Research Article

Received: 21 November 2013 Revised: 14 January 2014 Accepted: 21 January 2014 Published online in Wiley Online Library

Rapid Commun. Mass Spectrom. 2014, 28, 886–892

886

Elimination of isobaric interference and signal-to-noise ratioenhancement using on-line mobile phase filtration in liquidchromatography/tandem mass spectrometry

Mathieu Lahaie, Nikolay Youhnovski, Milton Furtado and Fabio Garofolo*Algorithme Pharma Inc., 575 Armand-Frappier, Laval, Quebec, Canada, H7V 4B3

RATIONALE: Liquid chromatography/tandem mass spectrometry (LC/MS/MS) instruments are selective and sensitivebut can still be affected by isobaric interference or chemical noise arising from multiple sources such as the mobile phase.In this study, a high-performance liquid chromatography (HPLC) on-line mobile phase filtration setup is described andused to remove interference to allow better detection of the analyte of interest.METHODS: For instance, a filtration device containing a chemical sorbent is installed at the HPLC outlet of the aqueoussolvent pump A or the organic solvent pump B. This manuscript reports different case scenarios under reversed-phaseand HILIC separations either in positive (ESI(+)) or negative electrospray ionization (ESI(–)) mode using selected reactionmonitoring (SRM) scans as well as additional Q1 MS scans.RESULTS: The filtration of the aqueous effluent of the mobile phase using a porous graphitic carbon filter eliminated theisobaric interferences and improved the detectability of gestodene and perindopril-D4. Also, a strong cation-exchangeguard column installed at the acetonitrile outlet pump was found helpful on reducing the baseline intensity andimproving significantly the signal-to-noise ratio (S/N) of methenamine. Moreover, the on-line mobile phase filtrationwas efficient at removing chemical background ions in full scan mode.CONCLUSIONS: This strategy demonstrated its usefulness by removing co-eluting isobaric interference, and reducingchemical background ions from themobile phase,while drastically improving S/N. Copyright © 2014 JohnWiley& Sons, Ltd.

(wileyonlinelibrary.com) DOI: 10.1002/rcm.6851

Liquid chromatography/tandem mass spectrometry(LC/MS/MS) is a widely used technology in qualitative andquantitative analysis due to its high sensitivity and selectivity.However, it is not uncommon to observe isobaric interferencesarising from the matrix affecting the selectivity of an assay.Moreover, contaminant(s) originating from the mobile phasecan also interfere with the analyte of interest. This type ofinterference may occur when gradient elution chromatographyis used and the contaminants accumulate at the head of theanalytical column during the initial conditions, and elute laterwith the gradient. Indeed, troubleshooting approaches to resolvethis issue have previously been reported or suggested.[1–4]

Impurities from the mobile phase may also affect thesensitivity of an analytical assay by increasing the chemicalnoise and affecting the analyte signal-to-noise ratio (S/N).For instance, chemical noise could result from cluster ions ofmobile phase constituents, clusters originating from thesolvation of contaminants and individual contaminants.[5,6]

Several approaches have been reported to reduce chemicalnoise such as selective ion-molecule reactions,[7–9] supplementalinfrared activation,[10] and ion mobility.[11] More intuitively,high-resolution mass spectrometer technology such as time-of-flight[12,13] or Orbitrap™[14] could be well suited for the samepurpose due to the high level of selectivity of these instruments.

* Correspondence to: F. Garofolo, Algorithme Pharma Inc., 575Armand-Frappier, Laval, Quebec, Canada, H7V 4B3.E-mail: fgarofolo@algopharm.com

Rapid Commun. Mass Spectrom. 2014, 28, 886–892

However, in an unequivocal way, impurities could also beremoved directly from the contaminated solution. In-houseoff-line (bench top) mobile phase cleanup has already beenshown to be effective by using an extraction disk as a purifyingprocedure.[15] The purification procedure is performed byfiltering the mobile phase through an activated-carbon orpoly(styrene–divinylbenzene (PS-DVB) cartridge, using avacuum manifold, and the purified solution is collected into aclean flask. Moreover, mobile phase purification has alsobeen demonstrated by installing a guard column to an high-performance liquid chromatography (HPLC) high-pressuremixing system between the outlet of pump A and the mixingchamber.[16,17] These approaches allow the trapping ofcontaminants and prevent their elution from the analyticalcolumn during gradient chromatographic conditions usingLC-UV applications. This approach and other ones used toclean upmobile phases have also been well reported by Snyderand Dolan in 2007.[2]

