Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: Environ Res. 2021 Oct 27;204(Pt C):112270. doi: 10.1016/j.envres.2021.112270

Effects of e-liquid flavor, nicotine content, and puff duration on metal emissions from electronic cigarettes

Di Zhao a,b, Vesna Ilievski a, Vesna Slavkovich a, Pablo Olmedo c, Arce Domingo-Relloso a, Ana M Rule d, Norman J Kleiman a, Ana Navas-Acien a, Markus Hilpert a
PMCID: PMC9140018  NIHMSID: NIHMS1797733  PMID: 34717948

Abstract

Vaping is the action of inhaling and exhaling aerosols from electronic cigarettes. The aerosols contain various amounts of toxic chemicals, including metals. The purpose of this study was to evaluate factors that can influence metal levels, including flavor and nicotine content in the e-liquid, and puff duration. Aerosols were collected from both closed-system (cartridge-based) and open-system e-cigarettes using e-liquids with different flavors (fruit, tobacco, and menthol), nicotine content (0, 6, 24, and 59 mg/mL), and different puff durations (1, 2, and 4 s). The concentrations of 14 metals in the collected aerosols were measured using inductively coupled plasma mass spectroscopy. Aerosol concentrations of As, Fe, and Mn varied significantly among fruit, tobacco, and menthol flavors in both closed-system and open-system devices. Concentrations of Al, Fe, Sn, and U were significantly higher in tobacco or menthol flavored aerosols compared to fruit flavors in closed-system devices. Aerosol W levels were significantly higher in tobacco flavored aerosols compared to fruit flavors in open-system devices. Concentrations of As, Fe, and Mn were higher in tobacco flavored aerosols compared to menthol flavors in both types of devices. The median Pb concentration decreased significantly from 15.8 to 0.88 μg/kg when nicotine content increased from 0 to 59 mg/mL, and median Ni concentration was 9.60 times higher in aerosols with nicotine of 59 mg/mL compared to 24 mg/mL (11.9 vs. 1.24 μg/kg) for closed-system devices. No significant differences were observed in aerosol metal concentrations for different puff durations. Aerosol metal concentrations varied widely between different flavors and nicotine content but not by puff duration. Flavor and nicotine content of the e-liquid could be potential factors in metal emissions. Some elements showed higher concentrations under certain conditions, highlighting the urgent need of developing strict product regulations, especially on e-liquid composition and nicotine content to inform e-cigarette users about metal exposure through vaping.

Keywords: E-cigarettes, metal emissions, flavor, nicotine, puff duration

1. Introduction

Metallic parts of electronic cigarette (e-cigarette) cartridges and devices can result in exposure of e-cigarette users to various metals including (but not limited to) chromium (Cr), lead (Pb), nickel (Ni), and tin (Sn), all of which can be found in e-cigarette hardware such as heating coils and solder joints (Aherrera et al., 2017; Goniewicz et al., 2014; Hess et al., 2017; Mikheev et al., 2016; Zhao et al., 2020). In a study of 64 e-cigarette users, Cr and Ni levels in saliva and urine were positively associated with e-cigarette aerosol metal concentrations (Aherrera et al., 2017), raising concern about potential chronic health effects of metals, including cardiovascular disease, respiratory disease and lung cancer (IARC 2012a, 2012b; Jaishankar et al., 2014; Navas-Acien et al., 2020). Of further concern, studies have shown that e-cigarette use may result in exceedance of chronic minimum risk levels (MRL) of Mn and Ni (Zhao et al., 2019). There is substantial variability in metal content in e-cigarette aerosols and e-liquids (Na et al., 2019; Gaur et al., 2019; Fowles et al., 2020). It is thus important to evaluate how different conditions can influence metal concentrations in aerosols inhaled by users.

Flavorings in e-liquids substantially contribute to youth e-cigarette use. Among 13,651 US youths aged 12–17 years, 81% of the e-cigarette users reported that they started vaping because of the availability of flavored products (Ambrose et al., 2015). In January 2020, the US Food and Drug Administration (FDA) announced action to remove sweet-flavored e-liquids from the market of cartridge-based e-cigarettes (closed-system) that appeal to children. The flavors that are now unauthorized include fruit, dessert and mint. However, menthol and tobacco-flavored products remain widely available and may also appeal to youth (FDA, 2020). One analysis showed higher exposures to benzaldehyde among users of cherry-flavored e-liquids compared to users of e-liquids with tobacco, sweet, mint, and other flavors (5.1–141.2 vs. 0.03–10.3 μg/30 puffs) (Kosmider et al., 2016). The potential impact of flavors, including menthol or tobacco-flavored products, on aerosol metal levels, however, is unknown and deserves further investigation.

For many e-cigarette users, nicotine content was found to be the third most important factor, after flavor and price, when choosing e-cigarettes or e-liquids (Laverty et al., 2016). Nicotine content varies between e-cigarette types; some e-cigarette devices allow the use of e-liquid with different nicotine content, including no nicotine, while other devices such as JUUL were available with only one high nicotine content in 2017 (Cameron et al., 2014; Kavuluru et al., 2019). One study, which examined mostly cigalikes, found that As, Cr, Cu, Ni, Sb, Sn, and Zn levels varied widely across aerosols with and without nicotine (Mikheev et al., 2016). Cigalike devices, however, are mostly first-generation devices and are not representative of e-cigarette devices currently on the market. The impact of nicotine content on aerosol metal levels from tank and mod devices, which allow users to refill the atomizer with e-liquids differing in nicotine content, is unknown.

Puff topography, especially puff duration, which is a known factor that influences the toxicity of traditional cigarettes (Maziak et al., 2011; Reilly et al., 2017), could also influence metal emissions from e-cigarettes. Puff duration varies substantially among e-cigarette users (range: 1.9–8.3 seconds), with a reported average puff duration of 4 seconds (Farsalinos et al., 2013a; Hua et al., 2013). Most previous studies evaluating metals in e-cigarette emissions have used 4 seconds. It has been argued that excess metal concentrations measured in the aerosol were related to puff duration (Williams et al., 2019). Evaluating vaping topography, especially puff duration, is important to assess potential differences in metal concentrations of the aerosols inhaled by e-cigarette users.

In a previous study, we found that device type and device power setting are determining factors of metal concentration in aerosol generated by e-cigarettes (Zhao et al., 2019). In this study, we evaluated the effects of e-liquid flavor and nicotine content as well as puff duration on aerosol metal emissions.

2. Methods

2.1. E-cigarette devices and e-liquid characteristics

Four commercial e-cigarettes including two closed-system devices (CD), both of them pods, and two open-system devices (OD), both of them mods, were studied. The CDs BLU and JUUL are identified as CD1 and CD2, respectively. The ODs Istick Pico 25 and SMOK Alien 220 are identified as OD1 and OD2, respectively.

A total of 16 e-liquids differing in flavor and nicotine content were used in this study (Table 1). For each device, e-liquids of three flavor categories were purchased: fruit, tobacco, and menthol. For CD1, each of the three flavors was purchased with a nicotine content of 24 mg/mL. Menthol flavored e-liquids with nicotine content of 0 mg/mL were also purchased. For CD2, only e-liquids with nicotine content of 59 mg/mL were used for each flavor, because this was the only nicotine content offered by the manufacturer at the time of purchase. The nicotine of CD1 is in the form of free-base nicotine, whereas nicotine salts are used in CD2. For open-system devices, each of the three flavors was purchased with three different nicotine contents: 0, 6, and 24 mg/mL, in the form of free-base nicotine. The same e-liquids were used for OD1 and OD2.

Table 1.

Device and e-liquid characteristics.

Device ID Commercial name Coil Shape Power (W) Resistance (Ohm) Battery voltage Flavor Nicotine content (mg/mL) Aerosol samples (n)
CD1 BLU Nichrome Cigalike N/A N/A 3.7 V Fruit (Blueberry) 24 7
Tobacco 24 20
Menthol 0 7
Menthol 24 9
CD2 JUUL Nichrome Pod N/A N/A N/A Fruit (Fruit medley) 59 3
Tobacco 59 9
Menthol (Cool mint) 59 3
OD1 Istick 25 Kanthal Tank 1–85 0.2 3.7 V Fruit (Blueberry) 0 3
Fruit (Blueberry) 6 3
Fruit (Blueberry) 24 3
Tobacco 0 3
Tobacco 6 3
Tobacco 24 9
Menthol 0 3
Menthol 6 3
Menthol 24 3
OD2 SMOK Stainless steel Tank 6–220 0.6 2 × 3.7 V Fruit (Blueberry) 0 3
Fruit (Blueberry) 6 3
Fruit (Blueberry) 24 3
Tobacco 0 3
Tobacco 6 3
Tobacco 24 9
Menthol 0 3
Menthol 6 3
Menthol 24 3
Open in a new tab

N/A=not available (Information not indicated directly on packages, not available on the internet, and not obtained after direct contact with the manufacturer). For open-system devices (OD1, OD2), aersosol samples (n) denotes the number of aerosol samples collected at intermediate power setting. For tobacco flavored e-liquid with nicotine content of 24 or 59 mg/mL, aerosols samples (n) denotes the number of aerosol samples collected from three puff durations.

