The FIRST and ONLY Sub-2 Core-Shell Columns on the Market © 2011 Phenomenex, Inc

The FIRST and ONLY Sub-2 core-shell columns on the market © 2011 Phenomenex, Inc. All rightsreserved. WE’LL MAKEANAPPFORYOU. HASN’T BEENCREATEDYET?JUST ASKUS. OVER 10,000APPLICATIONSBY: EASILY ACCESS WWW.PHENOMENEX.COM/APPS FIND YOURAPPNOW! "OENPSF t 5FDIOJRVF t 3FHVMBUPSZ.FUIPE t $PNQPVOE$MBTT t $PNQPVOE/BNF t

PA86990511_L SUPPLEMENT TO

June 2011 www.chromatographyonline.com

THE APPLICATION NOTEBOOK Tetra Images/Getty Images

e-learning for the analytical chemistry community Learn the fundamentals at your own pace Troubleshoot problems before they multiply

Join at www.CHROMacademy.com LCGC’s e-learning community for analytical chemists

VIEW SAMPLES & ENROLL ONLINE TODAY! $295/ e200 - 1 year full access Corporate and academic discounts available

Featuring a regularly refreshed syllabus: • Self paced modules with assessments • Laboratory tools and application notes • LCGC magazine articles • Web seminars, newsfeeds and more • CHROMmunity for collaboration

A service of www.CHROMacademy.com The intersection of great science and practical learning Our technology couldn’t be more state-of-the-art. Our business philosophy remains stuck in the 1950s.

As the only independent, family-owned light scattering company in the industry, we have no shareholders, investors or 3rd parties to please. We’ve been in business for nearly three decades, building the ﬁ nest, most versatile light scattering instruments available anywhere on earth. We’re passionate about your success and make sure our instruments are tools you can rely on every day. And we go to great lengths (and expense) to make sure you’re delighted. Take, for example, our Light Scattering University, where we bring you to Santa Barbara, CA for three days of intensive, hands-on training (all expenses paid*). While you’re here, you meet the people who design and build your instruments, learn about the software from the people who write it, and interact with the people who will be answering your questions. We don’t just build instruments, we build relationships that last for years. To learn more about how you can become a member of the Wyatt instrument family of customers, visit www.wyatt.com. We think you’ll appreciate our old-fashioned approach. Because whether you’re buying a light scattering instrument or a bag of groceries, it’s nice to be treated like you’re the most important person in the world. CORPORATION

DAWN HELEOS®. The most advanced Optilab T-rEX®. The refractometer with ViscoStar™. The viscometer with Eclipse. The ultimate system for the DynaPro Plate Reader. Automated, non-invasive multi-angle light scattering instrument for the greatest sensitivity and range. unparalleled signal-to-noise, stable separation of macromolecules and dynamic light scattering for proteins and nano- macromolecular characterization. baselines and a 21st-century interface. nanoparticles in solution. particles in 96 or 384 or 1536 well plates.

*Round trip airfare for US and Canadian customers only. ©2011 Wyatt Technology. DAWN, Optilab, DynaPro and the Wyatt Technology logo are registered trademarks, and ViscoStar and Eclipse are trademarks of Wyatt Technology Corporation. 4 Table of Contents THE APPLICATION NOTEBOOK – JUNE 2011

THE APPLICATION NOTEBOOK

Biological

10 Extraction of Sub-Nanogram Levels of Catecholamines from Human Plasma Using EVOLUTE® WCX Victor E. Vandell, Biotage, LLC

11 Novel C18 Stationary Phase Provides Exceptional Results for Metabolite Profiling Bhavana Verma and Itaru Yazawa, Imtakt USA Oscar Yanes, Ralf Tautenhahn, Gary J. Patti, and Gary Siuzdak, The Scripps Research Institute

12 BioPharma Compass: A Fully Automated Solution for Characterization and QC of Intact and Digested Proteins Christian Albers, Laura Main, Carsten Bäßmann, and Wolfgang Jabs, Bruker Daltonik GmbH

14 Analyzing Testosterone in Human Serum by UHPLC Using High Efficiency Kinetex® 1.7μm C18 Core-Shell Columns Seyed Sadjadi and Jeff Layne, Phenomenex, Inc.

15 Self-Aggregation of Gelatin Wyatt Technology Corporation

16 An Improved SPE-LC–MS–MS Platform for the Simultaneous Quantification of Multiple Amyloid β Peptides in Cerebrospinal Fluid for Preclinical or Biomarker Discovery Erin E. Chambers and Diane M. Diehl, Waters Corporation Mary E. Lame, Pfizer, Neuroscience Research Unit

18 Hydroxyethylstarches (HES) Wyatt Technology Corporation T H E APPLICATION NOTEBOOK – JUNE 2011 Table of Contents 5

Chiral

19 Improving SFC Chiral Separations by Employing a New Chlorinated Polysaccharide Chiral Stationary Phase: The RegisPack® CLA-1 Ted Szczerba, Regis Technologies, Inc.

Environmental

20 Determination of Explosives Using Fast and High Resolution Liquid Chromatography with the Agilent 1290 Infinity LC System Detlef Wilhelm, AnaTox GmbH & Co. KG Edgar Nägele and Udo Huber, Agilent Technologies R&D and Marketing GmbH & Co. KG

21 Direct Determination of Endothall in Water Samples by IC–MS Leo (Jinyuan) Wang and William C. Schnute, Dionex Corporation

22 Separation of Quaternary Ammonium Compounds Using a Bonded Polymeric Zwitterionic Stationary Phase Patrik Appelblad, Merck SeQuant AB

23 Automated One Step Solid Phase Extraction and Concentration of PCBs in Water Phil Bassignani, FMS, Inc.

24 Advantages of Multi-Pesticide Screening by GC–MS Kory Kelly, Phenomenex Inc.

25 Reducing Cycle Time for Analysis of 1,4-Dioxane Using an Automated Purge and Trap Sample Prep System Nathan Valentine, Teledyne Tekmar

26 Method 525.2 Update: Analyte Recoveries with New CCL3 Compounds Xiaoyan Wang, Michael Telepchak, Jeffery Hackett, and Don Shelly, UCT, LLC Food and Beverage

27 Screening for Pesticide Residues in Food and Identification with Highest Confidence Using High Resolution and Accurate Mass LC–MS-MS André Schreiber and David Cox, AB Sciex 6 Table of Contents THE APPLICATION NOTEBOOK – JUNE 2011

28 Simultaneous Quantitation of 2- and 4-Methylimidazole in Carbonated Drinks Leo (Jinyuan) Wang, Xiaodong Liu, Christopher Pohl, and William Schnute, Dionex Corporation

29 Fast Determination of Anthocyanins to Authenticate Pomegranate Juice Pranathi R. Perati, Brian De Borba, and Jeffrey S. Rohrer, Dionex Corporation

30 Fast Determination of Catechins in Tea Pranathi R. Perati, Brian De Borba, and Jeffrey S. Rohrer, Dionex Corporation

31 Sensitive Determination of Hydroxymethyl Furfural in Honey, Syrup, and Fructose Solution Lipika Basumallick, Deanna Hurum, and Jeff Rohrer, Dionex Corporation

32 Cleanup of Baby Food Samples Using Gel Permeation Chromatography (GPC) Elizabeth Badgett, Laura Chambers, and Gary Engelhart, OI Analytical

33 Analysis of Pesticides in Citrus Oil Using PTV Backflush with GC–MS-MS Triple Quadrupole for High Sample Throughput Charles Lyle, Eric Phillips, and Hans-Joachim Huebschmann, Thermo Fisher Scientific Inc.

