Method Development ofMixed-Mode Solid Phase Extractionfor Forensics Applications

Author

Mike ChangAgilent Technologies, Inc.

Application Note

Forensics & Toxicology

Abstract

Mixed-mode solid phase extraction has the versatility and power to extract targetcompounds from biological matrixes and is one of the most widely used samplepreparation methods in forensic laboratories. This application note is an overviewof sample preparation method development in its entirety, prior to LC/MS/MS orGC/MS, using Agilent Bond Elut Certify mixed-mode solid phase extraction toshorten the method development process or to improve an existing method.

Introduction Many workplace screening tests of drugs of abuse are done by mass spectrometrywith gas chromatography or liquid chromatography. Sometimes, scientists inforensic laboratories encounter the need to change existing standard operatingprocedures (SOPs) when there are requirements for lower detection limits,improved linearity in the lower or upper concentration range, calibration rangeextension, limited sample volume, or too-frequent replacement of the analyticalcolumn. When new solid phase extraction (SPE) method development is underconsideration, there is no straightforward shortcut to the new method. In this studywith phencyclidine (PCP) in urine, the complete workflow for solid phase extractionis reviewed, and the decision making process to the final method is highlighted.

2

Materials and MethodsAcetonitrile, methanol, and formic acid were LC/MS grade.Dichloromethane was LC grade. Ethyl acetate was GCresidue analysis grade, and hexane, NH4OH, and KH2PO4were reagent grade. Water was Milli-Q filtered or LC/MSgrade. Acetic acid was premium quality. PCP and PCP-D5analytes were purchased from Sigma-Aldrich, Corp. The QCsample was Liquichek urine toxicology control, level C2, fromBio-Rad Laboratories. The PCP concentration was 19 ng/mL.

Sample preparation was accomplished using Agilent BondElut Certify, 130 mg, 3 mL, 50/pk (p/n 12102051).

Instrument conditionsInstrument parameters of LC/MS/MS and GC/MS withpulsed splitless and pulsed split injections are discussed ingreater detail elsewhere [1].

Principles of solid phase extraction A wide range of solid phase extraction sorbents exist,including reversed-phase, cation-exchange, anion-exchange,or mixed-mode. The principle of SPE is to let the sample,with compounds of interest or interferences, or both, bind toa solid phase sorbent (usually chemically modified silica orpolymeric material packed in cartridges or well plates),followed by rinsing off interferences to waste, and finallyeluting and collecting the compounds of interest. All thesesteps take place selectively and so interferences andcompounds of interest will separate from sample matrixsuccessfully.

The first step in SPE method development is to understandthe characteristics of the target compound. PCP is relativelybasic and hydrophobic, with pKa = 8.29 and log P = 4.69.Ideally, using cation-exchange interaction and hydrophobicinteraction would be the optimal solid phase extractionchoice. Agilent Bond Elut Certify mixed-mode solid phaseextraction has a good balance of both hydrophobic andcation-exchange characteristics and is well suited for PCPanalysis.

The conventional solid phase extraction workflow consists ofconditioning, equilibration, sample loading, washing, elution,evaporation, and reconstitution, as shown in Figure 1. Someof the steps may be eliminated to reduce process time.

In this procedure, the solid phase extraction sorbent retainsthe compounds of interest until the elution step, while otherinterferences are removed in the wash step.

Condition

Equilibrate

Sample load

Wash

Elute

Evaporate and reconstitute

������������������������������������������

3

GC/MS

To waste

Collect

Evaporate

Reconstitute

ConditionEquilibrate

Sample loadRinse

Elute

LC/MS/MS

Figure 2. Schematic of forensic sample analysis using solidphase extraction.

From Figure 2, the first three steps (condition, equilibrate,and sample load) typically do not require much variation tooptimize the performance of solid phase extraction.Sometimes the conditioning and equilibration steps can beeliminated for some polymeric solid phase extractionsorbents. The most effective optimization can be done in thewashing and elution steps. During sample loading,compounds of interest bind to the sorbent material alongwith interferences. The purpose of the washing step is toremove these interferences. The elution step then recoversthe target compound by interrupting the interaction betweenit and the solid phase extraction sorbent. The maximumremoval of interference while maintaining optimal recovery ofthe target compound is the key to successful methoddevelopment in solid phase extraction. To verify methodoptimization, every step from sample loading to elution needsto be collected and analyzed by chromatography.