All e-cigarette products were purchased online in June 2017. Additional details of each product and online shopping websites are described in Zhao et al. (2019).

2.2. Aerosol generation and collection

Four devices and 16 e-liquids were used to study the impact of e-liquid flavor, nicotine content, and puff duration on metal release. Aerosol was collected from all the devices using the method developed by Olmedo et al. (2016), with minor modifications to collect aerosols from open-system devices to increase the aerosol recovery rate. Puff topography duration was microcomputer-controlled as described in Hilpert et al. (2019). The aerosol recovery rate ranged from 73 to 86%, with a mean (standard deviation) aerosol recovery rate of 72.9% (9.31%), 83.1% (0.45%), 84.2% (7.90%), and 85.6% (7.63%) for CD1, CD2, OD1, and OD2, respectively. Detailed information about aerosol collection for closed-system and open-system devices as well as the method for calculating aerosol recovery rate are reported in Zhao et al. (2019) and Hilpert et al. (2021).

Like in our previous studies, a puff duration of 4 s was used when generating and collecting aerosols from the e-cigarettes, with an inter puff time of 11 s for closed-system devices and 26 s for open-system devices (Zhao et al., 2019). The closed-system devices were operated at their non-adjustable power settings which were not disclosed by the manufacturer but were likely different between devices. The open-system devices were operated at intermediate power settings of 40 W and 120 W for OD1 and OD2, respectively. Different numbers of puffs were used for different devices to collect sufficient amounts of aerosol for laboratory analysis. A total of 50–100, 290–330, 8–92, and 6–72 puffs were required to collect aerosols from CD1, CD2, OD1, and OD2, respectively. The highest number of puffs was required for CD2 as this device releases much less aerosol than the other devices.

Puff duration varies among e-cigarette users, experienced users take longer puffs of an average 4 s, while inexperienced users take shorter puffs of 2 s (Farsalinos et al., 2015; Talih et al., 2015). To study the effect of puff duration on metal release, aerosols were also generated for puff durations of 1 and 2 s. Due to limited resources, these additional experiments were performed only with tobacco flavored e-liquid, because it can be expected to remain widely available in the market. Moreover, a nicotine content of 24 mg/mL was chosen (except for CD2, which used a nicotine content of 59 mg/mL), because it is the most common one among the four devices.

2.3. Metal analyses

Aerosols were diluted with diluent containing 2% HNO3, 1% methanol, and 0.02% Triton X-100 as previously described (Zhao et al., 2019), and analyzed using inductively coupled plasms mass spectrometry (ICP-MS, NexION 350S, PerkinElmer) at the Columbia University Trace Metal Core Lab. Levels of the following 14 metals were determined using ICP-MS: aluminum (Al), arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), iron (Fe), manganese (Mn), nickel (Ni), lead (Pb), antimony (Sb), tin (Sn), uranium (U), tungsten (W), and zinc (Zn). Gallium (Ga), rhodium (Rh), and iridium (Ir) were used as internal standards to control for long-term signal drift. A mixture solution of propylene glycol and glycerol was analyzed to account for matrix effects. Reference material SRM 1640a (Ultra-pure, ICN Biochemicals, USA) was used for quality assurance. The limits of detection (LOD) for metals were as follows (μg/kg): 6.7 for Al, 0.15 for As, 0.1 for Cd, 0.1 for Cr, 0.4 for Cu, 5.55 for Fe, 0.05 for Mn, 0.2 for Ni, 0.05 for Pb, 0.05 for Sb, 0.6 for Sn, 0.05 for U, 0.05 for W and 0.7 for Zn. The percentage of samples below the LOD ranged from 0% for Al, As, Cu, Ni, Pb, and Zn to 97.8% for U. For aerosols with metal concentrations below the LOD, the values were substituted by LOD/√2 for statistical analysis.

2.4. Statistical analyses

Since device type is a major factor that influences metal emissions (higher levels of toxic metals for open-system devices compared to closed-system devices (Zhao et al., 2019)), we calculated medians and interquartile ranges (IQR) of metal concentrations in e-cigarette aerosols with different flavors, nicotine content, and puff duration stratified by device type (OD and CD) to summarize quantitative variables. Overall differences in log-transformed metal concentrations across different flavors, nicotine content, and puff durations were analyzed by one-way ANOVA. P-values and beta coefficients [95% confidence intervals (95% CIs)] from linear regression were determined to show pairwise differences in metal concentrations by these variables. When p-values were less than 0.05, differences were considered significantly different. Box plots were used to visualize metal concentration distributions stratified by flavor, nicotine content, and puff duration. All statistical analyses were performed using the R software version 3.4.2.

3. Results

3.1. Aerosol metal concentrations by e-liquid flavor

For closed-system devices, we found statistically significant differences in As, Fe, Mn, Pb, Sb, Sn, and U concentrations among fruit, tobacco, and menthol flavored aerosols (Table 2; Fig. 1). When comparing tobacco to fruit flavors, median Al, Fe, Sn, and U aerosol concentrations were 2.71, 1.65, 1.38, and 1.5 times higher for the tobacco flavor, while the median Pb concentration was 2.6 times lower. When comparing menthol to fruit flavors, Mn concentrations were lower (0.53 vs. 1.39 μg/kg) for the menthol flavor, and Sn concentrations were higher (0.65 vs. 0.37 μg/kg). When comparing tobacco to menthol flavors, median As, Fe, Mn, Ni, and U concentrations were 1.44, 1.63, 1.45, 3.74, and 1.5 times higher, while Pb and Sb concentrations were 5.92 and 3.45 times lower.

Table 2.

Median (interquartile range) metal concentrations (μg/kg) in aerosols by flavor.

