particulate matter from the Cambridge filters, and gas chromatography/mass spectrometry detection. Deliveries of total PAHs in mainstream smoke of local brands were statistically significantly higher (p < 0.01) than Marlboros in seven countries. In four countries, Marlboro cigarettes had mainstream smoke total PAH levels that were statistically significantly higher (p < 0.01) than local brands. In the remaining three countries, the differences in PAH levels were not statistically significant. Under standard machine smoking conditions, PAH levels were negatively correlated with cigarette filter ventilation levels. We found that several local brands containing primarily fluecured tobacco filler had relatively high mainstream smoke PAH deliveries, in agreement with findings by previous researchers that flue-cured tobacco typically delivers more PAHs than other tobacco types. We also observed that PAHs were inversely correlated with total carcinogenic tobacco-specific nitrosamines and nitrate content, but these correlations were not statistically significant at the 95% confidence interval. The findings suggest that tobacco blend and nitrate levels may influence PAH deliveries, but other factors may confound this relation.
Introduction The polycyclic aromatic hydrocarbons (PAHs) are a diverse group of carcinogens formed during the incomplete combustion of organic materials such as tobacco. Various PAHs have long been known as environmental pollutants (1, 2). In laboratory animals, several PAHs are potent, locally acting carcinogens, inducing tumors of the upper respiratory tract * Corresponding author phone: (770)488-7638; fax: (770)488-0181; e-mail:
[email protected]. 10.1021/es0517320 Not subject to U.S. Copyright. Publ. 2006 Am. Chem. Soc. Published on Web 01/14/2006
and lung when administered by inhalation, installation in the trachea, or implantation in the lung (3). Benzo[a]pyrene is the most studied member of this class of compounds, and its ability to induce lung tumors has been well-documented (4). A full evaluation of PAH exposure from tobacco smoke to a smoker and the surrounding environment should consider the contribution from environmental tobacco smoke (ETS). Mainstream smoke, along with sidestream smoke and exhaled mainstream smoke, is a major contributor to ETS. Presently, there are no standard methods for measuring either sidestream smoke or exhaled mainstream smoke. Although a stochastic deposition model was developed to predict the deposition of tobacco smoke in the human respiratory tract, prediction and actual experimental differed by a factor of 4 (5). Measuring mainstream smoke using standardized machine smoking protocols has been well-established. Mainstream smoke from cigarettes is a complex aerosol containing more than 4000 compounds from multiple chemical classes (6). In a recent report, 69 tobacco smoke constituents were identified as carcinogens on the basis of IARC’s (International Agency for Research on Cancer) evaluations (3, 7). Among these 69 carcinogens, 10 fall within the PAH chemical classification (8). Therefore, accurate quantitative assessment of these compounds in cigarette smoke is essential to help estimate human and environmental exposure. Previously, we developed an analytical technique for measuring PAHs in tobacco smoke particulate by using gas chromatography with mass spectrometric detection (GC/ MS) (9). We determined the quantitative levels of 14 PAHs in mainstream smoke particulate from domestic commercial cigarette brands manufactured by the four major cigarette companies in the United States. We also observed the differences in PAH distribution profiles among various tobacco types (burley, bright, oriental, and reconstituted tobacco) (9). To help assess global public health risks associated with tobacco consumption, we designed this study to obtain PAH data on a diverse selection of cigarettes manufactured worldwide. We hope it provides valuable information to help ascertain whether regional differences in tobacco deliveries result from various cultivation practices or from blend preferences. We report results for low molecular weight PAHs, high molecular weight PAHs, and total PAH mainstream smoke levels generated by using a standardized machine smoking protocol from cigarettes purchased in 14 countries. The PAH content in the mainstream smoke total particulate matter (TPM) was determined using an automated cleanup step with GC/MS detection. We compared the levels of low and high molecular weight PAHs and total PAH content in mainstream smoke between local brand and Marlboro cigarettes for each of the 14 countries. We also examined the influence of cigarette filter ventilation, tobacco composition, and nitrate levels on the mainstream PAH deliveries.
