An Analytical Perspective on Determination of Free Base

Research ArticleAn Analytical Perspective on Determination of Free BaseNicotine in E-Liquids

Vinit V Gholap 1 Rodrigo S Heyder2 Leon Kosmider3 and Matthew S Halquist 1

1Department of Pharmaceutics School of Pharmacy Virginia Commonwealth University Richmond 23298 VA USA2Department of Pharmaceutics Pharmaceutical Engineering School of Pharmacy Virginia Commonwealth UniversityRichmond 23298 VA USA3Department of General and Inorganic Chemistry Medical University of Silesia Katowice FOPS in Sosnowiec Jagiellonska 441-200 Sosnowiec Poland

Correspondence should be addressed to Matthew S Halquist halquistmsvcuedu

Received 18 October 2019 Accepted 4 February 2020 Published 10 March 2020

Academic Editor Demetrio Milea

Copyright copy 2020 Vinit V Gholap et al -is is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

In electronic cigarette users nicotine delivery to lungs depends on various factors One of the important factors is e-liquid nicotineconcentration Nicotine concentration in e-liquids ranges from 0 to gt50mgmL Furthermore nicotine exists in protonated andunprotonated (ldquofree baserdquo) forms -e two forms are believed to affect the nicotine absorption in body -erefore in addition tototal nicotine concentration e-liquids should be characterized for their free base nicotine yield Two approaches are being used forthe determination of free base nicotine in e-liquids -e first is applying a dilution to e-liquids followed by two methodsHendersonndashHasselbalch theory application or a Liquid-Liquid Extraction -e second is the without-dilution approach followedby 1H NMR method Here we carried out controlled experiments using five e-liquids of different flavors using these twoapproaches In the dilution approach the HendersonndashHasselbalchmethod was tested using potentiometric titration-e accuracywas found to be gt98 for all five e-liquid samples (n 3) A Liquid-Liquid Extraction was carried out using toluene or hexane asextraction solvent -e Liquid-Liquid Extraction technique was found to be limited by solvent interactions with flavors Solventextractions resulted in flavor dependent inaccuracies in free base nicotine determination (5 to 277 of calculated values) -ewithout-dilution approach was carried out using 1H NMR as described by Duell et al -is approach is proposed to offer anindependent and alternative scale None of the methods have established a strong correlation between pre- and postvaporizationfree base nicotine yield Here we present comparative results of two approaches using analytical techniques Such a comparisonwould be helpful in establishing a standardized method for free base nicotine determination of e-liquids

1 Introduction

Nicotine is an alkaloid with a weakly basic nature Based onthe solvent nature it can exist as protonated and free basenicotine Historically tobacco companies have been usingalkaline chemical substances such as ammonia or its relatedbasic compounds in manufacturing of cigarettes [1 2]Although the tobacco companies have denied the effect ofsuch substances on the alteration of the absorption ofnicotine USFDA argued that using such alkaline substancesshifts the balance from protonated (NicH+) to free base form(Nic) of nicotine in tobacco to cause a rapid and efficient

absorption of nicotine in consumers [1 2] Based on thePankow theory and studies by tobacco companies [3] freebase nicotine is responsible for harshness or impact insmokers Here impact means ldquosudden sharp but short livedsensation which is noticed immediately [when] smokemakes contact with the back of the throatrdquo [3 4] -epossible reason for such harshness is the free base nicotine inaerosol which is readily volatile and can deposit quickly inupper respiratory track [1 3] Based on pH partition hy-pothesis of drug absorption [5] any drug easily penetratesbiological membrane barrier in an unionized form -ere-fore free base nicotine is believed to easily cross the

HindawiJournal of Analytical Methods in ChemistryVolume 2020 Article ID 6178570 12 pageshttpsdoiorg10115520206178570

biological membrane of the respiratory tract and lead tonicotinersquos rapid absorption Many authors based on their invitro and in vivo studies have confirmed that rate of nicotineabsorption is higher for basic than acidic nicotine solutionsand aerosols [6ndash10]

In contrast PAX Labs Inc in its patent has shown that asalt form of nicotine which is mostly protonated nicotinegave higher plasma nicotine concentration (Cmax) than freebase nicotine One possible explanation for this outcomecould be that the less volatile and particulate nature ofprotonated nicotine can have deeper penetration into thelungs where absorption is more rapid than upper respiratorytract Additionally since protonated nicotine does not causeimpact in throat [11] user may perform deep inhalationwhich can result in higher amount of nicotine reaching toalveolar region

Experimentally it is still not clear how nicotine crossesbiological membrane With the two conflicting theories andexperimental data about the absorption of free base vsprotonated nicotine independent and unbiased scientificresearch is still not available to determine the ldquorate of nicotineuptake in smokersvapers as a function of its free base doserdquoIn either case it is important to classify e-liquids based ontheir free base or protonated nicotine yield for two reasons

(a) Frequent high amounts of nicotine absorption canlead to its addiction Unlike European regulations[12] FDA does not specify a cap on maximum nic-otine content in e-liquids Nicotine content ine-liquids ranges from 0 to gt50mgmL [13] JUULproducts which have gained high popularity in recentyears have about 50mgmL content of nicotine [14]which is believed to yield mainly the protonated form(NicH+) Additionally the current 3rd and 4th gen-eration boxmod e-cigarette devices allow control overpower settings such that high power gives high nic-otine delivery-e question that arises here is whethersuch a high amount of nicotine in e-liquids is justi-fiable if protonated nicotine is suspected of higherabsorption Increased exposure of nicotine in youthcan affect the prefrontal cortex development leadingto a deficiency in attention and lasting effects oncognitive functions [15] -erefore to control theyouth addiction of nicotine it is important to addressthe free base or protonated nicotine content ine-liquids for their adequate regulation

(b) Unlike free base nicotine salt based nicotine causesless harshness in throat hit and thus improves pal-atability and smoothness of e-cig aerosol [11] Basedon popularity surveys and flavor studies of severalother e-liquids different flavors of e-liquids of equalstrength are perceived to give different nicotineimpacts [16ndash18] -us classifying e-liquids based ontheir free base nicotine delivery would help in un-derstanding the acceptance criteria of e-liquids byvapers

We have recently proposed a potential standard methodfor analyzing total nicotine content in e-liquids using peak

purity criteria by HPLC [19] However there is no stan-dardized method yet to quantify free base nicotine yieldfrom e-liquids

Figure 1 explains the two major approaches currentlybeing used for the determination of free base nicotine ine-liquids -e first is a dilution approach and the second iswithout-dilution approach A dilution approach has beenutilized in two methods HendersonndashHasselbalch methodand Liquid-Liquid Extraction method Without-dilutionapproach has been followed by 1H NMR method alone