Median (IQR) in CD Overall p-value Linear regression p-value Beta value (95% CI)
Fruit (n=10) Tobacco (n=17) Menthol (n=19) Fruit vs. Tobacc Fruit vs. Menthol Tobacco vs. Menthol Fruit vs. Tobacco Fruit vs. Menthol Tobacco vs. Menthol
Al 6.02 (4.29–8.65) 16.3 (12.3–35.6) 9.96 (6.56–16.0) 0.10 0.03 0.12 0.44 0.82 (0.07, 1.57) 0.57 (−0.16, 1.31) −0.25 (−0.87, 0.38)
As 0.09 (0.09–0.12) 0.13 (0.10–0.36) 0.09 (0.09–0.10) 0.03 0.13 0.41 0.008 0.56 (−0.18, 1.30) −0.30 (−1.03, 0.43) −0.86 (−1.48, −0.24)
Cd 0.04 (0.04–0.05) 0.06 (0.05–0.08) 0.05 (0.04–0.07) 0.19 0.10 0.10 0.98 0.36 (−0.07, 0.79) 0.35 (−0.07, 0.77) −0.004 (−0.36, 0.36)
Cr 0.05 (0.04–0.44) 0.48 (0.29–1.47) 0.43 (0.15–0.62) 0.18 0.08 0.12 0.78 1.18 (−0.15, 2.52) 1.03 (−0.28, 2.34) −0.16 (−1.28, 0.96)
Cu 19.9 (12.0–30.8) 15.4 (4.29–67.2) 33.4 (21.6–62.1) 0.43 1.00 0.32 0.25 0.003 (−1.01, 1.02) 0.5 (−0.50, 1.49) 0.49 (−0.36, 1.34)
Fe 3.46 (3.42–3.82) 5.72 (4.16–12.1) 3.50 (3.42–4.08) 0.03 0.02 0.47 0.04 0.73 (0.14, 1.31) 0.21 (−0.37, 0.79) −0.52 (−1.01, −0.02)
Mn 1.39 (0.59–2.91) 0.77 (0.33–1.58) 0.53 (0.11–0.78) 0.01 0.46 0.007 0.02 −0.40 (−1.47, 0.68) −1.48 (−2.54, −0.42) −1.09 (−1.99, −0.18)
Ni 0.73 (0.27–13.2) 6.73 (3.16–9.21) 1.80 (0.44–2.98) 0.05 0.12 0.61 0.02 1.07 (−0.30, 2.44) −0.34 (−1.69, 1.00) −1.41 (−2.56, −0.27)
Pb 6.28 (5.12–11.7) 2.38 (1.15–3.65) 14.1 (7.91–32.0) <0.001 0.02 0.21 <0.001 −1.18 (−2.20, −0.17) 0.62 (−0.37, 1.62) 1.81 (0.96, 2.66)
Sb 0.48 (0.39–0.91) 0.33 (0.13–0.64) 1.14 (0.66–1.76) 0.03 0.37 0.15 0.008 −0.39 (−1.25, 0.48) 0.61 (−0.24, 1.46) 1.00 (0.27, 1.72)
Sn 0.37 (0.35–0.47) 0.51 (0.39–0.70) 0.65 (0.35–1.00) 0.04 0.04 0.01 0.61 0.47 (0.01, 0.93) 0.57 (0.12, 1.02) 0.10 (−0.29, 0.48)
U 0.04 (0.04–0.05) 0.06 (0.05–0.08) 0.04 (0.04–0.05) 0.005 0.006 0.75 0.004 0.36 (0.11, 0.61) 0.04 (−0.21, 0.29) −0.32 (−0.53, −0.11)
W 0.05 (0.04–0.06) 0.06 (0.05–0.08) 0.04 (0.04–0.05) 0.50 0.29 0.76 0.36 0.24 (−0.22, 0.70) 0.07 (−0.38, 0.52) −0.18 (−0.56, 0.21)
Zn 1014 (699–2057) 1577 (916–2448) 704 (490–2366) 0.18 0.24 0.72 0.07 0.34 (−0.23, 0.92) −0.10 (−0.66, 0.46) −0.44 (−0.92, 0.04)
Median (IQR) in OD Overall p-value Linear regression p-value Beta value (95% CI)
Fruit (n=18) Tobacco (n=18) Menthol (n=18) Fruit vs. Tobacc Fruit vs. Menthol Tobacco vs. Menthol Fruit vs. Tobacco Fruit vs. Menthol Tobacco vs. Menthol
Al 11.6 (7.06–17.4) 16.7 (9.41–25.1) 14.8 (6.94–29.2) 0.24 0.16 0.13 0.90 0.42 (−0.17, 1.00) 0.46 (−0.13, 1.04) 0.04 (−0.55, 0.63)
As 1.58 (1.38–2.00) 2.07 (1.59–3.40) 1.17 (0.68–1.37) <0.001 0.07 0.04 <0.001 0.35 (−0.02, 0.71) −0.39 (−0.76, −0.02) −0.74 (−1.10, −0.37)
Cd 0.16 (0.11–0.29) 0.05 (0.05–0.18) 0.05 (0.05–0.28) 0.24 0.12 0.79 0.19 −0.68 (−1.54, 0.18) −0.11 (−0.97, 0.75) 0.57 (−0.29, 1.43)
Cr 4.48 (1.49–62.5) 7.99 (4.89–15.0) 3.04 (2.20–8.17) 0.37 0.94 0.21 0.24 −0.04 (−0.95, 0.88) −0.58 (−1.49, 0.34) −0.54 (−1.45, 0.37)
Cu 247 (150–447) 615 (134–1305) 278 (61.6–689) 0.16 0.06 0.47 0.24 0.78 (−0.03, 1.58) 0.29 (−0.52, 1.10) −0.48 (−1.29, 0.33)
Fe 159 (66.6–246) 194 (115.9–260) 38.5 (25.6–92.6) <0.001 0.19 0.005 <0.001 0.45 (−0.23, 1.12) −0.98 (−1.65, −0.30) −1.42 (−2.10, −0.75)
Mn 33.4 (20.3–57.2) 51.4 (23.6–91.3) 21.9 (14.0–48.3) 0.04 0.12 0.30 0.01 0.38 (−0.10, 0.87) −0.25 (−0.74, 0.23) −0.63 (−1.12, −0.15)
Ni 1181 (811–2757) 1032 (40.6–2807) 852 (466–1901) 0.13 0.04 0.36 0.26 −0.93 (−1.84, −0.02) −0.42 (−1.32, 0.49) 0.52 (−0.39, 1.42)
Pb 377 (166–588) 254 (120–676) 150 (67.7–519) 0.12 0.98 0.07 0.08 −0.01 (−0.70, 0.68) −0.63 (−1.32, 0.06) −0.62 (−1.31, 0.07)
Sb 3.14 (2.36–4.02) 3.54 (2.97–5.22) 4.29 (2.13–6.01) 0.41 0.49 0.18 0.52 0.15 (−0.28, 0.58) 0.29 (−0.14, 0.73) −0.14 (−0.29, 0.58)
Sn 154 (12.6–309) 208 (43.4–385) 99.5 (11.6–256) 0.44 0.26 0.97 0.27 0.58 (−0.44, 1.59) 0.02 (−1.00, 1.03) −0.56 (−1.57, 0.46)
U 0.05 (0.05–0.05) 0.05 (0.05–0.05) 0.05 (0.05–0.05) 0.16 0.21 0.52 0.06 −0.02 (−0.06, 0.01) 0.01 (−0.03, 0.05) 0.04 (−0.002, 0.08)
W 0.10 (0.05–0.17) 0.25 (0.11–0.58) 0.05 (0.04–0.13) <0.001 0.005 0.36 <0.001 0.99 (0.32, 1.65) −0.31 (−0.97, 0.36) −1.29 (−1.96, −0.62)
Zn 3420 (2207–4551) 3149 (2233–5142) 2558 (1751–4324) 0.51 0.88 0.36 0.29 0.04 (−0.53, 0.61) −0.26 (−0.83, 0.31) −0.31 (−0.88, 0.27)
Open in a new tab

Figure 1.

Figure 1.

Open in a new tab

Metal concentrations in aerosols by flavor in closed-system devices (left) and open-system devices (right). The horizontal lines within boxes indicate medians; boxes, interquartile ranges; whiskers, values within 1.5 times the interquartile range from boxes; solid circles outside the box, outlier data values.

For open-system devices, we found statistically significant differences in As, Fe, Mn, and W concentrations among fruit, tobacco, and menthol flavored aerosols, with tobacco flavor showing the highest median concentrations compared to the other two flavors (Table 2). When comparing tobacco to fruit flavors, the median Ni concentration was lower (1032 vs. 1181 μg/kg), and the median W concentration was higher (0.25 vs. 0.10 μg/kg). When comparing menthol to fruit flavors, median As and Fe concentrations were lower. When comparing tobacco to menthol flavors, median As, Fe, Mn, and W concentrations were 1.77, 5.03, 2.35, and 5 times higher.

3.2. Aerosol metal concentrations by e-liquid nicotine content

For closed-system devices, we found statistically significant differences in Al, Cd, Cu, Mn, Ni, Pb, Sb, U, and Zn concentrations among nicotine contents of 0, 24, and 59 mg/mL (Table 3; Fig. 2). When performing pairwise comparisons between nicotine levels, the median Mn concentration increased significantly from 0.30 to 1.02 μg/kg when nicotine content increased from 0 to 24 mg/mL, whereas the median Pb level decreased from 15.8 to 5.60 μg/kg. When nicotine content further increased from 24 to 59 mg/mL, the median Ni concentration increased 9.60-fold, while Al, Cd, Cu, Fe, Mn, Pb, Sb, Sn, U, and Zn concentrations decreased significantly.

Table 3.

Median (interquartile range) metal concentrations (μg/kg) in aerosols by nicotine content.