Experimental Section Materials. The World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC) collaborated to collect and analyze cigarettes from a popular transnational brand and from a locally popular brand from at least two nations in each of the six WHO regions (Africa, America, South-East Asia, Europe, Eastern Mediterranean, and Western Pacific). Countries were selected first on the basis of their total population. Other than the United States, the 10 most populous countries are China, India, Indonesia, Brazil, Pakistan, the Russian Federation, Japan, Bangladesh, Nigeria, and Mexico. In addition, cigarettes from three other VOL. 40, NO. 4, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1133
countries, Germany, Egypt, and Kenya, were added to ensure representation for two countries in each of the six WHO regions. In each country, a locally popular brand and, for comparison, a transnational cigarette brand (Marlboro) that was available in all 14 countries were selected. Both tobacco filler and mainstream smoke from these cigarettes have been previously analyzed for selected chemical components (1012). Neat PAHs used for calibration were obtained from Aldrich Chemical Co. (Milwaukee, WI). A 16-PAH standard (13C, 99%, 5 µg/mL) stock, used as a labeled internal standard, was purchased from Cambridge Isotope Laboratories, Inc. (Andover, MA). Nitrate stock solution (1000 ppm as nitrogen) and nitrate ion suppressor solution were purchased from Thermo Orion (Beverley, MA). All dilutions were prepared in deionized water. Cambridge filter pads (CFPs) used to collect mainstream smoke particulate matter were obtained from Whatman (Maidstone, United Kingdom). International cigarette samples were purchased by CDC personnel who were either permanently stationed abroad or on temporary travel status between October 2000 and March 2001. The cigarettes were forwarded to the CDC by overnight carrier, by diplomatic pouch, or were hand carried by the purchaser. Upon arrival, the cigarette packs were assigned unique identification numbers, were sealed in a plastic bag in their original packaging, and were stored at -70 °C until needed. Reference cigarettes used as quality control (QC) samples (2R4F and 1R4F) were from the University of Kentucky (Lexington, KY). Smoke Collection. Cigarettes and CFPs were conditioned at 22 °C and 60% relative humidity for at least 24 h before smoking. Mainstream smoke total particulate matter (TPM) generated under a standard machine smoking protocol (60-s puff interval, 2-s puff duration, and 35-mL puff volume) was collected on individual CFPs using a Cerulean ASM500 16port smoking machine (Milton Keynes, United Kingdom). The cigarettes were smoked to a butt length of 23 mm or to the length of the filter overwrap plus 3 mm, whichever was longer. Three cigarettes each from 9 to 15 different packs of each brand were smoked to obtain the average smoke particulate level for each of the 14 PAHs listed below. During each smoking run, 2R4F cigarettes were smoked as QC samples. After the three cigarettes were smoked per pad, each CFP was spiked with 25 µL of the 13C PAH internal standard solution prior to sample cleanup. Sample Preparation and GC/MS Analysis. The sample preparation scheme was primarily based on a published method (9), modified to use a Gilson 215 automated solidphase extraction system (SPE) (Middleton, WI) and Varian C-18 SPEC cartridges (Lake Forest, CA) to automate the SPE. The GC/MS analysis was performed using a single quadruple mass spectrometer (9). Individual PAH concentrations (naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benz[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[e]pyrene, and benzo[a]pyrene) were reported as ng analyte per cigarette. The total PAHs was obtained by summing the abundances of the 14 individual PAHs. In our previous study (9), we showed that tobacco composition can influence the distribution between low and high molecular weight (MW) PAHs. The lower molecular weight PAHs were more abundant in the mainstream smoke from a cigarette made with burley tobacco than with bright tobacco. The higher molecular weight PAHs were less abundant in the mainstream smoke from a cigarette made with burley tobacco than with bright tobacco. Because low MW PAHs are in greater abundance than high MW PAHs, the low MW PAHs have a higher relative contribution to the calculated total PAHs. To assess the separate contribution between these two PAH groups (low and high MW) more 1134
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 4, 2006
accurately, the 14 PAHs were divided into two groups, low MW PAHs and high MW PAHs. We summed levels of twoand three-ring PAHs (naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene, and fluoranthene) to obtain overall low molecular weight PAHs (low MW PAHs). Similarly, levels of four- and five-ring PAHs (pyrene, benz[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[e]pyrene, and benzo[a]pyrene) were summed to obtain overall high molecular weight PAHs (high MW PAHs). Tobacco Composition. Tobacco filler from each cigarette brand was removed for visual and optical examination. A digital image of the dispersed tobacco filler was analyzed using custom software developed in-house to determine whether the composition was primarily flue-cured or was a blended variety on the basis of its color characteristics relative to known tobacco types. Calibration of Nitrate Probe. A nitrate probe was calibrated over the range of 0.1-1000 ppm using serial dilutions of a stock nitrate standard (1000 ppm). At each concentration, 50 mL of standard was combined with a 3-mL aliquot of the interference suppressor and was analyzed by a nitrate-ion specific probe (Thermo Orion, Beverly, MA). To minimize short-term drift, the probe mV response was recorded after 3-min equilibration. Calibration curves were generated with each batch of samples analyzed. Determination of Nitrate Content. The nitrate content in the tobacco filler was measured in triplicate. The filler from individual cigarettes was placed in 50 mL of deionized water. After shaking for 1 h, the samples were filtered under vacuum and the filtrate was adjusted to 50 mL. To minimize background interference, 3 mL of nitrate ion interference suppressor solution was added to each sample. The filtrate, containing the interference suppressor solution, was analyzed with the nitrate-specific probe. A blank consisting of 50 mL deionized water and a QC sample (1R4F) were included with each sample batch. Determination of Nicotine. The level of nicotine was determined by a previously published method (11). Each brand was sampled twice, with three cigarettes smoked for each sampling event, for a total of six cigarettes for each data point. Each cigarette was pulled from a different pack. Statistical Analysis. Reconstructed ion chromatograms were processed using Xcalibur software (Thermo Electron), and results were exported to an Excel spreadsheet. Statistical analyses were performed using SAS/STAT software (SAS Institute, Inc., Cary, NC). We used two-sided t-tests to compare low MW, high MW, and total PAH levels between local and Marlboro brands for each country. Differences were considered statistically significant when the p-value was less than 0.05 for a two-sided t-test. The levels of tobacco-specific nitrosamines (TSNAs) taken as the sum of N-nitrosonornicotine (NNN) and 4-(N-nitrosomethylamino)-1-(3-pyridyl)1-butanone (NNK) from the cigarette smoke examined in this study were previously reported (12). To examine relations between PAH levels, filter ventilation, nitrate content, and TSNA levels, we fit regression models by using PAH levels as the dependent variable and by using filter ventilation, nitrate content, and TSNA levels as the independent variables. The sign of slope indicated the direction of correlation (negative vs positive), while the magnitude of the slope was tested for significant departure from zero.
