Detection of Metal-Containing Particles in FDA-authorized and New Disposable Electronic Cigarette Aerosols

Abstract

Use of electronic cigarettes (e-cigs) among youth raises concerns about nicotine addiction and health risks, highlighting the urgency of understanding e-cig aerosol composition. While metals in e-cig aerosols are well-documented, the presence of metals as particles versus ions in newer disposable devices remains unclear. We examined metal particles in aerosols from disposable e-cigs and FDA-authorized devices.

E-cigs analyzed included Vuse Alto (FDA-authorized), EB Create, and Orion Bar. Aerosol condensate was analyzed for arsenic (As), cadmium (Cd), chromium (Cr), iron (Fe), nickel (Ni), lead (Pb), and zinc (Zn) using single particle-inductively coupled plasma-mass spectrometry. Bulk metal content, particle size, mass, and one-month stability were also assessed. Particle size was determined based on the assumption that particles contained the detected single metal alone and not the oxide.

All aerosols analyzed contained metals, with Cr, Ni, Fe, Pb, and Zn detected as particles. The FDA-authorized Vuse Alto Golden Tobacco device consistently exhibited the highest particle count for all metals except Cr. EB Create aerosols had the highest amount of Cr particles. Particle sizes ranged from 14.21 nm to 281.3 nm.

The widespread presence of metal particles in e-cig aerosols raises public health concerns and underscores the need for research to inform regulatory standards.

Graphical Abstract:

1. Introduction

The most recent US National Youth and Tobacco Survey estimated that 1.21 million youth (<18 years old) reported electronic cigarette (e-cig) use in 2024, with disposable devices being the most common type of device used among this population¹. Given the addictive nature of nicotine and our current understanding of the effects of e-cig use, there are significant concerns for long-term adverse health outcomes. E-cig devices typically use a metallic heating element, commonly made of alloys such as Nichrome, Kanthal, and stainless steel, to generate the aerosol. Several studies have detected metals such as lead (Pb), chromium (Cr), zinc (Zn) and nickel (Ni) in e-cig liquids and aerosols^2–6 and have determined several variables, such as battery power, type of device, nicotine and flavor, can affect the generation of metals in e-cig aerosols and thus vary widely across devices^4,7.

Inhalation of metal particles has been associated with an increased risk of cardiovascular and respiratory diseases⁸. Among other factors, metal toxicity depends on where particles deposit in the lung, which is influenced by particle size⁹. Nanosized particles have been shown to reach the alveoli and enter the circulation by crossing the lung barrier^10,11. After absorption, particles have been shown to accumulate in the liver, kidneys, and brain^12–17.

Indeed, nanosized particles have been shown to cross the blood-brain barrier ^16,17. Another pathway for particles entering the brain is the olfactory bulb ^14,18. A recent study detected increased levels of several metals, such as Cr, Cu, Fe, Mn, and Pb, in the central nervous system of mice following a two-month-long e-cig exposure ¹⁹. The fact that metal particles can end up in the brain is concerning, particularly for young e-cig users, given the known neurotoxicity of certain metals.

The metals in e-cig aerosols can be found either in ionic or particle form ^20,21, which are each transported into cells through different mechanisms. Metal ions require a transporter to cross the plasma membrane and are stringently regulated. For example, greater expression of the divalent metal transporter 1 (DMT1) occurs when cells are iron deficient ²². Particles, however, are taken up through cytosis, pinocytosis in epithelial cells, and phagocytosis in macrophages ²³. Whether a metal is present as a solid particle or an ion can affect its bioavailability in the lungs, thereby impacting toxicity and risk of differing health effects^24–27. For example, greater metal uptake, DNA damage, and cell death were observed in lung epithelial cell lines after exposure to CuO nanoparticles compared to CuCl₂ ions ²⁸. Interestingly, in the same cell line, AgNO₃ ions displayed greater toxicity compared to Ag-NPs ²⁸. Other factors, such as surface charge, hardness, aggregation, and particle size, can affect metal particle bioavailability ^{11,12,29–31}. Accordingly, the physical characteristics of metals need to be considered to have a more nuanced understanding of the potential adverse health effects of e-cig aerosols.