Results and Discussion After conditioning (2 mL MeOH) and equilibration (2 mL of100 mM KH2PO4), spiked urine sample (1 mL) was loadedinto a pipette, as shown in Figure 2. During the loading step,any eluate was collected in a vial. Washing was split intothree aliquots of 1 mL each. The first wash was 1 mL 5 %acetic acid and the others were 1 mL MeOH. Each eluatewas collected separately in a vial. Elution was done thesame way, for example, 3 x 1 mL volumes, with ACN + 2 %NH4OH. Each eluate was collected individually in a vial. Thewash 2 vial in Figure 3 indicates that the majority ofinterference was removed by MeOH during the washing step.All collected eluates were injected into an LC/MS/MS andall chromatograms are shown in Figure 4.

Figure 3. Collection eluates from every step in the solidphase extraction method (from left to right, eluate collectionsof sample loading, wash 1, wash 2, wash 3, eluate 1,eluate 2, and eluate 3).

4

From the information in Figure 4, it was clear that the targetcompound PCP (RT = 1.35 minutes) was not coming throughthe sorbent in the sample loading step. The chromatogramsfrom washes 1 to 3 also confirmed that PCP was not washedoff during the washing steps. The elution step should havePCP and, as the chromatograms show from eluates 1 to 3,

Figure 4. Chromatograms of PCP by LC/MS/MS for all eluates collected during solid phase extraction.

target compounds were coming off of the sorbent.The wash 3 vial was visually clear and the chromatogramfrom wash 3 did not have the target compound peak; hence,wash 3 can be skipped. Eluate 3 did not have a significantamount of PCP, which indicated that one aliquot of 2 mL ofACN + 2 % NH4OH, by combining eluate 2 and 3, wassufficient for a single elution step.

0123

×104

0123

×104

0123

×104

0123

×104

0123

×104

0123

×104

00.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

1.353548.19

1.353411.12

1.352398.87

0.466973.10

1.35419,937.44

1.35265,903.91

1.4 1.5Acquisition time (min)

Sample loading

Wash 1

Wash 2

Wash 3

Eluate 1

Eluate 2

Eluate 3

Coun

tsCo

unts

Coun

tsCo

unts

Coun

tsCo

unts

Coun

ts

1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0 3.1 3.2 3.3 3.4

123

×104

5

Different elution solvents can be tested for further elutionstep optimization. A widely accepted elution solvent mixturein forensic applications is DCM:IPA:NH4OH at 78:20:2. Inaddition, a typical elution solvent for LC/MS/MSapplications is ACN or MeOH. When three different elutionsolvents were tested, different peak area counts and full-scan background information was obtained. Along withchromatograms, visual observation is also important andFigure 5 shows the effect of different elution solvents.

A) Agilent Bond Elut Certify cartridges after elution B) Elutions by three different solvents

C) After evaporation D) Reconstituted in initial mobile phase

Figure 5. Elution solvent variation study (from left to right, elutions made by ACN + 2 % NH4OH, MeOH + 2 % NH4OH, andDCM:IPA:NH4OH at 78:20:2).

6

Figure 6. LC/MS/MS chromatograms of PCP (A) and PCP-d5 (B) using three different elution solvents (top to bottom: ACN +2 % NH4OH, DCM:IPA:NH4OH at 78:20:2, and MeOH + 2 % NH4OH).

0

0.4

0.8

1.2

1.6

2.0

2.4 A

0.1

2.33894,618.21

2.33755,106.25

2.33742,700.06

Acquisition time (min)

Elution byACN + NH

4OH

Elution byMeOH + NH

4OH

Elution byDCM:IPA:NH

4OH at 78:20:2

Coun

ts (×

104 )

0

0.4

0.8

1.2

1.6

2.0

2.4

Coun

ts (×

104 )

0

0.4

0.8

1.2

1.6

2.0

2.4

Coun

ts (×

104 )

0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9

00.20.40.60.81.01.21.41.6 B

0.1

2.3296,684,059.94

2.3303,892,128.87

2.3303,108,574.38

Acquisition time (min)

Elution byACN + NH

4OH

Elution byMeOH + NH

4OH

Elution byDCM:IPA:NH

4OH at 78:20:2

Coun

ts (×

106 )

00.20.40.60.81.01.21.41.6

Coun

ts (×

106 )

00.20.40.60.81.01.21.41.6

Coun

ts (×

106 )

0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9

Elution by MeOH + 2 % NH4OH had the dirty eluate, whereasACN +2 % NH4OH and DCM:IPA:NH4OH at 78:20:2 hadvisually clean extracts. Chromatographically supporting dataconfirmed that MeOH-based elution was not appropriate,with lower recovery (Figure 6) and more LC/MS backgroundnoise (Figure 7).

Based on the data shown in Figures 5 to 8, ACN-basedelution performed the best, minimizing interference andmaximizing the peak area count of the target compound.

7

Figure 7. Full-scan LC/MS/MS chromatograms of blank urine prepared by Agilent Bond Elut Certify with different elution solvents.