Results Cigarette Physical Attributes. Cigarette brands purchased in 14 countries were analyzed for percentage filter ventilation, nitrate content of the filler, nicotine, and tobacco blend composition (Table 1). Most of the cigarettes were king-sized (∼85 mm in length) and contained a cellulose acetate filter. Cigarettes purchased from Japan (both Marlboro and Mild
TABLE 1. Ventilation and Tobacco Composition in U.S. and Non-U.S. Brand Cigarettes country Bangladesh Brazil China Egypt Germany India Indonesia Japan Kenya Mexico Nigeria Pakistan Russia United States
brand
number of packs
filter ventilationa
nitrate content (ppm)
nicotine (mg/cig)
tobacco composition
JP Gold leaf Marlboro Derby Marlboro Hongtashan Marlboro Cleopatra Marlboro West Marlboro Gold Flake Marlboro Ardath Marlboro Mild Seven Marlboro Sportsman Marlboro Boots Marlboro High Society Marlboro Embassy King Marlboro Prima Marlboro Doral Marlboro
15 15 15 14 13 14 10 14 14 15 14 12 14 11 10 9 13 11 15 15 15 10 14 11 15 11 14 14
4.7 16.5 16.9 18.3 13.5 18.1 2.1 16.2 19.8 20.5 3.8 5.3 3.3 15.8 15.9b 34.7b 4.1 14.9 2.3 2.8 3.7 8.6 3.0 13.0 no filter 16.9 16.9 14.8
28.2 138.4 221.5 394.5 51.4 106.3 38.1 90.7 84.2 130.0 9.4 165.0 36.2 184.8 64.2 87.9 27.6 91.4 63.2 428.1 29.6 121.2 29.1 124.5 84.4 120.3 132.0 102.2
1.6 1.1 0.9 (1.0)c 0.9 (1.1)c 0.9 (1.2)c 1.0 (1.0)c 1.0 1.0 0.9 (0.9)c 0.9 (0.9)c 1.3 1.2 1.3 (1.3)c 1.2 0.9 (0.9)c 0.9 (1.0)c 1.2 1.0 0.9 1.2 0.5 (1.0)c 1.1 1.8 1.0 1.0 1.1 1.0 1.0
blended blended blended blended flue-cured blended blended blended flue-cured blended blended blended blended blended flue-cured blended flue-cured blended blended blended blended blended blended blended blended blended blended blended
a Filter ventilation of cigarette was determined on a QTM5 ventilation measurement apparatus purchased from Filtrona (Richmond, VA). filter. c Numbers in parentheses are manufacturer provided values.
b
Charcoal
TABLE 2. Mean PAH Levels (ng per Cigarette) ( Standard Error in Mainstream Smoke country Bangladesh Brazil China Egypt Germany India Indonesia Japan Kenya Mexico Nigeria Pakistan Russia United States
brand
total PAHs
p value
low MWa PAHs
p value
high MWb PAHs
p value
JP Gold leaf Marlboro Derby Marlboro Hongtashan Marlboro Cleopatra Marlboro West Marlboro Gold Flake Marlboro Ardath Marlboro Mild Seven Marlboro Sportsman Marlboro Boots Marlboro High Society Marlboro Embassy King Marlboro Prima Marlboro Doral Marlboro
1414 ( 46 905 ( 20 992 ( 24 1106 ( 22 2036 ( 66 1052 ( 46 1481 ( 66 1040 ( 45 934 ( 36 934 ( 18 1525 ( 32 1875 ( 114 1390 ( 36 1294 ( 50 801 ( 14 840 ( 17 1380 ( 26 1127 ( 24 869 ( 35 1023 ( 28 1310 ( 21 1478 ( 4 2673 ( 70 1196 ( 52 1434 ( 70 837 ( 26 983 ( 18 806 ( 26