Currently, disposable e-cig devices dominate the market. Thus, there is an urgent need to characterize metals and other toxicants in the aerosols generated from these devices. Research characterizing metals in aerosols, however, has been conducted on older e-cig devices ^4,32–35. Moreover, no study has conducted single particle analysis of metals in newer disposable devices; detecting metal particles has only been done in older mod- and pod-style devices ^20,21.

Our study aims to fill this gap by utilizing single particle-inductively coupled plasma-mass spectrometry (SP-ICP-MS) to investigate the presence of metal particles in aerosol generated from several widely consumed disposable and FDA-authorized devices. This work explores potential differences in metal content, focusing on metal particles based on device type, flavor, and FDA-authorized status.

2. Methods

2.1. E-cig Devices and Collection

All e-cig devices were bought from local vape shops in Baltimore, Maryland. When the study was conducted, two brands, Orion Bar and EB Create (formerly known as Elf Bars), were selected as they were the two most popular e-cig brands, as identified by vape shop employees. Vuse Alto was chosen due to previous research on its metal particle content and its recent FDA-authorized status. Two flavors were identified for each brand by asking vape shop employees which flavors were most popular among their customers. The flavors chosen were Blueberry Raspberry and Cool Mint for Orion Bar, Blue Razz Ice and Gummy for EB Create, and Golden Tobacco and Menthol for Vuse Alto (only Golden Tobacco is FDA-authorized). Three devices were purchased for each brand and flavor combination, totaling eighteen samples.

Condensate was collected by vaping the e-cig devices as previously described³⁶. Briefly, a series of cut 250 μL pipette tips and 1.5 mm tubing were connected to 4.8 mm C-FLEX metal-free tubing (Masterflex Item # EW-06424–15), which in turn was connected to the e-cig device. A peristaltic pump was utilized to activate each device for 200 puffs, resulting in 1–2 g of condensate. Condensate was collected into a metal-free 15 mL plastic conical tube using a puff topography consisting of a 4-second inhalation at 1 L/min, followed by 30 seconds of rest. Condensate samples were prepared for metal analysis immediately after collection and stored at 4°C until ICP analysis. To quantify background levels, a 30:70 Propylene Glycol: Vegetable Glycerin mixture (PG/VG) was prepared and dripped into a metal-free 15 mL plastic conical tube via the same series of cut pipette tips and tubing used for aerosol collection described above (Figure S.1). Glass was not used at any point during sample collection, preparation or analysis of any of the samples.

2.2. Metal Analysis

Condensate samples were analyzed for seven metals: arsenic (As), cadmium (Cd), chromium (Cr), iron (Fe), nickel (Ni), lead (Pb), and zinc (Zn) using an Inductively Coupled Plasma-Mass Spectrometer (ICP-MS, 8900 triple quad (ICP-QQQ-MS) Agilent Technologies, US) operated in bulk analysis mode or single particle mode in combination with a single cell sample introduction kit (cytospray and cytoneb, Elemental Scientific, Inc., US) and autosampler with syringe-driven sample introduction (microFAST Single Cell, Elemental Scientific, Inc., US). These metals were chosen because of their known toxicity^37–43, their presence in heating elements ^35,44–47, and previous work by our lab^2,4,6,48. Condensate samples for bulk analysis were diluted 1:10 (w/v) with 2% Optima HNO₃ and stored at 4°C until analysis. A multi-element (Bi, Ge, In, Li-6, Rh, Sc, Tb, Y) internal standard from LGC Standards (Teddington, UK) was employed. Trace elements in water standard reference material (SRM 1643f) from the National Institute of Standards and Technology (NIST) were utilized as the quality control check, using a bracketed approach before and after sample analysis. In addition, seven blanks were measured at the beginning of each run to determine the limit of detection (LOD) calculations. All calibration curves were prepared using a multi-element standard from High Purity Standards (Charleston, SC).