00.51.01.52.02.53.03.54.04.5

0.1

*0.5041,534,294,985.23

*0.512210,074,000.99

*0.7433,406,397,970.28

Acquisition time (min)

Elution byACN + NH

4OH

Elution byMeOH + NH

4OH

Elution byDCM:IPA:NH

4OH at 78:20:2

Coun

ts (×

107 )

00.51.01.52.02.53.03.54.04.5

Coun

ts (×

107 )

00.51.01.52.02.53.03.54.04.5

Coun

ts (×

107 )

0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9

Figure 8. Full-scan GC/MS chromatograms of blank urine prepared by Agilent Bond Elut Certify with different elution solvents.

400450500550600650700750800850900950

1,000

Abu

ndan

ce

1,0501,1001,1501,2001,2501,3001,3501,4001,4501,5001,5501,6001,650

3501.20 1.40 1.60 1.80 2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80

Time4.00 4.20 4.40 4.60 4.80 5.00 5.20 5.40 5.60 5.80 6.00 6.20 6.40

8

From Figure 9, it was evident that for smaller injectionvolumes such as 2 µL, the percentage of organic solvent hadminimal impact on the peak shape and peak area. For largerinjection volumes, for example, 5 to 10 µL, having samplesolvent composition close to the initial mobile phaseimproved the peak shape and sometimes the peak areacount as well. These experimental data can differ, dependingon the LC/MS/MS conditions and the compounds ofinterest.

Final experimental conditions and results with human urinesamples, including sample preparation, and instrumentparameters, have been published [1].

Injection volumes for LC/MS/MS can vary from sub-µL to 20 µL. With UHPLC systems, injection volumes tend to besmaller. However, if large-volume injection is required forbetter detection limits, a sample solvent and injection volumestudy can be done. The general rule is to have initial mobilephase as the sample solvent. Typically, the higher the organicsolvent composition in the sample solvent, the higher theLC/MS/MS response, due to improved desolvation in theionization source. Different injection volumes and samplesolvent compositions were tested, as shown in Figure 9.

10 µL injection

0

0.4

0.8×105

0.1 0.3

1.382108,379.15

1.382204,253.20

1.381190,936.01

0.32315,623.43

1.38255,837.75

1.38267,790.22

1.388189,846.57

Acquisition time (min)

100 % ACN

75 % ACN

50 % ACN

30 % ACN

5 % ACN

5 % ACN + 0.1 % FA

Counts

0

0.4

0.8×105

Counts

0

0.4

0.8×105

Counts

0

0.4

0.8×105

Counts

0

0.4

0.8×105

Counts

0

0.4

0.8×105

Counts

0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.0 3.3

9

Figure 9. Injection volume and sample solvent effect in LC/MS/MS (injection volume is shown at the top and sample solventcomposition is shown in every chromatogram).

5 µL injection

2 µL injection

0

2

4

6

0.1 0.3

1.37995,266.81

1.37774,976.80

1.38194,965.26

1.381100,266.20

Acquisition time (min)

100 % ACN

75 % ACN

50 % ACN

30 % ACN

Coun

ts (×

104 )

0

2

4

6

Coun

ts (×

104 )

0

2

4

6

Coun

ts (×

104 )

0

2

4

6

Coun

ts (×

104 )

0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.0 3.3

00.51.01.52.22.5

0.1 0.3

1.37737,239.17

1.37728,980.10

1.37936,110.30

1.37939,062.03

Acquisition time (min)

100 % ACN

75 % ACN

50 % ACN

30 % ACN

Coun

ts (×

104 )

00.51.01.52.22.5

Coun

ts (×

104 )

00.51.01.52.22.5

Coun

ts (×

104 )

00.51.01.52.22.5

Coun

ts (×

104 )

0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.0 3.3

www.agilent.com/chemFor Forensic Use.

Information is subject to change without notice.

© Agilent Technologies, Inc., 2014, 2017Published in the USAJune 26, 20175991-4695EN

ConclusionsThe sample preparation process using mixed-mode solidphase extraction was illustrated with real samples andchromatographic data, producing optimized conditions forimproved quality of data for forensic applications. A goodbalance between the removal of interference and retention ofthe target compounds, followed by elution optimization forcleaner extracts, is essential for successful methoddevelopment in solid phase extraction.

Reference1. Mike Chang. Analysis of Phencyclidine in Urine to U.S.

SAMHSA Guidelines with LC/MS/MS and GC/MS.Application note, Agilent Technologies, Inc. Publicationnumber 5991-4575EN (2014).

For More InformationFor more information on our products and services visit ourWebsite at www.agilent.com/chem.