Median (IQR) in CD Overall p-value Linear regression p-value Beta value (95% CI)
0 mg/mL (n=7) 24mg/mL (n=30) 59 mg/mL (n=9) 0 vs. 24 mg/mL 0 vs. 59 mg/mL 24 vs. 59 mg/mL 0 vs. 24 mg/mL 0 vs. 59 mg/mL 24 vs. 59 mg/mL
Al 13.4 (6.93–14.4) 15.0 (9.18–32.2) 4.12 (4.05–4.40) <0.001 0.23 0.03 <0.001 0.42 (−0.27, 1.11) −0.96 (−1.79, −013) −1.38 (−2.00, −0.75)
As 0.09 (0.09–0.10) 0.11 (0.09–0.16) 0.09 (0.09–0.09) 0.56 0.31 0.68 0.57 0.43 (−0.41, 1.26) 0.21 (−0.80, 1.22) −0.22 (−0.98, 0.54)
Cd 0.05 (0.04–0.18) 0.06 (0.05–0.08) 0.04 (0.04–0.04) 0.04 0.28 0.01 0.04 −0.23 (−0.67, 0.20) −0.66 (−1.18, −0.14) −0.42 (−0.82, −0.03)
Cr 0.49 (0.28–0.69) 0.29 (0.04–0.88) 0.41 (0.34–0.57) 0.49 0.37 0.97 0.35 −0.65 (−2.09, 0.79) −0.03 (−1.76, 1.70) 0.62 (−0.69, 1.92)
Cu 33.4 (26.3–45.2) 30.6 (17.8–76.6) 6.02 (4.05–7.88) <0.001 0.90 0.002 <0.001 0.06 (−0.84, 0.95) −1.73 (−2.81, −0.66) −1.79 (−2.60, −0.98)
Fe 3.45 (3.42–3.90) 4.75 (3.66–7.55) 3.38 (3.36–3.42) 0.05 0.12 0.68 0.03 0.49 (−0.14, 1.11) −0.16 (−0.91, 0.59) −0.64 (−1.20, −0.08)
Mn 0.30 (0.10–0.68) 1.02 (0.57–1.99) 0.39 (0.20–0.53) 0.02 0.02 0.74 0.03 1.36 (0.21, 2.51) 0.23 (−1.15, 1.61) −1.13 (−2.17, −0.09)
Ni 1.80 (0.95–2.75) 1.24 (0.34–4.60) 11.9 (10.7–22.7) <0.001 0.98 <0.001 <0.001 −0.02 (−1.24, 1.21) 2.71 (1.24, 4.18) 2.73 (1.62, 3.84)
Pb 15.8 (11.0–35.5) 5.60 (2.55–13.8) 0.88 (0.57–1.15) <0.001 0.01 <0.001 <0.001 −1.33 (−2.35, −0.30) −2.96 (−4.19, −1.74) −1.64 (−2.56, −0.71)
Sb 1.61 (1.20–2.04) 0.67 (0.35–1.17) 0.15 (0.11–0.53) 0.002 0.07 <0.001 0.008 −0.79 (−1.65, 0.07) −1.87 (−2.90, −0.83) −1.08 (−1.85, −0.30)
Sn 0.53 (0.40–0.76) 0.64 (0.40–0.80) 0.35 (0.34–0.35) 0.05 0.75 0.10 0.01 0.08 (−0.41, 0.56) −0.48 (−1.06, 0.10) −0.56 (−1.00, −0.12)
U 0.04 (0.04–0.05) 0.05 (0.05–0.07) 0.04 (0.04–0.04) 0.01 0.05 0.64 0.007 0.25 (−0.005, 0.50) −0.07 (−0.38, 0.23) −0.32 (−0.55, −0.09)
W 0.04 (0.04–0.05) 0.05 (0.04–0.07) 0.04 (0.04–0.04) 0.32 0.45 0.65 0.16 0.18 (−0.30, 0.66) −0.13 (−0.70, 0.45) −0.31 (−0.75, 0.12)
Zn 1125 (787–2366) 1412 (668–2627) 683 (597–864) 0.04 0.98 0.06 0.02 −0.007 (−0.59, 0.58) −0.67 (−1.38, 0.03) −0.67 (−1.20, −0.14)
Median (IQR) in OD Overall p-value Linear regression p-value Beta value (95% CI)
0 mg/mL (n=18) 6 mg/mL (n=18) 24 mg/mL (n=18) 0 vs. 6 mg/mL 0 vs. 24 mg/mL 6 vs. 24 mg/mL 0 vs. 6 mg/mL 0 vs. 24 mg/mL 6 vs. 24 mg/mL
Al 14.7 (7.92–17.7) 13.4 (7.33–23.0) 15.3 (9.98–24.4) 0.80 0.91 0.53 0.61 0.03 (−0.57, 0.64) 0.19 (−0.42, 0.79) 0.16 (−0.45, 0.76)
As 1.47 (1.21–1.69) 1.57 (1.38–3.15) 1.86 (1.11–2.30) 0.61 0.47 0.82 0.34 0.15 (−0.27, 0.57) −0.05 (−0.47, 0.37) −0.20 (−0.62, 0.22)
Cd 0.05 (0.05–0.15) 0.23 (0.13–0.40) 0.05 (0.05–0.15) 0.002 0.001 0.73 0.003 1.34 (0.56, 2.12) 0.14 (−0.64, 0.92) −1.21 (−1.99, −0.42)
Cr 5.17 (1.84–9.51) 8.19 (3.40–35.2) 5.97 (2.49–9.66) 0.61 0.32 0.57 0.67 0.46 (−0.46, 1.38) 0.26 (−0.66, 1.19) −0.20 (−1.12, 0.73)
Cu 217 (95.7–428) 572 (110–1497) 370 (208–546) 0.19 0.07 0.40 0.32 0.75 (−0.07, 1.56) 0.34 (−0.47, 1.15) −0.41 (−1.22, 0.41)
Fe 103 (45.4–194) 182 (106–206) 65.7 (32.6–4534) 0.56 0.30 0.81 0.43 0.40 (−0.38, 1.19) 0.09 (−0.69, 0.88) −0.31 (−1.09, 0.47)
Mn 29.9 (20.2–48.3) 52.7 (17.7–91.4) 32.7 (20.3–54.8) 0.32 0.20 0.95 0.18 0.33 (−0.18, 0.83) −0.01 (−0.52, 0.49) −0.34 (−0.84, 0.16)
Ni 882 (501–1643) 920 (489–2358) 1793 (881–2970) 0.08 0.68 0.03 0.08 0.19 (−0.71, 1.09) 0.98 (0.08, 1.88) 0.79 (−0.11, 1.69)
Pb 175 (117–488) 619 (92.1–827) 195 (151–398) 0.47 0.34 0.86 0.26 0.34 (−0.37, 1.05) −0.06 (−0.77, 0.65) −0.41 (−1.12, 0.30)
Sb 3.07 (2.06–4.52) 4.64 (3.42–6.75) 2.59 (2.18–4.29) 0.007 0.005 0.82 0.009 0.59 (0.19, 0.99) 0.05 (−0.36, 0.45) −0.54 (−0.95, −0.14)
Sn 171 (14.3–290) 99.5 (51.3–319) 134 (9.85–369) 0.50 0.41 0.76 0.26 0.42 (−0.60, 1.44) −0.16 (−1.18, 0.86) −0.58 (−1.60, 0.44)
U 0.05 (0.05–0.05) 0.05 (0.05–0.05) 0.05 (0.05–0.05) 0.12 0.30 0.30 0.54 −0.02 (−0.05, 0.02) 0.02 (−0.01, 0.05) 0.03 (0.001, 0.06)
W 0.19 (0.05–0.30) 0.08 (0.05–0.15) 0.12 (0.05–0.17) 0.59 0.40 0.94 0.36 −0.32 (−1.08, 0.44) 0.03 (−0.73, 0.79) 0.35 (−0.41, 1.11)
Zn 2346 (1799–3804) 3913 (2624–6279) 3039 (2032–4240) 0.55 0.28 0.62 0.55 0.31 (−0.26, 0.88) 0.14 (−0.43, 0.71) −0.17 (−0.74, 0.40)
Open in a new tab

Figure 2.

Figure 2.

Open in a new tab

Metal concentrations in aerosols by nicotine content in closed-system devices (left) and open-system devices (right). The horizontal lines within boxes indicate medians; boxes, interquartile ranges; whiskers, values within 1.5 times the interquartile range from boxes; solid circles outside the box, outlier data values.

For open-system devices, we found significant differences in Cd and Sb concentrations among aerosols with nicotine content of 0, 6, and 24 mg/mL (Table 3). Median Cd and Sb concentrations increased 4.60 and 1.51-fold when the nicotine content increased from 0 to 6 mg/mL, and then decreased 4.60 and 1.79-fold when the nicotine content increased from 6 to 24 mg/mL.

3.3. Aerosol metal concentrations by puff duration

For closed-system devices, we found no significant differences in metal concentrations between puff durations (1, 2, and 4 s) (Table 4; Fig 3), except for Sb which decreased from 0.21 to 0.03 μg/kg when the puff duration increased from 1 s to 2 s, and then increased to 0.07 μg/kg when puff duration increased to 4 s. Similarly, there were no overall significant differences in metal concentrations between puff durations for open-system devices.

Table 4.

Median (interquartile range) metal concentrations (μg/kg) in aerosols by puff duration.