Condensate samples were prepared for single-particle analysis by diluting the condensate 1:10 (w/v) with MQ water. Dilution and analysis were performed within twenty-four hours and one month after collection. Standards for single-particle analysis included 46 nm platinum (Pt) at 1 ppb (nanoComposix San Diego, CA) prepared in MQ water, ionic Pt standard at 1 ppb (Elements Shasta Lake, CA), and multi-element standard at 1 ppb (High Purity Standards Charleston, SC), both prepared in 2% optima HNO₃. A 2% optima HNO₃ and MQ water blanks were run at the beginning of each run to determine the background. Additionally, the 46 nm Pt standard was run every six condensate samples with an MQ water blank before and after to ensure constant transport efficiency throughout the run. The mean and standard deviation of three technical replicates were computed for all standards and blanks.

2.3. Data analysis

To relate bulk metal analysis to human exposure, metal concentrations were converted from mg/kg to mg/m^{3 2,4}, as mass of metal per volume of air measurements are used in establishing regulations and in risk assessments. To perform these calculations, the total collected mass (mg) of metal was divided by the volume of air required for 200 puffs (Equation 2.1). The volume of air was determined by multiplying the puffing flow rate Q (L/minute), puff time (0.066 min/puff), and the number of puffs to collect each sample (Equations 2.2–2.3). To better estimate the amount of metal an individual would be exposed to, metal concentrations were converted to ng/puff by computing the total mass(ng) of metal in the sample and dividing it by the total number of puffs to generate the sample. Statistical significance was determined using a Kruskal-Wallis test ⁷. All statistical analysis was performed in GraphPad Prism 10 (La Jolla, CA).

$$Airconcentration\left(\frac{mg}{m^3}\right)=\frac{\left[metalconcentration\left(\frac{mg}{kg}\right)\mathrm{*}totalmassofcondensate\left(kg\right)\right]}{volumeofair\left(m^3\right)}$$ Equation 2.1

$$VolumeofAir\left(m^3\right)=Q\left(\frac{L}{\mathrm{m}\mathrm{i}\mathrm{n}}\right)\mathrm{*}pufftime\left(\mathrm{m}\mathrm{i}\mathrm{n}\right)\mathrm{*}(\frac{1m^3}{1,000L})$$ Equation 2.2

$$Pufftime\left(\mathrm{m}\mathrm{i}\mathrm{n}\right)=\#puffs\mathrm{*}puffduration(\frac{min}{puff})$$ Equation 2.3

SP-CAL python code⁴⁹ was used to analyze data from the SP-ICP-MS analysis, and statistical significance was computed in GraphPad Prism 10 (La Jolla, CA). Mean count values for the three replicates for each standard were used to input into SP-CAL. The number of particles detected was converted to particles per puff to relate particle content to human exposure. To perform these calculations, the number of detected particles, N, was divided by the mass of condensate per injection, M (mg), and then multiplied by the mass of condensate per puff, C (mg/puff) (Equation 2.4). The mass of condensate per injection was determined by multiplying the sample flow rate (10 uL/min), acquisition time (min), and concentration (mg/uL) (Equation 2.5). Normality was tested using a D’Agostino & Pearson test and a QQ plot. Statistical significance was determined using a Kruskal-Wallis test and Dunn’s multiple comparison test.

$$\frac{Particles}{puff}=\left(\frac{N}{M\left(mg\right)}\right)\mathrm{*}C(\frac{mg}{puff})$$ Equation 2.4

$$\begin{gathered} Massofcondensateperinjection\left(mg\right) \\ =flowrate\left(\frac{ul}{min}\right)\mathrm{*}acquisitiontime\left(min\right)\mathrm{*}\frac{MassCondensae\left(mg\right)}{voluemwater\left(uL\right)} \end{gathered}$$ Equation 2.5

3. Results

Every metal analyzed (As, Cd, Cr, Fe, Ni, Pb, Zn) was detected in the condensates as ions. The amount of metal in the condensate varied according to the flavor and brand (Figure 1). In the analysis of particles, five (Cr, Fe, Ni, Pb, and Zn) of the seven (As, Cd, Cr, Fe, Ni, Pb, Zn) metals analyzed were detected as particles. The size and mass of particles varied according to brand and flavor. Results are described in more detail below.

