B American Society for Mass Spectrometry, 2014 Understanding Gas Phase Modifier Interactions in Rapid Analysis by Differential Mobility-Tandem Mass Spectrometry Introduction

B American Society for Mass Spectrometry, 2014DOI: 10.1007/s13361-013-0808-5

J. Am. Soc. Mass Spectrom. (2014)

FOCUS: MASS SPECTROMETRY AND DNA DAMAGE: RESEARCH ARTICLE

Understanding Gas Phase Modifier Interactions in RapidAnalysis by Differential Mobility-Tandem MassSpectrometry

Amol Kafle ,4 Stephen L. Coy ,4 Bryan M.Wong ,1 Albert J. Fornace Jr.,2,3 James J. Glick ,4

Paul Vouros 4

1Department of Chemistry and Department of Materials Science and Engineering, Drexel University, Philadelphia, PA 19104, USA2Department of Biochemistry and Molecular and Cell Biology, Georgetown University, Washington, DC 20057, USA3King Abdulaziz University, Center of Excellence in Genomic Medical Research, Jeddah 21413, Saudi Arabia4Department of Chemistry and Chemical Biology and Barnett Institute, Northeastern University, Boston, MA 02115, USA

Abstract. A systematic study involving the use and optimization of gas-phasemodifiers in quantitative differential mobility-mass spectrometry (DMS-MS)analysis is presented using nucleoside-adduct biomarkers of DNA damage asan important reference point for analysis in complex matrices. Commonly usedpolar protic and polar aprotic modifiers have been screened for use against twodeoxyguanosine adducts of DNA: N-(deoxyguanosin-8-yl)-4-aminobiphenyl (dG-C8-4-ABP) and N-(deoxyguanosin-8-y1)-2-amino-l-methyl-6-phenylimidazo[4,5-b]pyridine (dG-C8-PhIP). Particular attention was paid to compensation voltage(CoV) shifts, peak shapes, and product ion signal intensities while optimizing theDMS-MS conditions. The optimized parameters were then applied to rapid

quantitation of the DNA adducts in calf thymus DNA. After a protein precipitation step, adduct levelscorresponding to less than one modification in 106 normal DNA bases were detected using the DMS-MSplatform. Based on DMS fundamentals and ab initio thermochemical results, we interpret the complexity ofDMS modifier responses in terms of thermal activation and the development of solvent shells. At very highbulk gas temperature, modifier dipole moment may be the most important factor in cluster formation andcluster geometry, but at lower temperatures, multi-neutral clusters are important and less predictable. Thiswork provides a useful protocol for targeted DNA adduct quantitation and a basis for future work on DMSmodifier effects.Key words: Differential ion mobility, DMS, FAIMS, DNA adducts, Gas phase interactions, Modifiers,Quantitation, Kinetics, Ion-polar molecule clustering

Received: 19 September 2013/Revised: /Accepted: 9 December 2013

Introduction

Differential mobility is a rapidly evolving new analyticaltechnology that offers rapid gas-phase ion separation/

filtration prior to mass analysis, and the advantages of thisplatform, such as improved signal to noise ratio, separationof closely related compounds, and removal of interferenceshave been well demonstrated [1–6].

The mobility of charged ion species in an applied electricfield forms the basis for ion separation in a DMS cell.Krylov and Nazarov [7] have studied three different modelsof ion-neutral interactions in the applied electric field: (1)rigid sphere scattering, (2) long-range ion–dipole attraction,and (3) clustering. They concluded that ion-neutral cluster-ing is the most relevant phenomenon to explain thedependence of ion mobility on field strength that is thesource of DMS selectivity. The mobility of ions between theelectrodes in a differential mobility cell has been welldescribed in a paper by Schneider et al. [4] and moreextensively in monographs by Shvartsburg [8], Eiceman and

Correspondence to: Stephen L. Coy; e-mail: [email protected],Paul Vouros; e-mail: [email protected]

Karpas [9], and in a discussion on the fundamentals of ionmobility by Mason and McDaniel [10]. We present here abrief discussion of these principles, which provides aframework for understanding the role of modifiers in DMSseparations as they particularly pertain to the analysis ofselected model DNA adducts, biomarkers indicative of DNAdamage from carcinogens.

Ion mobility as used in DMS is a high pressurephenomenon in which ions in an electric field quickly reacha limiting speed, described by a field-dependent ion mobilitycoefficient, K(E),

v!d Eð Þ ¼ K Eð Þ!E;K Eð Þ ¼ K 0ð Þ 1þ α Eð Þð Þ;

ð1Þ

where v!d Eð Þ is the drift velocity of the ion,E the electric fieldstrength, and α(E) contains the deviation from low-fieldbehavior, and is known as the alpha parameter, or thedifferential mobility. K(0) is the low-field mobility coefficient,with a dependence on the ion’s chemical identity, but alsodepending on pressure, temperature, and chemical environ-ment. Low field mobility, K(0), is the property measured in ionmobility spectrometry (IMS) under the same conditions.However, the ion separations used in differential mobilityspectrometry (DMS) are determined solely by the differentialmobility parameter, α(E), [11, 12] because of the DMS filtercondition that determines the relationship between the DMSseparation peak voltage, SV, and the DMS compensationvoltage, CoV, for a particular DMS waveform shape, f(t) [13].Ion transmission in a planar DMS geometry requires that theion remain along the axis of the DMS analytical region aftereach f(t) waveform cycle of period Tf,Z T f

0dt f tð ÞE tð Þ 1þ α E tð Þð Þð Þ ¼ 0;

where E tð Þ ¼ SV ˙ f tð Þ þ CoV :ð2Þ

As a result of this filter condition, there is completeequivalence between the observed DMS compensation

voltages (CoV) for a range of field amplitudes (SV) and thedifferential mobility, α(E) [4].

The mobility coefficient, K(E), acquires a dependence on

field strength, E ¼ E!��� ��� , as a result of interactions of the ion

with its chemical environment [4, 7, 14], and also varieswith pressure and temperature in the DMS analytical region[15, 16]. Ion mobility loses its dependence on pressure if thefield used in ion mobility is expressed in density-normalizedTownsend units for the electric field [10]:

ETd ¼ E

N;

KTd ETdð Þ ¼ NK Eð Þ ¼ vd ETdð ÞETd

;ð3Þ

where ETd=E/N is the ion mobility field in Townsend units[15] (1 Td=10–17 volt/cm2 ). Density scaling providespressure correction for analytical instruments such as theAB SCIEX SelexION system [17] , but is not useful fortemperature variation. The ion mobility coefficient, evenunder cluster-free conditions, still retains a dependence ontemperature that can be minimized but not eliminated bytemperature scaling. Temperature scaling leads to a reducedmobility, K0 (corrected to 0 °C, 1 atm) that still dependsweakly on temperature, but that variation is dwarfed in DMSby dynamic cluster size variation. Although it is notimportant in low pressure IMS, or in traveling wave IMS[18, 19], the importance of clustering is well known inatmospheric pressure IMS, where a cluster-free mass-mobility correlation can only be obtained by the use of highdrift tube temperatures, typically 200 °C or higher [9, 20–26]. Except in special cases, low pressure IMS and travelingwave IMS operate with pure, non-polar drift gases anddetermine accurate cross-sections for a wide range ofmolecules, whereas DMS modifier effects depend on polarmolecules with a long-range attractive potential [27].

The dependence on chemical composition of the transportgas has been found to be the largest contributor todifferential mobility. This is both a hindrance and an

Scheme 1. Structure if DNA adducts in this study

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

advantage. Ions of similar molecular weight and relatedstructure can usually be separated with the assistance of amodifier, and peak capacity is greatly increased, but theunderlying cause is still obscure. While useful for improvedresolution and reduced chemical noise, the modifier effect isthe most difficult to predict; however, it is important torecognize that the mobility coefficient is directly related tothe effective size of the ion, whether it is a bare ion or a ion-neutral cluster (effective size is referred to as the cross-section for collisions between the ionic species andmolecules in the neutral transport gas). The simplestexpression for the mobility coefficient which presents thatconnection in a way that is valid even at the very high DMSfield strengths is from the two-temperature theory inmomentum-transfer form (section 6-2.C (eq. 6-2-25), [10])

NK ¼ 3q

4

1þ αK

μvrel Teff

� �˙ Ω

�Teff

� �;

where vrel Teff

� �≡

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi8kBTeff

πμ

s;

and Teff ¼ T þ ςM

3kBNKð Þ2 E

N

� �2

ð4Þ

In this expression, μ is the ion-neutral reduced mass (μ=mM/(m+M), m the ion mass, andM the neutral mass), vrel is the well-known expression for the relative speed of particles in aMaxwell-Boltzmann distribution, whereas Ω is an average cross-section.Other symbols in Equation 4 are N (gas density), q (ion chargemagnitude), kB (Boltzmann constant), and αK (a small correctionG0.02). From this, it is clear that the ion mobility is inverselyrelated to the loss of momentum by collisions, since themomentum, μvrel(Teff), multiplies a cross-sectionΩ Teff

� �giving

a mass-scaled rate (see Equation 6). Both relative speed andcross-section are evaluated at a special temperature, Teff, thatvaries dynamicallywith theDMS field and depends jointly on thebulk temperature, pressure, and electric field.