During the course of the development of different LC/MS/MSbioanalytical assays, itwas observed thatmobile phase impuritiescan strongly impact selectivity and/or sensitivity of the analyteor its internal standard. The easiest application to overcomethese issues during LC/MS/MS analysis is to use differentquality of solvents, ionization process (atmospheric pressurechemical ionization (APCI), electrospray ionization (ESI)),polarities (positive, negative) or different mass transitions.However, these approaches cannot be used successfully in allcases. An alternative procedure based on trapping contaminantsand referred as a HPLC on-line mobile phase filtration is

Copyright © 2014 John Wiley & Sons, Ltd.

Elimination of interference using on-line filtration in LC/MS/MS

presented for LC/MS/MS applications. Examples of effectivecleanup of different commonly used solvents such as waterand acetonitrile are demonstrated to enable better selectivityand sensitivity.

EXPERIMENTAL

Chemicals and materials

Gestodene was supplied by Jinglong PharmaTech Inc.(Nanjing, Jiangsu, China). Gestodene-D6 was provided byToronto Research Chemicals Inc. (Toronto, ON, Canada).Perindopril-D4 was supplied by Synfine Research Inc.(Richmond Hill, ON, Canada). Methenamine was purchasedfrom United States Pharmacopeia (Rockville, MD, USA).The chemical structures of all analytes are presented in Fig. 1.Methanol (MeOH) OmniSolv® and acetonitrile (ACN)OmniSolv® from EMD Chemicals Inc. were supplied byVWR International (Ville Mont-Royal, QC, Canada). Glacialacetic acid (CH3COOH), formic acid (88% HCOOH),ammonium formate (HCOONH4), ammomium acetate(CH3COONH4) and water HPLC grade (submicron filtered)were provided by Fisher Scientific (Ottawa, ON, Canada).Acetic acid (≥99.99%) was supplied by Sigma-Aldrich (St.Louis,MO,USA). Type 1water (deionizedH2O)was dispensedin house from a Millipore water distribution system.

Solution preparation

Gestodene and gestodene-D6 stock solutions were bothprepared in MeOH at 10 μg/mL. Stock solutions were thendiluted to prepare a single solution at 600 pg/mL ofgestodene and 5 ng/mL of gestodene-D6 inACN/1%HCOOHin H2O (20:80, v/v) solution.Perindopril-D4 stock solution was prepared inMeOH/H2O

(50:50, v/v) at 100 μg/mL. The stock solution was dilutedto prepare a solution at 1 ng/mL of perindopril-D4 inMeOH/H2O (15:85, v/v) solution.Methenamine stock solution was prepared in MeOH/H2O

(25:75, v/v) at 10 mg/mL. The stock solution was diluted toprepare a solution at 200 ng/mL of methenamine inACN/20 mM ammonium formate pH 3 (75:25, v/v) solution.

Figure 1. Chemical structures and molecular weights of theanalytes.

Copyright © 2014 JoRapid Commun. Mass Spectrom. 2014, 28, 886–892

Chromatographic conditions

All separationswere achieved using a 1100 Series HPLC system(Agilent Technologies, Santa Clara, CA, USA) consisting of a1100 binary HPLC pump, autosampler, and thermostatedcolumn compartment.

Gestodene

A Zorbax SB-C18 column (2.1 × 50 mm, 3.5 μm; AgilentTechnologies) was set at 35 °C. The mobile phase consistedof 1% HCOOH in H2O (channel A) and ACN (channel B) ata constant flow rate of 600 μL/min with the followinggradient: (0.0–2.0 min: 20% to 45%B, linear, 2.0–3.5 min: 45% B,3.5–3.6 min: 45% to 95% B, linear, 3.6 to 4.5 min: 95%B, 4.5to 4.51 min: 95% to 20%B, linear, 4.51 to 6.5 min: 20% B). AHypercarb guard column (2.1 × 10 mm, 5 μm; Thermo FisherScientific Inc., Waltham, MA, USA) was used for the on-linemobile phase filtration.