Figure 1. Bulk metal concentration (ng/m³) in aerosol condensate from popular e-cig devices (n= 18).

Error bars indicate the geometric mean and geometric standard deviation. Statistical significance was determined using a Kruskal-Wallis test. (* P ≤ 0.05 ** P ≤ 0.01 *** P ≤ 0.001 **** P ≤ 0.0001). EB_BR = EB Create Blue Razz Ice; EB_G = EB Create Gumi; OR_CM = Orion Bar Cool Mint; OR_Br = Orion Bar Blueberry Raspberry; V_M = Vuse Menthol; V_GT = Vuse Golden Tobacco.

3.1. Bulk Metal ICP-MS Analysis

We first assessed the overall metal content in all e-cig devices analyzed. Considerable variability was observed in metal concentrations when comparing different brands and flavors (Figure 1). Differences in Ni were detected in condensates from each device, with the Vuse devices having the highest concentrations; however, these differences were not statistically significant. Condensates from both of the EB Create flavors (EB_BR=32,948 ng/m³, EB_G=66,791 ng/m³) and Vuse flavors (V_GT=11878 ng/m³, V_M=1,4921 ng/m³) contained higher levels of Cr than flavors from Orion Bar devices (OB_CM= 522 ng/m³), OB_BR= 686.6 ng/m³). In contrast, we found higher Pb (OB_CM=510,302 ng/m³, OB_BR=5,61,016ng/m³) and Cd concentration (OB_CM=3,966ng/m³, OB_BR=4054 ng/m³) in both Orion Bar flavors than in both EB Create and Vuse flavors. The EB Create Gummi flavor contained higher levels of As (EB_G= 40,026 ng/m³, EB_BR= 15,946 ng/m³) than all other devices analyzed. All devices had comparable amounts of Fe. Additionally, both Orion bar flavors had almost two orders of magnitude higher Zn (OB_CM= 5,592,191ng/m³, OB_BR=6,315,017ng/m³) than both EB Create flavors (EB_BR=560,926ng/m³, EB_G=482,863 ng/m³) Zn concentrations were <LOD in the Vuse aerosols. Zn data is not reported for Vuse devices due to high background. Metal concentrations in ng/puff are reported in (Figure S.2).

3.2. Single particle metal analysis

Number, size, and mass measurements of metal particles were obtained using single particle analysis (Figure 2 and S3; Table 1). Of the seven metals analyzed, five (Cr, Fe, Ni, Pb, and Zn) were detected as particles. The particle number, mass, and size distributions varied based on metal, brand, and flavor (Table 1). The metal with the highest number of particles varied depending on the device. Vuse Alto Golden Tobacco generated more Ni, Pb, Fe, and Zn particles, whereas condensate from the EB Create Gumi had the largest number of Cr particles compared to the other devices. Overall, lower numbers of Pb particles were detected (ranging from 1 to 19 particles) compared to the other metal particles in condensates from all devices (Figures 2; Table s 1). The size and mass of particles detected in Orion Bar devices are not reported due to an inability to run the samples within twenty-four hours of collection. Particle size and mass are reported for these devices after one month of storage (Figures S4 and S5)

Figure 2. Size(nm) distribution of Ni(A) Cr (B) Pb(C) Fe(D) Zn(E) particles detected in e-cig condensate (n=18).