34 Glyphosate Analysis in Soy Beans, Corn, and Sunflower Seeds by HPLC with Post-Column Derivatization and Fluorescence Detection Maria Ofitserova, Rebecca Smith, and Michael Pickering, Pickering Laboratories, Inc.

36 Simultaneous and Direct Analysis of Biogenic Amines in Food by LC–MS-MS Using Hydrophilic Chromatography Seiji Ito and Fumiya Nakata, Tosoh Corporation Industrial

38 Simultaneous Determination of Mineral Acids, Fluoride, and Silicate in Etching Baths by Ion Chromatography with Dual Detection German Bogenschütz, Thomas Kolb, Beni Galliker, Andrea Wille, and Alfred Steinbach, Metrohm International Headquarters

41 Determination of Phthalate Esters in Child Care Products and Children’s Toys by Gas Chromatography–Mass Spectrometry (GC–MS) Richard Whitney, Shimadzu Scientific Instruments T H E APPLICATION NOTEBOOK – JUNE 2011 Table of Contents 7

Pharmaceutical

42 Analysis of Pharmaceuticals in Whole Blood by Poroshell 120, Using a Modified Mini-QuEChERS Approach for Sample Preparation Joan Stevens, Agilent Technologies, Inc.

44 LC Analysis of Aminoglycoside Antibiotics Kanamycin and Amikacin Lipika Basumallick, Deanna Hurum, and Jeff Rohrer, Dionex Corporation

45 Enantiomeric Separation of Proton Pump Inhibitors Using Polysaccharide-Based Chiral Stationary Phases in Reversed-Phase HPLC Conditions Liming Peng, Michael McCoy, Jeff Layne, and Kari Carlson, Phenomenex, Inc. Polymer

46 Chemical Composition Analysis of Polyolefins by Multiple Detection GPC-IR5 Wallace W. Yau, Polymer Char Scientific Consultant Alberto Ortín and Pilar del Hierro, Polymer Char

General

48 Isolation of Benzylideneacetophenone from a Crude Reaction Mixture A. Talamona, BUCHI Corporation

50 Characterization of an Unknown Cannabinomimetic Compound in an Herbal “Incense” Sample by Gas Chromatography–High Resolution Time-of-Flight Mass Spectrometry Joe Binkley, LECO Corporation

51 Rapid Analysis of Amphetamines in Biological Samples Michael Rummel, Matthew Trass, and Jeff Layne, Phenomenex, Inc.

52 Improved Analysis of Preservatives in Cosmetics Using a Unique C18 Core-Shell Phase Terrell Matthews, Zeshan Aqeel, and Jeff Layne, Phenomenex, Inc.

53 Rapid and Streamlined Screening of Barbiturates Matthew Trass, Seyed Sadjadi, Jeff Layne, Sky Countryman, Michael Rummel, and Erica Pike, Phenomenex, Inc.

54 Resistive Glass Inlet Tubes Increase Ion Throughput Paula Holmes, Ph.D., and Bruce N. Laprade, Photonis USA

55 Selection of Optical Fiber for Chromatographic Detectors and Remote Sensing Applications Joe Macomber and Rick Timmerman, Polymicro Technologies 8 Table of Contents THE APPLICATION NOTEBOOK – JUNE 2011

56 Advantages of Shodex™ NH2P Series, Polymer-Based Amino HILIC Column, Over Silica-Based Amino HILIC Columns Kanna Ito, Shodex/Showa Denko America, Inc.

57 Fast and Accurate LC–MS Analysis of Vitamin D Metabolites Using Ascentis® Express F5 HPLC Columns Craig R. Aurand and David S. Bell, Supelco/Sigma-Aldrich

Articles

58 Aggregated Singletons for Automated Puri cation W o r k o w Bhagyashree A. Khunte and Laurence Philippe

63 A New Path to High-Resolution HPLC–TOF-MS — Survey, Targeted, and Trace Analysis Applications of TOF-MS in the Analysis of Complex Biochemical Matrices Je rey S. Patrick, Kevin Siek, Joe Binkley, Viatcheslav Artaev, and Michael Mason

70 25-Hydroxyvitamin D2/D3 Analysis in Human Plasma Using LC–MS Phil Koerner and Michael McGinley Departments

74 Call for Application Notes

Cover Photography: Getty Images THE APPLICATION NOTEBOOK – JUNE 2011 9

MANUSCRIPTS: For manuscript preparation guidelines, see LCGC 28(7), 560 Administration, Sales, and Editorial Offices (2010), or call The Editor, (732) 596-0276. LCGC welcomes unsolicited articles, Woodbridge Corporate Plaza, 485F US Highway 1 South, manuscripts, photographs, illustrations, and other materials but cannot be held Suite 100, Iselin, NJ 08830 responsible for their safekeeping or return. Every precaution is taken to ensure accuracy, but LCGC cannot accept responsibility for the accuracy of information tel. (732) 596-0276 fax (732) 596-0003 supplied herein or for any opinion expressed. Science Group Publisher Michael J. Tessalone [email protected] SUBSCRIPTIONS: For subscription and circulation information: LCGC, P.O. Box Associate Publisher Edward Fantuzzi 6168, Duluth, MN 55806-6168, or call (888) 527-7008 (7:00 a.m.–6:00 p.m. central [email protected] time). International customers should call +1-218-740-6477. Delivery of LCGC East Coast Sales Manager Stephanie Shaffer outside the United States is 14 days after printing. For single and back issues, [email protected] call (800) 598-6008 or (218) 740-6480. (LCGC Europe and LCGC Asia Pacific are available free of charge to users and specifiers of chromatographic equipment in Editorial Western Europe and Asia and Australia, respectively.) Editorial Director Laura Bush [email protected] CHANGE OF ADDRESS: Send change of address to LCGC, P.O. Box 6168, Duluth, Managing Editor Megan Evans MN 55806-6168; alternately, send change via e-mail to [email protected] [email protected] or go to the following URLs: Group Technical Editor Stephen A. Brown • Print: https://advanstar.replycentral.com/Default.aspx?PID=469 [email protected] • Digital: https://advanstar.replycentral.com/?PID=469&V=DIGI Associate Editor Cindy Delonas Allow four to six weeks for change. PUBLICATIONS MAIL AGREEMENT No. [email protected] 40612608. Return all undeliverable Canadian addresses to: Pitney Bowes, P.O. Box Art Director Dan Ward 25542, London, ON, N6C 6B2, CANADA. Canadian GST number: R-124213133RT001. [email protected]