Median (IQR) in CD Overall p-value Linear regression p-value Beta value (95% CI)
1 s (n=6) 2 s (n=6) 4 s (n=6) 1 vs. 2 s 1 vs. 4 s 2 vs. 4 s 1 vs. 2 s 1 vs. 4 s 2 vs. 4 s
Al 4.14 (3.38–14.7) 4.43 (4.18–4.89) 7.98 (4.89–32.8) 0.20 0.52 0.24 0.08 −0.32 (−1.37, 0.72) 0.60 (−0.45, 1.65) 0.92 (−0.13, 1.97)
As 0.21 (0.15–1.59) 0.21 (0.10–0.33) 0.11 (0.07–0.18) 0.12 0.23 0.04 0.35 −0.73 (−1.96, 0.51) −1.28 (−2.52, −0.05) −0.56 (−1.79, 0.68)
Cd 0.05 (0.03–0.15) 0.18 (0.09–0.25) 0.43 (0.20–0.51) 0.14 0.35 0.05 0.26 0.63 (−0.77, 2.04) 1.40 (−0.006, 2.81) 0.77 (−0.64, 2.18)
Cr 0.51 (0.16–2.50) 0.20 (0.09–0.42) 0.15 (0.12–0.25) 0.35 0.23 0.20 0.92 −0.91 (−2.46, 0.64) −0.98 (−2.54, 0.57) −0.07 (−1.63, 1.48)
Cu 27.7 (3.82–82.6) 13.3 (2.98–28.6) 20.7 (8.55–33.8) 0.83 0.55 0.82 0.71 −0.53 (−2.41, 1.34) −0.20 (−2.08, 1.68) 0.34 (−1.54, 2.21)
Fe 3.97 (3.69–4.46) 3.69 (3.66–3.75) 4.03 (3.70–4.14) 0.32 0.15 0.29 0.69 −0.31 (−0.75, 0.13) −0.23 (−0.67, 0.21) 0.08 (−0.35, 0.52)
Mn 0.25 (0.04–1.41) 0.03 (0.03–0.10) 0.07 (0.05–0.28) 0.39 0.18 0.43 0.55 −1.23 (−3.07, 0.62) −0.70 (−2.54, 1.14) 0.53 (−1.32, 2.37)
Ni 12.7 (8.14–18.0) 7.67 (2.91–17.1) 6.89 (0.57–17.7) 0.26 0.49 0.11 0.33 −0.62 (−2.52, 1.27) −1.52 (−3.41, 0.38) −0.89 (−2.79, 1.00)
Pb 6.39 (1.14–12.49) 14.6 (1.42–47.4) 5.30 (3.95–5.96) 0.77 0.55 0.98 0.53 0.55 (−1.37, 2.47) −0.02 (−1.95, 1.90) −0.57 (−2.50, 1.35)
Sb 0.21 (0.08–0.39) 0.03 (0.03–0.03) 0.07 (0.04–0.42) 0.03 0.01 0.43 0.05 −1.87 (−3.23, −0.51) −0.52 (−1.88, 0.84) 1.35 (−0.01, 2.70)
Sn 1.62 (0.74–3.09) 1.85 (0.51–3.04) 3.45 (2.14–4.55) 0.43 0.92 0.29 0.25 −0.06 (−1.28, 1.16) 0.63 (−0.59, 1.86) 0.69 (−0.53, 1.92)
U 0.03 (0.03–0.03) 0.03 (0.03–0.03) 0.04 (0.03–0.04) 0.25 0.64 0.24 0.11 −0.02 (−0.13, 0.08) 0.06 (−0.05, 0.17) 0.08 (−0.02, 0.19)
W 0.03 (0.03–0.22) 0.03 (0.03–0.03) 0.04 (0.04–0.04) 0.25 0.11 0.23 0.69 −0.66 (−1.49, 0.18) −0.49 (−1.33, 0.34) 0.16 (−0.67, 1.00)
Zn 306 (268–510) 371 (285–1391) 442 (340–488) 0.80 0.53 0.87 0.64 0.32 (−0.73, 1.36) 0.08 (−0.96, 1.13) −0.24 (−1.28, 0.81)
Median (IQR) in OD Overall p-value Linear regression p-value Beta value (95% CI)
1 s (n=6) 2 s (n=6) 4 s (n=6) 1 vs. 2 s 1 vs. 4 s 2 vs. 4 s 1 vs. 2 s 1 vs. 4 s 2 vs. 4 s
Al 9.34 (6.62–18.4) 7.87 (4.61–10.4) 4.18 (3.87–5.62) 0.27 0.41 0.11 0.41 −0.38 (−1.33, 0.57) −0.76 (−1.71, 0.19) −0.38 (−1.33, 0.57)
As 1.21 (0.12–2.78) 3.41 (2.27–3.86) 3.05 (2.64–3.38) 0.07 0.05 0.04 0.94 1.53 (0.01, 3.06) 1.59 (0.06, 3.11) 0.05 (−1.47, 1.58)
Cd 0.44 (0.24–0.54) 0.23 (0.17–0.77) 0.40 (0.19–0.62) 0.86 0.71 0.61 0.89 −0.22 (−1.44, 1.00) −0.30 (−1.52, 0.92) −0.08 (−1.30, 1.14)
Cr 26.0 (3.76–63.1) 19.3 (3.17–94.8) 25.3 (2.32–58.1) 0.93 0.86 0.71 0.84 −0.19 (−2.58, 2.19) −0.42 (−2.80, 1.97) −0.23 (−2.61, 2.15)
Cu 893 (164–2203) 735 (170–3157) 732 (120–1593) 0.88 0.92 0.63 0.71 −0.10 (−2.17, 1.96) −0.47 (−2.54, 1.59) −0.37 (−2.44, 1.69)
Fe 143 (15.5–377) 107 (14.9–569) 143 (6.0–364) 0.93 0.86 0.71 0.84 −0.24 (−3.14, 2.66) −0.52 (−3.42, 2.38) −0.28 (−3.18, 2.62)
Mn 39.5 (32.2–55.3) 33.5 (30.8–48.1) 24.4 (20.3–29.4) 0.24 0.39 0.10 0.40 −0.29 (−0.96, 0.39) −0.56 (−1.24, 0.12) −0.28 (−0.96, 0.40)
Ni 7848 (6143–10428) 4927 (4226–10241) 3914 (2487–5470) 0.09 0.28 0.03 0.23 −0.43 (−1.24, 0.38) −0.91 (−1.72, −0.10) −0.48 (−1.29, 0.33)
Pb 267 (235–367) 210 (208–556.3) 252 (140–366) 0.64 0.74 0.36 0.56 −0.15 (−1.07, 0.77) −0.40 (−1.32, 0.51) −0.26 (−1.18, 0.66)
Sb 2.37 (1.72–3.30) 5.17 (3.76–5.62) 4.73 (2.92–4.94) 0.39 0.17 0.47 0.51 0.46 (−0.23, 1.15) 0.24 (−0.45, 0.93) −0.22 (−0.91, 0.47)
Sn 43.8 (33.7–54.0) 30.1 (21.7–89.9) 39.4 (13.0–65.8) 0.65 0.66 0.36 0.63 −0.24 (−1.37, 0.90) −0.50 (−1.63, 0.63) −0.26 (−1.40, 0.87)
U 0.04 (0.04–0.04) 0.04 (0.04–0.04) 0.04 (0.04–0.04) 0.89 0.64 0.76 0.87 −0.009 (−0.05, 0.03) −0.006 (−0.05, 0.03) 0.003 (−0.04, 0.04)
W 0.04 (0.04–0.04) 0.04 (0.04–0.04) 0.04 (0.04–0.04) 0.38 0.23 0.24 0.98 −0.26 (−0.69, 0.18) −0.25 (−0.68, 0.18) 0.006 (−0.43, 0.44)
Zn 3543 (2995–4314) 2705 (2246–6640) 3674 (1793–5124) 0.78 0.69 0.49 0.78 −0.17 (−1.05, 0.71) −0.29 (−1.17, 0.59) −0.12 (−1.00, 0.76)
Open in a new tab

Figure 3.

Figure 3.

Open in a new tab

Metal concentrations in aerosols by puff duration in closed-system devices (left) and open-system devices (right). The horizontal lines within boxes indicate medians; boxes, interquartile ranges; whiskers, values within 1.5 times the interquartile range from boxes; solid circles outside the box, outlier data values.