Early on in the application of DMS technology, it was realizedthat the use ofmodifiers in the transport gas is an important step foranalytical applications [14, 28]. Prior works from our laboratoryand several others have demonstrated the advantages of introduc-ing drift gas modifiers to aid in DMS separations [29, 30]. In arecent publication by Schneider et al., the effects of transport gasmodifiers have been extensively documented with particularattention to resulting peak capacity [31], and other groups areactive as well in areas such as Green Chemistry [30, 32]. Theseresults are part of a growing interest in the use of modifiers withthe DMS-MS platform as it offers a promising alternative for rapidanalytical applications aswell as possible field applications such asthose proposed for NASA applications [33, 34].

Our group has long been involved in the development ofDMS technology for analytical applications [14, 29, 35] andin recent years, we have been particularly interested in theuse of DMS-MS for quantitative analysis with reduced

sample cleanup and minimal or zero LC time [36]. There area number of advantages to the use of DMS in rapidquantitative biomarker analysis, many previously discussed,as in Coy et al. [2], but one point is of special interest whenDMS-MS quantitative analysis is compared with LC-MS:LC-MS as well as drift-time or traveling-wave IMS requireintegration of an ion current over an elution or drift timeprofile, but DMS-MS intensities for quantitation by infusionare obtained continuously by monitoring the CoV for peakion transmission. Under conditions of flow injection or LC-DMS-MS, a time profile is integrated, but without scanningCoV [37, 38]. The DMS analytical region is a continuousion filter that transmits with minimum diffusion losses atcharacteristic (SV, CoV) values. Because the width of thepeak in CoV units is inversely related to the low field ionmobility of the ion, integrating over the DMS peak widthwould contaminate the observed data with an unknown scalefactor, one which has a variable environmental dependence.Thus, DMS at the peak CoV value provides a continuousmeasurement of ion concentration, limited in response timeonly by the transit time of the ion through the DMSanalytical region and the agility of the electronics of themass spectrometer.

Prior works from our laboratory have demonstrated thatthe judicious use of modifiers provided conditions for rapidquantitation of selected analytes in matrices of varyingcomplexities [35, 36]. We have shown that in addition to therapid analysis time afforded by use of the DMS, samplepreparation steps can be minimized or even eliminated bytaking advantage of the unique post-electrospray selectiveion filtration capabilities. As with any other analyticalplatform, proper method development is necessary toachieve optimal sensitivity and efficacy. In the workpresented here, we demonstrate the process of stepwiseevaluation of modifier effects from a method developmentperspective in order to establish the optimal conditions forquantitation of selected DNA damage biomarker analytesScheme 1 by differential mobility-mass spectrometry.

DNA adducts provide direct evidence of genetic exposureand damage in cells, and monitoring their levels from ahealth perspective is important. So-called bulky DNAadducts share common structural features in that, other thanthe carcinogen, they all contain a deoxyribose moiety alongwith the nucleobase, typically a guanine. In this regard, thecompounds selected for this study provide excellent modelsfor the purpose of evaluating the DMS conditions that mightbe appropriate for the DMS-MS analysis of this class ofanalytes. N-(2-deoxyguanosine-8-yl)-4-ABP (dG-C8-4-ABP), the deoxyguanosine adduct of the bladder carcinogen4-aminobiphenyl (4-ABP), is a known carcinogen found incigarette smoke, paints, food colors, hair dyes, and fumesfrom heated oils and fuels [39–42]. The heterocyclicaromatic amine 2-amino-1-methyl-6-phenylimidazo[4,5]pyridine (PhIP) is found in grilled meats [43]. PhiP is aknown foodborne carcinogen that has been implicated inmammary gland tumors in rodents [44]. These two adducts

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

have been related to lifestyles of people and we have usedthem as model analytes to explore and optimize the gasphase interactions in DMS and develop differential mobility-mass spectrometry as a rapid quantitative platform.

ExperimentalChemicals and Standards

Caution: 2-Hydroxyamino-1-methyl-6-phenylimidazo[4,5]pyridine and 4-aminobiphenyl and its derivatives arecarcinogenic and should be handled carefully.

Calf-thymus DNA, nuclease p1 from Penicilliumcitrinium, deoxyribonuclease 1 (DNase I) type 2 frombovine pancreas, alkaline phosphatase from Escherichia coli(type IIIs), ethanol, dimethyl sulfoxide (DMSO) werepurchased from Sigma-Aldrich Chemical Co. (St. Louis,MO, USA). Snake venom phosphodiesterase was purchasedfrom USB Corporation (Cleveland, OH, USA). N-(2-deoxyguanosine-8-yl)-4-ABP (dG-C8-4-ABP) was pur-chased from Toronto Research Chemicals (Toronto, ON,Canada, D239600, CAS 84283-08-9, C22H22N6O4). N-( d e o x y g u a n o s i n e - 8 - y l ) - 2 - am i n o - 1 -m e t h y l - 6 -phenylimidazo[4,5-β]pyridine (dG-C8-PhIP) were purchasedfrom Toronto Research Chemicals (Toronto, ON, Canada,D239630, CAS 142784-25-6, C23H23N9O4). Formic acidsolution was purchased from Sigma-Aldrich Chemical Co.(St. Louis, MO, USA). HPLC grade water and methanolwere purchased from Fisher Scientific (Fair Lawn, NJ,USA).

DNA Quantification, Enzymatic Digestion,and Protein Precipitation

DNA quantification was done using an Invitrogen Corpora-tion (Carlsbad, CA) Quant-IT double strand (ds) DNA BRAssay kit with a Qubit fluorometer. Aliquots containing 2 ugDNA (dissolved in 5 mM Tris-Cl/ 10 mM ZnCl2) wereremoved for digestion and analysis for each sample point.

Calf thymus DNA was hydrolyzed similarly to a methodpreviously described [45]. Samples were incubated at 98 °Cfor 3–5 min and chilled in the freezer down to roomtemperature; 0.3 units of nuclease P1 (0.3 units μL−1

solution of 5 mM Tris–Cl, pH 7.4) and 3.1 Kunits of DNaseI (1 μg μL−1 solution in 5 mM TRIS/10 mM MgCl2, pH 7.4)were then added per μg of DNA and incubated in a waterbath maintained at 37 °C. After 5 hours, 0.003 units ofphosphodiesterase (100 ng μL−1 in 5 mM TRIS/10 mMMgCl2, pH 7.4), and 0.002 units of alkaline phosphatase perμg of DNA were added and the mixture was furtherincubated at 37 °C for 18 h.

Protein precipitation was done by adding five volumesof ice cold ethanol and centrifuging at 10,000 rpm for 15min. The samples were dried down and stored at –80 °Cuntil analysis.

Instrumentation

The fundamental goals of the proposed approach to theoptimization of modifiers in quantitative analysis by DMS-MS was tested using two different planar DMS systems, oneattached to a 3-D ion trap (Thermo Finnigan LCQ Classic)(San Jose, CA, USA) and the second interfaced to an ABSCIEX API 3000 triple quadrupole MS [AB SCIEX,Framingham, MA, USA]. The DMS filters varied only interms of their dimensions, and this comparison provided anassessment of the general effect and applicability of theDMS configuration to such analyses.

DMS-Ion Trap Mass Spectrometer

A planar DMS developed by Sionex Corporation (Bedford,MA, USA) with a filter gap 0.5 mm high×3.0 mm wide×10.0 mm long, which was positioned at the entrance of theheated capillary of a Thermo-Finnigan, LCQ Classic massspectrometer was used for the work. Sionex Expert softwarewas used to set the DMS parameters. The SV could be set atzero or scanned in the range from 500 to 1500 V, and theCoV could be set or scanned from –43 to +15 V. Althoughthe system in use here is no longer commercially available,the underlying electronics technology has been described,and successor commercial instrumentation (SelexION, ABSCIEX) has become commercially available.

Electrospray was performed using coated 10 μmPicoTip emitters from New Objective (Woburn, MA,USA). The syringe was connected to the emitter tipusing a 150 um (i.d.) capillary tubing and the ESIvoltage was applied to the union at the liquid–liquidjunction between the capillary and the emitter tip. Sampleswere introduced at the rate of 300 nL/min using a HarvardApparatus syringe pump (Holliston, MA, USA). Thedesolvation gas (ultra high purity nitrogen) was intro-duced at a flow rate of approximately 100 cc/min intothe desolvation region at a temperature of 100 °C. Thevacuum drag of the mass spectrometer was measured tobe 500 cc/min. External air flow of approximately 400cc/min also merged in with the desolvation gas into theDMS. The bulk gas temperature was estimated to be 45ºC. Modifiers were introduced into the desolvation regionalong with the nitrogen gas. The electrospray emittervoltage was held at 2 kV throughout the analysis.

DMS-Triple Quadrupole Mass Spectrometer

A prototype DMS-API 3000 Triple Quadrupole massspectrometer (AB SCIEX, Concord, ON, Canada), whichhas an integrated DMS filter online in front of the vacuumorifice, was used for this study. The dimensions of the DMSanalytical region were 1 mm×10 mm×15 mm. A modifiedversion of the software Analyst ver. 1.5, which includedparameters for SV and CoV, was used. The SV could bevaried from 0 to 5000 V and the CoV could be varied from –

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

100 to +100 V. Electrospray was performed using stainlesssteel 30 μm i.d. emitter from Proxeon (Thermo Fisher, estPalm Beach, FL, USA). Samples were introduced at the rateof 400 nL/min using a Harvard Apparatus syringe pump.Electrospray was held constant at 3500 V throughout theanalysis. Modifiers were introduced into the curtain gas(nitrogen, 1.1 L/min, 600 cc/min vacuum drag 500 cc/mincurtain gas outflow at 40 ºC) using a second HarvardApparatus syringe pump.