DIRECT MAIL LIST RENTAL: Contact Tamara Phillips, tel. (440) 891-2773, e-mail Marketing [email protected]. Marketing Manager Anne Young [email protected] REPRINTS: Reprints of all articles in this issue and past issues of this publication are available (500 minimum). Call (800) 290-5460, ext. 100, or e-mail Market Development [email protected]. Classified/Recruitment Sales Representative Tod McCloskey [email protected] MARKETING DEVELOPMENT/CLASSIFIED: Contact Tod McCloskey, tel. (440) 891- Direct List Rental Sales Tamara Phillips 2739, fax (440) 826-2865. [email protected] Reprints [email protected] RECRUITMENT: Contact Tod McCloskey, tel. (440) 891-2739, fax (440) 826-2865. (800) 290-5460, x100

INTERNATIONAL LICENSING: Contact Maureen Cannon, tel. (440) 891-2742, fax Production and Audience Development Offices (440) 891-2650, or e-mail [email protected]. 131 West First Street, Duluth, MN 55802-2065 ©2011 Advanstar Communications Inc. All rights reserved. No part of this publication tel. (218) 740-7036 fax (218) 740-7223 may be reproduced or transmitted in any form or by any means, electronic or mechani- Production Manager David Erickson cal including by photocopy, recording, or information storage and retrieval without [email protected] permission in writing from the publisher. Authorization to photocopy items for internal/ Audience Development Manager Peggy Olson educational or personal use, or the internal/educational or personal use of specific clients is granted by Advanstar Communications Inc. for libraries and other users registered [email protected] with the Copyright Clearance Center, 222 Rosewood Dr. Danvers, MA 01923, (978) 750- Assistant Audience Development Manager Gail Mantay 8400, fax (978) 646-8700. For uses beyond those listed above, please direct your written [email protected] request to Permission Dept. fax (440) 891-2650 or email: [email protected].

LCGC and the logo appearing on the cover of this magazine are registered trademarks of Advanstar Communications Inc. Advanstar Communications provides certain customer contact data (such as customers’ names, addresses, telephone numbers, and e-mail addresses) to third parties who wish to promote relevant products, services, and other opportunities that may be of interest to you. If you do not want Advanstar Communica- Chief Executive Officer Joseph Loggia tions to make your contact information available to third parties for marketing purposes, Executive Vice-President, Finance & Chief Financial Officer Theodore S. Alpert call (866) 529-2922 between the hours of 7:30 a.m. and 5:00 p.m. central time and fol- Executive Vice-President, Exhibitions Tony Calanca low the instructions to remove your name from Advanstar’s lists. Outside the United Executive Vice-President, Licensing, Market Development & Europe Georgiann DeCenzo States, please call (866) 529-2922. LCGC does not verify any claims or other information Executive Vice-President, Fashion & President, MAGIC International Chris DeMoulin appearing in any of the advertisements contained in the publication and cannot take Executive Vice-President, Chief Administrative Officer Thomas Ehardt any responsibility for any losses or other damages incurred by readers in reliance on such content. Executive Vice-President, Corporate Development Eric I. Lisman Executive Vice-President, Powersports, Dental & Veterinary Daniel Phillips Executive Vice-President, Fashion & President, PROJECT Andrew Pollard Advanstar Communications Inc. (www.advanstar.com) is a leading worldwide media Executive Vice-President, Chief Marketing Officer Steve Sturm company providing integrated marketing solutions for the Fashion, Life Sciences and Executive Vice-President, Pharmaceutical/Science & CBI Ron Wall Powersports industries. Advanstar serves business professionals and consumers in Vice-President, Media Operations Francis Heid these industries with its portfolio of 91 events, 67 publications and directories, 150 Vice-President, Electronic Information Technology J. Vaughn electronic publications and Web sites, as well as educational and direct marketing products and services. Market leading brands and a commitment to delivering inno- Vice-President, Electronic Media Group Mike Alic vative, quality products and services enables Advanstar to “Connect Our Customers Vice-President, Human Resources Nancy Nugent With Theirs.” Advanstar has approximately 1000 employees and currently Vice-President, General Counsel Ward D. Hewins operates from multiple offices in North America and Europe. Vice-President, General Manager David C. Esola

All color separations, proofs, and film produced by Advanstar’s Scanning and Digital Prepress Departments. 10 Biological THE APPLICATION NOTEBOOK – JUNE 2011

Extraction of Sub-Nanogram Levels of Catecholamines from Human Plasma Using EVOLUTE® WCX

Victor E. Vandell, Biotage, LLC

A new, efﬁcient, effective, and sensitive sample preparation application for the extraction of the catechol- Max. 3100.0 cps amines epinephrine, norepinephrine, and dopamine 5954 Epinephrine (100pg/ml) 5500 from human plasma at signiﬁcantly low levels using Norepinephrine (1000 pg/ml) EVOLUTE WCX, a resin-based mixed-mode sorbent 5000 with an optimized combination of nonpolar (hydro- 4500 phobic) and weak cation exchange interactions. 4000 3500

atecholamines are standard biomarkers for the detection 3000 Intensity, cps Intensity, 2500 Dopamine Cof diseases such as hypertention, pheochromocytoma, and (200pg/ml) neuroblastoma. Detection of these analytes in biological matrices 2000 (serum, plasma, and urine) at sub-nanogram per milliliter levels is 1500 critical in the clinical diagnosis of these and other diseases. 1000 500