Violin plot of particle size detected in three technical replicates of three devices per flavor; each point indicates an individual detected particle. Error bars indicated the geometric means with the geometric standard deviation. Statistical significance was determined using a Kruskal – Wallis test and a Dunn’s multiple comparison test. (* P ≤ 0.05 ** P ≤ 0.01 *** P ≤ 0.001 **** P ≤ 0.0001). Data is log-transformed. Summary statistics are in supplementary table 2. EB_BR = EB Create Blue Razz Ice; EB_G = EB Create Gumi; OR_CM = Orion Bar Cool Mint; OR_Br = Orion Bar Blueberry Raspberry; V_M = Vuse Menthol; V_GT = Vuse Golden Tobacco.

Table 1.

Size(nm) and Mass(fg) summary statistics for total particles detected in all devices and technical replicates of e-cig condensate.

	Size (nm)				Mass(fg)
Cr	Count	Range	Geometric mean	Geometric standard deviation	Count	Range	Geometric mean	Geometric standard deviation
EB_BR	41	81.55–171.4	102.50	1.19	41	2.03–18.85	4.03	1.68
EB_G	1305	67.33–310.6	106.50	1.20	1305	1.140–112.2	4.53	1.71
OR_CM	1	97.09–97.09	97.09	1.00	1	3.43–3.43	3.43	1.00
V_GT	37	44.55–115.2	65.08	1.28	37	0.33–5.72	1.03	2.09
V_M	4	73.68–157.0	113.20	1.39	4	1.50–14.49	5.44	2.65
Fe	Count	Range	Geometric mean	Geometric standard deviation	Count	Range	Geometric mean	Geometric standard deviation
EB_BR	86	40.59–326.9	79.69	1.56	86	0.28–144.0	2.09	3.78
EB_G	27	39.43–245.4	64.94	1.55	27	0.25–60.90	1.13	3.71
OR_CM	44	39.43–359.1	73.36	1.96	44	0.25–190.9	1.63	7.50
V_GT	5151	42.75–218.3	67.34	1.21	5151	0.32–42.89	1.26	1.79
V_M	1144	29.22–294.4	65.41	1.19	1144	0.10–105.1	1.15	1.70
Ni	Count	Range	Geometric mean	Geometric standard deviation	Count	Range	Geometric mean	Geometric standard deviation
EB_BR	10	68.66–232.9	113.90	1.55	10	1.51–58.97	6.90	3.74
EB_G	26	59.07–327.8	88.25	1.61	26	0.96–164.4	3.21	4.18
OR_CM	34	45.18–219.3	59.26	1.32	34	0.43–49.21	0.97	2.28
V_GT	770	40.30–251.9	62.18	1.33	770	0.31–74.57	1.12	2.38
V_M	72	31.53–240.8	69.44	1.40	72	0.15–94.22	1.56	2.76
Pb	Count	Range	Geometric mean	Geometric standard deviation	Count	Range	Geometric mean	Geometric standard deviation
EB_BR	1	44.46–44.46	44.46	1	1	0.52–0.52	0.52	1.00
EB_G	1	24.77–24.77	24.77	1	1	0.09–0.09	0.09	1.00
OR_CM	4	61.47–110.2	79.55	1.35	4	1.38–7.96	2.99	2.46
V_GT	19	9.63–47.49	14.21	1.451	19	0.08–0.79	0.02	2.98
V_M	3	23.88–51	37.33	1.487	3	0.01–0.64	0.31	3.31
Zn	Count	Range	Geometric mean	Geometric standard deviation	Count	Range	Geometric mean	Geometric standard deviation
EB_BR	7	81.77–87.37	84.22	1.03	7	2.04–2.49	2.23	1.10
EB_G	70	72.82–118.6	79.09	1.11	70	1.44–6.23	1.85	1.36
OR_CM	0				0
V_GT	3007	99.07–818.4	279.7	1.37	3007	3.63–2047	81.74	2.57
V_M	34	74.12–374.7	118.40	1.66	34	1.52–1.96	47.43	3.10

EB_BR = EB Create Blue Razz Ice; EB_G = EB Create Gumi; OR_CM = Orion Bar Cool Mint; OR_Br = Orion Bar Blueberry Raspberry; V_M = Vuse Menthol; V_GT = Vuse Golden Tobacco.

The size geometric means (GM) of all detected particles ranged from 14.21 nm (Pb, V_GT) to 279.7 nm (Zn, V_GT), with a geometric standard deviation (GSD) ranging from 1 to 1.96 (Figure 2 and Table 1). The smallest particles detected were Pb particles in the Vuse Alto Golden Tobacco devices, with a GM of 14.21 nm (GSD=1.45), while the largest Pb particle was detected in the Orion Bar Cool Mint device, with a GM of 79.55 nm. The largest particles detected were Zn particles in the Vuse Alto Golden Tobacco aerosol, with a GM of 279.7 nm (GSD=1.37).

To estimate the exposure and potential risk for users and to compare with other studies, the number of particles was converted to particles per puff (Figure 3 & Table S 1). When comparing devices, Vuse Golden Tobacco flavor generated the highest number of Ni, Pb, and Fe metal particles per puff (~ 100,000 particles per puff), whereas EB Create Gummi generated the highest Cr, and Vuse Menthol generated the highest Zn particle counts.

Figure 3. Particles per puff concentrations of Ni(A) Cr (B) Pb(C) Fe(D) Zn(E) in e-cig condensate (n=18).

Violin plot of particles per puff in three technical replicates of three devices per flavor. Error bars indicated the geometric means with the geometric standard deviation. Zero values were removed to calculate geometric mean and geometric standard deviation. Statistical significance was determined using a Kruskal – Wallis test and a Dunn’s multiple comparison test. (* P ≤ 0.05 ** P ≤ 0.01 *** P ≤ 0.001 **** P ≤ 0.0001). Data is log-transformed. EB_BR = EB Create Blue Razz Ice; EB_G = EB Create Gumi; OR_CM = Orion Bar Cool Mint; OR_Br = Orion Bar Blueberry Raspberry; V_M = Vuse Menthol; V_GT = Vuse Golden Tobacco.

Lastly, the percentage (%) of the metals as an ion and as a particle was computed (Table S.2). With the exception of Fe, metals present as ions in all devices accounted for over 75% of the total metal detected, with nearly all of Ni, Pb, Zn, and Cr, found in ionic form. Fe particles accounted for 49.5%, 41.7%, and 34.7% of the Fe mass detected in the Orion Bar Cool Mint, Vuse Alto Menthol, and EB Create Blue Razz Ice, respectively.

3.3. Single particle analysis after one month of storage

We assessed the stability of particles after one month of storage (Figures S4 and S5 and Table S3). Particle size and mass were reduced after one month of storage for Ni in both the Vuse Golden Tobacco and Menthol devices. While the number of Ni particles increased in Vuse Golden Tobacco aerosols, they decreased in the Vuse Menthol devices. Cr and Fe particle size and mass were reduced in the EB Create Blue Razz Ice devices. The Vuse menthol aerosol had reduced size and mass of Cr particles, and the Vuse Golden tobacco aerosols had increased Fe particle size and mass. The number of Cr particles increased in both EB Create Blue Razz Ice and Vuse Menthol devices. Lastly, Zn particle size was reduced along with an increase in the number of particles detected in the Vuse Golden tobacco devices after one month of storage.

4. Discussion

The study’s objective was to assess and compare metal particles generated from different e-cig devices and flavors. This information is needed to determine the risk of using e-cigs, especially for a device that the FDA has authorized. Overall, we found metal particles generated from aerosols from every device and flavor, with no exception for FDA-authorized devices.

Indeed, the FDA-authorized Vuse Alto Golden Tobacco device exhibited the highest number of metal particles for each metal, except for Cr. The EB Create Gummi device generated the highest Cr particle count. The count, size, and mass distribution of particles varied according to the device type and flavor. The variation detected between brands and flavors suggests that both the devices’ metal composition and the e-liquid’s chemical composition and properties affect particle emissions. These findings align with previous research highlighting the impact of flavor, type of nicotine, and device brand on metal content ^2–4,7,50.

Bulk metal content and the particle count varied significantly across devices. Vuse devices consistently generated higher particle counts compared to other devices, which could be related to their smaller e-liquid volume of 1.8 mL compared to the 13 mL of disposable devices, potentially leading to a higher ratio of metal components to liquid contact and increased degradation of metal components. Additionally, the type of nicotine salt can affect metal content in aerosols generated from e-cig devices ⁵⁰. Specifically, devices that use lactic acid salts, such as Vuse Alto, have been shown to generate aerosols with higher Ni compared to devices that use benzoic acid or levulinate acid nicotine salts⁵⁰. Furthermore, the specific metals and quality of the metal components may contribute to differences in particle emissions. Several studies have investigated the metal content of filaments and solder joint components in e-cig devices and have found that they are composed of metals similar to those found in this study, such as Fe, Ni, and Zn, suggesting that metal particles may originate from these device components ^35,51–53. Lastly, storage conditions and duration of e-liquid contact with the metal heating elements can increase the concentration of metals transferred to the e-liquid and aerosol ^54,55. Given the rapid rise in popularity and demand of EB Create, it is reasonable to assume there is a short window between manufacturing and consumption, and thus less time for metals to leach from the heating element. However, because the Vuse devices are less popular, these devices might spend more time on the shelf, with the e-liquid having a longer time in contact with the heating element, resulting in higher metal leaching. It is thus likely that a smaller e-liquid volume, type of nicotine salt, and longer shelf-life contribute to the increase in the particle number we report for the Vuse devices. However, elucidating the contribution of each variable is outside the scope of this paper, and further research is needed to fully understand how they can affect the metal particle count.

A reduction in particle size and mass was observed after one month of storage for metals such as Ni, Cr, and Zn in the Vuse or the EB Create devices. This observation could be due to compounds in the e-liquid that have been shown to degrade the coil, such as types of nicotine salt, resulting in degradation of the particles. paper by Pappas et al. (2024)⁵⁰, demonstrated that the type of nicotine salts, lactic acid vs. benzoic acid, affected the concentrations of Ni and Cr in the condensate. It is possible that these acids may affect particle size even after the particles have been generated.

Except for Fe and Cr, metal particles constituted less than 5% of the total metal mass detected (Table S2.4). A higher number of Fe particles was detected in the Orion Bar Cool Mint aerosol, with nearly 50% of Fe in particle form. However, there was considerable variability among devices; for instance, the EB Create Gummi had 9.3% of its total Fe mass as particles. Total Cr particle content was highest in the EB Create Blue Razz Ice, making up approximately 24% of the total mass, while Vuse Menthol had the lowest contribution of Cr particles to the total mass at 8.3%. Although the overall contribution of metals coming from particles is low (below 5%) for all metals except Fe and Cr, the number of particles per puff (Table S2.2) a user is exposed to is substantial; this is important as the adverse health effects of inhaling metal particles are widely known.

Inhaling metal ultrafine and nanosized particles is particularly concerning because their small size facilitates deposition deep in the lungs and translocation to secondary organs such as the heart and brain ^9,15 and exposure, even at low doses, can contribute to adverse outcomes, including oxidative stress, systemic inflammation, and disruption of cardiovascular and pulmonary function ^56–58. In the case of e-cig exposure, Pb inhalation has been shown to accumulate in the brains of mice after a two-month exposure ¹⁹, raising significant concern given the large youth population ¹ that uses these devices and the well-documented cognitive impacts of Pb on brain development^39,59–61. Pb exposure can also result in cardiovascular⁶² and kidney disease ^63,64. Inhalation of Cr particles can cause nasal irritation and dermatitis, and, with chronic exposure, has been implicated in asthma ^65–68 and carcinogenicity ⁶⁹. Ni inhalation has been linked to sensitization and allergic responses, even in the case of e-cig use ^66,70–75. Other metal particles, such as Fe and Zn, have also been shown to result in both acute and chronic inflammation after exposure, and cause metal fume fever and fibrosis in individuals exposed to iron- and steel-containing aerosols ^{25,42,76–78}. All of this taken together demonstrates that the metal particle exposure quantified in our study is not only analytically significant but also toxicologically relevant, emphasizing the need for further evaluation of metal exposures from e-cigs and their relation to human health.

Prior to the pandemic, the FDA moved to ban flavors for e-cig liquids, except for tobacco and mint-flavored POD products, due to their attraction among youth⁷⁹. This ban was successful in reducing the use of re-usable POD-style⁸⁰ devices (i.e. JUUL); however, due to a loophole in the FDA policy, disposable devices were left out of the ban⁸¹, which ultimately resulted in a shift in the market towards disposable-style devices^80,82 Here, we investigated the metal particle content of two of the most popular brands of disposable devices in four different exotic flavors, investigating the risk for continued metal exposure to users despite the FDA’s regulatory efforts.

A major advantage of the studies described here is the use of the SP-ICP-MS to analyze particles. SP-ICP-MS has the capability to analyze multiple samples quickly. Considering how frequently new devices appear on the market, a fast validated method is needed. Nonetheless, SP-ICP-MS has several limitations. First, it only allows the detection of one metal per particle, making it unable to determine if a single particle contains multiple metals. This is a limitation when estimating particle size, as particle size is estimated using metal density. Additionally, in this manuscript, particle size was determined based on the assumption that particles contained the detected single metal alone and not the oxide. Pappas and colleagues ²⁰ have considered particles as oxides when evaluating particle size, as particulate metals may result from oxidation of metal components in e-cig devices; our particle size results fall within the range of those reported by Pappas.

Another limitation of ICP-MS is that it does not provide information on particle shape and charge, which are relevant for assessing potential toxicity. These limitations suggest that further research is needed to fully understand the health implications of metal particles in e-cig aerosols and to explore the effects of different devices and e-liquid compositions.

Lastly, it is important to note a limitation of blank samples for this study. While we employed the same collection tubing and plastic conical vials for our blank samples, and we use metal-free tubing, we did not utilize a peristaltic pump to generate the blanks. As such, it is possible that the use of the peristaltic pump could crack and cause wear on the tubing during collection, leading to a possibility for background not captured with these experiments. Additionally, it is important to note that the PG:VG ratios of our blank samples were 30:70; however, the PG: VG ratios of the disposable devices used in this paper are not reported by the manufacturers, nor have they been reported in published work. Based on internet searches and blog posts ⁸³, the PG/VG ratios used for our blanks are most similar to those found in the EB Create devices. For the Vuse Alto Devices, some researchers have determined that a ~50:40 ratio is used ⁸⁴.

Notwithstanding these limitations, the study reported here confirms the presence of metal particles in all e-cigarette aerosols tested. The FDA-authorized Vuse Alto devices generated the highest particle counts for Fe, Ni, Pb, and Zn, while the Elf Bar Gumi generated aerosols with the highest number of Cr particles. Our findings contribute to the understanding of metal particle generation in e-cigarette aerosols and highlight the need for further research to better assess exposure risks and potential health effects.

Supplementary Material

Supplementary Information: Additional figures, including background blanks, bulk metal in ng/puff, particle mass, particle mass and size after one month of storage, and tables, including geometric mean particles per puff concentrations, percent particle and soluble ionic metals, and mass and size summary statistics.

Synopsis:

This is the first study to examine the metal particle composition in disposable e-cigarettes and compare it to that of FDA-authorized devices. All devices contained Cr, Fe, Ni, Pb, and Zn, particles, mostly <100 nm, with the highest particle counts, except Cr, from the FDA-authorized Vuse Alto Golden Tobacco Device.

Data Availability

Data will be made available on request.