Perindopril-D4

Gradient chromatography was done on a Zorbax SB-C18column (2.1 × 50 mm, 5 μm; Agilent Technologies) set at70 °C with mobile phase consisting of 0.01% CH3COOH inH2O (channel A) and MeOH (channel B) at a constant flowrate of 500 μL/min with the following gradient: (0.0–0.2 min:13% to 70% B linear, 0.2–2.0 min: 70% B, 2.0–2.01 min: 70%to 13%B, linear, 2.01–3.50 min: 13% B). A Hypercarb guardcolumn (4 × 10 mm, 7 μm) was used for the on-line mobilephase filtration.

Methenamine

Separation was achieved on an Atlantis® HILIC silica column(2.1 × 50 mm, 3 μm; Waters Corp., Milford, MA, USA) set at35 °C. The mobile phase consisted of 20 mM ammoniumformate in H2O pH 3 (25% channel A) and ACN (75% channelB) at a flow rate of 800 μL/min. A Zorbax SCX guard column(4.6 × 12.5 mm, 5 μm; Agilent Technologies) was used for theon-line mobile phase filtration.

Q1 MS scan

The mobile phase consisted of 5 mM ammonium acetate inH2O (50% channel A) andMeOH (50% channel B) at a flow rateof 800 μL/min. A Hypercarb guard column (2.1×10 mm,5 μm) or a XBridge™ Shield RP18 (4.6 × 20 mm, 5 μm) (bothWaters, Milford, MA, USA) was used for the on-line mobilephase filtration.

88

HPLC on-line mobile phase filtration setup

TheHPLCon-linemobile phasefiltration setup is schematicallyrepresented in Fig. 2. The filters used are commercially packedguard columns. This filtration device includes two purifyingmechanisms where the frits eliminate particulates and thechemical sorbent removes dissolved contaminants from themobile phase. The filter is installed to the HPLC high-pressuremixing system either between the outlet of pump A or pump Band the mixing chamber.

wileyonlinelibrary.com/journal/rcmhn Wiley & Sons, Ltd.

7

Figure 2. HPLC on-line mobile phase filtration setup.

1.0 2.0 3.0 4.0Time, min

0.0

2000.0

4000.0

6000.0

8000.0

1.0e4

1.2e4

1.4e4

1.6e4

1.0 2.0 3.0 4.00.0

5000.0

1.0e4

1.5e4

2.0e4

2.5e4

3.0e4

3.5e4

2500

3000

3440

Inte

nsity

, cps

Inte

nsity

, cps

A

B

C

Gestodene

Interference

Interference

Gestodene-D6

Time, min

3.69

3.66

3.52

3.55

M. Lahaie et al.

888

Detection conditions

Detection was performed on an API 5000TM triple quadrupolemass spectrometer (AB SCIEX, Concord, ON, CAN) equippedwith a Turbo V™ Ionspray source.The detection conditionswere operated in positive ionization

mode for gestodene and gestodene-D6 (IS). The SRMtransitions of m/z 311.2→ 109.1, 311.2→ 91.0, 311.2→ 81.0,311.2→ 79.0, 311.2→ 77.0 for gestodene and m/z 317.2→ 114.2for gestodene-D6 were monitored. Also the SRM transitionsused for methenamine were m/z 141.1→ 112.1, 141.1→ 98.1,141.1→ 85.1, 141.1→ 71.1 and 141.1→ 58.0. In a different way,the TurboIonspray® was operated in negative ionizationmode for the detection of perindopril-D4 with SRM transitionof m/z 371.3→ 169.0.Finally, the Q1 MS acquisitions were performed in positive

ionization from 50–1250m/z by way of ten MCA scans.

1.0 2.0 3.0 4.00

500

1000

1500

2000

Inte

nsity

, cps

Time, min

RESULTS AND DISCUSSION

With the goal of eliminating interferences originating fromthe mobile phase, an HPLC on-line filtration approach wasapplied for different circumstances. Each case scenario whichis related to improve selectivity and/or sensitivity is discussed.However, it should be noted that this study is not intended toidentify impurities interfering with the analytes.

Figure 3. Representative chromatogram when using theinitial HPLC conditions (standard HPLC setup) of (A)gestodene 600 pg/mL solution monitoring SRM m/z311.2> 109.1, (B) gestodene-D6 5 ng/mL solution, and (C)blank gradient monitoring SRM m/z 311.2→ 109.1.