DMS Electric Fields

The principal types of waveforms in use for DMS aredescribed in a technical instrumentation paper by Krylovet al. [13], and include a flyback waveform consisting ofa truncated sinusoidal pulse connected to a flat baseline,and a two-harmonic design in which phased first andsecond harmonic voltages are applied to opposingelectrodes. The waveform applied on the ion trap MSwas of the flyback type at 1.25 MHz where the voltagesare applied across a 0.5 mm gap, and are measured interms of zero to peak separation voltages with CoV=0.The waveform applied in the DMS-API-3000 and on thecommercial AB SCIEX SelexION systems is a two-harmonic design applied across a 1.0 mm gap, withfrequencies 3 and 6 MHz in 2:1 voltage ratio and theseparation voltage measured as peak to peak. Because ofthe differing gap dimension, and the reporting of flybackSV as zero-to-peak and two-harmonic SV as peak-to-peak, there is a factor of three between the same-fieldflyback and two-harmonic SV voltage values. Forinstance, the ion trap system can apply a peak voltageof up to 1500 V (zero-to-peak) over the 0.5 mm ga. TheDMS-API 3000 at a setting of 4500 V SV (peak-to-peak)applies the same peak field to the 1.0 mm DMS region.Flyback and two-harmonic shapes are the most feasibleelectronics designs for DMS and have similar DMSefficiencies, with the flyback shape slightly superior,depending on α(E). Flyback and two-harmonic fields atidentical peak voltage values are illustrated in Figure 6ain the Discussion section, where clustering mechanismsare considered in more detail.

Results and DiscussionEffect of Modifiers on CoV Shifts

In an effort to understand and characterize the gas phasemolecular interactions, varying percentages of modifierswere introduced into the transport gas and the CoV shiftswere recorded with increasing separation voltage. Voltagescorresponding to the apex of extracted ion chromatogramsare reported here as CoV values. The goal of using themodifiers in this work is to shift the analyte of interestselectively to a CoV value free of interferences where DMS

can be exploited to perform rapid quantitation in thepresence of a complex matrix.

Effect of Modifier on CoV of dG-C8-4-ABP

The clustering effects of two gaseous modifiers of differingsize and polarity, ethyl acetate (C4H8O2, mw 88.11, μ=4.325 D, Q=–6.619 D·Ǻ) and isopropanol (C3H8O, mw60.10, μ=1.560 D, Q=3.242 D·Ǻ) [46], on dG-C8-4-ABPat varying concentrations were investigated by monitoringthe CoV shift as a function of separation voltage (SV). Asshown in Figure 1a, in the absence of a modifier, the analyteion does not exhibit any shifts in CoV at SV values up to2000 V. Even when the SV is increased beyond 2000 V, theCoV starts shifting only slightly towards the negativevoltage reaching a maximum of –2 V at SV=4500 V.

With ethyl acetate as a modifier (Figure 1a(ii)), dG-C8-4-ABP shifts rapidly to negative CoV values, as expected forcharge–dipole interactions [7], reaching –5 V at only 0.30%ethyl acetate. Interestingly, at higher ethyl acetate concen-trations of 0.6%, 1.20%, and 2.50%, identical CoVresponses are observed, as indicated by the overlappingCoV curves. This strongly suggests that steric and thermo-dynamic effects limit the maximum number of coordinatedethyl acetate molecules to a value which is reached between0.3% and 0.6% by volume, under the AB-SCIEX 3000triple-quad conditions. It can also be noted that the CoVshifts for all SV values were intermediate between those ofno modifier and higher modifier concentrations.

With isopropanol as a modifier, in comparison to ethylacetate, the CoV value of the adduct shifted to even greaternegative values with isopropanol than for ethyl acetate, andthe effect was even more substantial with increasingconcentrations [Figure 1a(i)]. For example, going from nomodifier to progressively higher modifier concentrations of0.60%, 1.10%, and 2.20%, the respective CoV values atSV=4500 V increased to –4.6, –10.4, and –14.2 V. Forisopropanol, the greatest CoV shift observed at 2.20% was –14.2 V compared with ethyl acetate’s maximum shift of –9V. The observation of a greater shift toward negative CoVvalues with isopropanol than with ethyl acetate, even thoughisopropanol is of lower molecular weight and of smallergeometric cross-section, can be interpreted in different ways.On one hand, it might indicate that the limiting coordinationnumber with isopropanol is greater than with ethyl acetateor, alternatively, thermochemical effects related to the freeenergy changes on cluster formation, and the related role ofhydrogen bonding in cluster formation in the two systemsmay be important. This is discussed further below.

It has been reported that the modifier effect is sensitive tosmall variations in ion structure, so that even closely relatedstructures may be separated (citrate/isocitrate, [2]; ephedrine/pseudoephedrine [3], and others [47]). Prior works in thefield of DMS have established the importance of modifiersto not only enhance CoV shifts (i.e., selectivity) but, usingthe analogy to chromatography, to improve peak capacity in

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

the DMS analysis [4] as well as to suppress chemical noise.We, thus, next examined the peak shapes of the analyte ofinterest dG-C8-4-ABP at different DMS conditions. Selectedextracted ion chromatograms from the CoV scan are shownin Figure 2. For dG-C8-4-ABP, the improvements in peakshape by the introduction of modifiers can be understood bycomparing the FWHM (full width at half maximum) at aconstant SV of 3500 V in Figure 2 (no modifier: Figure 2b,ethyl acetate modifier at 0.6%: Figure 2c, and isopropanolmodifier at 0.6%: Figure 2e). The FWHM without modifieris ~4 V compared with ~3 V in presence of ethyl acetate and~3 V in presence of isopropanol, which highlights the role ofDMS and gas phase clustering in the presence of modifierson peak shape improvements during DMS analysis. It shouldalso be pointed out that that the FWHM increased initiallyby 1.5 V when the SV was set to 3500 V (Figure 2b)compared with zero SV (Figure 2a), which was subsequentlyimproved by introducing modifiers. Much of this effect maybe due to DMS suppression of chemical noise due to ions ofhigher molecular weight, which are detected with a widerpeak width due to lower mobility.

It is interesting to point out the effect of increasingmodifier percentages on peak shape. Specifically, whenethyl acetate modifier is introduced at 2.5% (Figure 2d),

which is roughly 4× the amount of the same modifierintroduced at 0.6% (Figure 2c), not only do the CoV shiftsoverlap, but the peak shapes look very similar, whichsupports our notion of “saturation effect” as discussedabove. However, a 4-fold increase in the isopropanolconcentration (Figure 2e and f) also show that the peakshape can begin to suffer with larger CoV shifts. It isreasonable to surmise that formation of larger clusters withincreasing isopropanol concentration may have caused thetransmission of analyte ion dG-C8-4-ABP over a slightlylarger CoV range as expected from the higher clustermolecular weight (DMS peak width~[1/K(E)]. The broad-ening effect is seen to be greater near the baseline than athalf-height, as would be expected for some fraction of largerclusters with broader peaks, which are in rapid equilibriumwith single adducts. It is, therefore, deemed necessary tooptimize the modifier percentage being introduced to avoidexcessive peak broadening during DMS-MS analysis.

Effect of Modifier on CoV of dG-C8-PhIP

The principal DNA adduct of PhIP, a diet-related carcino-gen, is also formed by covalent bonding at the C-8 positionof guanine. In that sense, this adduct provides an interesting

Figure 1. (a) Effect on compensation voltage of varying concentrations of modifiers (i) isopropanol and (ii) ethyl acetate on dG-C8-4-ABP CoV shifts. (b) Effect of varying concentrations of modifiers (i) isopropanol, (ii) ethyl acetate, (iii) 1-butanol on dG-C8-PhIP CoV shifts

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

example for comparison with dG-C8-4-ABP since the twocompounds share the 2′-deoxyguanosine structure(C10H13N5O4, mw=267.2413) and differ since the PhIPmoiety is larger, richer in heteroatoms, and more polar thanABP. In evaluating the modifier effects on CoV, ethylacetate and isopropanol were tested as before, and 1-butanol(C4H10O, mw=74.1216, μ=1.584 D (1.660 D expt.) [46] )was also screened for use in DMS-MS analysis of the adductand the results are presented in Figure 1b.

In the absence of a modifier, the dG-C8-PhIP CoV valuesstarted shifting in the positive direction reaching +3 V at SVvalue of 4500 V (Figure 1b). As discussed by Krylov, apositive CoV shift is associated with short-range collisions,whereas long-range, charge–dipole interactions result innegative CoV shifts. Upon the introduction of modifiers,the CoV values reversed direction to the negative side andthis trend increased significantly as a function of modifierconcentrations and SV. Ethyl acetate was the least effectivein shifting the CoV to a negative direction, whereas 1-butanol exhibited larger shifts than isopropanol, reaching amaximum peak value of –10 V at 2.20%. It is conceivablethat the stronger action of 1-butanol compared withisopropanol might be attributable to the larger geometric

size of 1-butanol, to the end position of the –OH, whichincreases the cluster profile, and to lower steric inhibition forthe end-attached –OH by the alpha methyl group, especiallysince the dipole moments of the two alcohols are nearlyidentical and, thus, have the same long-range attractiveinteraction with an ion.