A dilution approach followed by HendersonndashHasselbalchmethod has been used by many authors [13 20ndash22] Howeverdue to various concerns about the effect of flavors on theaccuracy of results obtained by HendersonndashHasselbalchequation a Liquid-Liquid Extraction method was proposed byEl-Hellani et al

It can be argued that the dilution approach causeschange in e-liquidsrsquo original solvent system which is majorlynonaqueous Duell et al argued that this change in solventsystem may not give a true picture of free base nicotine ine-liquids Additionally different dilution factors might in-fluence the results and make comparisons challenging Toaddress this issue Duell at el proposed a without-dilutionapproach by which free base nicotine is determined using 1HNMR

To compare and present critique of the two approacheswe carried out controlled experiments using five e-liquids ofdifferent flavors We present results using various analyticaltechniques that will facilitate the establishment of a stan-dardized method by regulatory bodies for classifyinge-liquids based on their free base nicotine yield

2 Materials and Methods

21 Instrumentation Analysis of nicotine by potentiometrictitration was carried out using a TruLab pH 1310P (YSIIncorporated Xylem Inc USA) potentiometric pH meterwith TruLine 15 glass electrode selective to H+ ions andcontaining silver chloride reference electrode Liquid-LiquidExtraction was carried out using a Waters Alliance 2695quaternary pump HPLC equipped with Waters 996 PDADetector Hypersil Gold Phenyl column (150mmtimes 46mm3 μm -ermo Scientifictrade USA) and a Security GuardCartridge Phenyl (4mmtimes 20mm Phenomenex USA)Waters Empower 2 software was used for processing dataNMR analysis was carried out using Bruker NanoBayAVANCE III 400MHz NMR spectrometer (Bruker Cor-poration USA)

22 Chemicals and Reagents (minus )-Nicotine liquid standard(purityge 99) was purchased from Sigma Aldrich USAHPLC grade acetonitrile methanol and water were pur-chased from BDH Chemicals VWR USA Ortho phos-phoric acid (85) was purchased fromMerck USA Triethylamine tert-butyl amine and hydrochloric acid (37) werepurchased from Sigma Aldrich USA Glacial acetic acid waspurchased from Macron Fine Chemicals USA Sodiumhydroxide (10N) was purchased from BDH Chemicals

2 Journal of Analytical Methods in Chemistry

VWR USA Propylene glycol was purchased from AmrescoLLC VWR USA USP grade vegetable glycerin was pur-chased from JT Baker USA e-Liquid flavors (withoutnicotine) of menthol tobacco fruit sweet and coffee werepurchased from Direct Vapor online vape shop USA NMRanalysis was carried out using coaxial inserts for 5mmNMRprecision sample tube (WGS-5BL-SP Wilmad-LabGlassUSA) Dimethyl sulfoxide-d6 was purchased from SigmaAldrich USA

23 Preparation of Reagents e-Liquids were prepared bydissolving liquid nicotine standard in each flavor Similarlyquality control (QC) samples were prepared by dissolvingliquid nicotine standard in unflavored matrix of propyleneglycol and vegetable glycerin (1 1 vv) Standards mobilephase and diluent for HPLC analysis were prepared asdescribed earlier [19] In potentiometric titration experi-ments pH adjustment and titration were carried out using01N NaOH and 01N HCl Water used for all preparationswas HPLC grade purified water Sample preparation forNMR analysis was carried out based on the method sug-gested by Duell et al [23]

24 Dilution Approach

241 HendersonndashHasselbalchMethodHendersonndashHasselbalchmethod for determination of free base nicotine yield ofe-liquids was tested using potentiometric titration

(1) Potentiometric Titration Potentiometric titration is atechnique based on measuring the change in potential afteraddition of titrant to a solution containing counterions tothat of the titrant Equivalence point is the point at whichthe rate of change in potential per each incremental ad-dition of titrant is maximum Change in potential wasmeasured for each incremental addition of titrant 01NNaOH

(a) Determination of pKa of pyrrolidine group ofnicotineNicotine standard solution of 08mgmL was pre-pared in water at three different pHs 7 8 and 9 -epH adjustments were carried out using 01NHCl and01N NaOH Potentiometric titration of these

solutions was carried out against 01N NaOHEquivalence point was determined by a first deriv-ative plot of change in potential vs volume of NaOHadded pKa was calculated using Hender-sonndashHasselbalch equation [5]

pH pKa + log[Nic]NicH+[ ]

(1)

where Ka is the nicotinersquos acid dissociation constant[Nic] is the free base nicotine concentration and[NicH+] is the protonated nicotine concentration

(b) Determination of free base nicotine from e-liquidsPotentiometric titration was carried out to quantifyNicH+ in e-liquid solutions (10x diluted) e-Liquidscontaining 80mgmL of nicotine were preparedusing five different flavors menthol tobacco fruitsweet and coffee A control e-liquid was prepared inPG VG (1 1 vv) Using the dilution approach eache-liquid was diluted 10x in water to achieve finalnicotine concentration of 8mgmL Due to sensi-tivity of the potentiometer and for ease of the ti-tration nicotine concentration of diluted sampleswas set as 8mgmL -e pH of the diluted e-liquidswas measured and potentiometric titration wascarried out against 01N NaOH Similarly placeboflavors and placebo control all without nicotinewere analyzed after by 10x dilution in water Firstthe pH of each diluted placebo was measuredSubsequently the pH of each diluted placebo wasadjusted to the pH of their respective diluted e-liquidsample using 01N NaOH Finally potentiometrictitration was carried out similar to that of samplesSchematic representation of the methodology isdescribed in Figure 2 Flavor cations generated ifany after pH adjustment of placebo by 01N NaOHare subtracted from total titrated cations generatedfrom sample to give the NicH+ -e percentage ()of free base nicotine (Nic) was calculated usingequations (2)ndash(4)

Since both the sample and placebo were diluted in thesame grade of water (HPLC grade with constant pH) effectof dissolved CO2 in water was nullified and was not con-sidered for calculations

NicH+1113858 1113859 [total cations] minus [flavor cations] (2)

[Nic] Nic + NicH+1113858 1113859 minus NicH+

1113858 1113859 (3)

Nic [Nic]

Nic + NicH+[ ]times 100 (4)

242 Liquid-Liquid Extraction Method Based on the hy-pothesis proposed by El-Hellani et al [24] organic solventsshould demonstrate selective extraction of free base nicotinefrom an aqueous solution of nicotine -erefore the ex-traction method was developed based on the earlier research