Gestodene

During the quantification of gestodene, an isobaric interferencewas observed in the chromatogram. The interference retentiontime was observed at 3.55 min at SRM m/z 311.2→ 109.1,which eluted close to the retention time of gestodene of3.69 min (Fig. 3(A)). The interference affected only gestodeneas it was not observed at the SRM transition of gestodene-D6(Fig. 3(B)). A blank gradient analysis (0 μL injection volume)demonstrated that the interfering peak was present, thereforesuggesting that the interference originated from the mobilephase and not the sample (Fig. 3(C)).Unfortunately, none ofthe SRM transitions for gestodene was found selective sincethe interference had similar product ions as gestodene (Fig. 4)thus requiring chromatographic separation or elimination.

wileyonlinelibrary.com/journal/rcm Copyright © 2014 John Wi

An investigation by modifying the mobile phase contentrevealed similar results via lowering the acid content from1% to 0.1% HCOOH or substituting to CH3COOH or usingtype 1 H2O instead of HPLC-grade H2O. No improvementwas observed when the organic solvent was changed tomethanol. To eliminate the source of contamination, the acidicmobile phase was prepared directly into the HPLC-gradesupplier bottle in order to avoid the use of any other labware.Unfortunately, these precautions did not eliminate theinterference.

ley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2014, 28, 886–892

1 2 3 4Time, min

0.00

2000.00

4000.00

6000.00

8000.00

1.00e4

1.20e4

3.71

3.66

3.53

1 2 3 4Time, min

0.00

2000.00

4000.00

6000.00

8000.00

1.00e4

1.20e4

3.69

3.513.83

1 2 3 4Time, min

0.00

2000.00

4000.00

6000.00

8000.00

1.00e4

1.19e4 3.68

3.53

1 2 3 4Time, min

0.00

2000.00

4000.00

6000.00

8000.00

1.00e4

1.20e43.68

3.53

3.79

A

C

B

D

Inte

nsity

, cps

Inte

nsity

, cps

Inte

nsity

, cps

Inte

nsity

, cps

Figure 4. Representative chromatogram of 600 pg/mL gestodene solution using the initial HPLCconditions (standard HPLC setup) monitoring SRM m/z (A) 311.2→ 91.0, (B) 311.2→ 81.0, (C)311.2→ 79.0, and (D) 311.2→ 77.0.

1 2 3 4

Time, min

0.0

2000.0

4000.0

6000.0

8000.0

1.0e4

1.2e4

1.4e4

1.6e4

1.8e4

1 2 3 4

Time, min

0.0

2000.0

4000.0

6000.0

8000.0

1.0e4

1.2e4

1.4e4

1.6e4

1.8e43.65

A

B

Inte

nsity

, cps

Inte

nsity

, cps

Gestodene

Figure 5. Representative chromatogram when using the HPLCon-linemobile phasefiltration setup for pumpA (100%aqueous)and monitoring SRM m/z 311.2→109.1 for (A) gestodene600 pg/mL pure solution and (B) blank run injection #250.

Elimination of interference using on-line filtration in LC/MS/MS

88

Since the water supply was the probable cause ofcontamination, a different strategy was then assessed toeliminate the interference. As indicated in the HPLC on-linemobile phase filtration setup schematic, a guard columnwas incorporated at the outlet of the HPLC pump A(100% aqueous) prior to mixing with pump B (100% organic),which allowed the mobile phase to be filtered and avoidingthe interference from flowing to the analytical column. In thiscase, a Hypercarb porous graphitic carbon guard column waschosen to be used for this purpose as this type of phase isrecognized to be retentive to a wide range of compoundpolarities.[18–20] The molecular structure of the Hypercarbphase consists of flat sheets of hexagonally arranged carbonatoms with fully satisfied valence and is stable across thepH range 0–14.[20] Moreover, the phase should not collapsewhen used under 100% aqueous conditions. These attributesmake it an ideal choice to purify aqueous mobile phase.The guard columnwas pre-wetted with 20 column volumes

of MeOH before installation to allow better conditioning andperformance under 100% aqueous solvent.This on-line mobile phase filtration was able to eliminate

the interference eluting near the retention time of gestodene.Furthermore, the filtering guard column was not found tohave any impact on the chromatographic peak shape orretention time of gestodene (Fig. 5(A)).This approach was evaluated by performingmultiple injections

with the same gradient configuration. No interference wasobserved during method development for at least 250 injectionsandno increase in guard columnpressurewas observed (Fig. 5(B)).