4. Discussion

Aerosol metal concentrations depended on e-cigarette flavor and nicotine content but not on puff duration. Among fruit, tobacco, and menthol flavors in both closed-system and open-system devices, concentrations of As, Fe, and Mn varied significantly. In closed-system devices, concentrations of Al, Fe, Sn, and U were significantly higher for tobacco or menthol flavors compared to fruit flavors; in open-system devices, aerosol W concentrations were significantly higher for tobacco flavors compared to fruit flavors. Even though actions have been taken to remove some fruit flavors appealing to youth from the market, we found a higher risk of Al, Fe, Sn, U, and W exposure associated with tobacco or menthol flavored products, which are still available in the market. Tobacco flavored e-liquids may result in higher doses of As, Fe, and Mn compared to menthol flavored e-liquids when using both types of devices. Aerosol metal concentrations varied with nicotine content; however, no universal trend was found for different metals. Lead concentrations decreased significantly when the nicotine content increased from 0 to 59 mg/mL, while Ni aerosol concentrations were higher for 59 mg/mL nicotine content compared to 24 mg/mL for closed-system devices. However, this comparison by nicotine level is hampered as it is based on data from two different devices; it is possible that device design, rather than nicotine concentration, is related to Pb emissions. For both types of devices, puff duration had no effect on metal emissions.

In January 2020, the US FDA announced action to ban the sale of flavored e-cigarette cartridges that appeal to children, such as fruit and mint flavors, but FDA exempted menthol and tobacco flavors as well as flavored e-liquids sold for use in open tank systems (FDA, 2020). Based on our analysis, users of tobacco or menthol flavors could experience exposure to higher metal emissions, especially for Al, Fe, and Sn, which could result in a range of adverse health effects. Inhalation of Al has been related to adverse effects on lungs and the nervous system; respiratory effects including impaired lung function and fibrosis have been observed in workers exposed to Al (ATSDR, 2008). Iron inhalation can result in respiratory irritation, metal fume fever, siderosis, and fibrosis (Johnson et al., 1985). High levels of Fe were commonly detected in the aerosol of e-cigarette devices with a coil made of Kanthal, which contains mainly Al, Cr, and Fe (Williams et al., 2017). High levels of Sn have been detected in aerosol when solder joints and copper wires were coated with Sn (Williams et al., 2015). Exposure to Sn oxide dust and fumes has been related to non-fibrosing pneumoconiosis, among other health effects (Güllü et al., 2005).

Lead levels in the banned fruit flavored aerosols were higher than those in tobacco flavors. Thus, the action of FDA to remove fruit flavored e-cigarette from the market could help with reducing exposure to some metals including Pb, which is neurotoxic, particularly for youth and young adults with developing brains (Lin et al., 2006).

Our results suggest that e-cigarette users who use tobacco flavored e-liquid may be exposed to higher As, Fe, and Mn levels than users of menthol flavors, a finding that is consistent for both types of devices evaluated (open-system and closed-system devices). Thus, there is high possibility that elevated metal levels in tobacco flavored aerosols are due to flavorings themselves. Both tobacco and menthol are available in the US market, although possible regulation of menthol is considered by the FDA. Inhalation of As is related to multiple adverse health effects, such as cardiovascular diseases and lung cancer (Smith et al., 2009; ATSDR, 2007). The unused e-liquid has been reported to be the main source of e-cigarette As emissions, because As concentrations in e-liquid and aerosol samples were found to be similar (Olmedo et al., 2018). However, a recent study focusing on As speciation in e-cigarettes found that the concentration of arsenite (AsIII) was significantly higher in the aerosol condensate than the e-liquid (3.27 vs. 1.08 μg/kg) (Liu et al., 2020), with higher toxicity of AsIII compared to other As species. Inhaling Mn can result in permanent neurological disorder with symptoms including tremors, difficulty walking, and facial muscle spasms (ASTDR, 2012). Stainless steel contains various elements (e.g. Mn, Si, Cr, Ni) in varying proportions to manipulate the steel’s properties; coils made of stainless steel (used in OD2) can therefore be a source of Mn. Thus, users of tobacco flavored e-cigarettes with stainless steel coils could potentially be at an increased risk of developing neurological disorders (Re et al., 2021). However, whether these health effects are relevant at the detected emission levels is still uncertain. In a previous study, we estimated that MRLs for metal inhalation were only exceeded for open-system devices (mods) but not for closed-system devices. Also, in a recent study of e-cig users in Spain, metal aerosol levels were not correlated with metal levels in urine or hair (Olmedo et al., 2021).

Metal emission variability within the same flavor group could depend on the e-liquid brand, likely due to differences in formulas and ingredient purity used to produce the flavor. This is consistent with two studies which analyzed chemical composition of e-liquids from different manufacturers, showing differences in e-liquid composition (Bahl et al., 2012; Behar et al., 2014). Differences in in-vitro cytotoxicity of certain flavored e-liquids also support that e-liquid ingredients for a given flavor group can be variable (Bahl et al., 2012; Farsalinos et al., 2013b). Physico-chemical interactions at the heating surface could also be the potential cause for this variation. Another issue is that we grouped similar flavors together based on product labels; however, flavors with the same marketing names may not be chemically similar. Thus, flavor by itself may not be a good predictor for aerosol metal levels; the highly variable composition of e-liquids and brand-specific toxicological effects suggest a need for better regulation of the use of flavoring agents in the production of e-liquids.

Aerosol Ni emissions increased significantly when nicotine content increased from 24 to 59 mg/mL. Notably, among the nicotine contents examined for the CDs, the highest Ni levels occurred for the JUUL device, which is currently the most popular e-cigarette among young adults (Krishnan-Sarin et al., 2019), for the highest nicotine content of 59 mg/mL. These Ni levels could be attributed to the Ni-containing heating coil of the JUUL; however, our data suggest that nicotine levels modify Ni aerosol levels, even though it is not clear which chemical processes are responsible for the potential modification of Ni levels by nicotine.

Similarly, one study reported that levels of metals including As, Cr, Ni, Cu, Sb, Sn, and Zn varied widely across nicotine and non-nicotine containing aerosols of cigalike devices, but different metals showed unique trends with increasing nicotine content (Mikheev et al., 2016). The unclear effect of nicotine on metal emissions could be caused by inaccurate labeling of nicotine content by manufacturers. E.g., analysis of 35 popular e-juices and refill cartridges showed that nearly half were inaccurately labeled for nicotine content, even for e-liquids labeled as nicotine-free (Goniewicz et al., 2013). A recent study showed that measured nicotine contents were 1.2 times higher than those provided by the manufacturers (Lee et al., 2020). The effects of nicotine content on aerosol metal levels we observed are supported by a study of environmental e-cigarette aerosol exposure that found increased aerosol FeNO levels (a marker of eosinophilic airway inflammation) when e-liquids with higher nicotine content were vaped (Schober et al., 2014). To protect e-cigarette users, e-liquid labels should contain accurate information about nicotine content and appropriate warnings about potential health effects, particularly with regard to toxicity and addiction.

In 2014, the European Union (EU) Tobacco Product Directive instructed its member states to ensure that the nicotine content of e-liquids sold in the EU may not exceed 20 mg/mL (European Commission, 2014). The JUUL products we examined were purchased in the US and have ~3 fold higher nicotine levels than those sold in the EU (Hammond et al., 2021); however, JUUL products with lower nicotine content were not included in this study.

Puff duration may vary by user, with inexperienced users taking shorter puffs on the order of 2 s and experienced users taking longer puffs of an average 4 s (Farsalinos et al., 2015; Talih et al., 2015). To our knowledge, this is the first study to evaluate the effect of puff duration on metal release from e-cigarettes. We did not observe a significant difference in metal concentrations in aerosols for different puff durations. In contrast, one study, which evaluated the effect of puff duration on organic compounds, found that longer puff duration (4 vs. 2 s) generated more particles and higher concentrations of organic species, since longer puff duration led to higher heating coil temperature (Zhao et al., 2018). More studies are needed to confirm that organic and inorganic chemicals behave differently during the heating process depending on puff duration.

The detection of cases of e-cigarette, or vaping, product use-associated lung injury (EVALI) in 2019–2020 attracted considerable attention of the public (Kalininskiy et al., 2019). As of January 2020, the US Centers for Disease Control (CDC) had identified 2668 hospitalized patients diagnosed with EVALI in the US (Krishnasamy et al., 2020). Even though most cases were associated with vaping of e-liquids containing tetrahydrocannabinol (THC) and the carrier liquid vitamin E acetate which is typically not used in the traditional nicotine-containing e-liquids, a small number of cases were reported to not have used these compounds. E.g., one study described a patient who developed giant cell interstitial pneumonia because of cobalt inhalation from e-cigarette use, demonstrating that insidious diseases can arise from traditional e-cigarette use (Elliott et al., 2019). Increasing evidence also shows associations between metal emissions from e-cigarettes and long-term adverse health effects like lung and sinonasal cancer (Badea et al., 2018; Gaur et al., 2019; Fowles et al., 2020), highlighting the need for further studies to determine potential chronic health effects related to metal exposure from e-cigarette use.