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Extraction Conditions Time, min Tis application note outlines the procedure using the EVOLUTE Figure 1: Extracted ion chromatogram for sub-nanogram levels WCX 50 mg 3 mL column (part #612-0005-B) optimized for a of epinephrine (100 pg/mL), dopamine (200 pg/mL), and norepi- 500 μL plasma sample volume, but other formats are available. nephrine (1000 pg/mL) extracted from spiked plasma with EVO- Method parameters and dilution factors have been optimized to LUTE WCX. The peaks denoted by * are plasma artifact peaks observed at the LLOD. maximize recoveries and minimize ion suppression. Sample pretreatment: Dilute 500 μL of plasma 1:1 (v/v) with tion monitoring mode (MRM); norepinephrine 170.2>152.2, 50 mM ammonium acetate, pH 7. Spike epinephrine 184.1>166.1, and dopamine 154.3>137.3. sample with internal standard. Column conditioning: Condition column with methanol (1 mL) Results Column equilibration: Rinse column with 50 mM ammonium Sample recoveries of over 85% with RSDs below 10% were achieved acetate, pH 7 (1 mL). at pg/mL levels for epinephrine and dopamine and 1 ng/mL for nor- Sample loading: Load pretreated plasma sample (1mL) at a flow epinephrine. Figure 1 shows the extracted ion chromatogram. rate of 1–2 mL/min. Interference elution 1: Remove polar and ionic interferences with Conclusions methanol:water (10:90) (2 mL). Tis method demonstrates a robust and effective extraction of a range of Interference elution 2: Remove nonpolar and phospholipid catecholamines at significantly low levels from the challenging biological interferences with isopropanol (2 mL). matrix of human plasma. Some artifact peaks from the plasma were ob- Analyte elution: Elute catecholamines with methanol containing served in the extracted ion chromatogram but these had no effect on the formic acid (95:5, v/v) (1 mL). ions of interest, resulting in a clean and interference free chromatogram Post extraction: Evaporate to dryness and reconstitute sample in mobile phase (200 uL). References Te extracted samples were run on an Agilent 1200 Liquid (1) Victor E. Vandell, Extraction of Sub-Nanogram Levels of Catechol- Handling System with a Restek Ultra PFP Propyl 3-μm analytical amines from Human Plasma Using EVOLUTE® WCX. Application column (50 × 2.1 mm id) at ambient temperature. Te mobile Note AN 739 (available from www.biotage.com ). phase was isocratic, 0.1% formic acid aq/methanol (98/2, v/v) at a flow rate of 0.200 mL/min for 5 min. Te detector was an Biotage, LLC Applied Biosystems/MDS Sciex 4000 Q-Trap triple quadrupole 10430 Harris Oaks Blvd., Suite C, Charlotte, NC 28269 mass spectrometer equipped with a Turbo Ionspray® interface for Website: www.biotage.com mass analysis. Negative ions were acquired in the multiple reac- Email: [email protected] THE APPLICATION NOTEBOOK – JUNE 2011 Biological 11

Novel C18 Stationary Phase Provides Exceptional Results for Metabolite Proﬁling

Bhavana Verma1, Itaru Yazawa1, Oscar Yanes2, Ralf Tautenhahn2, Gary J. Patti2, and Gary Siuzdak2, 1Imtakt USA and 2The Scripps Research Institute

etabolomics, the study of small molecule metabolites that Mare found within a biological sample, is an emerging field of study. Progress in this field depends upon technological advancement in the fields of LC–MS and separation technology. In untargeted metabolomics, simultaneous detection of the largest number of metabolites is desired. Reversed phase C18 columns are generally used for this type to study, however they provide limited retention of polar compounds. In this paper, we compare the retention of 31 model compounds characterized by diflerent polarities using a Cogent Bidentate C18, a Waters XBridge C18, and an Imtakt Scherzo SM-C18 column. ffe Imtakt Scherzo column showed the best retention of polar compounds, with only 10% of the polar compounds eluting within the first 2 min as compared to 22% by the Cogent column and 45% by the Waters column.

Experimental All data were collected on an HPLC system coupled to mass spec detection. ffe mobile phase used for the column comparison of the 31 model compounds was A = water, 0.1% formic acid, B = acetonitrile, 0.1% formic acid. ffe linear gradient used started at 100%A (0–5 min) and finished at 100%B (35–40 min) (see Figure 1). Figure 1: 1. alanine 19. s-(p-azidophenacyl)glutathione Discussion and Results 2. N,N-dimethylglycine 20. Taurocholic acid 3. serine 21. O-beta-D-glucuronosyl-naphthol AS-BI ffe study shows that Imtakt’s Scherzo column, consisting of a combi- 4. fumaric acid 22. Leucine enkephalin nation of C18, anion, and cation ligands, outperforms other C18 col- 5. succinic acid 23. Arg-Pro-Pro-Gly-Phe 6. cysteine (Bradykinin fragment 1-5) umns in the retention of polar metabolites. ffis result is important in the 7. oxaloacetic acid 24. Reserpine advancement of untargeted metabolomics, as traditional C18 columns 8. Gly-Gly 25. GSSG 9. malic acid 26. Arg-Lys-Asp-Val-Tyr struggle to retain polar compounds and can limit the detection of a wide 10. alpha-ketoglutarate (Thymopoietin II fragment 32-36) 11. citric acid 27. Coenzyme A polarity range of metabolites. 12. 2-methylhippuric acid 28. Acetyl CoA 13. gamma-D-glutamylglycine 29. Ala-Ser-Thr-Thr-Thr-Asn-Tyr-Thr (peptide T) 14. salbutomal 30. Arg-Pro-ProGly-Phe-Ser-Pro-Phe Reference 15. AMP (bradykinin fragment 1-8) (1) Oscar Yanes, Ralf Tautenhahn, Gary J. Patti, and Gary Siuzdak, Ana- 16. ribo-avin 31. Tyr-Tyr-Tyr-Tyr-Tyr-Tyr 83 17. Phe-Gly-Phe-Gly XBridge C18 (150 x 1.0 mm) lytical Chemistry (6), 2152–2161 (2011). 18. s-(p-nitrobenzyl)glutathione Scherzo SM-C18, 150 x 2 mm A: 0.1% formic acid in water B: 0.1% formic acid in acetonitrile The linear gradient elution used started at 100% A (time 0-5 min) and finished at 100% B (35-40 min) 250 uL/min, room temp. ESI-MS in positive and negative ionization mode

Imtakt USA 6703 Germantown Avenue, Suite 240, Philadelphia, PA 19119 tel. (888) 456-HPLC, (215) 665-8902, fax (501) 646-3497 Email: [email protected], Website: www.imtaktusa.com 12 Biological THE APPLICATION NOTEBOOK – JUNE 2011

BioPharma Compass: A Fully Automated Solution for Characterization and QC of Intact and Digested Proteins

Christian Albers, Laura Main, Carsten Bäßmann, and Wolfgang Jabs, Bruker Daltonik GmbH