Perindopril-D4

Another case scenario was also observed during thedevelopment of perindopril in human plasma using ESInegative ionization. In this case, an interference was found

Copyright © 2014 JoRapid Commun. Mass Spectrom. 2014, 28, 886–892

to elute in the retention time zone of the stable-isotope-labeled internal standard perindopril-D4 (1.91 min) andaffecting its response (Fig. 6(A)). A blank gradient (0 μL

wileyonlinelibrary.com/journal/rcmhn Wiley & Sons, Ltd.

9

1 2 3Time, min

0

1000

2000

3000

4000

5000

6000

7000

8000

1.91

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Time, min

0.00

2000.00

4000.00

6000.00

8000.00

1.00e4

1.18e41.90

A

B

Perindopril-D4Interference

Interference

Perindopril-D4

Inte

nsity

,cps

Inte

nsity

,cps

Figure 6. Representative chromatogram of perindopril-D41 ng/mL pure solution monitoring SRM m/z 371.3→ 169.0(A) using standard HPLC setup and (B) using the HPLCon-linemobile phase filtration setup for pumpA, 100% aqueous.

0.5 1.0 1.5 2.0 2.5

Time, min

0.0

1.0e5

2.0e5

3.0e5

4.0e5

5.0e5

6.0e5

6.7e51.26

S/N = 18 Without Filter

S/N = 450 With Filter

Inte

nsity

, cps

Figure 7. Representative chromatograms (overlay) ofmethenamine 200 ng/mL pure solution monitoring SRMm/z 141→ 112with andwithout using theHPLC on-linemobilephase filtration setup for pump B (100% ACN).

M. Lahaie et al.

890

injection volume) test revealed that the interferenceoriginated from the mobile phase as for another precedingcase study. The use of different concentration and quality ofacetic acid (≥99.99%) or water did not eliminate theinterference affecting the peak integration of perindopril-D4.The solution was resolved by inserting a Hypercarb guard

column to the HPLC outlet of pump A (100% aqueous). Thiswas found to be an effective solution in eliminating theinterference and allowing selective detection of perindorpil-D4with improved signal-to-noise ratio (S/N) (Fig. 6(B)). Thisdecontamination process was incorporated into the finalanalyticalmethod andwas validated in a regulated environmentfor the quantification of perindopril in human plasma.

Methenamine

A different scenario was investigated using isocratic HILICHPLC conditions where the interference originated fromthe organic portion of the mobile phase. The interferencecreated high chemical noise affecting the S/N ofmethenamine. Under the methenamine SRM transitions, anelevated baseline was observed between 2.5 × 104 countsper second (cps) and 2.3 × 105 cps. The high baselines didnot allow practical detection of methenamine. Conversely,no intense background was observed for the stable-isotope-labeled internal standard methenamine-D12 (data notshown).With the goal of trapping the interference, a Zorbax SCX

(strong cation-exchange) guard column was installed at theoutlet of the HPLC channel pump B (100% ACN). This

wileyonlinelibrary.com/journal/rcm Copyright © 2014 John Wi

sorbent was chosen as it is likely to be more effective to retaincontaminants over reversed-phase packing under 100% ACNflow condition. Throughout a continuous ACN filtration, thisapproach significantly reduced the baseline intensity andincreased S/N for methenamine. As such, the baseline wasreduced by a factor of 100-fold allowing a gain of 25 timesS/N for methenamine (Fig. 7).

The mobile phase filtration has shown to have no impacton peak shapes and retention times for methenamine.However, the high level of contamination of the mobilephase allowed a gain in S/N to be effective for only20 injections.

Q1 MS scan

The usefulness of the HPLC on-line mobile phase filtrationstrategy was further evaluated to demonstrate its efficacy atremoving chemical noise across the full scan m/z range ofthe mass spectrometer. Q1 MS acquisitions from 50 to1250m/z under ESI(+) was done for this assessment.Furthermore, two different packed guard columns wherechosen as filter in regard to their high degree of retentionand stability under 100% aqueous conditions. A Hypercarbguard column or a reversed-phase XBridge™ Shield RP18(with embedded polar group) guard column was installedat the outlet of the channel pump A (100% aqueous).

The on-line filter when applied on the aqueous channelsignificantly reduced the MS ion background across thescanned range between m/z 50 and 700 (Fig. 8). However,with the XBridge™ Shield RP18 guard column an interferenceat m/z 185.6 was observed indicating that contaminant(s)could leach from the filter guard column (Fig. 8(C)).Nevertheless, the mass spectra suggest that porous graphiticcarbon or reversed-phase with embedded polar group guardcolumn can effectively remove contaminants by filteringthe aqueous flow portion of the mobile phase. Moreover, itdemonstrates the flexibility of the approach by usingalternative chemical sorbent phases to efficiently removeinterference.