Free

bas

e nic

otin

e in

e-liq

uids

Dilution

HendersonndashHasselbalch method

Liquid-liquid extraction method

Without dilution 1H NMR method

Figure 1 Schematic of methods for free base nicotine determi-nation in e-liquids

Journal of Analytical Methods in Chemistry 3

published by El-Hellani et al [24] Briefly nicotine stan-dard solutions (08mgmL) of five different pHs namely 57 9 11 and 13 were prepared in water El-Hellani et alhave used toluene as extracting solvent To assess the ex-traction efficiency of toluene we compared it with anotherwell studied organic solvent hexane which had been usedfor recovery of nicotine from tobacco extracts [25] Ex-traction efficiency was determined for single and doubleextractions for each solvent Nicotine in aqueous and or-ganic layer was quantified by an HPLC method [19] Asimilar procedure was followed for nicotine extractionusing e-liquids (8mgmL) of five different flavors namelymenthol tobacco fruit sweet and coffee Extractionmethod involves 10x dilution of e-liquids in aqueous layer-us final concentration of e-liquids achieved in aqueouslayer was similar to standards ie 08mgmL of nicotineAdditionally pH measurement was carried out for bothstandards and samples (10x diluted) to calculate free baseusing HendersonndashHasselbalch equation Figure 3 describesthe methodology of extraction

25 Without-Dilution Approach

251 1H NMR Spectroscopy Based on methodology byDuell et al [23] free base nicotine in e-liquids was deter-mined by 1H NMR spectroscopy Briefly e-liquids wereprepared at concentration of 8mgmL using the same fivedifferent flavors (menthol tobacco fruit sweet and coffee)-e control e-liquid was prepared in propylene glycol vegetable glycerin (PG VG 54 46 vv) Free base andprotonated standards were prepared by combining the al-iquots of the control sample with base (t-butylamine nicotine 1 1mol mol) and with acid (acetic acid nicotine5 1 mol mol)

NMR spectroscopy was carried out using precisioncoaxial NMR inserts with experimental parameters as de-scribed by Duell et al [23] Free base nicotine was calculatedusing equations (5) and (6) [23] and (7) based on differencein chemical shifts between aromatic hydrogens and hy-drogens of the methyl (-CH3) group which connects toprotonable nitrogen (N) atom of the pyrrolidine ring

Δδ [δHaromatic proton(ieHa throughHd)] minus [δHe] (5)

αfb [(Δδ commercial sample) minus (Δδmonoprotonated standard)]

[(Δδ free base standard) minus (Δδmonoprotonated standard)] (6)

Nic αfb times 100 (7)

e-Liquid sample (80mgmL)

e-Liquid flavor (0mgmL)

Measure pH

10times dilution 10times dilution

Measure pH and mV

Potentiometric titration against 01N NaOH

Equivalence point

[Nic] = [H+] [H+] = [OHndash]

[OHndash] = [NicH+] + [Flavor+] [OHndash] = [Flavor+]

01N NaOH

Figure 2 Schematic of methodology of potentiometric titration to measure protonated nicotine [NicH+]


3 Results


311 HendersonndashHasselbalch Method

(1) Potentiometric Titration

(a) Determination of pKa of pyrrolidine group ofnicotineTo confirm the accuracy of potentiometric titrationthe pKa was determined using a nicotine standardsolution (08mgmL) at pHs 7 8 and 9 Each so-lution was titrated against 01N NaOH and equiv-alence point was determined by a first derivative plotof change in potential vs volume of NaOH addedUsing equation (1) the pKa of pyrrolidine group wasfound to be 817plusmn 012 at 190plusmn 1degC

(b) Determination of free base nicotine from e-liquidsUsing potentiometric titration e-liquids were ana-lyzed for free base nicotine yield as described in (b)in Section 241 and Figure 2 Using equations(2)ndash(4) free base nicotine was calculated andcompared with the theoretical results obtained fromHendersonndashHasselbalch equation -e mean per-centage difference between experimental and theo-retical results (HH-PT) was found to be 112plusmn 064for all e-liquids (Table 1 and Figure 4)

312 Liquid-Liquid Extraction Method As discussed in themethodology free base nicotine quantification was carriedout using hexane and toluene to selectively extract free basenicotine from aqueous solutions at pH 5 7 9 11 and 13Figure 5 describes the hexane and toluene extraction effi-ciency for nicotine standard solutions at various pHs Sincenicotine is expected to follow HendersonndashHasselbalch

equation in its standard solution in water free base nicotineextraction results were compared with the theoretical valuesobtained from the equation

Based on the difference between the theoretical valuesand experimental results toluene was found to have higherextraction efficiency (9632plusmn 351) for free base nicotine ascompared to hexane (7734plusmn 372) (Tables 2 and 3 andFigure 5)

We also calculated nicotinersquos partition coefficient intoluene-water system (KTW) as 692 after a single extractionIn contrast the partition coefficient of nicotine in hexane-water system (KHW) was found to be 125 -erefore basedon the results of nicotine standard solutions toluene wasused as an extracting solvent for e-liquids e-Liquids of fivedifferent flavors namely menthol tobacco fruit sweet andcoffee were used for toluene extraction as described inSection 24

Since the pKa of any molecule depends on the solvent itis dissolved in [26] the pKa of nicotine can be perturbed by asolvent environment due to presence of flavoring chemicalsPG or VG However it was confirmed from the potentio-metric experiment (Section 241 and Figure 4) that 10xdilution of e-liquids in water smooths out the effect of otherchemicals on pKa of nicotine in water

-erefore similar to the standard solution Hender-sonndashHasselbalch equation was used to calculate the free basenicotine yield from e-liquid samples -e mean percentagedifference between theoretical results and experimentalresults (HH-LLE) was found to be variable ranging from 574to 27702 for five e-liquids (Table 4 and Figure 6)


321 1H NMR Spectroscopy As described in Section 251the 1H NMR experiment was performed on e-liquidswithout dilution Using equations (5)ndash(7) the free base

Liquid-liquid extraction Toluene Water

WT

T

W

W

WT

T

T WH+

WH+

TT

Discard

2nd extraction2nd extraction

WH

H

W

W

WH

H

1st extraction 1st extraction

Liquid-liquid extraction hexane Water

H Evaporation and reconstitution with

diluent

HPLC

HPLCReconstitution with diluent

Figure 3 Schematic of methodology of measurement of free base and protonated nicotine by extraction using hexane (H) or toluene (T) inwater (W) or acid (WH+)


nicotine was calculated for five e-liquid flavors (Table 5) Asthe e-liquids were not diluted in water the results werefound to be different from the results obtained by Hen-dersonndashHasselbalch method (Table 6) with a deviationfrom 370 to 2693

4 Discussion

Considering the addictive nature of nicotine e-liquids needto be regulated for the total nicotine content as well as freebase nicotine yield -ere are published methods for de-termining total nicotine content in e-liquids using gaschromatography and liquid chromatography coupled withmass detectors [27ndash29] Recently we have proposed a HPLC

method based on peak purity criteria for accurate quanti-fication of total nicotine content [19] In contrast methodsfor quantification of free base nicotine in e-liquids are stillbeing debated for their accuracy