Also especially interesting was the limited effect exhib-ited by ethyl acetate with a maximum peak value of –2.2 Vat 2.50%. This may be interpreted in two radically differentways: (1) difference in hydrogen-bonding donor–acceptorcharacteristics may cause ethyl acetate (acceptor) to clusterless strongly than the alcohols (donor), or (2) the muchgreater dipole moment of ethyl acetate and the greaterpolarity of the PhIP part of dG-C8-PhIP may cause theclustering with dG-C8-PhIP to be so strong that the complexremains fully saturated throughout the DMS waveform,quenching the differential mobility effect. Ab initio calcula-tions of Gibbs free energies of cluster formation do showsignificantly stronger binding with ethyl acetate than withisopropanol as expected from the charge-dipole interaction(manuscript in preparation). The stronger effect of 1-butanolcompared with isopropanol may be due to the largergeometric size of 1-butanol so that the cross-section

Figure 2. Comparison of peak shapes of dG-C8-4-ABP with and without modifier. Improvements in peak shape by theintroduction of modifiers can be realized by comparing the FWHM (full width at half maximum) at a constant SV of 3500 V

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

increases more on binding 1-butanol. Steric inhibition of themethyl groups located alpha to the hydroxylic carbon mightalso play a role in determining the free energy change oncomplex formation since isopropanol is disadvantaged bythe mid-chain position of the hydroxyl group.

Use of Modifiers in Separation

The ability of gaseous modifiers to control the compensationvoltage of an analyte provides the means for enhancingseparations during DMS analysis. Even more importantly, ina targeted analysis, control of the CoV shift can provide ameans for isolating the analyte of interest into a transmissionzone that is free of matrix interferences, thereby improvingsignal to noise ratio and sensitivity. In the studies of thebiomarkers of DNA damage, the matrix is a DNA enzymaticdigest comprised of normal deoxynucleosides and enzyme-related breakdown products. Ensuring the separation of theadduct from normal (unmodified) nucleosides is of particularimportance since they have the common deoxynucleosidefunctional group.

As an initial test, a mixture of 1 ng of eachdeoxyguanosine (dG), deoxycytosine (dC), deoxyadenosine(dA), and 100 pg of the DNA adduct dG-C8-4-ABP wasprepared to investigate the role of modifiers on separation.Two modifiers, ethyl acetate and isopropanol, were evalu-ated for separation of the dG-C8-4-ABP adduct from theexcess unmodified DNA bases. CoV scans at a fixed SVvalue were performed and the analytes of interest wereextracted from the TIC. It should be noted that effectiveseparations were not observed in the absence of a modifier.

From Figure 3a, we can see that the four-componentnucleoside mixture is separated well by the DMS using ethylacetate as the modifier. As expected, dG-C8-4-ABP beingthe bulkiest molecule in the mixture, moves the leasttowards the negative compensation voltage compared withother unmodified nucleosides because its mobility (~1/cross-section) will be less modified by cluster formation. dAappears as a double peak whose characteristics have notbeen investigated further. Ethyl acetate and isopropanol havebeen successfully applied as effective modifiers for otherapplications with DMS, so we decided to screen isopropanolfor separation purposes as well with dG-C8-4-AB. Introduc-tion of isopropanol as the modifier provided the necessaryconditions for separation of the normal nucleosides dA anddC but failed to resolve dG from dG-C8-4-ABP as the lattertwo were transmitted at the same CoV value. It was,therefore, decided to ascertain if an increase in modifierconcentration could be used in order to achieve desiredseparations. As can be seen from (Figure 3b(i) and b(ii)), dGand dG-C8-4-ABP could not be separated without modifieror when the modifier isopropanol was introduced at a lowerrate of 0.6% into the curtain gas. However, when themodifier was introduced at the rate of 1.1%, separation wasachieved between dG and dG-C8-4-AB The small dG signalat the dG-C8-4-ABP CoV is interpreted as arising from

fragmentation of dG-C8-4-ABP to dG after the DMS and inthe mass spectrometer transition to vacuum [Figure 3b(iii)].Using the example of dG-C8-4-ABP and unmodified dG, wehave demonstrated here that the percentage of modifierbeing introduced can be used as an another tool in achievingor enhancing separations using a DMS based platform.

Effect of Separation Voltages and Modifierson Signal Intensities in MRM Mode

In the experiments discussed above, the focus was on theCoV shifts of dG-C8-4-ABP and dG-C8-PhI. These studiesestablished that the use of organic modifiers can have adramatic effect on analytical performance. Given our goal ofusing the DMS-MS platform to perform rapid quantitation ofthese biomarkers, we examined next the influence ofmodifiers on signal intensity in order to integrate all ofthese variables into a comprehensive analytical protocol. Therapid scanning features of the DMS-MS platform forperforming rapid quantitation can be best appreciated byusing it in the CoV stepping mode and the signal intensitiesmonitored in the MRM mode. Under these conditions, oncethe CoV apex for any given analyte has been identified, thisvoltage value can be set for subsequent quantitative analysis.Setting the CoV value allows the user to filter out matrixcontaminants and selectively transmit analyte(s) of interestcontinuously into the mass spectrometer without physicallyadding time for the mass spectral acquisition. With theDMS-on, signal intensities in the MRM mode were thusmonitored for a fixed concentration of each compound tooptimize the best DMS conditions for selected analyte iontransmission.

As mentioned above, in light of the variations in CoVshifts as a function of modifier concentration, it was deemeduseful to monitor the effect of modifier choice on theproduct ion signal intensity. Accordingly, samples contain-ing 0.100 pmol of dG-C8-PhIP were analyzed and, initially,CoV scans were performed at different values of separationvoltages and modifier percentages in order to identify thepeak apex values. Once these values had been identified,product ion counts were monitored in the MRM mode at therespective operating values of separation voltage, compen-sation voltage, and modifier percentage (only combinationsof separation voltages and modifier percentages that exhib-ited shift away from the zero compensation voltage wereinvestigated). It is shown that with increasing separationvoltages at all three isopropanol modifier percentages (0.6%,1.1%, and 2.2%) examined, the general trend on theintensity of the product ion is the same (Figure 4a, b, c,and d). Specifically, the product ion intensity value increasesinitially, peaks off at SV=3500 V and then starts decreasingwith application of higher separation voltage. The initialincrease with the application of the DMS field is generallydescribed as a DMS focusing effect, but may also involveenhanced desolvation of the electrospray plume due to rfheating by the DMS field. The transmission of the analyte

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

ion through the DMS electrodes is optimal at the separationvoltage value of 3500 V.

Further increase in separation voltage results in areduction in analyte ion count, which is evident at eachmodifier concentration, but becomes more pronounced asmodifier concentration increases. This has previously beenillustrated through the use of a dispersion plot [35], wherethe drop in signal intensity is shown in 2-D DMS SV-CoVscanning mode. Figure 4d shows the comparison of theproduct ion intensity at separation voltages of 3500, 4000,and 4500 V with varying isopropanol modifier percentagesof 0.6%, 1.1%, and 2.2%. Signal intensity of the product ionof dG-C8-PhIP falls both with increased modifier concen-tration and with SV above 3500 V. This effect can beunderstood in terms of two contributions: (1) highermodifier contributions result in more frequent ion-modifiercollisions, leading to higher probability of loss of chargefrom the ion or ion cluster, and (2) higher SV results inhigher velocity and more energetic and more destructivecollisions between the ion and the transport gas, againleading to a higher rate of charge loss, or even to reactivecollisions [48].

Referring back to Figure 1b, it may be noted that in thecase of dG-C8-PhIP, the shifts have been significantlyinfluenced by both IPA and 1-butanol with the lattergiving larger CoV shifts. It was, thus, important in thecontext of quantitative analysis to also compare theeffect of these two modifiers on the signal intensities.

Samples of dG-C8-PhIP (20 fmol), using isopropanol and1-butanol at 0.6% were screened at the optimal separa-tion voltage value of 3500 V (Figure 4e and f). Signalintensities were 830 for isopropanol and 400 for 1-butanol (2.1:1). For dG-C8-PhIP, use of isopropanol asthe modifier gave much better signal intensity than 1-butanol by more than a factor of two. For dG-C8-4-ABP,a sample containing 58 fmol of dG-C8-4-ABP wasanalyzed in the MRM mode at a separation voltagevalue of 3500 V at 0.6% concentration of modifiersisopropanol and ethyl acetate in two separate runs.Signal intensities were 570 for IPA and 910 for ethylacetate (1:1.60) Thus, for dG-C8-4-ABP, ethyl acetate asthe modifier produced 60% higher signal intensity in theMRM mode compared with isopropanol. The sameexperiment was repeated at different concentrations ofboth the analytes (dG-C8-4-ABP and dG-C8-PhIP) andthe same trend was observed for all concentrations.