This study has several limitations. First, even though we studied the most commonly available flavors including tobacco, menthol, and fruit, there is high possibility that different manufacturers use different formulas and different ingredient purity to produce the same flavor. E-liquids with the same flavor but from different manufacturers should be compared to examine whether certain flavors are systematically associated with elevated metal levels. Second, we used nicotine content provided by manufacturers directly. E-liquid nicotine content should be measured in future studies on the effect of nicotine content on metal emissions. Third, we did not conduct metal speciation in our study. This is especially important for understanding the potential health hazards of Cr, with Cr(VI) being an established carcinogen while Cr(III) is an essential metal. Last, there could be country-specific differences between e-cigarette devices and e-liquids even if the manufacturer model number is the same. Therefore, results from this study, which was performed with e-cigarette products purchased on the US market, should be generalized to other countries with caution.

Despite some limitations, our study systematically evaluated some of the factors that can affect metal concentrations including e-liquid flavor, nicotine content, and puff duration, representing a variety of scenarios related to human vaping. The findings support that e-cigarette use is of concern due to toxic metals exposure, although aerosol from some devices had markedly lower levels of certain metals, suggesting that it is possible to reduce metal levels in e-cigarette aerosol. Future research should evaluate whether these emissions result in metal exposure levels and health effects in human populations, either as a result of primary or secondhand exposure.

5. Conclusions

Overall, e-cigarettes are a source of exposure to a wide variety of toxic metals. Tobacco and menthol flavored aerosols are still available in the US market and can result in potential higher exposures to Al, Fe, Sn, U, and W compared to other flavors. Metal levels vary widely across different nicotine contents, even though no clear trends in metal emissions were found with increasing nicotine content. Our findings do not support a significant impact of puff duration on aerosol metal levels. Due to potential toxicity of metal exposure in e-cigarette emissions, strict product regulations should be developed to inform policy at national and international levels, especially on e-liquid composition and nicotine content.

Acknowledgments

This study was supported by NIEHS/FDA grant 1R21ES029777–01 and R01ES030025, NIEHS grants P30 ES009089, R01ES029967, 1R01ES032954, and a Johns Hopkins University Technology Transfer Seed Award. DZ was supported by the China Scholarship Council (201706190116).