411 HendersonndashHasselbalch Method Many authors haveused this approach by diluting e-liquids in fixed amount ofwater followed by pH measurement [13 20ndash22] -is di-lution method has been used for analytical characterizationof smokeless tobacco [30] However e-liquids are complexformulations with variety of flavoring chemicals Consid-ering the autoprotolysis constant theory pKa value of anymolecule depends upon solventrsquos acidbase properties andpolarity [26] -e use of HendersonndashHasselbalch method bypH measurement of diluted e-liquids and calculation of freebase nicotine yield raises concerns about the effect of flavoringchemicals PG and VG on pKa of nicotine in water-erefore using a fixed dilution factor (10x) we tested theHendersonndashHasselbalch method using potentiometry

-e accuracy of the potentiometric titration was con-firmed by calculating pKa of the pyrrolidine group of nic-otine as described in (a) in Section 311 Gonzalez et al [31]had extensively studied pKa of nicotine against temperatureand reported the value of pyrrolidine pKa as 820 and 806 at15degC and 20degC respectively Based on our results the pKa ofnicotine was found to be 817plusmn 012 at 19degCplusmn 1 which isclose to the value reported by Gonzalez and Monge [31]

We carried out potentiometric study to test Hender-sonndashHasselbalch method and we used 80mgmL concen-tration of nicotine -e reason for using high concentrationof nicotine was to overcome the low sensitivity of the po-tentiometer and for ease of titration

Table 1 Determination of free base nicotine in e-liquids by potentiometric titration (HendersonndashHasselbalch method)

e-Liquid flavor(80mgmL)

e-Liquidcategory

pH of 10xdilution in

water

free base nicotine byHendersonndashHasselbalch equation

Experimental freebase nicotine

difference

Mean difference

Stddev

Unflavored PG VG1022 9912 10000 088

052 0331020 9907 9947 0411015 9897 9921 025

Maui menthol Menthol948 9531 9442 093

067 037941 9452 9475 024915 9046 9121 083

Elementtobacco Tobacco

890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019

Motley brew Fruit889 8396 8161 280

197 108887 8343 8146 236884 8239 8176 076

Milkman milkycloud Sweet

887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199

Hazelnut coffee Coffee963 9668 9538 134

090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064

difference of Nic yield by HH equation and PT

d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005

Figure 4 Percentage difference of free base nicotine (Nic) in e-liquids by HH equation vs potentiometry (ns Pgt 005 statisticalanalysis was carried out using GraphPad Prism 70)


Hexane water single extraction

nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)

Hexane water double extraction

n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)

Toluene water single extraction

n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)

Toluene water double extraction

n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)

Figure 5 Extraction of free base (Nic) and protonated nicotine (NicH+) by (a) hexane-water single extraction (b) hexane-water doubleextraction (c) toluene-water single extraction and (d) toluene-water double extraction Note Percent recovery of hexane and tolueneextractions is 9804plusmn 166 and 9710plusmn 347 respectively All results are in triplicate with RSDlt 20

Table 2 Determination of free base nicotine in standard solutions by hexane-water double extraction

pH free base nicotine byHendersonndashHasselbalch equation

experimental freebase nicotine difference Mean difference Std dev

Nicotine std (08mgmL)

501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768

Avg 2266plusmn 372Note Extraction efficiency was calculated as ldquo100 ndash mean differencerdquo (7734plusmn 372) NA not applicable since the free base nicotine values are lt1


Table 3 Determination of free base nicotine in standard solutions by toluene-water double extraction




504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139


Table 4 Determination of free base nicotine in e-liquid samples by toluene-water double extraction

e-Liquidflavor

e-Liquidcategory

pH of 10xdilution inwater (08mgmL)

free base nicotine byHendersonndashHasselbalch

equation

Experimental free basenicotine

difference Mean difference Std dev

Maui menthol Menthol 851 68686500 536

574 4586774 1376147 1050

Element tobacco Tobacco 801 41007325 7864

7690 3487346 79167089 7289

Motley brew Fruit 782 30938071 16098

16028 1048065 160788012 15908

Milkman milkycloud Sweet 757 1993

7602 2814827702 3897461 27438

7477 27518

Hazelnut coffee Coffee 882 81818879 853

876 0208904 8848910 891

difference of Nic yield by HH equation and LLE

d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast

Figure 6 Percentage difference of free base nicotine (Nic) yield of e-liquid samples by HH equation vs LLE (toluene-water extraction)(lowastPlt 005 statistical analysis was carried out using GraphPad Prism 70)


Based on results described in (b) in Section 311 themean percentage difference between potentiometric titra-tion and HendersonndashHasselbalch method results for eachsample was lt20 Similarity between these two methods isuniform across all flavors (KruskalndashWallis Test Pgt 005)(Figure 4)-us pKa of nicotine is not found to be perturbedby flavoring chemicals PG or VG after 10x dilution inwater In other words 10x dilution smooths out the effect offlavoring chemicals PG or VG on pKa of nicotine

Since pKa of a molecule is independent of its concen-tration in a particular solvent the dilution approach usingHendersonndashHasselbalch method can also be applied toe-liquids of lower nicotine concentrations for quantifyingfree base nicotine yield

412 Liquid-Liquid Extraction Method Under the dilutionapproach El-Hellani et al proposed alternative method whichwas Liquid-Liquid Extraction -e method proposed by El-Hellani et al [24] is based on their hypothesis that ldquofree basenicotine is selectively extracted in organic solvent toluenerdquo Weperformed a detailed study of this method with additional pa-rameters such as partition coefficients and extraction efficiency

It is well established fact that the protonated from ofnicotine always remains in aqueous phase [32] -is fact hasbeen reconfirmed by our study (Figure 5) where almost100 protonated nicotine in a standard solution of pH 5remained in aqueous phase during extraction phase As

nicotine starts to exist in protonated as well as free base formsat pHgt 5 the distribution of nicotine in aqueous and organicphase becomesmuchmore complex and is primarily governedby the partition theory of nicotine in aqueous and organicsolvents [32] Partition coefficient is the ratio of unionizedmolecules of a compound in two immiscible liquids [5] Basedon our extraction study (Section 312) we are reportingnicotinersquos partition coefficient in toluene (KTW) as 692 whichis higher than its partition coefficient in hexane (KHW) 125-us extraction efficiency of toluene is about 87 higher thanhexane -erefore after a second extraction toluene showsalmost 98 extraction efficiency for free base nicotine

Nicotine is a weak base with two nitrogen atoms capableof accepting protons Out of the two rings pyrrolidine ring(pKa 820) is more basic than the pyridine ring (pKa 341)[32] Based on the pKa values and HendersonndashHasselbalchtheory the pyridine ring ionizes at lower pH (lt50) ascompared to pyrrolidine ring -us the contribution ofpyridine ring towards ionization of nicotine can be neglectedat higher pH Based on the ionization theory of weak acidsand weak bases [5] nicotine in aqueous solution at differentpH (5ge pHge 13) should ionize as per Hender-sonndashHasselbalch theory Considering toluenersquos extractionefficiency the free base nicotine obtained after toluenedouble extraction should also match the theoretical valuesobtained from HendersonndashHasselbalch equation -is hy-pothesis has been confirmed from data obtained by El-Hellani et al [24] and our results are described in Figure 5