Since generation of ion signal is directly dependent on theabundance of the respective [M+H]+ions, the simplestassumption is to attribute these differences in signal intensityto a competition for the proton between the analyte ion orion-neutral clusters and the modifiers. The proton affinity ofethyl acetate is 835.7 kJ/mol, isopropanol is 793.0 kJ/mol,and 1-butanol is 789.2 kJ/mol, respectively [49]. Had thedifference in modifier proton affinity been the dominantfactor, ethyl acetate with a greater proton affinity value thanisopropanol would have produced lower signal intensity for

Figure 3. (a) Separation of a mixture of dC, dA, dG, and dG-C8-4-ABP in the presence of ethyl acetate modifier at a fixedconcentration done on the DMS- ion trap (b) Separation of dG and dG-C8-4-ABP (i) without modifier, and using IPA modifier attwo different modifier concentration done on DMS-triple quadrupole (ii) 0.60%, and (iii) 1.10%

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

dG-C8-4-AB For dG-C8-PhIP, although the proton affinityvalue of 1-butanol is almost the same as that of isopropanol,the difference in signal intensity between the two modifiersis large, with isopropanol giving higher signal intensity.Methyl groups on the isopropanol introduce some sterichindrance that reduces the contribution of H-bonding to theclustering process, whereas the hydroxyl group is moreexposed in 1-butanol, which may have facilitated clustering.In an alternative model, the fact that isopropanol and 1-butanol have very similar dipole moments while ethylacetate dipole moment is much larger could lead to

alternative conclusions. None of these approximations canlead to a full and accurate prediction of modifier behavior;only accurate thermochemical free energy values based onall these effects can be expected to be of help. Thus,screening of potential modifiers against the analyte(s) ofinterest is necessary and can provide users a betterunderstanding of the modifier selection process and optimi-zation of the DMS parameters for analysis. Signal intensitiesin detail must be examined experimentally, although thegeneral trends with modifier concentration and with SVamplitude are understood.

Figure 4. Effect of separation voltage and modifier isopropanol (IPA) concentration on the MS/MS signal intensity of the DNAadduct dG-C8-PhIP (a) at 0.6%, (b) 1.1%, (c) 2.2%, (d) Trends of AB SCIEX DMS API 3000 signal intensities at separationvoltages of 3500, 4000, 4500 V with increasing modifier percentages. Effect of modifiers on the product ion intensity at a fixedseparation voltage of 3500 V; (e) modifiers isopropanol (IPA) and 1-butanol on the DNA adduct dG-C8-PhIP, and (f) modifiersisopropanol (IPA) and ethyl acetate (EtoAc) on the DNA adduct dG-C8-4-AB Calibration curves (g) dG-C8-4-ABP with ethylacetate (0.6%), SV(3500V), and (h) dG-C8-PhIP with IPA (0.6%), SV 3500 V. Selection of optimum conditions of SV andmodifier allow accurate calibration of low adduct concentrations as indicated by calibration curves

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

Quantitation of DNA Adducts in Calf-ThymusDNA

The gas phase modifier selection and optimization studieswere performed in order to facilitate rapid quantitation ofDNA adducts: dG-C8-4-ABP and dG-C8-PhIP, with theresults shown in Figure 4g and h. Ethyl acetate at 0.6% wasestablished as the modifier of choice for dG-C8-4-ABPanalysis and isopropanol at 0.6% was established as themodifier of choice for dG-C8-PhIP analysis. Separationvoltage value at 3500 V was used for analysis of both theadducts. For dG-C8-4-ABP quantitation, the transition fromthe precursor ion [M+H]+ of dG-C8-4-ABP (m/z 435) to theproduct ion (m/z 319) [M+H – 116]+ was monitored for theanalyte and the transition from the precursor ion [M+H]+ ofdG-C8-4-ABP-d9 (m/z 444) to the product ion (m/z 328)[M+H – 116]+ was monitored for the internal standard,respectively. For dG-C8-PhIP quantitation, the transitionfrom the precursor ion [M+H]+ of dG-C8-PhIP (m/z 490) tothe product ion (m/z 374) [M+H – 116]+ for the analyte andtransition from the precursor ion [M+H]+ of dG-C8-4-ABP-d9 (m/z 493) to the product ion (m/z 377) [M+H – 116]+ forthe internal standard were monitored. Two μg of calf thymusDNA was used as the matrix for each sample point andcalibration curves were prepared in the DMS-on mode. Theutility of DMS to remove matrix ion interferences to producea linear calibration curve has previously been demonstrated[36]. DNA modifications of less than 1 in 106 nucleosideswere detected with both of the calibration curves showingexcellent linearity (R290.99) through 1000 modifications in107 nucleosides after a simple protein precipitation step. It isinteresting to note that whereas the calibration curve for dG-C8-PhIP had essentially a zero intercept, the one for dG-C8-4-ABP crossed the y-axis at a higher value, most likely dueto background interferences at the ion masses monitored.

In reviewing the calibration curves in Figure 4, it isimportant to further consider the broader significance of theresults and the implications not only to the field of DNAadduct analysis but also to other related analytical applica-tions that may be conducted by DMS-MS/MS. To beginwith, these calibration curves were generated in approxi-mately 3 h and consisted of 7–9 points each analyzed intriplicate. In addition, they included several blanks in orderto account for any carryover contamination. This is only afraction of the time normally consumed using our traditionalLC-MS/MS approach in which additional valuable time isalso expended waiting for column equilibration before eachchromatographic run. Moreover, the value of DMS-MS/MSshould also be considered in the context of the broaderpicture of an analysis of biological specimens, whichtypically involves several sample preparation steps, beforeinjection into the LC-MS. The procedure for the DNAadducts of interest here is outlined in Figure 5 and alsoshows the approximate time period associated with each oneof the steps. As indicated, following a simple proteinprecipitation, lyophilized samples can be reconstituted and

directly analyzed by DMS/MS bypassing other preparativesteps. In the DMS-on mode, the analyte of interest can beselectively introduced into the mass spectrometer byselecting the appropriate modifier and setting the CV to aunique voltage for that particular compound. Once theparameters for DMS/MS analysis have been set, data wereacquired in just 30 s and, since DMS is a continuousmethod, subsequent samples can be introduced without theneed for re-equilibration. In summary, the example present-ed here for the analysis of DNA adducts from a biologicalmatrix demonstrates the high throughput analytical capabil-ities of DMS and the time savings that can be realized in abroad range of bioanalytical applications.

Thermochemical Kinetics of Modifier-AssistedDMS

Ion-neutral clustering has been the subject of considerablestudy, although mechanisms are complex and not yet fullyelucidated. A useful early review is that of Castleman andKeesee [50], but there are a number of more recent studies,including those from the Leone group [51], and others. Aquantitative analysis of comparative modifier effects appearspossible based on thermochemical free energy values fromcomputational chemistry, optimized conformations consid-ering energetics and steric hindrance, combined withcomputations of mobility [18, 22, 52] from predictedstructures, and further investigations are in progress. Theschema and thermochemistry for ion-cluster/modifier equi-libria is summarized in Appendix A.

In order to understand the observed variation in compen-sation voltage with DMS field strength and modifierconcentration, it is necessary to consider three effects ofthe DMS field on the ion and on ion-modifier clusters: theheating of the ion, the change in the frequency of collisionbetween the ion and the bulk gas, and the change in collisionenergy. The rapidly varying DMS field does not affect thetransport gas because the ions are very dilute in the total gasmixture. However, the DMS field does change the internaland translational energy of the ion/ionic cluster, as well asthe collision energy and collision rate for collisions betweenthe ion or ion cluster and the transport gas composition. Theheating effect on the ion can be written in terms of an ioneffective temperature, Teff, which is different from the bulkgas temperature, T, as follows [7, 10, 16]:

Teff ffi T þ ςM

3kBNKð Þ2 E

N

� �2

ffi T þ MN20

3kBςK2

0

E

N

� �2

¼ T þ MkBN20

3P2

� ςK2

0

� �T2E2

ð5Þ

In this expression, P is the ambient pressure, N0 the gasnumber density at 0 °C, 1 atm, M the mass of the transportgas neutral molecule. We have first used the reduced

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

mobility relationship NK=N0K0 (exact for pressure varia-tion, approximate for temperature [Ch. 5-1.B, Mason andMcDaniel]) [10], then the ideal gas law, and we havefollowed Krylov [16] in introducing ζ, an inelasticity factornecessary for polyatomic ions. Without an independentmeasurement of K0, DMS data can determine only theςK0

2 product. For an understanding of the modifier effects inDMS, the most important aspect of Teff is the DMS heatingterm (Teff−T) which is quadratic in both bulk temperature, T,and applied electric field, E; these effects jointly allow ionsin DMS to be strongly heated and become unclustered.

The kinetic limit to the ability of clustering number totrack the dynamic effective temperature can be estimatedfrom the collision rate in the two-temperature theory, andcomparing it to the amount of time spent at each DMSvoltage level (Figure 6a). There are different ways toestimate the collision rate, but the different approachesproduce similar results. For instance, equation (4) can berearranged to estimate the A-C collision rate, ξA-C, as

ξA−C ¼ C½ �˙ vrel Teff

� �˙ Ω Teff

� �

ffi C½ �˙3q

4

1þ αKμACN0K0

¼ XCT 0

T

3q

4

1þ αKμACN0K0

ð6Þ

(XC is the neutral mole fraction). A form that differs onlyin numerical coefficient is given by Equation 37 inGoeringer and Viehland [53]. For collisions between 1%isopropanol in a non-polar transport gas and molecular ionsin the range from methyl histamine to deoxyguanosine

cations, assuming K0=1.5 cm2/V∙s, and T=200 °C, themomentum transfer collision rate is generally between 50and 100 per microsecond. From Figure 6a, the 2-Hwaveform spends about 50% of the period near the lowfield value and 10% near the highest field. For the ABSCIEX Selexion (3 MHz frequency) this gives 0.167 μs(G17 collisions) at low field, and 0.033 μs (~3 collisions) athigh field.

The ability of the internal energy of the ion and itsclustered state to remain in dynamic equilibrium with theDMS effective temperature, as determined by the varyingelectric field, bulk pressure, and temperature, is based on thefollowing considerations:

1. Cluster formation rate. Ion mobility is inversely relatedto the ion-neutral collisional cross-section. The ionmobility cross-section was first derived by Langevin[54] and more fully developed for ion–polar moleculeinteractions by Su and coworkers using trajectorycalculations and dipole orientation [55–58], and byTurulski and Forys using transition state theory and othermethods [27]. The long-range interaction between an ionand a polar molecule is through the ion - dipoleinteraction and is strongly attractive (r-2 , [59, 60]).Cluster formation, when the free energy of association isnegative, can occur in a very small number of collisions,or even in a single collision, as long as the clustercontains more than a few atoms so that the rovibrationalstate density allows the collision energy to be retained asinternal energy (see especially Lias and Ausloos [61]).

Figure 5. Comparison of LC-MS and DMS-MS speed of analysis for DNA adducts. For SPE-LC-MS, a second lyophilizationstep is required

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

Thus, cluster formation between an ion and a polarmodifier can proceed at a nearly gas-kinetic rate.Redistribution and stabilization of the internal degreesof freedom within the cluster [62–64] is enhanced byhydrogen bonding between the polar modifier and the ionafter they have been attracted by the long range forces.

2. Thermal equilibration rate. The cluster ion acquireskinetic energy through field acceleration between colli-sions but loses this energy in collisions, largely with thepredominant non-polar bulk gas (N2 in our case).Although modifier collisions can lead to cluster forma-tion, the more frequent bulk-gas collisions lead tomodifier loss and thermal equilibration at the effectivetemperature, affecting both the internal state and the sizeof the cluster ion. This effective temperature is deter-mined with reasonable accuracy by momentum transfertheory, as modified for polyatomic ions by Krylov [7].The theory developed by Goeringer and Viehland [65,66] for an analysis of ion swarms in an ion trap can beapplied to follow the oscillatory motion of the dynamiccluster ion. Thus, the collision rate that determinesequilibration of the cluster at the effective temperature isapproximately 100 times faster than the rate that controlscluster formation (for 1% modifier). Thermal equilibra-tion of an existing cluster ion at the effective temperaturecorresponding to the electric field, bulk temperature, andpressure occurs rapidly because it is controlled by bulk gascollisions. Because collisions with the polar modifier occurmuch less frequently, the dynamic change in cluster numberfrom the DMS high-field, de-clustered condition to the low-field, clustered condition can be limited by kinetic effects.

3. Saturation behavior. After the first modifier is in place,modifier shell development is controlled and limited bygeometric compatibility, charge and dipole shielding,polarity localization, and hydrogen bonding. It is possiblefor a second strongly dipolar modifier to add to anexisting cluster canceling the dipole moment of the first,but this behavior is dependent on the lock and key

relationships between ion and modifier. We have ob-served such saturation behavior with ethyl acetate, buthave not yet performed any related calculations on thedecrease in cluster free energy with size. Multipleclustering makes performance less predictable, andappears to be a primary cause of the need for com-pound-specific method development.

As an example of the cooperative effect of gas temper-ature and DMS separation field in modulating the clusternumber, we extract one result from an extensive set of ab-initio calculations for the change in free energy, enthalpy,and entropy on cluster formation. The full set results will bereported in detail, but are beyond the scope of the currentpublication. Ab initio structural optimizations and thermo-chemical calculations were performed for several smallorganic molecules with and without the modifiersisopropanol or ethyl acetate. The calculations were donewith a series of new and intensive G4 “compound” methods,which have had good success in predicting complexationreactions [67]. For all molecules studied, ΔG0 values showthat isopropanol is less strongly bound than ethyl acetate,much as might be expected from the difference in dipolemoment values [IPA (1.560 D), EtOAc (4.325 D)]. For R-alpha-methylhistamine (C6H11N3, mw 125.17164,PubChem 156615), thermochemical values for clustering ofthe protonated methyl-histamine molecule with IPA weredetermined to be (ΔG0=–7.45 kcal/mol, ΔH0=–16.62 kcal/mol, ΔS0=–30.76 cal/mol/K). This compound has somestructural similarity to deoxyguanosine in that it has nitrogenheteroatoms both in a 5-membered ring and in an aminogroup. In order to examine access to the unclustered ionconfiguration, and in recognition of the kinetic limitdiscussed above, we have applied a limit of NC=3 for themaximum IPA count. Assuming K0=1.5 cm2/V∙s, 1.5%isopropanol, ζ=0.7, the cluster number dependence both onapplied electric field and on bulk gas temperature for(methyl histamine∙H+ ∙isopropanol) can be calculated, with

Figure 6. (a) Flyback and two-harmonic waveforms. The two-harmonic shape is used by AB SCIEX in the DMS API 3000, andthe flyback shape on the ion trap. (b) Using ab-initio thermochemical values, the mean number of bound neutrals is shown as afunction of DMS field and transport gas temperature at 1 atm. The maximum neutral count is limited to 3 because shellthermochemistry has not been calculated. These results for R-alpha-methylhistamine, which has a polar core similar to dG, with1.5% isopropanol, show that bulk gas temperature is an essential controlling parameter in modifier DMS selectivity and thatmultiple clustering is common even at relatively high bulk gas temperatures

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

the very interesting result of Figure 6b. The requirement ofhigh combined field and/or bulk gas temperature for iondeclustering is not unexpected because of the earlierexperience with atmospheric pressure drift-time IMS [22,25, 26, 68], but it indicates that both bulk gas temperatureand DMS field are critical control parameters

We can also use these results to interpret intensity data. InFigure 4d, we saw that the DMS intensity fell off at higherDMS field values. An additional mechanism for signal lossin the higher SV range is collision-induced chemistry, eitherdissociation (CID) or charge detachment. CID in DMS hasbeen reported in DMS [48] and charge detachment insystems like SF6

– is well known [69]. At lower DMS field,the heating is quite a bit less so that less declustering willoccur, preserving the ion intensity regardless of the modifierconcentration. At very high fields very high energycollisions are effective at declustering but can also bedestructive, more so for larger clusters because the protonis more likely to go with the cluster of modifiers. Thus, bothDMS field and modifier concentration affect the signalintensity in similar ways, as our data shows.

ConclusionsIn the past few years, efforts have been made to understandand characterize the gas phase cluster formation between theanalyte ion and the modifiers. However, an explicit methodof characterizing the interactions and predicting the CoVshifts is not available to date. Screening modifiers againstthe analyte of interest at different conditions has been themost feasible way to perform method development usingdifferential mobility spectrometry and contribute to theunderstanding on the use of modifiers.

In this report, we propose theoretical considerations for thephenomena associated with compensation voltage shifts inDMS that explain step-wise clustering of modifiers including“saturation effect” and variations in signal intensities. Theseresults show that high transport gas temperatures increase theDMS effect and that some combinations of temperature andDMS field lead to sharp changes in cluster number with DMSSV field (Figure 6). Conversely, with moderate DMStemperatures, DMS is modulating an already clustered ion,making the effect less predictable and requiring a methoddevelopment phase. This process is illustrated by the approachwe have taken for the selective quantitation of modifiednucleosides, dG-C8-4-ABP and dG-C8-PhIP, used as modelanalytes, from a complex mixture of normal nucleosides usinga differential mobility-mass spectrometry-based platform. Thepotential to eliminate or minimize sample preparation stepswhile bringing the analysis time down to less than a minute persample point makes this method highly attractive for furtherbioanalytical applications. Optimizing the DMS operationalparameters including modifier concentration gives us themaximum selectivity and sensitivity for selected applications.

Modifier DMS at lower temperatures is generally mod-ulating the size of an already clustered ion species for which

thermochemical information is not available and few calcula-tions have been done. We believe that further progress can bemade in understanding and predicting DMS modifier perfor-mance by approaching future experimental work from the hightemperature limit with modifiers of differing dipole momentsand structure, and by performing calculations on the develop-ment of solvent shell clusters.

AcknowledgmentsThe authors consider it an honor to be included in thisspecial issue of JASMS honoring Professor Yinzheng Wangin his receipt of the Biemann Medal. The authors gratefullyacknowledge support from NIH: RO1 CA 069390-16 (P.V.)and R01 AI101798 (A.J. F. Jr.).

Appendix A. Ion-modifier ClusteringKineticsThe kinetic analysis of ion-neutral clustering can be based onthe following scheme describing the chemical equilibria amongclustering polar modifier, C, protonated ion, MH+, and ion-neutral clusters of different sizes, (Cn)∙MH+ in the gas phase.