BioPharma Compass™ is a fully automated solution efficacy or safety of the protein drug. In this application note, for the rapid characterization of biopharmaceutical we will demonstrate the BioPharma Compass QC workflow (as products such as proteins, peptides, RNA, and DNA. shown in Figure 1) in combination with two industry leading plat- This push button solution assists nonspecialist op- forms, the Dionex UltiMate 3000 UHPLC system and the maXis erators to generate high quality, accurate data for UHR-TOF as samples of intact IgG1 and digested transferring automatic comparison with laboratory reference are analyzed. standards. Automated, visual reports are then generated for each sample and important information Experimental regarding a products purity and identity can be ob- For protein proflling experiments, human IgG1 from Sigma-Al- served at a glance. In this application note, we will drich was used without further puriflcation. A batch of 20 sam- apply the BioPharma Compass workow to the QC ples were prepared; 18 vials contained IgG1 with a concentration characterization of two proteins including intact of 1 µg/µL, 1 vial contained a mixture of IgG1 and IgG4 and 1 IgG1 and digested transferrin. vial only IgG4. 4 µL were injected from each vial.

haracterization of protein therapeutics and comparison with UHPLC Dionex RSLC Creference standards for the analysis of biotherapeutics usu- Guard cartridge: Dionex Acclaim 120, C8, 5 µm, 2 × 10 mm ally includes accurate analysis of the intact protein mass followed Analytical column: Dionex Acclaim 120 C8 column, 3 µm, by complete amino acid sequence coverage, usually obtained by 2.1 × 150 mm digestion of the protein by an enzyme and analysis by mass spec- Column oven: 70 ∘C trometry. These two complementary workﬂows ensure that the Flow rate: 300 µL/min correct product has been produced with the correct modiﬁca- Solvent A: 0.1% FA in H2O tions and detects any unexpected impurities that may affect the Solvent B: 0.1% FA in ACN Run time: 15 min applying a gradient from 5 to 90% B followed by 3 min equilibration with 2% B

Mass spectrometer: Bruker maXis UHR-TOF For peptide mapping experiments, transferrin was reduced, alkylated, and digested using trypsin according to the standard protocols. 48 vials per protein digest were prepared; the peptide concentration was 1 pmol/µL. 5 µL were injected from each vial.

LC–MS instrumentation for peptide mapping experiments: UHPLC Dionex RSLC Analytical column: Dionex Acclaim RS LC120 C18 column, 2.2 µm, 2.1 × 100 mm Column oven: 40 ∘C Flow rate: 300 µL/min Figure 1: Overview of BioPharma Compass, describing the steps Solvent A: 0.1% formic acid in H2O involved in a typical protein QC characterization workow. Solvent B: 0.1% formic acid in ACN/H2O 90:10 These steps are completely automated. After comparison with reference standards and the setting of individual QC criteria, Results reports are automatically generated. The reports shown here display the deconvoluted intact protein mass, protein sequence QC analysis of Intact IgG1 and digested transferrin were per- coverage, BPC annotated with peptide sequences and a rapid formed in a fully automated fashion with BioPharma Compass. QC screening report. As soon as the IgG1 batch acquisition is started, a QC result table THE APPLICATION NOTEBOOK – JUNE 2011 Biological 13

Figure 2: Table view summarizing the QC result of 20 IgG1 intact protein samples. Each row contains the result of one LC–MS Figure 4: Rapid view, QC result of digested Transferrin protein experiment; in the columns are values of different quality con- samples. In this example the mass deviation was chosen as qual- trol criteria. Two samples are highlighted due to the presence of ity criterion. The sample is represented in green if the deviation an unexpected impurity, another sample highlighted in red has is smaller than 3 ppm, yellow when the deviation is in the range failed to meet the QC criteria based on mass accuracy. This table between 3 and 5 ppm and red when it is above 5 ppm. Genera- also provides access to more-detailed pdf reports. tion of the QC table begins as soon as the ﬁrst sample has been acquired.

shown as amber, and samples that successfully meet QC criteria are shown in green. In this instance, the ﬁrst 48 wells correspond to the analysis of digested transferrin.

Conclusion Biopharma Compass provides a fully automated workTow for the characterization of biopharmaceutical products such as proteins, peptides, RNA, and DNA. Sample acquisition, processing, comparison with reference standards, and report generation are achieved in parallel thus dramatically increasing productivity and sample throughput. In combination with the outstanding sensitivity, mass accuracy, and resolution of the maXis UHR-TOF, BioPharma Compass assists nonspecialist operators to rapidly generate high quality data to Figure 3: Part of the automatically generated pdf QC report conﬁrm the identity and purity of biopharmaceutical compounds from an intact IgG1 protein sample. The identity of the sample is displayed in the header. In the second section the total ion to ensure drug safety and eﬂcacy. chromatogram (TIC) and the chromatographic peaks (Com- pound List) are reported. The spectrum, deconvoluted with the Maximum Entropy algorithm, indicates the different glycosyl- ated isoforms of the antibody. The deconvoluted mass peaks are compared with a reference standard and the result of this comparison is reported in the Result List. is generated as shown in Figure 2. Each row in the QC table in Figure 2 represents one IgG sample analysis. In this example the pass/fail criteria was based simply on mass accuracy. However, QC criteria may be adjusted to suit individual analysis requirements. Figure 3 represents an example of the automated report output available with BioPharma Compass. For rapid throughput, samples can also be viewed in a 96 well format (Figure 4). Again, samples that have failed QC criteria Bruker Daltonics are highlighted in red, those requiring further investigation are Billerica, MA tel. (978) 663-3660, fax (978) 667-5993 Email: [email protected] Website: bruker.com/biopharma 14 Biological THE APPLICATION NOTEBOOK – JUNE 2011

Analyzing Testosterone in Human Serum by UHPLC Using High Efﬁciency Kinetex® 1.7µm C18 Core-Shell Columns

Seyed Sadjadi and Jeff Layne*, Phenomenex, Inc.

Testosterone was extracted from human serum by Table I: LC Gradient Program strong anion exchange polymeric SPE and analyzed Step Total Time(min) Flow Rate(µl/min) B (%) using a Kinetex C18, 30 x 2.1 mm, 1.7 µm column and 00 400 10 positive polarity ESI LC–MS-MS system. Kinetex sub- 1 2.5 400 90 2 µm core-shell technology offers higher efﬁciencies 2 3.5 400 90 than traditional sub-2 µm columns, producing greater 3 3.6 400 10 chromatographic resolution, sensitivity, and higher 45 400 10 peak capacities. Table II: MRM Transitions Used for Data Analysis estosterone is an androgenic steroid responsible for the de- Peak Name MRM Channel Tvelopment of male reproductive organs, maintaining (or Testo (1) 304.3 → 124.0 increasing) muscle mass and bone density. As anabolic steroids, Testo (2) 304.3 → 112.0 testosterone has been used (or abused) to increase muscle mass IS (Testo-D3 1) 307.3 → 124.0 and enhance the athletic performance. Te concentration of tes- IS (Testo-D3 2) 307.3 → 112.0 tosterone is lower in the female population than the male and in general diminishes with advancing age. Monitoring body concen- function (1). A short-length 30 mm, 1.7 um Kinetex C18 col- tration of testosterone is an aid in diagnosing and treating disease umn eﬂciently separates testosterone from its isomeric form epit- state related to the hormonal imbalance. estosterone (Figure 1). Te analysis is based on a simple extraction method using strong anion exchange SPE (Strata-X-A) to produce a clean ex- Experimental Conditions tract from human serum. Following the extraction, testosterone Te mobile phase consisted of 0.1% formic acid with 1 mM am- is derivatized to form an oxime which is then analyzed in posi- monium formate with no pH adjustment, in water (MP A) and tive mode ESI LC–MS-MS under multiple-reactions-monitoring acetonitrile (MP B). A typical LC gradient is used (Table I) for the separation. An AB Sciex API 5000 triple-quadrupole tandem mass spectrometer is used for analysis equipped with an ESI probe operating