As a suggestion in conjunction with these presentedanalytical situations, the mobile phase can be pre-filteredbefore use (off-line). However, the HPLC on-line mobilephase filtration offers a better advantage in eliminating

ley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2014, 28, 886–892

100 200 300 400 500 600 700

m/z, Da

0.0

1.0e7

2.0e7

3.0e7

4.0e7

4.9e7

100 200 300 400 500 600 700

m/z, Da

0.0

1.0e7

2.0e7

3.0e7

4.0e7

4.7e7

Inte

nsity

, cps

100 200 300 400 500 600 7000.0

1.0e7

2.0e7

3.0e7

4.0e7

4.7e7

Inte

nsity

, cps

Inte

nsity

, cps

m/z, Da

185.6

Figure 8. Representativemass spectra (ESI(+)) ofMeOH/5mMammonium acetate in H2O (zoom m/z 50–700): (A) withoutfilter, (B) with Hypercarb filter, and (C) with XBridge ShieldRP18 filter for pump A, 100% aqueous.

Elimination of interference using on-line filtration in LC/MS/MS

89

contamination originating from labware or HPLCinstrument components. Moreover, when necessary bothHPLC effluents (aqueous and organic) may perhaps bedecontaminated simultaneously using proper chemicalsorbents for each channel. Each method is a case by casescenario and the scientist is required to determine thecapacity of the filtration device in order to avoidcontaminant breakthrough.

Copyright © 2014 JoRapid Commun. Mass Spectrom. 2014, 28, 886–892

On-line mobile phase filtration was shown effective toreduce chemical background and to remove chromatographicinterferences thus facilitating the detection and quantificationof compounds. The on-line mobile phase filtration can also beused for a wide range of analytical applications such aspharmaceutical and environmental analysis.

In addition, it will be convenient to equip the HPLC systemwith a flow meter, which will measure the volume of mobilephase delivered by each separated pump channel. Thisfeature might be useful to measure filter lifetime and to setproper maintenance cycle.

CONCLUSIONS

TheHPLC on-linemobile phase filtration approach is adaptableby selecting the appropriate chemical sorbent to removecontaminants from either the aqueous or organic portion ofthe mobile phase. It was found to be helpful when reversed-phase or HILIC was used as HPLC separation mode.

This strategy, as demonstrated, could lead towardsimprovement of analyte detection when used with both ESIpositive or negative ionization. Furthermore, it was confirmedto be efficient to eliminate co-eluting isobaric interferenceduring gradient elution. Moreover, it is effective to enhancesignal-to-noise ratio by reducing chemical noise. This simple,cost-effective and versatile concept has the potential to beapplied to multiple liquid chromatography/massspectrometry domains.

AcknowledgementsThe authors would like to thank Jean-Nicholas Mess andAnnik Bergeron at Bioanalytical Services Department atAlgorithme Pharma Inc. for their help and for offeringvaluable advice during the review of this manuscript.

REFERENCES

[1] S. Williams. Ghost peaks in reversed-phase gradient HPLC:a review and update. J. Chromatogr. A 2004, 1052, 1.

[2] L.R. Snyder, J.W. Dolan. High-Performance Gradient Elution:The Practical Application of the Linear-Solvent-Strength Model,John Wiley, New York, 2007, pp. 182–185, 200–203, 217–220.

[3] H. M. Herath, P. N. Shaw, P. Cabot, A. K. Hewavitharana.Effect of ionization suppression by trace impurities inmobile phase water on the accuracy of quantification byhigh-performance liquid chromatography/mass spectrometry.Rapid Commun. Mass Spectrom. 2010, 24, 1502

[4] S. Sadikin, D. Dee Zhang, R. Inloes, S. Redkar, J. W. Dolan.Ghost peak investigation in a reversed-phase gradientLC system. LCGC North America 2011, 29, 394. Available:http://www.chromatographyonline.com/lcgc/Column%3A+LC+Troubleshooting/Ghost-Peak-Investigation-in-a-Reversed-Phase-Gradi/ArticleStandard/Article/detail/725248.