MHþ þ C ←→k−1

k1Cð Þ1˙MHþ

Cð Þ1˙MHþ þ C ←→k−2

k2

Cð Þ2˙MHþ

� � �

K1 ¼ k1k−1

¼ Cð Þ1˙MHþ �MHþ½ � C½ � ¼ exp −ΔG0; 1ð Þ=RT

� �

¼ exp −ΔH0; 1ð Þ=RT þ ΔS0; 1ð Þ=R� �

K2 ¼ k2k−2

¼ Cð Þ2˙MHþ �Cð Þ1˙MHþ �

C½ �¼ exp −ΔG0; 2ð Þ=RT

� �¼ exp −ΔH0; 2ð Þ=RT þ ΔS0; 2ð Þ=R

� �� � �

ðA:1Þ

Using an abbreviated notation for concentrations relativeto the standard state of 273.15 K, 1 atm, we writeconcentrations and the total ion concentration as follows:

Aj≡ Cð Þ j˙MHþh i

;C≡ C½ �;Atotal≡ MHþ½ �totalAtotal ¼ A0 þ A1 þ A2 þ…

ðA:2Þ

The Gibbs free energy of cluster formation is expected todecrease with increasing cluster size. Because cluster shellthermodynamic properties for ions of significant size are not

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

available, we use a constant value, ΔG0,(j)=ΔG0, independentof cluster number, j. Instead, we limit the maximum number ofclustered modifiers to a specified number, NC. This is a modifiershell-size limit corresponding to a number of modifier moleculesgeometrically and electrically compatible with the conditions.There is also a kinetic limit in DMS to the maximum change incluster number between low and high field cases, determined bythe ratio of the ion-neutral collision rate to the waveformfrequency, as discussed in the text. The two approaches consistingof (1) assuming a slowly declining free energywith cluster size, or(2) taking a discrete maximum shell size, are similar, but differqualitatively in one important way. Only the shell model can leadto the dramatic solvent saturation effect we observed in dG-C8-4-ABP with ethyl acetate, but not with isopropanol.

The results for the populations of each ion-neutral clustersize, Aj, and the average cluster number, nC , in terms of theproduct KC=K1C (the equilibrium constant times the neutralconcentration relative to the standard state, giving thedensity ratio of larger cluster to smaller) are shown here.

Aj ¼ AtotalKjC

1−KC

1−K1þNCC

; and

n�C ¼ KC 1− 1þ Nð ÞKNCC þ NCK

1þNCC

� �1−KCð Þ 1−K1þNC

C

� � ; if KC≠1;

or n�C ¼ NC

2; if KC ¼ 1:

ðA:3Þ

As a first approximation, the stepwise equilibriumconstants are taken to be independent of cluster number,Kj=K1 for 1≤ j≤NC, where NC is the maximum number ofclustered neutrals. The concentrations are relative to 1 atm,at T0=273.15 K, so that 1% modifier at 1 atm corresponds toC≅0.01 N0, or at a different temperature or pressure toC≅0:01N 0

PT0P0T

¼ 0:01N , where N is the total number

density, with either N0 under standard conditions (T0=273.15 K, 1 atm), or N at a different pressure andtemperature.

References1. Schneider, B.B., Covey, T.R., Coy, S.L., Krylov, E.V., Nazarov, E.G.:

Control of chemical effects in the separation process of a differentialmobility mass spectrometer system. Eur. J. Mass Spectrom. 16(1), 57–71 (2010)

2. Coy, S.L., Krylov, E.V., Schneider, B.B., Covey, T.R., Brenner, D.J.,Tyburski, J.B., Patterson, A.D., Krausz, K.W., Fornace, A.J., Nazarov,E.G.: Detection of Radiation-exposure biomarkers by differentialmobility prefiltered mass spectrometry (DMS-MS). Int. J. MassSpectrom. 291(3), 108–117 (2010)

3. Schneider, B., Covey, T., Coy, S., Krylov, E., Nazarov, E.: Planardifferential mobility spectrometer as a pre-filter for atmosphericpressure ionization mass spectrometry. Int. J. Mass Spectrom. 298(1/3), 45–54 (2010)

4. Schneider, B.B., Covey, T.R., Coy, S.L., Krylov, E.V., Nazarov, E.G.:Chemical effects in the separation process of a differential mobility/mass spectrometer system. Anal. Chem. 82(5), 1867–1880 (2010)

5. Hall, A., Coy, S.L., Kafle, A., Glick, J., Nazarov, E., Vouros, P:Extending the dynamic range of the ion trap by differential mobilityfiltration. J. Am. Soc. Mass Spectrom. 24(9), 1428–1436 (2013)

6. Hall, A.B., Coy, S.L., Nazarov, E.G., Vouros, P.: Rapid separation andcharacterization of cocaine and cocaine cutting agents by differentialmobility spectrometry-mass spectrometry. J. Forensic Sci. 57(3), 750–756 (2012)

7. Krylov, E.V., Nazarov, E.G.: Electric field dependence of the ionmobility. Int. J. Mass Spectrom. 285(3), 149–156 (2009)

8. Shvartsburg, A.A.: Differential Ion Mobility Spectrometry: NonlinearIon Transport and Fundamentals of FAIMS, p. 322. CRC Press, Taylorand Francis Group, Boca Raton (2008)

9. Eiceman, G.A., Karpas, Z.: Ion Mobility Spectrometry, p. 350. Taylorand Francis/CRC Press, Boca Raton (2005)

10. Mason, E.A., McDaniel, E.W.: Transport Properties of Ions in Gases.Wiley-Interscience, New York (1988)

11. Buryakov, I.A., Krylov, E.V., Makas, A.L., Nazarov, E.G., Pervukhin,V.V., Rasulev, U.K.: Drift spectrometer for the control of amine tracesin the atmosphere. J. Anal. Chem. 48(1), 114–121 (1993)

12. Buryakov, I.A., Krylov, E.V., Nazarov, E.G., Rasulev, U.K.: A newmethod of separation of multi-atomic ions by mobility at atmosphericpressure using a high-frequency amplitude-asymmetric strong electricfield. Int. J. Mass Spectrom. Ion Processes 128(3), 143–148 (1993)

13. Krylov, E., Coy, S., Vandermey, J., Schneider, B., Covey, T., Nazarov,E.: Selection and generation of waveforms for differential mobilityspectrometry. Rev. Sci. Instrum. 81(2), 024101 (2010)

14. Levin, D.S., Vouros, P., Miller, R.A., Nazarov, E.G., Morris, J.C.:Characterization of gas-phase molecular interactions on differentialmobility ion behavior utilizing an electrospray ionization-differentialmobility-mass spectrometer system. Anal. Chem. 78(1), 96–106 (2006)

15. Nazarov, E.G., Coy, S.L., Krylov, E.V., Miller, R.A., Eiceman, G.A.:Pressure effects in differential mobility spectrometry. Anal. Chem.Anal. Chem. 78(22), 7697–7706 (2006)

16. Krylov, E.V., Coy, S.L., Nazarov, E.G.: Temperature effects indifferential mobility spectrometry. Int. J. Mass Spectrom. 279(2/3),119–125 (2009)

17. ABSCIEX. selexION technology. Avai lable at : ht tp: / /www.absciex.com/Documents/Downloads/Literature/SelexION-Technology-New-Solution-Selectivity-Challenges-TN%20P-2960211-01.pdf. Accessed August 2013

18. Campuzano, I., Bush, M.F., Robinson, C.V., Beaumont, C., Richard-son, K., Kim, H., Kim, H.I.: Structural characterization of drug-likecompounds by ion mobility mass spectrometry: comparison oftheoretical and experimentally derived nitrogen collision cross sections.Anal. Chem. 84(2), 1026–1033 (2012)

19. Salbo, R., Bush, M., Naver, H., Campuzano, I., Robinson, C.,Pettersson, I., Jorgensen, T., Haselmann, K.: Traveling-wave ionmobility mass spectrometry of protein complexes: accurate calibratedcollision cross-sections of human insulin oligomers. Rapid Commun.Mass Spectrom. 26(10), 1181–1193 (2012)

20. Kim, H., Johnson, P, Beegle, L., Beauchamp, J., Kanik, I.: Electrosprayionization ion mobility spectrometry of carboxylate anions: ion mobilitiesand amass-mobility correlation. J. Phys. Chem.A 109(35), 7888–7895 (2005)

21. Johnson, P., Kim, H., Beegle, L., Kanik, I.: Electrospray ionization ionmobility spectrometry of amino acids: ion mobilities and a mass-mobility correlation. J. Phys. Chem. A 108(27), 5785–5792 (2004)

22. Kim, H., Kim, H., Johnson P., Beegle, L., Beauchamp, J., Goddard, W.,Kanik, I.: Experimental and theoretical investigation into the correlationbetween mass and ion mobility for choline and other ammonium cationsin N-2. Anal. Chem. 80(6), 1928–1936 (2008)

23. Berant, Z., Karpas, Z.: Mass-mobility correlation of ions in view of newmobility data. J. Am. Chem. Soc. 111(11), 3819–3824 (1989)

24. Berant, Z., Karpas, Z., Shahal, O.: Effects of temperature and clusteringon mobility of ions in CO2. J. Phys. Chem. 93(21), 7529–7532 (1989)

25. Berant, Z., Shahal, O., Karpas, Z.: Correlation between measured andcalculated mobilities of ions - sensitivity analysis of the fittingprocedure. J. Phys. Chem. 95(19), 7534–7538 (1991)

26. Karpas, Z., Berant, Z., Shahal, O.: Effect of temperature on the mobilityof ions. J. Am. Chem. Soc. 111(16), 6015–6018 (1989)

27. Turulsk i , J . , Forys , M.: Ion-pola r molecule co l l i s ionfrequency—thermodynamic treatment. J. Phys. Chem. 83(22), 2815–2817 (1979)

28. Eiceman, G.A., Krylov, E.V., Krylova, N.S., Nazarov, E.G., Miller,R.A.: Separation of ions from explosives in differential mobility

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation

spectrometry by vapor-modified drift gas. Anal. Chem. 76(17), 4937–4944 (2004)