4.0e4 1 & 2 in positive polarity mode. Under an MRM mode, two channels 3.8e4 3.6e4 were monitored for testosterone and testosterone-D3 (Table II). 3.4e4 3.2e4 3.0e4 2.8e4

Intensity, cps 2.6e4 Results and Discussions 2.4e4 2.2e4 As is demonstrated from the chromatogram, the Kinetex column 2.0e4 1.8e4 provides a high degree of selectivity even in small dimensions to 1.6e4 provide superior chromatographic separation. For further details 1.4e4 1.2e4 or questions, contact your Phenomenex sales representative. 1.0e4 3 8000.0 6000.0 4000.0 References 2000.0 (1) M.M. Kushnir et al, Clinical Chemistry 52:1, 120–128 (2006). 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 min. 49 98 146 194 243 291 340 388 436

Figure 1: The separation of 100 pg/mL standard of testosterone Phenomenex, Inc. and epitestosterone extracted from human serum on Kinetex 411 Madrid Avenue, Torrance, CA 90501 C18, 30 × 2.1 mm, 1.7 um using the LC gradient proﬁle listed in Table I. Testosterone retention time is 2.62 min, epitestosterone tel. (310) 212-0555, fax (310) 328-7768 2.77 min, and int. std. is 2.61 min. Website: www.phenomenex.com THE APPLICATION NOTEBOOK – JUNE 2011 Biological 15

Self-Aggregation of Gelatin

Wyatt Technology Corporation

elatin is the denatured form of collagen, the main compo- Gnent of skin, bone, and connective tissues of animals. Its special properties are a sharp sol-gel transition, low sol viscosity, and high efiectiveness as a protective colloid. For properties of aqueous mixtures of gelatin with other biopolymers, the character of aggregates of α-chains is important: Te entropy of mixing is reduced when gelatin chains are limited in their mobility by aggregation. SEC-MALS was used to study the aggregation behavior of gelatin solutions as a function temperature without using denaturing agents, such as urea or SDS. In addition to the usual concentration detection by refractive index, optical rotation detection was applied to detect any changes in the degree of helicity of the gelatin chains. After soaking overnight at 4 °C, gelatin was heated for 30 min at 80 °C and subsequently kept for 60 min at the temperature of the measurement (50–80 °C). Te eluent was a 0.01-M potas- sium phosphate bufier of pH 6.7 with 0.125 M LiNO3. Te col- Figure 1: Concentration as a function of molar mass. umn set consisted of a precolumn PWH, 6000 PW, and 3000PW (all of TSK). Figures 1 and 2 show a reversible change of the bimodal molar mass distribution of two difierent porcine gelatins at temperatures between 50 °C and 80 °C (number in legend corresponds to temperature), well above the gelatin temperature of about 30 °C. Te efiect of temperature on the molar mass distribution sug- gests the existence of a monomer-dimer equilibrium between gelatin chains (the shoulder corresponding to the monomer was 90–100 kDa, and the one corresponding to the dimer was approximately 200 kDa). Te results explain the difierent behavior of these gelatins in mixtures with other biopolymers. Te ratio of the refractive index and the optical rotation was constant for MM <~400 kDa and is independent of temperature. No involvement of helix formation could be detected.

Figure 2: Another sample’s concentration as a function of molar mass.

Wyatt Technology Corporation 6300 Hollister Avenue, Santa Barbara, CA 93117 tel. (805) 681-9009, fax (805) 681-0123 E-mail: [email protected], Website: www.wyatt.com 16 Biological THE APPLICATION NOTEBOOK – JUNE 2011

An Improved SPE-LC–MS–MS Platform for the Simultaneous Quantiﬁcation of Multiple Amyloid β Peptides in Cerebrospinal Fluid for Preclinical or Biomarker Discovery Erin E. Chambers1, Mary E. Lame2, and Diane M. Diehl1, 1Waters Corporation and 2Pﬁzer, Neuroscience Research Unit

ast, flexible platforms for peptide quantification are need- Experimental Conditions Fed, particularly for a discovery setting. This type of meth- SPE-LC–MS–MS Conditions odology would be especially advantageous in the case of amy- LC system: Waters ACQUITY UPLC System loid beta (aβ) peptides. The deposition/formation of insoluble Column: ACQUITY UPLC BEH C18 300 Å, 2.1 × 150 aggregates, or plaques, of aβ peptides in the brain is consid- mm, 1.7 μm, Peptide Separation Technology ered to be a critical event in the progression of Alzheimer’s disease (AD) and thus has the attention of many researchers. A previous Waters application note (720003682en) described Table I: Sequence, MW, and pI Information for Amyloid β Peptides in detail the development of a fast, flexible SPE-LC–MS–MS Amyloid β 1-38 platform for the quantification of multiple aβ peptides from DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGG human or monkey CSF for use in a biomarker or preclinical MW 4132 pI 5.2 discovery setting. In this work, the mass spectrometry plat- Amyloid β 1-40 form has been updated from the Xevo TQ MS to the Xevo DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV TQ-S mass spectrometry system. This change facilitated both MW 4330 pI 5.2 a 4× reduction in required sample size and a 4–5× increase in Amyloid β 1-42 assay sensitivity. This work focuses on methods for the 1–38, DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA 1–40, and 1–42 aβ, Table I. MW 4516 pI 5.2

Daughters of 1130ES+ 1129.44 4.09e6 100

b40

1078.99 1078.85 b41

% b39 .43

1054.00

b36 b37 b38

b35 1029.22

1200.65 1000.73 b34 1257.35 1333.78 1404.21 b33 942.97 914.61

0 m/z 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700

Figure 1: Representative ESI+ MS-MS spectrum for amyloid β 1-42 with fragment sequence ions labeled. THE APPLICATION NOTEBOOK – JUNE 2011 Biological 17