References

  1. Aherrera A, Olmedo P, Grau-Perez M, Tanda S, Goessler W, Jarmul S, Chen R, Cohen JE, Rule AM, Navas-Acien A. 2017. The association of e-cigarette use with exposure to nickel and chromium: A preliminary study of non-invasive biomarkers. Environ Res 159:313–320. [DOI] [PubMed] [Google Scholar]
  2. Ambrose BK, Day HR, Rostron B, Conway KP, Borek N, Hyland A, Villanti AC. 2015. Flavored tobacco product use among US youth aged 12–17 years, 2013–2014. JAMA 314:1871–1873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. ATSDR (Agency for Toxic Substances and Disease Registry), 2007. Toxicological profile for Arsenic. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA. https://www.atsdr.cdc.gov/ToxProfiles/tp2.pdf, Accessed date: 1 November 2020 [Google Scholar]
  4. ATSDR (Agency for Toxic Substances and Disease Registry), 2008. Toxicological Profile for Aluminum. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA. https://www.atsdr.cdc.gov/toxprofiles/tp22.pdf, Accessed date: 1 November 2020 [Google Scholar]
  5. ATSDR (Agency for Toxic Substances and Disease Registry), 2012. Toxicological Profile for Manganese. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA. http://www.atsdr.cdc.gov/ToxProfiles/tp.asp?id=102&tid=23, Accessed date: 1 November 2017. [Google Scholar]
  6. Badea M, Luzardo O, Gonzales-Antuna A, Zumbado M, Rogozea L, Floroian L, Alexandrescu D, Moga M, Gaman L, Radoi M. 2018. Body burden of toxic metals and rare earth elements in non-smokers, cigarette smokers and electronic cigarette users. Environ Res 166: 269–275. [DOI] [PubMed] [Google Scholar]
  7. Bahl V, Lin S, Xu N, Davis B, Wang YH, Talbot P. 2012. Comparison of electronic cigarette refill fluid cytotoxicity using embryonic and adult models. Reprod Toxicol 34:529–537. [DOI] [PubMed] [Google Scholar]
  8. Behar RZ, Davis B, Wang Y, Bahl V, Lin S, Talbot P. 2014. Identification of toxicants in cinnamon-flavored electronic cigarette refill fluids. Toxicol In Vitro 28:198–208. [DOI] [PubMed] [Google Scholar]
  9. Cameron JM, Howell DN, White JR, Andrenyak DM, Layton ME, Roll JM. 2014. Variable and potentially fatal amounts of nicotine in e-cigarette nicotine solutions. Tob Control 23:77–78. [DOI] [PubMed] [Google Scholar]
  10. Elliott DRF, Shah R, Hess CA, Elicker B, Henry TS, Rule AM, Chen R, Golozar M, Jones KD. 2019. Giant cell interstitial pneumonia secondary to cobalt exposure from e-cigarette use. Eur Respir J 54(6):1901922. [DOI] [PubMed] [Google Scholar]
  11. European Commission. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing directive 2001/37/EC; contract No. 1, 2014. Available: www.eur-lex.europa.eu.
  12. Farsalinos KE, Romagna G, Tsiapras D, Kyrzopoulos S, Voudris V. 2013a. Evaluation of electronic cigarette use (vaping) topography and estimation of liquid consumption: Implications for research protocol standards definition and for public health authorities’ regulation. Int J Environ Res Public Health 10:2500–2514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Farsalinos KE, Romagna G, Allifranchini E, Ripamonti E, Bocchietto E, Todeschi S, Tsiapras D, Kyrzopoulos S, Voudris V. 2013b. Comparison of the cytotoxic potential of cigarette smoke and electronic cigarette vapour extract on cultured myocardial cells. Int J Environ Res Public Health 10:5146–5162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Farsalinos KE, Spyrou A, Stefopoulos C, Tsimopoulou K, Kourkoveli P, Tsiapras D, Kyrzopoulos S, Poulas K, Voudris V. 2015. Nicotine absorption from electronic cigarette use: comparison between experienced consumers (vapers) and naïve users (smokers). Sci Rep 5:11269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. FDA, 2020. FDA finalizes enforcement policy on unauthorized flavored cartridge-based e-cigarettes that appeal to children, including fruit and mint. https://www.fda.gov/news-events/press-announcements/fda-finalizes-enforcement-policy-unauthorized-flavored-cartridge-based-e-cigarettes-appeal-children.
  16. Fowles J, Barreau T, and Wu N. 2020. A review and assessment of cancer and non-cancer risks from metal concentrations in e-cigarette liquids and aerosols. Int J Environ Res Public Health 17:2146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gaur S, Agnihotri R. 2019. Health effects of trace metals in electronic cigarette aerosols-a systematic review. Biol Trace Elem Res 188: 295–315. [DOI] [PubMed] [Google Scholar]
  18. Goniewicz ML, Kuma T, Gawron M, Knysak J, Kosmider L. 2013. Nicotine levels in electronic cigarettes. Nicotine Tob Res 15:158–166. [DOI] [PubMed] [Google Scholar]
  19. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, Prokopowicz A, Jablonska-Czapla M, Rosik-Dulewska C, Havel C, Jacob P, Benowitz N. 2014. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control 23(2):133–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Güllü E, Karnak D, Kayacan O, Beder S. 2005. A tinner with stannosis and tuberculosis. Case Rep Clin Pract Rev 6:73–76. [Google Scholar]
  21. Hammond D, Reid JL, Burkhalter R, O’Connor RJ, Goniewicz ML, Wackowski OA, Thrasher JF, Hitchman SC. 2021. Trends in e-cigarette brands, devices and the nicotine profile of products used by youth in England, Canada and the USA: 2017–2019. Tob Control 0:1–11. doi: 10.1136/tobaccocontrol-2020-056371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hess CA, Olmedo P, Nbavas-Acien A, Goessler W, Cohen JE, Rule AM. 2017. E-cigarettes as a source of toxic and potentially carcinogenic metals. Environ Res 152:221–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hilpert M, Ilievski V, Coady M, Andrade-Gutierrez M, Yan B, Chillrud SN, Navas-Acien A, Kleiman NJ. 2019. A custom-built low-cost chamber for exposing rodents to e-cigarette aerosol: practical considerations. Inhal Toxicol 31(11–12):399–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hilpert M, Ilievski V, Hsu SY, Rule AM, Olmedo P, Drazer G. 2021. E-cigarette aerosol collection using converging and straight tubing sections: Physical mechanisms. J Colloid Interface Sci 584:804–815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hua M, Yip H, Talbot P. 2013. Mining data on usage of electronic nicotine delivery systems (ENDS) from YouTube videos. Tob Control 22(2):103–106. [DOI] [PubMed] [Google Scholar]
  26. IARC (International Agency for Research on Cancer). 2012a. Chromium (VI) compounds. IARC Monographs 100C:147–167. [Google Scholar]
  27. IARC, 2012b. Nickel and nickel compounds. IARC Monographs 100C: 169–218. [Google Scholar]
  28. Jaishankar M, Tseten T, Anbalagan N, Mathew BB, Beeregowda KN. 2014. Toxicity, mechanism and health effects of some heavy metals. Interdiscip Toxicol 7(2):60–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Johnson A, Moira CY, MacLean L, Atkins E, Dybuncio A, Cheng F, Enarson D. 1985. Respiratory abnormalities among workers in an iron and steel foundry. Br J Ind Med 42(2): 94–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kalininskiy A, Bach CT, Nacca NE, Ginsberg G, Marraffa J, Navarette KA, McGraw MD, Croft DP. 2019. E-cigarette, or vaping, product use associated lung injury (EVALI): Case series and diagnostic approach. Lancet Respir Med 7:1017–1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kavuluru R, Han S, Hahn EJ. 2019. On the popularity of the USB flash drive-shaped electronic cigarette JUUL. Tob Control 28:110–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kosmider L, Sobczak A, Prokopowicz A, Kurek J, Zaciera M, Knysak J, Smith D, Goniewicz ML. 2016. Cherry-flavoured electronic cigarettes expose users to the inhalation irritant, benzaldehyde. Thorax 71:376–377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Krishnan-Sarin S, Jackson A, Morean M, Kong G, Bold KW, Camenga DR, Cavallo DA, Simon P, Wu R. 2019. E-cigarette devices used by high-school youth. Drug Alcohol Depend 194:395–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Krishnasamy VP, Hallowell BD, Ko JY. 2020. Update: Characteristics of a nationwide outbreak of e-cigarette, or vaping, product use–associated lung injury-United States, August 2019–January 2020. MMWR Morb Mortal Wkly Rep 69:90–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Laverty AA, Vardavas CI, Filippidis FT. 2016. Design and marketing features influencing choice of e-cigarettes and tobacco in the EU. Eur J Public Health 26:838–841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lee YJ, Na CJ, Botao L, Kim KH, Son YS. 2020. Quantitative insights into major constituents contained in or released by electronic cigarettes: Propylene glycol, vegetable glycerin, and nicotine. Sci Total Environ 703:134567. [DOI] [PubMed] [Google Scholar]
  37. Lin JL, Lin-Tan DT, Li YJ, Chen KH, Huang YL. 2006. Low-level environmental exposure to lead and progressive chronic kidney diseases. Am J Med 119(8):707.e1–9. [DOI] [PubMed] [Google Scholar]
  38. Liu Q, Huang C, Chris Le X. 2020. Arsenic species in electronic cigarettes: Determination and potential health risk. J Environ Sci 91:168–176. [DOI] [PubMed] [Google Scholar]
  39. Maziak W, Rastam S, Shihadeh AL, Bazzi A, Ibrahim I, Zaatari GS, Ward KD, Eissenberg T. 2011. Nicotine exposure in daily waterpipe smokers and its relation to puff topography. Addict Behav 36:397–399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mikheev VB, Brinkman MC, Granville CA, Gordon SM, Clark PI. 2016. Real-time measurement of electronic cigarette aerosol size distribution and metals content analysis. Nicotine Tob Res 18:1895–1902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Na C, Jo S, Kim K, Sohn J, Son Y. 2019. The transfer characteristics of heavy metals in electronic cigarette e-liquid. Environ Res 174: 152–159. [DOI] [PubMed] [Google Scholar]
  42. Navas-Acien A, Martinez-Morata I, Hilpert M, Rule A, Shimbo D, LoIacono NJ. 2020. Early cardiovascular risk in e-cigarette users: the potential role of metals. Curr Environ Health Rep 7(4):353–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Olmedo P, Navas-Acien A, Hess C, Jarmul S, Rule A. 2016. A direct method for e-cigarette aerosol sample collection. Environ Res 149:151–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Olmedo P, Goessler W, Tanda S, Grau-Perez M, Jarmul S, Aherrera A, Chen R, Hilpert M, Cohen JE, Navas-Acien A, Rule AM. 2018. Metal concentrations in e-cigarette liquid and aerosol samples: the contribution of metallic coils. Environ Health Perspect 126(2):027010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Olmedo P, Rodrigo L, Grau-Pérez M, Hilpert M, Navas-Acién A, Téllez-Plaza M, Pla A, Gil F. 2021. Metal exposure and biomarker levels among e-cigarette users in Spain. Environ Res 202:111667. [DOI] [PubMed] [Google Scholar]
  46. Re DB, Hilpert M, Saglimbeni B, Strait M, Ilievski V, Coady M, Talayero M, Wilmsen K, Chesnais H, Balac O, Glabonjat RA, Slavkovich V, Yan B, Graziano J, Navas-Acien A, Kleiman NJ. 2021. Exposure to e-cigarette aerosol over two months induces accumulation of neurotoxic metals and alteration of essential metals in mouse brain. Environ Res 202: 111557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Reilly SM, Goel R, Bitzer Z, Elias RJ, Foulds J, Muscat J, Richie JP. 2017. Effects of topography-related puff parameters on carbonyl delivery in mainstream cigarette smoke. Chem Res Toxicol 30:1463–1469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Schober W, Szendrei K, Matzen W, Osiander-Fuchs H, Heitmann D, Schettgen T, Jörres RA, Fromme H. 2014. Use of electronic cigarettes (e-cigarettes) impairs indoor air quality and increases FeNO levels of e-cigarette consumers. Int J Hyg Environ Health. 217(6):628–637. [DOI] [PubMed] [Google Scholar]
  49. Smith AH, Ercumen A, Yuan Y, Steinmaus CM. 2009. Increased lung cancer risks are similar whether arsenic is ingested or inhaled. J Expo Anal Nv Epid 19:343–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Talih S, Balhas Z, Eissenberg T, Salman R, Karaoghlanian N, El Hellani A, Baalbaki R, Saliba N, Shihadeh A. 2015. Effects of user puff topography, device voltage, and liquid nicotine concentration on electronic cigarette nicotine yield: measurements and model predictions. Nicotine Tob Res 17(2):150–157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Williams M, To A, Bozhilov K, Talbot P. 2015. Strategies to reduce tin and other metals in electronic cigarette aerosol. PloS one 10:e0138933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Williams M, Bozhilov K, Ghai S, Talbot P. 2017. Elements including metals in the atomizer and aerosol of disposable electronic cigarettes and electronic hookahs. PloS one 12:e0175430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Williams M, Li J, Talbot P. 2019. Effects of model, method of collection, and topography on chemical elements and metals in the aerosol of tank-style electronic cigarettes. Sci Rep 9:13969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zhao D, Navas-Acien A, Ilievski V, Slavkovich V, Olmedo P, Adria-Mora B, Domingo-Relloso A, Aherrera A, Kleiman NJ, Rule AM, Hilpert M. 2019. Metal concentrations in electronic cigarette aerosol: Effect of open-system and closed-system devices and power settings. Environ Res 174:125–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhao D, Aravindakshan A, Hilpert M, Olmedo P, Rule AM, Navas-Acien A, Aherrera A. 2020. Metal/metalloid levels in electronic cigarette liquids, aerosols, and human biosamples: a systematic review. Environ Health Perspect 128 (3): 036001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhao J, Nelson J, Dada O, Pyrgiotakis G, Kavouras IG, Demokritou P. 2018. Assessing electronic cigarette emissions: Linking physico-chemical properties to product brand, e-liquid flavoring additives, operational voltage and user puffing patterns. Inhal Toxicol 30:78–88. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES