Portland State University Portland State University

PDXScholar PDXScholar

University Honors Theses University Honors College

3-1-2018

Evaluation of Synthetic Dyes and Food Additives in Evaluation of Synthetic Dyes and Food Additives in

Electronic Cigarette Liquids: Health and Policy Electronic Cigarette Liquids: Health and Policy

Implications Implications

Tetiana Korzun Portland State University

Follow this and additional works at: https://pdxscholar.library.pdx.edu/honorstheses

Let us know how access to this document benefits you.

Recommended Citation Recommended Citation Korzun, Tetiana, "Evaluation of Synthetic Dyes and Food Additives in Electronic Cigarette Liquids: Health and Policy Implications" (2018). University Honors Theses. Paper 499. https://doi.org/10.15760/honors.502

This Thesis is brought to you for free and open access. It has been accepted for inclusion in University Honors Theses by an authorized administrator of PDXScholar. Please contact us if we can make this document more accessible: pdxscholar@pdx.edu.

Evaluation of synthetic dyes and food additives in electronic cigarette liquids:

Health and policy implications

by

Tetiana Korzun

An undergraduate honors thesis submitted in partial fulfillment of

the requirements for the degree of

Bachelor of Science

in

University Honors

and

Biology

Thesis Advisers

Professor Robert M. Strongin, PhD Dr. R. Paul Jensen, PhD

Portland State University

2018

1

TABLE OF CONTENTS

ABSTRACT .................................................................................................................................... 4

INTRODUCTION .......................................................................................................................... 5

METHODS AND INSTRUMENTS .............................................................................................. 8

Materials ..................................................................................................................................... 8

Sample Preparation ..................................................................................................................... 8

Samples for HPLC-DAD analysis: ......................................................................................... 8

Samples for ESI-HRMS validation: ........................................................................................ 9

Samples for IC analysis .......................................................................................................... 9

HPLC-DAD Analysis ................................................................................................................ 10

ESI-HRMS Validation .............................................................................................................. 12

Ion Chromatography ................................................................................................................. 12

RESULTS AND DISCUSSION ................................................................................................... 14

CONCLUSION ............................................................................................................................. 19

ACKNOWLEDGMENTS ............................................................................................................ 19

SUPPLEMENTARY INFORMATION ....................................................................................... 20

HPLC-DAD chromatograms .................................................................................................... 20

ESI-HRMS spectra of synthetic dye standards ......................................................................... 21

ESI-HRMS validation of synthetic dyes in e-liquid samples ................................................... 23

Ion chromatograms of vaporized samples ................................................................................ 25

REFERENCES ............................................................................................................................. 26

2

INDEX OF TABLES Table 1. Quantitative features of the HPLC-DAD method for the selected dye standards .......... 11 Table 2. Quantitative features of the ESI-HRMS method for the selected dye standards. .......... 12 Table 3. Seven synthetic dyes used in foods. Chemical classes and ingestion ADIs. ................. 14 Table 4. Identification and quantification of unknown dyes in the samples of commercially

available e-liquids. ................................................................................................................ 15 Table 5. ESI-HRMS validation of unknown dyes in commercially available e-liquids. ............. 16 Table 6. Synthetic dye concentrations in selected foodstuffs. ..................................................... 16 Table 7. Sulfate and chloride concentration in vaporized e-liquid samples. ................................ 18 TABLE OF FIGURES

Figure 1. Structures of synthetic dyes. ........................................................................................... 6 Figure 2. Representative chromatograms of synthetic dye standards .......................................... 11 Figure 3. Chromatograms of unknown dyes.. .............................................................................. 15

SUPPLEMENTARY INFORMATION

Figure S1. Additional chromatograms of dyes standards ............................................................ 20 Figure S2. ESI-HRMS spectrum of Allura Red AC Standard. .................................................... 21 Figure S3. ESI-HRMS spectrum of Erythrosine Standard. ......................................................... 21 Figure S4. ESI-HRMS spectrum of Fast Green FCF Standard .................................................... 21 Figure S5. ESI-HRMS spectrum of Tartrazine Standard ............................................................. 22 Figure S6. ESI-HRMS spectrum of Sunset Yellow FCF Standard .............................................. 22 Figure S7. ESI-HRMS spectrum of Brilliant Blue FCF Standard ............................................... 22 Figure S8. ESI-HRMS spectrum of Allura Red AC in Red E-liquid .......................................... 23 Figure S9. ESI-HRMS spectrum of Tartrazine in Green E-liquid ............................................... 23 Figure S10. ESI-HRMS spectrum of Brilliant Blue FCF Dye in Green E-liquid ........................ 23 Figure S11. ESI-HRMS spectrum of Brilliant Blue FCF in Blue E-liquid. ................................. 24 Figure S12. IC of vaporized Red E-liquid. .................................................................................. 25 Figure S13. IC of vaporized Blue E-liquid. ................................................................................. 25 Figure S14. IC of vaporized Green E-liquid ................................................................................ 25

3

To beloved chemists and the physicist

Зробити щось, лишити по собі, а ми, нічого, – пройдемо, як тіні,

щоб тільки неба очі голубі цю землю завжди бачили в цвітінні. Щоб ці ліси не вимерли, як тур, щоб ці слова не вичахли, як руди. Життя іде і все без коректур, і як напишеш, так уже і буде.

Ліна Костенко

4

ABSTRACT

Prior studies of e-liquid thermal degradants do not reflect many of the potential health

hazards related to e-cigarettes. Although current studies have focused on solvents and flavoring

additives in e-cigarette formulations, there have been no prior reports on the identity and levels

of synthetic dye additives. Therefore, the purpose of this study was to identify and quantify these

compounds to enhance understanding of the risks associated with the inhalation of colored e-

liquids. Furthermore, e-liquids were subjected to thermal degradation under normal vaping

conditions to quantify the sulfur oxides (SOx) content indicating dye decomposition. The dyes

were analyzed by a combination of high-performance liquid chromatography and high resolution

mass spectroscopy. The thermal decomposition of dyes in vaporized e-liquid samples was

studied by ion chromatography. The findings of this investigation revealed that e-liquid

manufacturers added synthetic dyes in concentrations comparable to those used in the food

industry. In addition, SOx were present in the aerosolized e-liquids suggesting that dyes undergo

thermal degradation. The aerosol samples contained a substantial amount of free chloride, which

could be associated with a breakdown of the sucralose molecules, whose presence in the e-

liquids was confirmed by nuclear magnetic resonance.

5

INTRODUCTION

Electronic cigarettes (e-cigarettes) may pose dangers to consumers, due to a lack of

formal oversight regarding their regulation and manufacturing. In addition, the long-term effects

of vaping on human health remain to be unknown. However, e-cigarettes are frequently lauded to

be a healthier alternative to tobacco products. For decades, tobacco manufacturers have faced

heavy advertising restrictions, in large part to avoid encouraging tobacco use among children. It

is known that advertising has a positive correlation with youth cigarette smoking.1 The bright

packaging designs of e-liquids, as well as their pleasant aromas (and potentially colors), are

known to be enticing to young people.2,3 These attractive products, readily available on the

largely unregulated market,4 are manufactured and advertised in a relaxed regulatory

environment. As a result, in recent years, e-liquid poisoning amongst children has increased by

1500%.5 This includes child fatalities associated with e-liquid nicotine ingestion overdoses.6 If

appealing color and flavoring additives in e-liquid formulations remain, there are substantive

health risks to our nation’s youth.

The current lack of regulations allows manufacturers to add new ingredients that have no

associated inhalation toxicological data. The yields of toxic aldehydes and related compounds

that are produced via the thermally–induced degradation of propylene glycol and glycerol (the e-

liquid solvents) are enhanced by the decomposition of additives that are often not listed as e-

liquid ingredients and are considered to be the manufacturing secrets.7,8

Food additives influence the consumers’ perception of food flavor and flavor identity.9

Such additives are classified as generally recognized as safe (GRAS) for ingestion.10 However,

their inhalation toxicity is unknown. Two categories of color additives used in food and drugs in

the US include those certified and those exempt from certification. They are categorized based

6

on the US Food and Drug Administration (FDA) testing requirements.11,12 Certified food dyes

are synthetic compounds that are widely used because of their uniform color and shelf-life

stability, as well as their ability to encompass a full spectrum of colors when mixed, while not

impacting or altering food taste.13 The synthetic dyes shown in Figure 1 are certified by the FDA

for use in food, as well as in drugs and cosmetics (FD&C). Other FD&C colorants, not shown

here, are dyes that are certified for usage only in specific foods (i.e. Orange B and Citrus Red

No. 2 are used to color the surfaces of sausages and oranges, respectively).14

Figure 1. Structures of synthetic dyes.

Although disclosing the identity of synthetic dyes is mandatory on the labels of

foodstuffs,12 this information is absent on the vast majority of e-liquid labels. Additionally,

acceptable daily intake (ADI) values for dye inhalation are not provided by the FDA, as synthetic

dyes had not been used or intended for inhalation previously.

S NN

S

OHO

-OO

O

OO O-

Allura Red AC (FD&C Red No. 40)

NN

HO

SO-

OO

S-OO

O

Sunset Yellow FCF (FD&C Yellow No. 6)

NN

O

O

O-

NN

SO

O

O-

SO

OO-

Tartrazine (FD&C Yellow No. 5)

N+

NS

O-

O

OHO

SO-

SO-

OOOO

Fast Green FCF (FD&C Green No. 3)

N+NS-O O

O

S O-

O

O

SO-

O

O

Brilliant Blue FCF (FD&C Blue No. 1)

O

O

O

I

O-

I

I

-OI

Erythrosine (FD&C Red 3)

NH

NH

O

O

S

S

-OO

O

O-

O

O

Indigotine(FD&C Blue No. 2)

7

Herein, the purpose of this study was to identify and quantify the specific coloring

compounds in commercially available e-liquids in order to enhance our understanding of their

potential risks associated with inhalation and to help raise awareness about the inhalation of

substances originally intended for ingestion.

8

METHODS AND INSTRUMENTS

Materials

Three e-liquids were ordered from the manufacturer’s online shop. All HPLC-grade

solvents (acetonitrile, methanol and water) were from Fisher Scientific (Fisher Scientific

Co., Chicago, IL, USA). Dibasic ammonium phosphate (ACS reagent, ≥98%) and potassium

hydroxide (ACS reagent grade, >85%, pallets) were used for HPLC buffer preparation and

were obtained from Sigma-Aldrich (Sigma Aldrich, St. Louis, MO, USA). Perchloric acid

(70 % aqueous solution; Acros Organics) and hydrogen peroxide (30 % aqueous solution;

Sigma Aldrich, St. Louis, MO, USA) were used for absorption solution preparation in ion

chromatography experiments. Analytical standards of Allura Red AC (FD&C Red 40),

Sunset Yellow FCF (FD&C Yellow 6), Tartrazine (FD&C Yellow 5), Brilliant Blue FCF

(FD&C Blue 1), Fast Green FCF (FD&C Green 3), Erythrosine (FD&C Red 3) and Indigo

Carmine (FD&C Blue 2) were obtained from Sigma Aldrich (Sigma Aldrich, St. Louis, MO,

USA). Water for sample preparation was purified using Thermo Scientific Barnstead

GenPure xCAD Plus UV–TOC Water Purification System (Thermo Scientific, Waltham,

MA, USA).

Sample Preparation

Samples for HPLC-DAD analysis: E-liquids were diluted with water (1:5 dilution)

and sonicated for 20 minutes. The standard solutions of Allura Red AC, Fast Green FCF,

Sunset Yellow FCF, Erythrosine, Tartrazine, Brilliant Blue FCF were prepared in water by

dissolving the solid powders to obtain a stock concentration of 1mg/mL. Four to eight

calibration concentrations were prepared from stock solutions and used for calibration curves

construction with curves forced through the origin. The concentration ranges for standard

9

solutions were 20-80 µg/mL for Allura Red AC, Fast Green FCF, Sunset Yellow FCF and

Erythrosine, and 2.5-80 µg/mL for Tartrazine and Brilliant Blue FCF. Each unknown and

standard solution were filtered (PVDF, 0.22 µm pore size) and subjected to HPLC-DAD

analysis.

Samples for ESI-HRMS validation: dye samples from three e-liquids were

concentrated using SPE cartridges (Oasis HLB cartridges, Waters, Milford, MA, USA),

connected to the vacuum manifold system (Waters, Milford, MA, USA) and the vacuum inlet.

Cartridges were conditioned using 3 ml of methanol and 5 ml of deionized water at a flow rate of

about 1 drop per minute. E-liquid samples diluted in water were transferred to the SPE

cartridges. The loaded cartridges were rinsed with 5 ml of water allowing e-liquid solvent

separation at a flow rate of about 2 drops per minute. The cartridges were eluted with two 3-mL

methanol rinses at the same flow rate. The eluent sample volumes containing dyes were reduced

by rotary evaporation (R-210 Rotavap, Buchi, New Castle, De, USA). The residues were

reconstituted in 1 ml of methanol. SPE of red and blue e-liquids yielded two samples containing

red and blue dyes, respectively; SPE of the green e-liquid yielded a combination of two samples

of yellow and blue dyes. The standard solutions of Allura Red AC, Fast Green FCF, Sunset

Yellow FCF, Erythrosine, Tartrazine, Brilliant Blue FCF dyes were prepared in water to give a

final concentration of 20 Μm. Note, Indigotine was not used for ESI-HRMS analysis.

Samples for IC analysis: the vapor was generated using the electronic cigarette

consisted of Tesla Invader III battery unit (Teslacigs, Shenzhen, China) with two 18650

HG2 batteries (3.7 V, 3000 mAh) (LG Chem, Holland, MI) and KangerTech SubTank Mini

atomizer with horizontal kenthal coil (1.2 Ω resistance) (KangerTech, Shenzhen, China). The e-

cigarette was operated at 17 W. The aerosol was drawn into a -78 ºC cold trap (dry ice/acetone)

10

followed by the impinger containing acidified hydrogen peroxide absorption solution. The setup

was connected to the SCSM-STEP single cigarette-smoking machine, simulating inhalation

(CH Technologies, Westwood, NJ). Aerosol was produced using CORESTA vaping mode (3-

s puff with 1-s power button activation prior vaping, 30-s puff intervals, 55-mL puff volume).15

The sample from the cold trap was combined with the impinger solution. The samples were

immediately subjected to analysis by ion chromatography. Triplicate experiments were

performed using each e-liquid; each replicate represents the session of 40 puffs (20 puffs – 20

minutes cooling – 20 puffs).

HPLC-DAD Analysis

The samples of commercially available e-liquids were analyzed using adapted HPLC-

DAD method.16 The HPLC-DAD set-up included 1525 Binary HPLC pump, 2996 Photodiode

Array Detector and column heater (Waters, Milford, MA). The chromatographic separation was

perfumed using Acclaim PA2 column with 3 µm particle size, 3×75 mm dimensions and

injection volume of 5 µL (Thermo Scientific, Waltham, MA). The DAD detector was set to

210–650 nm spectral range. The column was kept at 30 °C. The mobile phase consisted of

buffers A (20 mM (NH4)2HPO4, pH 8.8) and B (20 mM (NH4)2HPO4: CH3CN = 50:50, v/v).

The analysis of standard solutions was performed using the gradient of 12% B from 0.00

to 3 min, ramping to 100% from 3.00 to 3.50 min with hold for 1.0 min, and return to 12% B in

0.1 min (flow rate of 0.71 mL/min). The e-liquid samples followed the same gradient program

and flow, except that the 100% B hold was extended to 29.50 min and returned to 12% in 1 min.

11

Table 1. Quantitative features of the HPLC-DAD method for the selected dye standards

Experiments and the construction of calibration curves were performed in triplicates.

Quantitative features of the HPLC-DAD method and chromatograms of the dye standards are

presented in Table 1 and Figure 2 (additional chromatograms: Supplementary information).

Empower 2 Chromatography Data Software was used for data collection and processing.

Figure 2. Representative chromatograms of synthetic dye standards. Standards at 40 µg/mL concentrations.

Dye Standard

Absorption Maxima (nm)

Regression (%)

LOD (µg/mL)

LOQ (µg/mL)

RSD (%)

Allura Red AC 506 99.998 0.27 0.88 1.37 Fast Green FCF 619 99.988 0.97 3.23 2.39

Sunset Yellow FCF 485 99.972 1.60 5.35 0.36 Erythrosine 530 99.972 1.57 5.23 2.45 Tartrazine 426 99.991 0.57 1.91 0.81

Brilliant Blue FCF 630 99.973 0.10 0.34 1.49

12

ESI-HRMS Validation

Validation of identified analytes was performed using a high-resolution (30,000

resolution power) Thermo LTQ-Orbitrap Discovery hybrid mass spectrometry instrument

equipped with Ion Max source with an electrospray ionization probe (Thermo Fisher Scientific,

San Jose, CA, USA). The ionization interface was operated in the negative mode using the

following settings: source voltage, 4 kV; sheath and aux gas flow rates, 50 and 5 units,

respectively; tube lens voltage, 90 V; capillary voltage, 49 V; and capillary temperature, 300 °C.

The Orbitrap mass analyzer was externally calibrated prior to analysis. ESI-HRMS spectra of

standards (molecular ions: Allura Red AC [M]2-, Brilliant Blue FCF [M]2-, Tartrazine [M]3-,

Sunset Yellow FCF [M]2-, Erythrosine [M]2- and Fast Green FCF [M]2-) revealed the molecular

ion masses with accuracy within ±7 ppm (Table 2 and Supplemental information).

Table 2. Quantitative features of the ESI-HRMS method for the selected dye standards.

Standard Calculated Mass of Standards (m/z)

Observed Mass of Standards (m/z)

Allura Red AC 225.00903 225.00924 Brilliant Blue FCF 373.07077 373.07067

Tartrazine 154.99315 154.99263 Sunset Yellow FCF 202.99592 202.99515

Erythrosine 834.64667 834.64394 Fast Green FCF 381.06823 381.06713

Ion Chromatography

Anion analysis of the vaped samples was conducted using an Dionex ICS-5000 ion

chromatography system (Dionex, Sunnyvale, CA), outfitted with a conductivity detector cell and

electrolytically regenerated suppressor (AERS 500, 4mm; Dionex, Sunnyvale, CA). 25 µL

aliquots of the filtered (0.2 µm pore size) samples were injected onto the system for each run.

The separation was carried out on an IonPac-AS15 with an IonPac-AG15 guard columns

13

(Dionex, Sunnyvale, CA) and a flow of 0.75 mL/min. Gradient elution was used to obtain

separation of the organic and inorganic peaks. The eluent concentrations were set as follows: 3

mM KOH for 36.5 minutes, 45 mM KOH for 16.5 minutes, and 3 mM KOH for 7 minutes.

Calibration curves were created using seven calibration standards with concentrations of chloride

and sulfate ranging from 2.50 to 40.0 mg/L and 0.125 to 2.00 mg/L, respectively

(Supplementary Information). Data acquisition and analysis was performed using Chromeleon

workstation. Linear calibration curves (without forcing the intercept through zero) were created

based on peak height. R-squared values for all calibration curves were greater than 0.99; LOD

values for chloride and sulfate were 0.063 and 0.079 mg/L, and LOQ values were 0.19 mg/L and

0.24 mg/L, respectively.

14

RESULTS AND DISCUSSION

Three e-liquids were randomly chosen to represent the primary additive colors (red, green

and blue). The manufacturers were contacted but declined the requests for the identity of

ingredients, including the synthetic dyes used in the formulations.

Three dyes, Allura Red AC, Brilliant Blue FCF and Tartrazine, were identified in the e-

liquid samples. According to their structures, Allura Red AC and Tartrazine are azo dyes, and

Brilliant Blue FCF is a triarylmethane (Table 3). Although these artificial food colorants are

generally regarded as safe (GRAS) for human consumption, batches of Allura Red AC, Sunset

Yellow FCF and Tartrazine are known to contain the human carcinogen benzidene and other

aromatic amines17-20, while Tartrazine is a known potent allergenic agent.21-24

Table 3. Seven synthetic dyes used in foods. Chemical classes and ingestion ADIs.

Synthetic Dye Chemical Class European ADI (mg/kg/day)

ADI25

(mg/p/day)a

Per Capita Exposure25

(mg/p/day)a

Allura Red AC (FD&C Red No. 40) Azo 0-7 (2016) 26 420 17.91

Sunset Yellow FCF (FD&C Yellow No. 6) Azo 0-2.5 (1982) 13 225 10.74

Tartrazine (FD&C Yellow No. 5) Azo 0-10 (2016) 26 300 12.06 Brilliant Blue FCF

(FD&C Blue No. 1) Triarylmethane 0-12.5 (1969) 13 720 1.72 Fast Green FCF

(FD&C Green No. 3) Triarylmethane 0-25 (1986) 13 150 0.038 Erythrosine

(FD&C Red No. 3) Xanthene 0-0.1 (1990) 13 150 0.61 Indigotine

(FD&C Blue No. 2) Sulfonated Indigo 0-5 (1974) 13 150 1.95 a Per 60-kg person

The identification and quantification of the dyes in the samples was performed by high-

performance liquid chromatography-diode array method (HPLC-DAD). The chromatograms of

three e-liquid samples were compared to chromatograms of the standard solutions of Allura Red

AC, Sunset Yellow FCF, Tartrazine, Brilliant Blue FCF, Fast Green FCF, Erythrosine and

15

Indigotine (Figure 3). The green e-liquid contained two dyes, Brilliant Blue FCF and

Tartrazine, at concentrations and proportions consistent with those used in the food industry to

afford neon green solutions.27 The blue and red e-liquids contained Brilliant Blue FCF and

Allura Red AC food colorants, respectively (Table 4).

Table 4. Identification and quantification of unknown dyes in the samples of commercially available e-liquids.

Sample Identified Dye

Absorption Maxima (nm)

Retention time (min)

Concentration (µg/g)a

Red Allura Red AC 505 2.38 66.71 Blue Brilliant Blue FCF 630 3.01 51.42

Green Brilliant Blue FCF 630 3.00 1.85 Green Tartrazine 426 1.38 29.43

a µg of synthetic dye per g of e-liquid

Figure 3. Chromatograms of unknown dyes. Notice that the green sample is made of a combination of yellow and blue dyes. Detector wavelengths were set at 426 nm for Tartrazine, 505 nm for Allura Red AC and 630 nm for Brilliant Blue FCF.

Follow-up validation by high resolution electrospray ionization mass spectroscopy (ESI-

HRMS) confirmed the results obtained by HPLC-DAD method. Allura Red AC, Brilliant Blue

FCF and Tartrazine were identified with mass accuracies within ±7 ppm (Table 5 and

Supplementary information).

16

Table 5. ESI-HRMS validation of unknown dyes in commercially available e-liquids.

E-liquid Sample Dye Identified Observed mass

(m/z) Calculated mass

(m/z) Red Allura Red AC 225.00829 225.00903 Blue Brilliant Blue FCF 373.06815 373.07077

Green Brilliant Blue FCF 373.07184 373.07077 Green Tartrazine 154.99319 154.99315

The overall data shows that manufacturers are using the three FDA certified dyes at

concentrations similar to those used in the food industry (Table 6).

Table 6. Synthetic dye concentrations in selected foodstuffs.

Foods Tartrazine (µg/g)

Brilliant Blue FCF (µg/g)

Allura Red AC (µg/g)

Soft drinks28 4.7-5.2 1.0-1.2 49.8 Confectioneries28 4.0-154.8 1.0-6.3 ND

Water soluble foods29 0.5 0.5-4.8 18.1-27.8 Gummy candy30 1.7-80.4 0.7-9.6 1.0-47.8

Jellies30 2.2-20.2 ND ND Juices31 0.06-121.82 2.75-71.44 5.77-7.15

Cookies31 0.13-17.36 0.90-0.99 ND Fruit jam32 ND ND 17.9-33.4

Salted fish32 136.0-292.5 ND ND Soft drinks33 158 0.063-12.9 0.107-0.14

Juice and jelly powder34a 1.3-56.2 2.9-6.4 30.2-53.8 a Units for this study are µg/mL; ND = not disclosed or not discovered in specified foods.

The justification for colorant addition to e-liquids in the same proportions relative to

foodstuffs is likely to enable a vaper’s identification of e-cigarette flavoring additives with

corresponding foods. The color was shown to influence food sensory characteristics and

acceptability of products due to possible learned associations of particular colors with

foodstuffs.35,36 Other reasons for using artificial dyes in e-liquids could include the masking of

unattractive natural colors of flavorings or other e-liquid constituents, or protecting e-liquids via

the sunscreen effect13 to extend their shelf life. Although these reasons are valid for justifying

17

dye addition to foods, the benefits of e-liquid color enhancement may be undermined due to the

unknown health risks associated with the inhalation of dyes and their potential degradation

byproducts.

Several conundrums arise regarding the lack of (i) certified dye labeling on e-cigarette

packaging and (ii) acceptable daily intake (ADI) values for their inhalation. In agreement with the

requirements of CFR 21, §70.25, certified synthetic dye identities are required on the labels of not

only food products, but also on drugs and cosmetics.12 Although e-liquids are intended for human

consumption, the identities of any dyes used are not required to be disclosed.

More importantly, there are no inhalation ADI values for these dyes, as inhalation is just

emerging as an intended method of their consumption. The European ADI values for dye ingestion

are defined by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) and the

European Food Safety Authority (EFSA). In the US, the ADI levels are regulated by the FDA and

are listed in terms of dye intake in mg per 60-kg person (Table 3). To achieve an ADI upper limit

of Brilliant Blue FCF (ADI: 720 mg/person/day), a 60-kg person would have to inhale 14 kg of

the blue e-liquid per day. In order to achieve an ADI value for Allura Red AC (ADI: 420

mg/person/day), one would have to inhale 6 kg of red e-liquid. For Tartrazine (ADI: 300

mg/person/day), the value would be 1 kg of green e-liquid. To reach average per-capita exposure

levels listed by FDA (Table 3), a person would have to consume 34, 268 and 410 g of blue, red

and green e-liquids, respectively (per-capita exposure levels for Brilliant Blue FCF, Allura Red

AC and Tartrazine are 1.72, 17.91 and 12.06 mg/60-kg person/day (Table 3).

In addition to the dilemma of inhaling these dyes, food colorants in e-liquids may produce

toxic byproducts via thermal degradation. Some of the potential degradants can include dye

precursors and aromatic amines that are considered to have carcinogenic potential.37

18

In vaped samples in the investigation herein, dye desulfonation was observed at a moderate

operational power of the e-cigarette (Table 7 and Supplementary Information). The presence of

sulfate (SO42-) ions in the collected aerosols indicates that sulfur dioxide (SO2) or/and sulfur

trioxide (SO3) are released from the sulfonated dyes (and oxidized to SO42- according to the method

described by Makkonen et al. for analysis).38 This suggests that the dyes start to degrade at

temperatures as low as those needed to form e-liquid aerosols.7 The loss of sulfonate groups via

thermal destruction in sulfonated dyes has been previously reported.39,40 Rehorek showed that the

extent of desulfonation is a function of temperature due to the different positions of sulfonic group

in the molecules (i.e., ortho, meta or para to the bonds adjoining the aryl rings), with para-

sulfonated compounds having the lowest optimal pyrolysis temperature.40 Accordingly, the green

e-liquid showed the highest level of SO42- in the vaporized samples, as the yellow dye, tartrazine,

has two para-sulfonyl groups.

In addition, the samples contained relatively high levels of free chloride (Cl-) (Table 7

and Supplementary Information). This could be derived from the breakdown of sucralose. The

presence of sucralose in the e-liquids was confirmed by nuclear magnetic resonance.41 Cl- or

SO42- were not observed in the air samples, nor in the negative (absorption solution) or positive

controls (e-liquid samples diluted in the absorption solution).

Table 7. Sulfate and chloride concentration in vaporized e-liquid samples.

Mass concentration [SO4

2-] (µg/g)a,b

Mass concentration [Cl-] (µg/g)a

Replicate 1 2 3 1 2 3

Sam

ple Red >LOD >LOD >LOD 23.38 56.34 36.48

Blue >LOD 0.84 1.96 20.5 33.5 32.82 Green >LOD 2.72 2.81 17.27 218.51 184.19

a analyte yield (µg) per g of e-liquid consumed; b SO42- was calculated back to SO2

19

While the in-depth investigation of sucralose decomposition was not an objective of the

current study, finding chloride in the absorption solution suggested the fragmentation of

sucralose molecules. The loss of hydrogen chloride (HCl) from sucralose is a known initial step

in its breakdown, and can ultimately lead to the release and/or formation of potentially toxic by-

products such as chloropropanols.42 Recent studies demonstrate that sucralose decomposition,

including HCl release and the formation of chlorinated derivatives, occurs at relatively mild

temperatures.43-44 The formation of chlorinated compounds from sucralose in the presence of

glycerol (one of the two common e-liquid solvents) and metal oxides is well-precedented.45,46

CONCLUSION

The data presented herein suggests that the infusion of e-cigarette formulations with food

additives, including synthetic dyes and sweeteners, creates potential health risks since their

inhalation toxicity as well as the toxicity of their byproducts is largely unknown. We hope the

results of this study will increase awareness and inform decisions about the regulation of e-

cigarettes, whether as inhalable foodstuffs, or as nicotine delivery matrices primarily intended for

smoking cessation.

ACKNOWLEDGMENTS

I am deeply grateful to the Strongin Research Group for the immense support, practical

guidance and the enormous contribution to this work. We thank the NIH and the FDA for their

support via award R01ES025257. The content is solely the responsibility of the authors and does

not necessarily represent the views of the NIH or the FDA. Support from the National Science

Foundation (Grant number 0741993) for purchase of the LTQ-Orbitrap Discovery is gratefully

acknowledged.

20

SUPPLEMENTARY INFORMATION

HPLC-DAD chromatograms

Figure S1. Additional chromatograms of dyes standards. Standards at 40 µg/mL concentrations.

-0.5

0

0.5

1A

505

time (min)

Erythrosine

-0.5

0

0.5

1

A61

8

time (min)

Fast Green FCF

-0.2

0

0.2

0.4

A48

2

time (min)

Sunset Yellow FCF

21

ESI-HRMS spectra of synthetic dye standards

Figure S2. ESI-HRMS spectrum of Allura Red AC Standard. [M]2- ion. Theoretical mass: 225.00903. Observed mass: 225.00924. Δ m/z: 0.93 ppm

Figure S3. ESI-HRMS spectrum of Erythrosine Standard. [M]2- ion. Theoretical mass: 834.64667. Observed mass: 834.64394. Δ m/z: 3.27 ppm

Figure S4. ESI-HRMS spectrum of Fast Green FCF Standard [M]2- ion. Theoretical mass: 381.06823. Observed mass: 381.06713. Δ m/z: 2.89 ppm

K:\Strongin LAB\...\08-31-2017-AR1 8/31/2017 9:46:51 AM 08-31-2017-AR

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce

0

50

1000.14

0.150.130.20

0.110.21

0.10 0.250.280.07 0.31 0.37 0.45 0.50 0.57 0.63 0.67 0.810.73 0.940.90 1.04 1.421.09 1.12 1.34 1.471.19 1.24

0.170.140.11

0.22

0.250.38 0.45 0.53 1.460.730.60 0.94 1.28 1.350.68 1.140.83 1.10 1.180.91 0.99 1.40

0.14

0.200.11

0.230.25

0.28 0.31 0.45 0.570.54 0.60 0.67 0.81 0.940.76 0.90 1.04 1.421.09 1.12 1.34 1.471.19 1.29

NL: 6.43E7TIC MS 08-31-2017-AR1

NL: 9.04E6TIC F: ITMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-AR1

NL: 6.43E7TIC F: FTMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-AR1

210 215 220 225 230 235 240 245 250 255 260 265 270 275 280m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

225.00924

226.50864

217.49825 234.99332

220.99128212.96994 252.30813 274.66849242.83986 264.71285

225.00903

228.50955

NL:2.40E708-31-2017-AR1#10-37 RT: 0.07-0.19 AV: 14 F: FTMS - p ESI Full ms [100.00-500.00]

NL:7.22E5

c18 h14 n2 o8 s 2: C18 H14 N2 O8 S2c (gss, s /p:40)(Val) Chrg 2R: 20000 Res .Pwr . @FWHM

K:\Strongin LAB\...\08-25-2017-ER 8/25/2017 7:22:22 PM 08-25-2017-ER

08-25-2017-ER #9-33 RT: 0.08-0.27 AV: 12 NL: 7.08E4F: FTMS - p ESI Full ms [200.00-1000.00]

300 400 500 600 700 800 900 1000m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

834.64394

416.81803

248.95950 560.83080 856.62571316.94628 377.00322 708.74744628.82092 924.61278802.12402452.92050 520.90705 976.35045

825 830 835 840 845 850m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

834.64394

835.64739

836.64996828.79191 843.22814823.27429 847.10615831.55430

833.78089

850.06508839.06697 854.81226834.64667

835.65003

836.65338 839.66098

NL:7.08E408-25-2017-ER#9-33 RT: 0.08-0.27 AV: 12 F: FTMS - p ESI Full ms [200.00-1000.00]

NL:7.96E5

c20 h6 i4 o5 +H: C20 H7 I4 O5pa Chrg 1

K:\Strongin LAB\...\08-25-2017-FG 8/25/2017 7:24:50 PM 08-25-2017-FG

08-25-2017-FG #9-33 RT: 0.07-0.28 AV: 25 NL: 1.12E3T: ITMS - p ESI Full ms [200.00-1000.00]

200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

381.00001

785.00002690.50001296.00000238.83333 834.50002588.50001544.58334 964.83336416.75001 897.50002

377 378 379 380 381 382 383 384 385 386 387 388 389 390 391m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

381.06713

381.56860

382.06364377.00330 384.93299380.67883378.91697 390.89010383.56740 387.94161

381.06823

381.56991

382.06613383.06948 384.56906 385.57118

NL:2.18E508-25-2017-FG#9-33 RT: 0.08-0.27 AV: 12 F: FTMS - p ESI Full ms [200.00-1000.00]

NL:5.54E5

c37 h34 n2 o10 s3: C37 H34 N2 O10 S3pa Chrg 2

22

Figure S5. ESI-HRMS spectrum of Tartrazine Standard [M]3- ion. Theoretical mass: 154.99315. Observed mass: 154.99263. Δ m/z: 3.35 ppm

Figure S6. ESI-HRMS spectrum of Sunset Yellow FCF Standard. [M]2- ion. Theoretical mass: 202.99592. Observed mass: 202.99515. Δ m/z: 3.79 ppm

Figure S7. ESI-HRMS spectrum of Brilliant Blue FCF Standard. [M]2- ion. Theoretical mass: 373.07077. Observed mass: 373.07067. Δ m/z: 0.27 ppm

K:\Strongin LAB\...\08-31-2017-TA-NP 8/31/2017 4:47:20 PM 08-31-2017-TA-NP

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce 0

50

1000.15

0.14 0.160.11 0.19

0.220.24 0.260.09 0.32 0.40 0.47 0.56 0.60 0.73 0.79 0.83 0.94 0.97 1.05 1.09 1.17 1.23 1.32 1.41

0.170.13

0.110.20

0.240.36 0.46 0.56 0.61 0.750.71 0.82 0.89 1.02 1.21 1.311.110.93 1.41 1.46

0.15

0.160.11 0.19

0.24 0.260.40 0.43 0.56 0.60 0.73 0.79 0.86 0.94 1.05 1.09 1.171.02 1.23 1.32 1.41

NL: 7.74E7TIC MS 08-31-2017-TA-NP

NL: 1.13E7TIC F: ITMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-TA-NP

NL: 7.74E7TIC F: FTMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-TA-NP

100 150 200 250 300 350 400 450 500m/z

0

20

40

60

80

1000

20

40

60

80

100R

elat

ive

Abun

danc

e154.99263

210.99789

197.98478

232.99244170.99774140.32963343.29217303.75496 488.97568434.12692376.54812274.73291

154.99315

NL:8.78E608-31-2017-TA-NP#10-43 RT: 0.08-0.23 AV: 17 F: FTMS - p ESI Full ms [100.00-500.00]

NL:1.71E4

c16 h9 n4 o9 s 2: C16 H9 N4 O9 S2p (gss, s /p:40) Chrg 3R: 20000 Res .Pwr . @FWHM

K:\Strongin LAB\...\08-31-2017-SY-NP 8/31/2017 4:49:57 PM 08-31-2017-SY-NP

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce 0

50

1000.12 0.16 0.20

0.110.21

0.10 0.220.240.09 0.31 0.36 0.45 0.50 0.57 0.62 0.70 0.76 0.86 1.071.010.96 1.241.19 1.34 1.451.40

0.150.18

0.120.19

0.100.23

0.29 0.35 0.43 0.53 0.58 0.66 0.69 1.171.020.79 0.84 0.95 1.281.231.07 1.431.380.12 0.16 0.20

0.40 0.44 0.57 0.65 0.70 0.76 0.86 1.071.010.96 1.241.19 1.34 1.451.40

NL: 1.13E8TIC MS 08-31-2017-SY-NP

NL: 1.75E7TIC F: ITMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-SY-NP

NL: 1.13E8TIC F: FTMS - p ESI Full ms [100.00-500.00] MS 08-31-2017-SY-NP

190 195 200 205 210 215 220 225m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

202.99515

203.99338

206.99679 213.98532199.27224 211.09484

201.01408

221.99647193.08564 228.53435218.51252202.99592

203.99507

205.49560

NL:3.26E708-31-2017-SY-NP#6-47 RT: 0.04-0.24 AV: 21 F: FTMS - p ESI Full ms [100.00-500.00]

NL:1.74E4

c16 h10 n2 o7 s 2: C16 H10 N2 O7 S2p (gss, s /p:40) Chrg 2R: 20000 Res .Pwr . @FWHM

K:\Strongin LAB\...\08-25-2017-BB 8/25/2017 7:19:54 PM 08-25-2017-BB

08-25-2017-BB #6-21 RT: 0.05-0.17 AV: 8 NL: 4.42E5F: FTMS - p ESI Full ms [200.00-1000.00]

300 400 500 600 700 800 900 1000m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

373.07067

769.13041225.00958 333.09229 837.11770561.11336481.15675

384.93403

889.08438422.05928

961.23393616.84228 700.85510

368 370 372 374 376 378 380 382 384m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

373.07067

373.57197

374.06995375.07189 377.00351372.69431 384.93403368.88981 378.91762 380.83104370.35115

373.07077

373.57245

374.06867375.07203 376.57160 378.06994

NL:4.42E508-25-2017-BB#6-21 RT: 0.05-0.17 AV: 8 F: FTMS - p ESI Full ms [200.00-1000.00]

NL:5.56E5

c37 h34 n2 o9 s3: C37 H34 N2 O9 S3pa Chrg 2

23

ESI-HRMS validation of synthetic dyes in e-liquid samples

Figure S8. ESI-HRMS spectrum of Allura Red AC in Red E-liquid. Theoretical mass: 225.00903. Observed mass: 225.00829. Δ m/z: 3.29 ppm

Figure S9. ESI-HRMS spectrum of Tartrazine in Green E-liquid. Theoretical mass: 154.99315. Observed mass: 154.99319. Δ m/z: 0.26 ppm

Figure S10. ESI-HRMS spectrum of Brilliant Blue FCF Dye in Green E-liquid. Theoretical: 373.07077. Observed: 373.07184. Δ m/z: 2.87 ppm

K:\Strongin LAB\...\AR 9/13/2017 7:04:25 PM AR+pnitrophenol

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce 0

50

1000.22 0.23 0.28

0.21 0.35 0.390.07 0.10 0.48 0.780.51 0.72 0.84 0.890.61 1.21 1.270.90 1.050.12 1.08 1.37 1.43

0.20 0.270.08

0.36 0.39 0.43 0.48 0.57 0.740.64 0.88 1.000.78 1.130.97 1.411.22 1.271.07 1.38

0.22 0.280.35 0.390.07 0.10 0.41 0.780.51 0.57 0.72 0.84 0.890.61 1.21 1.271.051.00 1.08 1.37 1.43

NL: 1.12E7TIC MS AR

NL: 2.80E6TIC F: ITMS - p ESI Full ms [100.00-500.00] MS AR

NL: 1.12E7TIC F: FTMS - p ESI Full ms [100.00-500.00] MS AR

224.5 225.0 225.5 226.0 226.5 227.0 227.5m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

225.00829

225.50980226.00616

226.50769224.83192 225.18503 227.05444225.00903

225.51048226.00837

226.50937 227.00828 227.50897

NL:1.38E6AR#32 RT: 0.18 AV: 1 F: FTMS - p ESI Full ms [100.00-500.00]

NL:1.69E4

c18 h14 n2 o8 s 2: C18 H14 N2 O8 S2p (gss, s /p:40) Chrg 2R: 20000 Res .Pwr . @FWHM

K:\Strongin LAB\...\GREEN+TA 9/13/2017 7:06:54 PM greenTA+pnitrophenol

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce 0

50

1000.300.250.22 0.330.08

0.350.440.13 0.49 0.580.07 0.730.63 0.840.76 1.020.90 1.360.94 1.421.11 1.271.17 1.49

0.390.08 0.37 0.44 0.530.300.25 0.58 0.64 0.810.19 0.73 1.450.88 1.08 1.211.06 1.240.93 1.281.16 1.40

0.300.250.22 0.330.080.38 0.440.13 0.49 0.58 0.730.63 0.840.76 1.020.90 1.360.94 1.421.11 1.271.17 1.49

NL: 8.20E7TIC MS GREEN+TA

NL: 8.98E6TIC F: ITMS - p ESI Full ms [100.00-500.00] MS GREEN+TA

NL: 8.20E7TIC F: FTMS - p ESI Full ms [100.00-500.00] MS GREEN+TA

153.5 154.0 154.5 155.0 155.5 156.0 156.5 157.0 157.5 158.0m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

154.99319

154.03465157.12236

155.32773155.65848

154.99315

155.32736155.65932

156.32589 156.65970 157.32673 157.66028

NL:1.20E5GREEN+TA#178 RT: 0.99 AV: 1 F: FTMS - p ESI Full ms [100.00-500.00]

NL:1.71E4

c16 h9 n4 o9 s 2: C16 H9 N4 O9 S2p (gss, s /p:40) Chrg 3R: 20000 Res .Pwr . @FWHM

K:\Strongin LAB\...\BB+TA_171029212949 10/29/2017 9:29:49 PM BB+TA

RT: 0.00 - 1.50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4Time (min)

0

50

1000

50

100

Rel

ativ

e Ab

unda

nce 0

50

1000.120.10 0.13

0.440.16 0.270.26 0.29 0.40 0.540.500.320.24 0.550.07 0.61 0.62 0.65 1.120.76 1.020.90 1.110.79 0.980.87 0.94 1.17 1.18 1.27 1.391.300.02 1.41 1.47

0.090.13 0.15 0.29 0.41 0.520.430.18 0.38 0.54

0.630.58 0.67 1.120.770.74 1.031.000.89 1.150.94 1.060.81 1.17 1.401.24 1.421.280.03 1.36

0.120.100.440.27 0.40 0.540.50

0.200.07 0.61 0.64 1.120.760.72 1.020.90 0.980.87 1.171.07 1.23 1.27 1.391.30 1.47

NL: 2.39E7TIC MS BB+TA_171029212949

NL: 1.13E6TIC F: ITMS - p ESI Full ms [300.00-400.00] MS BB+TA_171029212949

NL: 2.39E7TIC F: FTMS - p ESI Full ms [300.00-400.00] MS BB+TA_171029212949

372.5 373.0 373.5 374.0 374.5 375.0 375.5 376.0 376.5 377.0 377.5 378.0 378.5m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

373.07184

373.57358

374.57083

373.07077

373.57231

374.07094

374.57151 375.07090 375.57109 376.07096 376.57112 377.07130 377.57160 378.07188

NL:2.24E5BB+TA_171029212949#94 RT: 0.52 AV: 1 F: FTMS - p ESI Full ms [300.00-400.00]

NL:1.30E4c 37 h 34 n 2 o 9 s 3: C 37 H34 N2 O 9 S3p (gss, s /p:40) Chrg 2R: 20000 Res .Pwr . @FWHM

24

Figure S11. ESI-HRMS spectrum of Brilliant Blue FCF in Blue E-liquid. Theoretical mass: 373.07077. Observed mass: 373.07220. Δ m/z: 3.83 ppm

K:\Strongin LAB\...\bb_180105094228 1/5/2018 9:42:28 AM BB

RT: 0.00 - 1.49

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4Time (min)

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

0.230.210.24

0.200.25 0.28

0.17

0.300.06

0.090.32

0.350.370.40 0.43

0.47 0.510.56

0.64 0.69 0.77 0.84 0.92 1.03 1.12 1.16 1.23 1.31 1.36 1.48

NL:2.02E7TIC MS bb_180105094228

373 374 375 376 377m/z

0

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

373.07220

373.57342

374.07160374.57127 377.10868375.07295372.69606 375.67752 376.27824 377.60504

373.07077

373.57231

374.07094

374.57151 375.07090 376.07096 376.57112 377.57160

NL:9.04E5bb_180105094228#1-180 RT: 0.00-1.49 AV: 180 T: FTMS - p ESI Full ms [300.00-400.00]

NL:1.30E4

c37 h34 n2 o9 s3: C37 H34 N2 O9 S3p (gss, s /p:40) Chrg 2R: 20000 Res .Pwr . @FWHM

25

Ion chromatograms of vaporized samples

Figure S12. IC of vaporized Red E-liquid. Bottom to top: replicates 1 – 3 and the spiked sample.

Figure S13. IC of vaporized Blue E-liquid. Bottom to top: replicates 1 – 3 and the spiked sample.

Figure S14. IC of vaporized Green E-liquid. Bottom to top: replicates 1 – 3 and the spiked sample.

26

REFERENCES

(1) Slater, S. J.; Chaloupka, F. J.; Wakefield, M.; Johnston, L. D.; O'Malley, P. M. The impact of retail cigarette marketing practices on youth smoking uptake. Arch. Pediatr. Adolesc. Med. 2007, 161, 440-445.

(2) Bassett, R. A.; Osterhoudt, K.; Brabazon, T. Nicotine poisoning in an infant. N. Engl. J. Med. 2014, 370, 2249-2250.

(3) Measham, F.; O’Brien, K.; Turnbull, G. “Skittles & Red Bull is my favourite flavour”: E-cigarettes, smoking, vaping and the changing landscape of nicotine consumption amongst British teenagers–implications for the normalisation debate. Drugs 2016, 23, 224-237.

(4) Zhu, S. H.; Sun, J. Y.; Bonnevie, E.; Cummins, S. E.; Gamst, A.; Yin, L.; Lee, M. Four hundred and sixty brands of e-cigarettes and counting: implications for product regulation. Tob. Control 2014, 23 Suppl 3, iii 3-9.

(5) Kamboj, A.; Spiller, H. A.; Casavant, M. J.; Chounthirath, T.; Smith, G. A. Pediatric exposure to e-cigarettes, nicotine, and tobacco products in the united states. Pediatrics 2016, 137, e20160041.

(6) Talbot, P. Potential health effects of electronic cigarettes: a systematic review of case reports. Prev. Med. Rep. 2016, 4, 169-178.

(7) Jensen, R. P.; Strongin, R. M.; Peyton, D. H. Solvent chemistry in the electronic cigarette reaction vessel. Sci. Rep. 2017, 7, 42549.

(8) Tierney, P. A.; Karpinski, C. D.; Brown, J. E.; Luo, W.; Pankow, J. F. Flavour chemicals in electronic cigarette fluids. Tob. Control 2016, 25, e10-5.

(9) Spence, C.; Levitan, C. A.; Shankar, M. U.; Zampini, M. Does food color influence taste and flavor perception in humans? Chemosens. Percept. 2010, 3, 68-84.

(10) Burdock, G. A.; Carabin, I. G. Generally recognized as safe (GRAS): history and description. Toxicol. Lett. 2004, 150, 3-18.

(11) Thorngate, J. Synthetic Food Colorants. In Food additives; 2nd ed.; Branen, A. L., Ed.; Marcel Dekker, Inc.: New York, 2002, p 422-500.

(12) Code of Federal Regulations, Title 21, Parts 70, 73, 74, and 82 and Section 101.22(k). (2012) U.S. Government Printing Office; Available from: http://www.fda.gov/ForIndustry/ColorAdditives/ColorAdditiveInventories/default.htm. (Accessed September 20, 2017).

(13) Nollet, L. M. L. Handbook of food analysis. Residues and other food component analysis.; Marcel Dekker: New York, 2004; Vol. 2.

(14) Petigara Harp, B.; Miranda-Bermudez, E.; Barrows, J. N. Determination of seven certified color additives in food products using liquid chromatography. J. Agric. Food Chem. 2013, 61, 3726.

(15) Recommended method No. 81: Routine analytical machine for e-cigarette aerosol generation and collection – definitions and standard conditions. CORESTA; Available from: https://www.coresta.org/routine-analytical-machine-e-cigarette-aerosol-generation-and-collection-definitions-and-standard. (Accessed September 1, 2016).

(16) Fast HPLC analysis of dyes in foods and beverages, Dionex Corporation, 2010. (17) Peiperl, M. D.; Prival, M. J.; Bell, S. J. Determination of combined benzidine in

FD&C Yellow No. 6 (Sunset Yellow FCF). Food Chem. Toxicol. 1995, 33, 829-839.

27

(18) Lancaster, F. E.; Lawrence, J. F. Determination of benzidine in the food colours tartrazine and sunset yellow FCF, by reduction and derivatization followed by high-performance liquid chromatography. Food Addit. Contam. 1999, 16, 381-90.

(19) Lancaster, F. E.; Lawrence, J. F. Determination of total non-sulphonated aromatic amines in tartrazine, sunset yellow FCF and allura red by reduction and derivatization followed by high-performance liquid chromatography. Food Addit. Contam. 1991, 8, 249-263.

(20) Bailey Jr, J. E.; Bailey, C. J. Determination of aromatic amines in FD&C Yellow No. 5 by diazotization and coupling followed by reversed-phase HPLC. Talanta 1985, 32, 875-882.

(21) Robinson, G. Tartrazine - the story so far. Food Chem. Toxicol. 1988, 26, 73-76. (22) Hurst, W. J.; Mckim, J. M.; Martin, R. A. Determination of tartrazine in food-

products by hplc. J. Food Sci. 1981, 46, 419. (23) Juhlin, L.; Michaelsson, G.; Zetterstrom, O. Urticaria and asthma induced by food-

and-drug additives in patients with aspirin hypersensitivity. J. Allergy Clin. Immunol. 1972, 50, 92-98.

(24) Settipane, G. A.; Chafee, F. H.; Postman, I. M.; Levine, M. I.; Saker, J. H.; Barrick, R. H.; Nicholas, S. S.; Schwartz, H. J.; Honsinger, R. W.; Klein, D. E. Significance of tartrazine sensitivity in chronic urticaria of unknown etiology. J. Allergy Clin. Immunol. 1976, 57, 541-546.

(25) Cheeseman, M. Certified color additives and childhood hyperactivity Office of Food Additive Safety. Food Advisory Committee; Available from: https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/FoodAdvisoryCommittee/UCM273030.pdf. (Accessed September 20, 2017).

(26) Evaluation of certain food additives: eighty-second report of the Joint FA; FAO, World Health Organization: Expert Committee on Food Additives, 2016.

(27) Reineccius, G.; McGorrin, R. J. Source book of flavors In Trends in Food Science and Technology; 2 ed. 1994; Vol. 5, p 798.

(28) Chou, S. S.; Lin, Y. H.; Cheng, C. C.; Hwang, D. F. Determination of synthetic colors in soft drinks and confectioneries by micellar electrokinetic capillary chromatography. J. Food Sci. 2002, 67, 1314-1318.

(29) de Andrade, F. I.; Florindo Guedes, M. I.; Pinto Vieira, I. G.; Pereira Mendes, F. N.; Salmito Rodrigues, P. A.; Costa Maia, C. S.; Marques Avila, M. M.; de Matos Ribeiro, L. Determination of synthetic food dyes in commercial soft drinks by TLC and ion-pair HPLC. Food Chem. 2014, 157, 193-198.

(30) Yang, Y.; Zhang, J.; Shao, B. Quantitative analysis of fourteen synthetic dyes in jelly and gummy candy by ultra performance liquid chromatography. Anal. Meth. 2014, 6, 5872-5878.

(31) Hajimahmoodi, M.; Afsharimanesh, M.; Moghaddam, G.; Sadeghi, N.; Oveisi, M. R.; Jannat, B.; Pirhadi, E.; Zamani Mazdeh, F.; Kanan, H. Determination of eight synthetic dyes in foodstuffs by green liquid chromatography. Food Addit. Contam. Part A 2013, 30, 780-785.

(32) Fuh, M. R.; Chia, K. J. Determination of sulphonated azo dyes in food by ion-pair liquid chromatography with photodiode array and electrospray mass spectrometry detection. Talanta 2002, 56, 663-671.

(33) Feng, F.; Zhao, Y.; Yong, W.; Sun, L.; Jiang, G.; Chu, X. Highly sensitive and accurate screening of 40 dyes in soft drinks by liquid chromatography-electrospray tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 2011, 879, 1813-1818.

28

(34) Alves, S. P.; Brum, D. M.; Branco de Andrade, E. C.; Pereira Netto, A. D. Determination of synthetic dyes in selected foodstuffs by high performance liquid chromatography with UV-DAD detection. Food Chem. 2008, 107, 489-496.

(35) Clydesdale, F. M. Color as a factor in food choice. Crit. Rev. Food Sci. Nutr. 1993, 33, 83-101.

(36) Johnson, J. L.; Dzendolet, E.; Clydesdale, F. M. Psychophysical relationship between sweetness and redness in strawberry-flavored drinks. J. Food Prot. 1983, 46, 21-25.

(37) Chequer, F. M.; Lizier, T. M.; de Felicio, R.; Zanoni, M. V.; Debonsi, H. M.; Lopes, N. P.; Marcos, R.; de Oliveira, D. P. Analyses of the genotoxic and mutagenic potential of the products formed after the biotransformation of the azo dye Disperse Red 1. Toxicol. In Vitro 2011, 25, 2054-2063.

(38) Makkonen, U.; Juntto, S. Field comparison of measurement methods for sulphur dioxide and aerosol sulphate. Atmos. Environ. 1997, 31, 983-990.

(39) Vlase, T.; Vlase, G.; Modra, D.; Doca, N. Thermal behaviour of some industrial and food dyes. J. Therm. Anal. Calorim. 2007, 88, 389-393.

(40) Rehorek, A.; Plum, A. Characterization of sulfonated azo dyes and aromatic amines by pyrolysis gas chromatography/mass spectrometry. Anal. Bioanal. Chem. 2007, 388, 1653-1662.

(41) Duell, A. Peyton Lab. Portland State University. Department of Chemistry. (42) Damodaran, S. Amino acids, peptides and proteins. In Fennema's Food Chemistry;

4th ed.; CRC Press: Boca Raton, FL, 2008; Vol. 4, p 217-329. (43) Bannach, G.; Almeida, R. R.; Lacerda, L.; Schnitzler, E.; Ionashiro, M. Thermal

stability and thermal decomposition of sucralose. Eclet. Quim. J. 2009, 34, 21-26. (44) de Oliveira, D. N.; de Menezes, M.; Catharino, R. R. Thermal degradation of

sucralose: a combination of analytical methods to determine stability and chlorinated byproducts. Sci. Rep. 2015, 5, 9598.

(45) Rahn, A.; Yaylayan, V. A. Thermal degradation of sucralose and its potential in generating chloropropanols in the presence of glycerol. Food Chem. 2010, 118, 56-61.

(46) Dong, S. J.; Wu, J. J.; Liu, G. R.; Zhang, B.; Zheng, M. H. Unintentionally produced dioxin-like polychlorinated biphenyls during cooking. Food Control 2011, 22, 1797-1802.