SPE: Oasis MCX μElution 96-well plate, 50 μL 2, and 6 ng/mL. Accuracy and precision values met the reg- human or animal CSF ulatory criteria for LC–MS–MS assays. Results from QC MS system: Waters Xevo TQ-S, ESI+ sample analysis are shown in Table IV. Average deviation from expected is 2.3%. Results and Conclusions • The method described herein eliminates time-consuming • An improved SPE-UPLC–MS–MS bioanalytical method immunoassays or immunoprecipitation steps for preclinical was developed and validated for the simultaneous quantifi- work. cation of multiple amyloid β peptides in human CSF. • The use of a single UPLC–MS–MS assay represents a • MS was performed in positive ion mode since CID of the significant advantage over an ELISA assay, which would 4+ precursor ion yielded several distinct product ions corre- require multiple assays with multiple antibodies to quantify sponding to inherently specific b sequence ions (representa- each of the relevant peptides. tive spectrum shown in Figure 1). • UPLC separation of the three amyloid β peptides is shown Copyright 2010: ACQUITY UPLC, Oasis, ﬁe Science of What’s Possible, in Figure 2. Xevo are trademark of Waters Corporation. • The increased sensitivity of the Xevo TQ-S triple quadrupole mass spectrometer facilitated the use of 4× less sample × and a 4–5 improvement in quantification limits (Table II). Table II: Comparison of Standard Curve and QC Range Using • Average basal levels and RSD values for all 3 aβ peptides in Xevo TQ and TQ-S MS two sources of human CSF are shown in Table III and are 200 µL sample 50 µL sample lower or equal to 5%. Xevo TQ Xevo TQ-S • Overspiked QC samples were prepared in triplicate in two Standard Curve sources of pooled human CSF at 0.04, 0.075, 0.15, 0.2, 0.8, 0.1 to 10 ng/mL 0.025 or 0.05 to 10 ng/mL Range QC Range 0.2 to 6 ng/mL 0.04 to 6 ng/mL

Amyloid ß 1-40 5.85 4.60e4 Table III: Baseline Levels of Amyloid β Peptides in Two Sources of Pooled Human CSF

5.60 Avg. Basal % RSD of IS % RSD Amyloid ß 1-42 Level Basal Level

Amyloid ß 1-38 6.03 Amyloid β 1-38 1.396 5.3 6.4 Human CSF 1

% Amyloid β 1-38 0.702 1.7 Human CSF 2 Amyloid β 1-40 5.429 3.3 4.7 Human CSF 1 6.87 Amyloid β 1-40 7.65 2.611 2.7 6.31 7.97 4.89 5.30 Human CSF 2

2 Time 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 Amyloid β 1-42 0.458 5.2 6.6 Human CSF 1 Figure 2: Representative UPLC–MS–MS analysis of amyloid β Amyloid β 1-42 1-38, 1-40, and 1-42 peptides extracted from articial CSF + 5% 0.226 1.9 rat plasma. Human CSF 1

Table IV: Average Deviation Values for all Overspike QC Samples QC 0.04 QC 0.075 QC 0.15 QC 0.2 QC QC 2 QC 6 ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL Amyloid β 1-38 Human 2.3 5.8 -3.2 7.3 14.8 5.1 13.1 CSF 1 and 2

Amyloid β 1-40 Human -0.8 -3.2 -1.9 2.5 -2.6 -4.2 -3.8 CSF 1 and 2

Amyloid β 1-42 Human 1.3 13.4 -3.6 5.6 2.0 -0.6 -0.2 CSF 1 and 2

Waters Corporation 34 Maple Street, Milford, MA 01757 tel. (508) 478-2000, fax (508) 478-1990 Website: www.waters.com 18 Biological THE APPLICATION NOTEBOOK – JUNE 2011

Hydroxyethylstarches (HES)

Wyatt Technology Corporation

ydroxyethylstarches (HES) are used increasingly as plasma Hexpanders in medical applications. fie HES’s circulation time in the blood depends strongly on its molar mass distributions. Historically, polysaccharide characterization by gel permeation chromatography (GPC) has been problematic, especially if high molar mass components are present. Because of its superior separation capability, especially on molecules exceeding 50 kDa, Tow fleld-Tow fractionation (Flow-FFF) is an excellent separation alternative. By coupling this technology to a multi angle light scattering (MALS) detector, such as a DAWN or miniDAWN, absolute values can be determined without making any assump- tions. We characterized 0.2% (w/v) HES solutions in doubly-dis- tilled water. fie HES types were 200/0.5, 130/0.42, and 70/0.5 from Serumwerk (Bernburg, Germany). fie FFF system was an Eclipse connected to a DAWN EOS and an RI detector (Shodex 101). Volumes of 100 μL were injected into the 350-μm spacer channel containing a 10 kDa regenerated cellulose membrane. A Figure 1: The fractograms of the three samples with their molar channel Tow of 1 mL/min was kept constant while the cross Tow mass values overlaid. decreased linearly from 2 mL/min to 0 mL/min within 20 min. Data were evaluated using Wyatt’s ASTRA software package. fie aF-FFF/MALS fractograms of the HES types are compared in Figure 1. Corresponding to normal mode Flow-FFF theory, samples with smaller average molar mass elute faster. In Figure 2, the molar mass distributions are given as cumulative weight fraction plots. As is evident in the plots, monomodal distributions were successfully achieved. fie values covered a range from approximately 20 kDa to approximately 600 kDa and up to 2 GDa in size, depending on the characterized HES type. fius, the average molar mass value was mainly inTuenced by high molar mass fractions. fiis was also indicated by higher polydispersity values for HES types with higher molar mass. Asymmetrical FFF/MALS is, therefore, an excellent method for characterizing medical polysaccharides such as HES. fie main advantage of this technique is that molar mass distributions can be determined from absolute values over an extremely wide range of masses. Figure 2: Three different HES samples shown on the cumulative weight fraction plot of ASTRA, indicating the large differences among them.

Wyatt Technology Corporation 6300 Hollister Avenue, Santa Barbara, CA 93117 tel. (805) 681-9009, fax (805) 681-0123 E-mail: [email protected], Website: www.wyatt.com THE APPLICATION NOTEBOOK – JUNE 2011 Chiral 19

Improving SFC Chiral Separations by Employing a New Chlorinated Polysaccharide Chiral Stationary Phase: The RegisPack® CLA-1

Ted Szczerba, Regis Technologies, Inc.