29. Levin, D.S., Vouros, P., Miller, R.A., Nazarov, E.G.: Using ananoelectrospray-differential mobility spectrometer-mass spectrometersystem for the analysis of oligosaccharides with solvent selected controlover ESI aggregate ion formation. J. Am. Soc. Mass Spectrom. 18(3),502–511 (2007)

30. Rorrer III, L.C., Yost, R.A.: Solvent vapor effects on planar high-fieldasymmetric waveform ion mobility spectrometry. Int. J. Mass Spectrom.300(2), 173–181 (2011)

31. Schneider, B., Nazarov, E., Covey, T.: Peak capacity in differentialmobility spectrometry: effects of transport gas and gas modifiers. Int. J.Ion Mobility Spectrom. 15(3), 141–150 (2012)

32. Tsai, C.W., Yost, R.A., Garrett, T.J.: High-field asymmetricwaveform ion mobility spectrometry with solvent vapor addition:a potential greener bioanalytical technique. Bioanalysis 4(11),1363–1375 (2012)

33. Coy, S.L., Krylov, E.V., Nazarov, E.G., Fornace, A.J.Jr., Kidd, R.D.:Differential mobility spectrometry with nanospray ion source as acompact detector for small organics and inorganics. Int. J. Ion MobilitySpectrom. 16, 217–227 (2013)

34. Johnson, P, Beegle, L., Kim, H., Eiceman, G., Kanik, I.: Ion mobilityspectrometry in space exploration. Int. J. Mass Spectrom. 262(1–2), 1–15 (2007)

35. Levin, D.S., Miller, R.A., Nazarov, E.G., Vouros, P.: Rapid separationand quantitative analysis of peptides using a new nanoelectrospray-differential mobility spectrometer-mass spectrometer system. Anal.Chem. 78(15), 5443–5452 (2006)

36. Kafle, A., Klaene, J., Hall, A.B., Glick, J., Coy, S.L., Vouros, P.: Adifferential mobility spectrometry/mass spectrometry platform for therapid detection and quantitation of DNA adduct dG‐AB. RapidCommun. Mass Spectrom. 27(13), 1473–1480 (2013)

37. Varesio, E.: Le Blanc, Hopfgartner, G.: Real-time 2D separation byLCx differential ion mobility hyphenated to mass spectrometry. Anal.Bioanal. Chem. 402(8), 2555–2564 (2012)

38. Jasak, J., Le Blanc, Y., Speer, K., Billian, P., Schoening, R.M.:Analysis of triazole-based metabolites in plant materials using differen-tial mobility spectrometry to improve LC/MS/MS selectivity. J. AOACInt 95(6), 1768–1776 (2012)

39. Garrigos, M.C., Reche, F., Pernias, K., Sanchez, A., Jimenez, A.:Determination of some aromatic amines in finger-paints for children'suse by supercritical fluid extraction combined with gas chromatography.J. Chromatogr. A 819(1,2), 259–266 (1998)

40. Oh, S.W., Kang, M.N., Cho, C.W., Lee, M.W.: Detection ofcarcinogenic amines from dyestuffs or dyed substrates. Dyes Pigments33(2), 119–135 (1997)

41. Tokiwa, H., Nakagawa, R., Horikawa, K.: Mutagenic/carcinogenicagents in indoor pollutants; the dinitropyrenes generated by keroseneheaters and fuel gas and liquefied petroleum gas burners. Mutat. Res.157(1), 39–47 (1985)

42. Turesky, R.J., Freeman, J., Holland, R.D., Nestorick, D.M., Miller,D.W., Ratnasinghe, D.L., Kadlubar, F.F.: Identification ofaminobiphenyl derivatives in commercial hair dyes. Chem. Res.Toxicol. 16(9), 1162–1173 (2003)

43. Sinha, R., Rothman, N., Salmon, C., Knize, M.G., Brown, E.D.,Swanson, C.A., Rhodes, D., Rossi, S., Felton, J.S., Levander, O.A.:Heterocyclic amine content in beef cooked by different methods tovarying degrees of doneness and gravy made from meat drippings. FoodChem. Toxicol. 36(4), 279–287 (1998)

44. Sugimura, T., Wakabayashi, K., Nakagama, H., Nagao, M.: Heterocy-clic amines: mutagens/carcinogens produced during cooking of meatand fish. Cancer Sci. 95(4), 290–299 (2004)

45. Gangl, E.T., Turesky, R.J., Vouros, P.: Determination of in vitro-and in vivo-formed DNA adducts of 2-amino-3-methylimidazo[4,5-f]quinoline by capillary liquid chromatography/microelectrospray massspectrometry. Chem. Res. Toxicol. 12(10), 1019–1027 (1999)

46. Fathi, F., Tabrizchi, M., Farrokhpour, H.: Two-dimensionaloptogalvanic spectroscopy. J. Quant. Spectrosc. Radiat. Transfer.113(3), 226–231 (2012)

47. Campbell, J., Le Blanc, J., Schneider, B.: Probing electrospray ionizationdynamics using differential mobility spectrometry: the curious case of 4-aminobenzoic acid. Anal. Chem. 84(18), 7857–7864 (2012)

48. Kendler, S., Lambertus, G.R., Dunietz, B.D., Coy, S.L., Nazarov, E.G.,Miller, R.A., Sacks, R.D.: Fragmentation pathways and mechanisms ofaromatic compounds in atmospheric pressure studied by GC-DMS andDMS-MS. Int. J. Mass Spectrom. 263(2/3), 137–147 (2007)

49. NIST Chemistry Webbook. 2012, National Institute of Standards andTechnology (NIST). Available at: http://webbook.nist.gov/chemistry/Accessed 20 June 2013

50. Castleman, A.W., Keesee, R.G.: Clusters - properties and formation.Annu. Rev. Phys. Chem. 37, 525–550 (1986)

51. deGouw, J.A., Krishnamurthy, M., Leone, S.R.: The mobilities of ionsand cluster ions drifting in polar gases. J. Chem. Phys. 106(14), 5937–5942 (1997)

52. Kim, H., Johnson, P., Beegle, L., Beauchamp, J., Kanik, I.: Electrosprayionization ion mobility spectrometry of carboxylate anions: Ionmobilities and a mass-mobility correlation. J. Phys. Chem. A 109(35),7888–7895 (2005)

53. Viehland, L., Goeringer, D.: Moment theory of ion motion in traps andsimilar devices: I. General theories. J. Phys. B 38(22), 3987–4009 (2005)

54. Langevin, P.: A fundamental formula of kinetic theory. Annale Chim.Physique 5, 245–288 (1905)

55. Su, T.: Kinetic-energy dependence of ion-polar molecule collision rateconstants by trajectory calculations. J. Chem. Phys. 82(4), 2164–2166 (1985)

56. Su, T., Chesnavich, W.J.: Parametrization of the ion-polar moleculecollision rate-constant by trajectory calculations. J. Chem. Phys. 76(10),5183–5185 (1982)

57. Su, T., Viggiano, A.A., Paulson, J.F.: The effect of the dipole-induceddipole potential on ion polar molecule collision rate constants. J. Chem.Phys. 96(7), 5550–5551 (1992)

58. Su, T., Su, E.C.F., Bowers,M.T.: Ion-polar molecule collisions—conservationof angular-momentum in average dipole orientation theory—AADOtheory. J. Chem. Phys. 69(5), 2243–2250 (1978)

59. Barker, R.A., Ridge, D.: Ion-polar neutral momentum-transfer collisionfrequencies - theoretical approach. J. Chem. Phys. 64(11), 4411–4416(1976)

60. Chesnavich, W.J., Su, T., Bowers, M.T.: Collisions in a non-centralfield - variational and trajectory investigation of ion-dipole capture. J.Chem. Phys. 72(4), 2641–2655 (1980)

61. Lias, S.G., Ausloos, P.: Comments on entropy change in ion-moleculeequilibria. J. Am. Chem. Soc. 99(14), 4831–4833 (1977)

62. Rice, S.A.: An overview of the dynamics of intramolecular transfer ofvibrational energy. Advances in Chem. Phys.: Photoselective Chem.Part 1, 47, 117–200 (1981)

63. Gruebele, M., Bigwood, R.: Molecular vibrational energy flow: beyondthe Golden Rule. Int. Rev. Phys. Chem. 17(2), 91–145 (1998)

64. Gruebele, M., Wolynes, P.G.: Vibrational energy flow and chemicalreactions. Acc. Chem. Res. 37(4), 261–267 (2004)

65. Goeringer, D., Viehland, L.: Moment theory of ion motion in traps andsimilar devices: III. Two-temperature treatment of quadrupole ion traps.J. Phys. B 38(22), 4027–4044 (2005)

66. Viehland, L., Danailov, D., Goeringer, D.: Moment theory of ion-neutral reactions in traps and similar devices. J. Phys. Chem. A 111(15),2820–2829 (2007)

67. Wong, B., Lacina, D., Nielsen, I., Graetz, J., Allendorf, M.: Thermochem-istry of alane complexes for hydrogen storage: a theoretical andexperimental investigation. J. Phys. Chem. C 115(15), 7778–7786 (2011)

68. Beegle, L., Kanik, I., Matz, L., Hill, H.: Effects of drift-gaspolarizability on glycine peptides in ion mobility spectrometry. Int. J.Mass Spectrom. 216(3), 257–268 (2002)

69. Lifshitz, C.: Energy entropy trade-offs in the unimolecular decomposi-tions of SF6. J. Phys. Chem 87(18), 3474–3479 (1983)

A. Kafle et al: Understanding Modifier-DMS-MS Quantitation