Table 5 Determination of free base nicotine in e-liquids by 1H NMR spectroscopy (e-liquids without dilution)

e-Liquid flavor (80mgmL) e-Liquid category experimental free base nicotine Mean free base nicotine Standard deviation

Maui menthol Menthol7156

6941 18768346832

Element tobacco Tobacco7483

7360 10773017295

Motley brew Fruit7587

7440 12773647370

Milkman milky cloud Sweet7579

7580 00275817579

Hazelnut coffee Coffee9176

9288 09793489340

Table 6 Comparison between free base nicotine in e-liquids by dilution approach (HH) and by without-dilution approach (1H NMR) (e-liquids without dilution)


e-Liquidcategory

Mean free base nicotine byHendersonndashHasselbalch equation (with dilution)

Mean free base nicotine by 1HNMR (without dilution)

deviation

Maui menthol Menthol 9499 6941 2693Element tobacco Tobacco 833 736 1164Motley brew Fruit 8301 744 1037Milkman milkycloud Sweet 8245 758 807



After confirming the toluenersquos extraction efficiency usingnicotine standard solutions at different pHs toluene waschosen as extracting solvent for determining free basenicotine yield of e-liquids Toluene double extraction wasperformed on e-liquids with five distinct flavors Extractionmethod involves 10x dilution of e-liquids in aqueous layer-erefore as described in Sections 242 and 312 pH ofe-liquid samples (10x diluted) was also measured and Nicwas calculated using HendersonndashHasselbalch equationAfter extraction the mean percentage difference betweenexperimental and theoretical free base nicotine yield (HH-LLE) was found to be significantly variable (574 to 27702Table 4 and Figure 6) -e possible reason for such highvariation could be that flavors in e-liquid samples interactwith toluene and affect toluenersquos extraction efficiency fornicotine Variation between Liquid-Liquid Extraction andHendersonndashHasselbalch method values can be found acrossall flavors (KruskalndashWallis Test Plt 005)

In addition e-liquids diluted in water can be a mixture ofweak acids or bases According to Le Chatelierrsquos principleafter single extraction of Nic in toluene there can be achange in equilibrium between Nic and NicH+ before thesecond extraction -us values in Liquid-Liquid Extractioncan be overestimated -erefore it can be concluded thatunder dilution approach Liquid-Liquid Extraction is not anaccurate method for determining free base nicotine yield ofe-liquids

413 Critique of Dilution Approach

(i) Using the dilution approach only Hender-sonndashHasselbalch method based on a fixed dilutionwas found to be accurate and not the Liquid-LiquidExtraction

(ii) -e dilution approach is a simple way to determinefree base nicotine yield of e-liquids

(iii) -e major critique of the dilution approach is thatdilution changes the solvent environment ofe-liquids which is mostly nonaqueous liquidsHowever it can be argued that the purpose of thedilution approach is to provide a pH relevant scalefor classifying e-liquids based on their free basenicotine yields To estimate the accuracy and rele-vance of the scale to postvaporization exposure ofnicotine it is important to establish pre- andpostvaporization correlation Such correlation canbe established if stable medium is used for pre- andpostvaporization analysis Aerosol is a highly un-stable phase to analyze as it is for free base nicotineDilution approach provides a stable ldquomediumrdquo forcollecting aerosol for analysis

(iv) -e ratio in which postvaporized aerosol interactswith lung surface fluid cannot be determined -usthe dilution factor eg 10x in current study is justan arbitrary number However such dilution factorshould be studied for effect of solvent environmenton pKa of nicotine Based on our results we haveshown that pKa of nicotine in water is not perturbed

by solvent environment of flavoring chemicals PGor VG after a 10x dilution factor -erefore wewould like to propose 10x dilution factor in fol-lowing a dilution approach

(v) -is scale would be uniform across all studies only ifthe dilution factor and water grade (for consider-ation of dissolved CO2) are fixed

(vi) As descried earlier using dilution approach it isnecessary to establish a correlation between pre-and postvaporization free base nicotine yield-erefore additional research is required

42 Without-Dilution Approach Based on the above-mentioned critique of the dilution approach Duell et alhave argued that water should not be considered whilemeasuring protonated and free base nicotine in e-liquids-e authors have stressed on measuring free base nicotinefrom e-liquids without dilution for measuring free basenicotine

421 1H NMR Spectroscopy Duell et al [23] proposedwithout-dilution approach and used 1H NMR spectroscopyto quantify free base nicotine in e-liquids -e method isbased on quantifying the difference between chemical shiftsof -CH3 group connected to protonable N atom of thepyrrolidine ring as described in Section 321 We replicatedthe procedure to compare the results of the same fivee-liquids used earlier in the extraction and potentiometrystudy Based on the results described in Table 5 the per-centage of free base nicotine in the five e-liquids givesdifferent values from the results obtained from Hender-sonndashHasselbalch method (Table 6) with a percentage dif-ference ranging from 370 to 2693

422 Critique of Without-Dilution Approach

(i) Without-dilution approach measures e-liquids intheir original nonaqueous state If the method isconsidered to be accurate the 1H NMR method canprovide free base nicotine determination of e-liq-uids in its original state (without dilution) How-ever there are some concerns for the method to beconsidered as accurate

(ii) Selectivity and resolution Since e-liquids are amixture of several chemicals in different concen-trations and 1H NMR detects all hydrogens fromthese molecules there is a possibility of ldquo-CH3 peakregionrdquo overlap of nicotine and other flavoringchemicals In some of the e-liquid samples analyzedNMR method was found to be limited by selectivityand resolution of nicotinersquos -CH3 peak For a betteridentification of the peaks relative to nicotine two-dimensional NMR (2D-NMR) such as HSQCHMQC could be applied However as the samplehas several components with many unknowns thisapproach would be complex andmay not be enough


for this identification in variety of e-liquids whenpeaks are overlappeddistorted

(iii) Limit of detection In case of 1H NMR the peaks arerecorded as relative intensities compared to thehighest intensity peak In the case of high con-centrations of flavoring chemicals and a low con-centration of nicotine in e-liquids nicotinersquos -CH3peak can merge in noisy baseline or other low in-tensity peaks ldquo-CH3rdquo peak for nicotine was notdetected for samples with nicotine concentrationlt30mgmL -is compromise with the limit ofdetection has not been addressed yet

(iv) Baseline As a mixture of several compounds andhigh ratio of PG VG to nicotine some e-liquidshave shown a drift in a baseline leading to incorrectintegration of the peaks with small intensity Suchincorrect integration can cause incorrect identifi-cation of nicotinersquos -CH3 peak

(v) Another challenge of without-dilution approach isthat it does not provide a medium to collectaerosol To estimate the accuracy and relevance ofwithout-dilution approach to postvaporizationexposure of nicotine it is necessary to establish acorrelation between pre- and postvaporization freebase nicotine yield Earlier research has analyzedpostvaporized aerosol by collecting it in NMRsample tube [23] Postvaporized aerosol is highlyunstable phase to analyze free base nicotine inaerosol form Such analysis can lead to overesti-mation or underestimation of free base nicotine-e data generated is too limited to address thisconcern [23]

5 Conclusions

Currently there are two approaches being used for deter-mination of free base nicotine in e-liquids -e first is di-lution approach and the second is without-dilutionapproach Each method has some advantages andshortcomings

Dilution approach provides a pH relevant scale forclassifying e-liquids based on their free base nicotine yieldsUsing dilution approach HendersonndashHasselbalch method isfound to bemore accurate than the Liquid-Liquid Extractionmethod For dilution approach to be uniform across allstudies dilution factor and water grade need to be fixedBased on our study we would like to propose 10x dilutionfactor

Although the dilution approach is being used by severalresearchers some authors have argued about the change insolvent system To address this issue without-dilution ap-proach has been proposed to measure free base nicotine ine-liquids without changing solvent environment-e data sofar generated using this approach is too limited to addresssome critical concerns such as overlap of nicotinersquos -CH3peak region with that of flavoring chemicals selectivityresolution limit of detection baseline drift and lack ofalternate method to overcome these concerns

Both of the approaches need to establish a strong cor-relation between free base nicotine yield of pre- and post-vaporized e-liquids -e actual delivery of free base nicotineto vapers is highly dependent on vaping profile -erefore itis equally important to consider variables such as batteryvoltage coil temperature puff duration puff frequency puffvolume and relative humidity while carrying out such astudy

Data generated from such correlation would eventuallyhelp in selecting appropriate approach for determination offree base nicotine in e-liquids and regulating the e-liquids inmarket

Abbreviations

NA Not applicableLLE Liquid-liquid extractionPT Potentiometric titrationHH HendersonndashHasselbalchNic Free base nicotineNicH+ Protonated nicotinee-cig Electronic cigarettePG Propylene glycolVG Vegetable glycerinens Not significant

Data Availability

-e data used to support the findings of this study areavailable from the corresponding author upon request

Disclosure

-e content is solely the responsibility of the authors anddoes not necessarily represent the views of the NIH or theFDA

Conflicts of Interest

-e authors declare no conflicts of interest regarding thepublication of this article

Authorsrsquo Contributions

All authors contributed significantly to the study and haveread and approved the final manuscript

Acknowledgments

-is work was supported by the National Institute on DrugAbuse of the National Institutes of Health (P30DA033934-05S1 U54DA036105) and by the Center for TobaccoProducts of the US Food and Drug Administration -eauthors would also like to acknowledge the Virginia YouthTobacco Projects Small Grants Program for researchfunding -e authors would also like to thank Center for theStudy of Tobacco Products at Virginia CommonwealthUniversity USA for providing e-liquids for analysis


References

[1] B E Garrett J E Henningfield and J F Pankow ldquoAmmoniaand other chemical base tobacco additives and cigarettenicotine delivery issues and research needsrdquo Nicotine andTobacco Research vol 6 no 2 pp 199ndash205 2004

[2] US Department of Health and Human Services How TobaccoSmoke Causes Disease e Biology and Behavioral Basis forSmoking-Attributable Disease US Department of Health andHuman Services Washington DC USA 2010

[3] J F Pankow ldquoA consideration of the role of gasparticlepartitioning in the deposition of nicotine and other tobaccosmoke compounds in the respiratory tractrdquo Chemical Re-search in Toxicology vol 14 no 11 pp 1465ndash1481 2001

[4] D E Creighton e Significance of pH in Tobacco and To-bacco Smoke BAT Warsow Poland 1987

[5] A Martin ldquoDiffusion and dissolutionrdquo in Physical Pharmacypp 324ndash355 Lea and Febiger Philadelphia PA USA 1993

[6] M Takano M Nagahiro and R Yumoto ldquoTransportmechanism of nicotine in primary cultured alveolar epithelialcellsrdquo Journal of Pharmaceutical Sciences vol 105 no 2pp 982ndash988 2016

[7] M Takano H Kamei M Nagahiro M Kawami andR Yumoto ldquoNicotine transport in lung and non-lung epi-thelial cellsrdquo Life Sciences vol 188 pp 76ndash82 2017

[8] X M Shao B Xu J Liang X Xie Y Zhu and J L FeldmanldquoNicotine delivery to rats via lung alveolar region-targetedaerosol technology produces blood pharmacokinetics re-sembling human smokingrdquo Nicotine amp Tobacco Researchvol 15 no 7 pp 1248ndash1258 2013

[9] C L Adrian H B D Olin K Dalhoff and J Jacobsen ldquoInvivo human buccal permeability of nicotinerdquo InternationalJournal of Pharmaceutics vol 311 no 1-2 pp 196ndash202 2006

[10] S G Burch L P Gann K M Olsen P J AndersonF C Hiller and M L Erbland ldquoEffect of pH on nicotineabsorption and side effects produced by aerosolized nicotinerdquoJournal of Aerosol Medicine Deposition Clearance and Effectsin the Lung vol 6 no 1 pp 45ndash52 1993

[11] E Keamy-Minor J McQuoid and P M Ling ldquoYoung adultperceptions of JUUL and other pod electronic cigarette de-vices in California a qualitative studyrdquo BMJ Open vol 9no 4 Article ID e026306 2019

[12] -e European Parliment and the Council of European UnionDirective 201440EU of the European Parliment and of theCouncil -e European Parliment and the Council of Euro-pean Union Brussels Belgium 2014

[13] -e National Academies of Sciences Engineering and Med-icine Public Health Consequences of E-Cigarettes NationalAcademies Press Washington DC USA 2018

[14] E Omaiye K J McWhirter W Luo J F Pankow andP Talbot ldquoToxicity of JUUL fluids and aerosols correlatesstrongly with nicotine and some flavor chemical concentra-tionsrdquo 2018 httpswwwbiorxivorgcontent101101490607v1full

[15] N A Goriounova and H D Mansvelder ldquoShort- and long-term consequences of nicotine exposure during adolescencefor prefrontal cortex neuronal network functionrdquo Cold SpringHarbor Perspectives in Medicine vol 2 no 12 2012

[16] S Zare M Nemati and Y Zheng ldquoA systematic review ofconsumer preference for e-cigarette attributes flavor nicotinestrength and typerdquo PLoS One vol 13 no 3 Article IDe0194145 2018

[17] H Kim J Lim S S Buehler et al ldquoRole of sweet and otherflavours in liking and disliking of electronic cigarettesrdquo To-bacco Control vol 25 no Suppl 2 pp ii55ndashii61 2016

[18] G S Helen D A Dempsey C M Havel P Jacob andN L Benowitz ldquoImpact of e-liquid flavors on nicotine intakeand pharmacology of e-cigarettesrdquo Drug and Alcohol De-pendence vol 178 pp 391ndash398 2017

[19] V V Gholap L Kosmider and M S Halquist ldquoA stan-dardized approach to quantitative analysis of nicotine ine-liquids based on peak purity criteria using high-perfor-mance liquid chromatographyrdquo Journal of Analytical Methodsin Chemistry vol 2018 Article ID 1720375 11 pages 2018

[20] J G Lisko H Tran S B Stanfill B C Blount andC H Watson ldquoChemical composition and evaluation ofnicotine tobacco alkaloids pH and selected flavors ine-cigarette cartridges and refill solutionsrdquo Nicotine amp TobaccoResearch vol 17 no 10 pp 1270ndash1278 2015

[21] I Stepanov and N Fujioka ldquoBringing attention to e-cigarettepH as an important element for research and regulationresearch letter Table 1rdquo Tobacco Control vol 24 no 4pp 413-414 2015

[22] J-F Etter and A Bugey ldquoE-cigarette liquids constancy ofcontent across batches and accuracy of labelingrdquo AddictiveBehaviors vol 73 pp 137ndash143 2017

[23] A K Duell J F Pankow and D H Peyton ldquoFree-basenicotine determination in electronic cigarette liquids by 1HNMR spectroscopyrdquo Chemical Research in Toxicology vol 31no 6 pp 431ndash434 2018

[24] A El-Hellani R El-Hage R Baalbaki et al ldquoFree-base andprotonated nicotine in electronic cigarette liquids and aero-solsrdquo Chemical Research in Toxicology vol 28 no 8pp 1532ndash1537 2015

[25] D L C Millen and W R Murphy ldquoDistribution coefficientsof nicotine between hexane and waterrdquo Industrial amp Engi-neering Chemistry Research vol 33 no 12 pp 3238ndash32401994

[26] A Kutt S Selberg I Kaljurand et al ldquopKa values in organicchemistrymdashmaking maximum use of the available datardquoTetrahedron Letters vol 59 no 42 pp 3738ndash3748 2018

[27] J W Flora C T Wilkinson K M Sink D L McKinney andJ H Miller ldquoNicotine-related impurities in e-cigarette car-tridges and refill e-liquidsrdquo Journal of Liquid Chromatographyamp Related Technologies vol 39 no 17-18 pp 821ndash829 2016

[28] X Liu P Joza and B Rickert ldquoAnalysis of nicotine andnicotine-related compounds in electronic cigarette liquids andaerosols by liquid chromatography-tandem mass spectrome-tryrdquo Beitrage zur Tabakforschung InternationalContributionsto Tobacco Research vol 27 no 7 pp 154ndash167 2017

[29] J Aszyk P Kubica A Kot-Wasik J Namiesnik andA Wasik ldquoComprehensive determination of flavouring ad-ditives and nicotine in e-cigarette refill solutions Part I liquidchromatography-tandem mass spectrometry analysisrdquo Jour-nal of Chromatography A vol 1519 pp 45ndash54 2017

[30] US Department of Health and Human Services FederalRegister Vol 64 US Department of Health and HumanServices Washington DC USA 1999

[31] E Gonzalez and CW J Monge ldquoIonization constants of 55prime-dimethyl-24-oxazolidinedione (DMO) and nicotine at tem-peratures and NaCl concentrations of biological interestrdquoActa Cientıfica Venezolana vol 31 pp 128ndash130 1980

[32] J Banyasz ldquo-e physical chemistry of nicotinerdquo in AnalyticalDetermination of Nicotine and Related Compounds and eirMetabolites pp 176-177 Elsevier Science AmsterdamNetherlands 1999


biological membrane of the respiratory tract and lead tonicotinersquos rapid absorption Many authors based on their invitro and in vivo studies have confirmed that rate of nicotineabsorption is higher for basic than acidic nicotine solutionsand aerosols [6ndash10]

In contrast PAX Labs Inc in its patent has shown that asalt form of nicotine which is mostly protonated nicotinegave higher plasma nicotine concentration (Cmax) than freebase nicotine One possible explanation for this outcomecould be that the less volatile and particulate nature ofprotonated nicotine can have deeper penetration into thelungs where absorption is more rapid than upper respiratorytract Additionally since protonated nicotine does not causeimpact in throat [11] user may perform deep inhalationwhich can result in higher amount of nicotine reaching toalveolar region

Experimentally it is still not clear how nicotine crossesbiological membrane With the two conflicting theories andexperimental data about the absorption of free base vsprotonated nicotine independent and unbiased scientificresearch is still not available to determine the ldquorate of nicotineuptake in smokersvapers as a function of its free base doserdquoIn either case it is important to classify e-liquids based ontheir free base or protonated nicotine yield for two reasons

(a) Frequent high amounts of nicotine absorption canlead to its addiction Unlike European regulations[12] FDA does not specify a cap on maximum nic-otine content in e-liquids Nicotine content ine-liquids ranges from 0 to gt50mgmL [13] JUULproducts which have gained high popularity in recentyears have about 50mgmL content of nicotine [14]which is believed to yield mainly the protonated form(NicH+) Additionally the current 3rd and 4th gen-eration boxmod e-cigarette devices allow control overpower settings such that high power gives high nic-otine delivery-e question that arises here is whethersuch a high amount of nicotine in e-liquids is justi-fiable if protonated nicotine is suspected of higherabsorption Increased exposure of nicotine in youthcan affect the prefrontal cortex development leadingto a deficiency in attention and lasting effects oncognitive functions [15] -erefore to control theyouth addiction of nicotine it is important to addressthe free base or protonated nicotine content ine-liquids for their adequate regulation

(b) Unlike free base nicotine salt based nicotine causesless harshness in throat hit and thus improves pal-atability and smoothness of e-cig aerosol [11] Basedon popularity surveys and flavor studies of severalother e-liquids different flavors of e-liquids of equalstrength are perceived to give different nicotineimpacts [16ndash18] -us classifying e-liquids based ontheir free base nicotine delivery would help in un-derstanding the acceptance criteria of e-liquids byvapers

We have recently proposed a potential standard methodfor analyzing total nicotine content in e-liquids using peak

purity criteria by HPLC [19] However there is no stan-dardized method yet to quantify free base nicotine yieldfrom e-liquids

Figure 1 explains the two major approaches currentlybeing used for the determination of free base nicotine ine-liquids -e first is a dilution approach and the second iswithout-dilution approach A dilution approach has beenutilized in two methods HendersonndashHasselbalch methodand Liquid-Liquid Extraction method Without-dilutionapproach has been followed by 1H NMR method alone