[5] X. Guo, A. P. Bruins, T. R. Covey. Characterization of typicalchemical background interferences in atmospheric pressureionization liquid chromatography-mass spectrometry. RapidCommun. Mass Spectrom. 2006, 20, 3145.

[6] M. Trötzmüller, X. Guo, A. Fauland, H. Köfeler, E. Lankmayr.Characteristics and origins of common chemical noise ionsin negative ESI LC-MS. J. Mass Spectrom. 2011, 46, 553.

wileyonlinelibrary.com/journal/rcmhn Wiley & Sons, Ltd.

1

M. Lahaie et al.

892

[7] M. J. Jarvis, G. K. Koyanagi, X. Zhao, T. R. Covey, D. K. Bohme.Scrubbing ionswithmolecules: kinetic studies of chemical noisereduction in mass spectrometry using ion-molecule reactionswith dimethyl disulfide. Anal. Chem. 2007, 79, 4006.

[8] X. Guo, A. P. Bruins, T. R. Covey. Method to reducechemical background interference in atmospheric pressureionization liquid chromatography-mass spectrometry usingexclusive reactions with the chemical reagent dimethyldisulfide. Anal. Chem. 2007, 79, 4013.

[9] X. Guo, A. P. Bruins, T. R. Covey, M. Trötzmüller, E. Lankmayr.Alternative reagents for chemical noise reduction in liquidchromatography-mass spectrometry using selective ion-molecule reactions. J. Am. Soc. Mass Spectrom. 2009, 20, 105.

[10] M. W. Gardner, J. S. Brodbelt. Reduction of chemical noise inelectrospray ionization mass spectrometry by supplementalIR activation. J. Am. Soc. Mass Spectrom. 2009, 20, 2206.

[11] A. B. Kanu, P. Dwivedi, M. Tam, L. Matz, H. H. Hill Jr. Ionmobility-mass spectrometry. J. Mass Spectrom. 2008, 43, 1.

[12] J.-N. Mess, L.-P. Morin, S. Ramagiri, M. Aiello, J. Cardenas,M. Furtado, F. Garofolo. Advanced application of a highresolution quadrupole time-of-flight mass spectrometer toresolve chemical noise in regulated bioanalysis. Proc. 59th

ASMS Conference on Mass Spectrometry and Allied Topics,Denver, CO, USA, 2011.

[13] L.-P. Morin, J.-N. Mess, S. Ramagiri, M. Aiello, J. Cardenas,M. Furtado, F. Garofolo. Elimination of selectivity and

wileyonlinelibrary.com/journal/rcm Copyright © 2014 John Wi

quantification issues during ursodiol regulated bioanalysisby using a high resolution quadrupole time-of-flight massspectrometer. Proc. 59th ASMS Conference on MassSpectrometry and Allied Topics, Denver, CO, USA, 2011.

[14] C. Wei, J. E. Grace Jr, T. A. Zvyaga, D. M. Drexler. Utility ofhigh-resolution accurate MS to eliminate interferences in thebioanalysis of ribavirin and its phosphate metabolites.Bioanalysis 2012, 4, 1895.

[15] M. C. Ringo, J. R. Allen, D. M. Mattocks. Mobile-phase cleanupusing solid-phase extraction disks. LCGC North America 2003,21, 168. Available: http://www.chromatographyonline.com/lcgc/data/articlestandard//lcgc/092003/47802/article.pdf.

[16] P.-L. Zhu, L. R. Snyder, J. W. Dolan. Improved baselines ingradient elution. J. Chromatogr. A 1995, 718, 429.

[17] J. W. Dolan, J. R. Kern, T. Culley. Water problems – A casestudy. LCGC 1996, 14, 202. Available: http://www.lcresources.com/resources/TSWiz/pdf/14031996.pdf.

[18] U. D. Neue. HPLC Columns: Theory, Technology, and Practice.Wiley-VCH, Weinheim, 1997, pp. 206–207.

[19] L. R. Snyder, J. J. Kirkland, J. L. Glajch. Practical HPLCMethod Development, (2nd edn.). John Wiley, New York,1997, pp.184–185.

[20] Thermo Scientific Hypercarb Columns, Thermo FisherScientific Inc., Waltham, MA, USA. Applications Notebook,Issue 1, June 2009. Available: http://www.thermo.com/eThermo/CMA/PDFs/Product/productPDF_51906.pdf.

ley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2014, 28, 886–892