A dilution approach followed by HendersonndashHasselbalchmethod has been used by many authors [13 20ndash22] Howeverdue to various concerns about the effect of flavors on theaccuracy of results obtained by HendersonndashHasselbalchequation a Liquid-Liquid Extraction method was proposed byEl-Hellani et al

It can be argued that the dilution approach causeschange in e-liquidsrsquo original solvent system which is majorlynonaqueous Duell et al argued that this change in solventsystem may not give a true picture of free base nicotine ine-liquids Additionally different dilution factors might in-fluence the results and make comparisons challenging Toaddress this issue Duell at el proposed a without-dilutionapproach by which free base nicotine is determined using 1HNMR

To compare and present critique of the two approacheswe carried out controlled experiments using five e-liquids ofdifferent flavors We present results using various analyticaltechniques that will facilitate the establishment of a stan-dardized method by regulatory bodies for classifyinge-liquids based on their free base nicotine yield

2 Materials and Methods

21 Instrumentation Analysis of nicotine by potentiometrictitration was carried out using a TruLab pH 1310P (YSIIncorporated Xylem Inc USA) potentiometric pH meterwith TruLine 15 glass electrode selective to H+ ions andcontaining silver chloride reference electrode Liquid-LiquidExtraction was carried out using a Waters Alliance 2695quaternary pump HPLC equipped with Waters 996 PDADetector Hypersil Gold Phenyl column (150mmtimes 46mm3 μm -ermo Scientifictrade USA) and a Security GuardCartridge Phenyl (4mmtimes 20mm Phenomenex USA)Waters Empower 2 software was used for processing dataNMR analysis was carried out using Bruker NanoBayAVANCE III 400MHz NMR spectrometer (Bruker Cor-poration USA)

22 Chemicals and Reagents (minus )-Nicotine liquid standard(purityge 99) was purchased from Sigma Aldrich USAHPLC grade acetonitrile methanol and water were pur-chased from BDH Chemicals VWR USA Ortho phos-phoric acid (85) was purchased fromMerck USA Triethylamine tert-butyl amine and hydrochloric acid (37) werepurchased from Sigma Aldrich USA Glacial acetic acid waspurchased from Macron Fine Chemicals USA Sodiumhydroxide (10N) was purchased from BDH Chemicals










(1)






1113858 1113859 (3)

Nic [Nic]



Free

bas

e nic

otin

e in

e-liq

uids

Dilution
















Measure pH


Measure pH and mV


Equivalence point



01N NaOH



3 Results















WT

T

W

W

WT

T

T WH+

WH+

TT

Discard


WH

H

W

W

WH

H




diluent

HPLC





4 Discussion









e-Liquidcategory


water



difference

Mean difference

Stddev


052 0331020 9907 9947 0411015 9897 9921 025


067 037941 9452 9475 024915 9046 9121 083


890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019


197 108887 8343 8146 236884 8239 8176 076


887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199


090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064


d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005




nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References










































(1)






1113858 1113859 (3)

Nic [Nic]



Free

bas

e nic

otin

e in

e-liq

uids

Dilution
















Measure pH


Measure pH and mV


Equivalence point



01N NaOH



3 Results















WT

T

W

W

WT

T

T WH+

WH+

TT

Discard


WH

H

W

W

WH

H




diluent

HPLC





4 Discussion









e-Liquidcategory


water



difference

Mean difference

Stddev


052 0331020 9907 9947 0411015 9897 9921 025


067 037941 9452 9475 024915 9046 9121 083


890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019


197 108887 8343 8146 236884 8239 8176 076


887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199


090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064


d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005




nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References












































Measure pH


Measure pH and mV


Equivalence point



01N NaOH



3 Results















WT

T

W

W

WT

T

T WH+

WH+

TT

Discard


WH

H

W

W

WH

H




diluent

HPLC





4 Discussion









e-Liquidcategory


water



difference

Mean difference

Stddev


052 0331020 9907 9947 0411015 9897 9921 025


067 037941 9452 9475 024915 9046 9121 083


890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019


197 108887 8343 8146 236884 8239 8176 076


887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199


090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064


d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005




nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References


































3 Results















WT

T

W

W

WT

T

T WH+

WH+

TT

Discard


WH

H

W

W

WH

H




diluent

HPLC





4 Discussion









e-Liquidcategory


water



difference

Mean difference

Stddev


052 0331020 9907 9947 0411015 9897 9921 025


067 037941 9452 9475 024915 9046 9121 083


890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019


197 108887 8343 8146 236884 8239 8176 076


887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199


090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064


d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005




nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References



































4 Discussion









e-Liquidcategory


water



difference

Mean difference

Stddev


052 0331020 9907 9947 0411015 9897 9921 025


067 037941 9452 9475 024915 9046 9121 083


890 8424 8253 203077 109884 8239 8231 009

882 8185 8169 019


197 108887 8343 8146 236884 8239 8176 076


887 8343 8058 342187 161872 7801 7785 021

872 7805 7650 199


090 038955 9601 9528 076954 9589 9530 061

Avg 112plusmn 064


d

iffer

ence

of N

ic y

ield

e-Liquids

Unfl

avor

ed

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

1

2

3

4

5 ns

ns P gt 005




nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References



































nic

otin

e

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(a)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(b)


n

icot

ine

6 8 10 12 144pH

0

20

40

60

80

100

NicNicH+

(c)


n

icot

ine

0

20

40

60

80

100

6 8 10 12 144pH

NicNicH+

(d)






501 007033 NA

NA NA026 NA025 NA

706 722881 2201

2137 071877 2150871 2059

900 87146333 2732

2692 0356388 26706383 2675

1100 99857520 2469

2412 0597637 23527574 2415

1300 100008108 1892

1823 0638190 18108232 1768







504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References






































504 007127 NA

NA NA097 NA112 NA

703 676643 483

885 442767 1358730 813

906 88568616 271

290 0198601 2888582 310

1102 99869830 156

161 0079818 1689828 158

1300 100009864 136

139 0029859 1419861 139



e-Liquidflavor

e-Liquidcategory



equation




574 4586774 1376147 1050


7690 3487346 79167089 7289


16028 1048065 160788012 15908


7602 2814827702 3897461 27438

7477 27518


876 0208904 8848910 891


d

iffer

ence

of N

ic y

ield

e-Liquids

Men

thol

Toba

cco

Frui

t

Swee

t

Coffe

e

0

100

200

300

lowast P lt 005

lowast












6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References











































6941 18768346832


7360 10773017295


7440 12773647370


7580 00275817579


9288 09793489340



e-Liquidcategory



deviation
























5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References






















































5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References






































5 Conclusions






Abbreviations


Data Availability


Disclosure






Acknowledgments



References


































References


































Documents

An Analytical Perspective on Determination of Free Base