As the number of chiral columns and chiral separations achieve partial or no separation on some analytes. Te new Reg- has grown, it has become difﬁcult to organize the infor- isPack® CLA-1 can exhibit unique selectivity in separation when mation for use in developing new methods. The success existing CSP’s fail. Tis CSP is highly recommended to be added of chiral method development depends on the intro- to the screening process. duction of appropriate columns. The RegisPack® polysaccharide-based chiral stationary phase (CSP) is most Note successful in achieving a majority of chiral separations. All work was performed on a WATERS THAR SFC Method Station. However, there are cases in which this proven chiral selector leads to partial or no separation. Excellent separations using the newly introduced RegisPack® CLA-1 CSP have been obtained for a range of compounds, either by improving or complementing RegisPack’s selectivity.

his study used the well-established RegisPack® CSP and the Tnew chlorinated RegisPack® CLA-1 CSP. Both phases were based on 5-µm silica. Columns were both 25 cm x 4.6 mm in size. Te RegisPack® and RegisPack® CLA-1 are coated with tris- (3,5-dimethylphenyl) carbamate of amylose and tris-(5-chloro-2- methylphenyl) carbamate of amylose, respectively. Tis study investigated several compounds that had poor or no separation on a RegisPack® column. Tey were screened on the new RegisPack® CLA-1 column. Following are some chromatographic illustrations and a tabular summary of our results. Figure 2: Chromatogram Set 2. Sample: Verapamil. Mobile Phase: CO2/IPA (75/25) + 0.5% DEA. Flow Rate: 4.0 mL/min. Temp: 40 ºC. Pressure: 125 bar. UV: 290 nm. Conclusion Te RegisPack® is a well-established CSP that does a superior job Table I: Various other samples screened showing an of separating the majority of compounds submitted for chiral improvement in separation with RegisPack® CLA-1. screening. Tere are cases where the RegisPack® CSP can only RegisPack® RegisPack® CLA-1

Sample k’1/k’2 alpha k’1/k’2 alpha Propafenone 2.15/2.49 1.16 2.61/5.12 1.96 Verapamil 0.83/1.04 1.25 1.91/2.67 1.40 Bupivacaine 3.41/3.88 1.14 5.19/7.48 1.44 Disopyramide 2.07/2.40 1.16 2.07/4.43 2.15 Cromakalim 3.45/4.35 1.26 2.11/2.96 1.40 Ketamine 1.04/1.04 No Sep 2.23/2.49 1.11 Acebutolol 0.51/0.51 No Sep 1.10/2.70 2.45 Huperzine 2.06/3.08 1.50 1.74/4.09 2.35 Naproxen 2.10/2.10 No Sep 1.79/2.39 1.34

Regis Technologies, Inc. 8210 Austin Ave, Morton Grove, IL 60053 Figure 1: Chromatogram Set 1. Sample: Propafenone. Mobile tel. (847) 583-7661, fax (847) 967-1214 Phase: CO2/IPA (70/30) + 0.5% DEA. Flow Rate: 4.0 mL/min. Temp: 40 ºC. Pressure: 125 bar. UV: 254 nm. E-mail: [email protected], Website: www.registech.com/chiral 20 MasterEnvironmental THE APPLICATION NOTEBOOK – JUNE 2011

Determination of Explosives Using Fast and High Resolution Liquid Chromatography with the Agilent 1290 Inﬁnity LC System

Detlef Wilhelm1, Edgar Nägele2, and Udo Huber2, 1AnaTox GmbH & Co. KG and 2Agilent Technologies R&D and Marketing GmbH & Co. KG

This application note describes the method for detection of nanogram levels of explosive constituents in mAU

seawater samples (e.g., after detonation of unexplod- 30

Tetryl ed ordnance devices). The improved chromatographic

25 TNT

method was developed using the Agilent 1290 Inﬁnity

LC Method Development system. 20 2,4-DNT HMX 2,6-DNT 15 RDX ong after World War II, unexploded ordnance devices are some- 10 Ltimes found, not only in the ground but also near the coasts, 5 Nitropenta such as in the Baltic Sea. Tese old munitions are often corroded Nitroglycerin and cannot be deactivated; sometimes, the only way to deactivate 0 is with a controlled detonation. Te residues of the detonation that 0 1 2 3 4 5 6 7 min can be found in the water must be monitored because of their tox- Figure 1: Detection of the 5 µg/ml standard with the 60-mm icity. Tis application note describes the method for detection of high sensitivity cell at 235 nm. Chromatographic conditions: Sta- nanogram levels of explosive constituents in seawater samples. Te tionary phase: Stable Bond CN-RRHT (2.1 × 100 mm, 1.8 µm); improved chromatographic method was developed using the Agi- Mobile phase A: Water; Mobile phase B: Acetonitrile; Gradient: at 0 min 20 % B, at 6 min 40 % B, at 7 min 40 % B; Flow rate: 0.5 lent 1290 Infinity LC Method Development system. mL/min; Column temp.: 30 °C; Inj. vol.: 1 µl; UV Detection: 214 nm/4, Ref.: 450 nm/80; Data rate: 20 Hz. Experimental LC System critical pair 2,4- and 2,6-dinitrotoluol is only completely separated For method development and analysis of the samples, an Agilent on the Bonus RP and the SB-AQ, but with the disadvantage of co- 1290 Infinity LC system was used. Te system consists of: 1290 elution of TNT and Tetryl. Te best separation with a simple gradient Infinity Binary pump with integrated vacuum degasser, 1290 In- was achieved with the Stable Bond CN column, and therefore, this finity high performance autosampler, 1290 Infinity Termostat- phase was used for further optimization. Te final method separates ted column compartment, and 1290 Infinity Diode Array Detec- all explosives within 7 min. A further improvement in sensitivity was tor with 10- and 60-mm Max Light Cartridge flow cell. achieved by using the new 60-mm high sensitivity flow cell. Figure 1 shows the chromatograms of the 5 µg/ml standard with the 10 mm Analytes cell (data not shown) and the 60 mm high sensitivity cell. If the detec- Components: Octogen (HMX), Hexogen (RDX), 2,4-Dintro- tion wavelength is changed from 235 nm to 214 nm, the influence of toluol (2,4-DNT), 2,6-Dintrotoluol (2,6-DNT), 2,4,6-Trintro- the eluent might be increased. However, the sensitivity for 2,3-DND- toluol (TNT), Nitropenta (PETN), Nitroglycerine, 2,4,6-Trini- MB, nitropenta, and nitroglycerine is further increased. trophenylmethylnitramine (Tetryl), 2,3-Dinitrodimethylbutane (2,3-DNDMB-internal standard, eluting at 3.35 min in Figure 1). Conclusion Te new Agilent 1290 Infinity LC is designed to provide the highest Results and Discussion speed, resolution, and sensitivity. Te transfer of a standard method to Te explosives analyzed in this application are very polar but not determine eight explosives on sub-2-µm RRHD or RRHT columns water-soluble substances, and are not well separated on typical RP18 could be done “overnight” by identifying the best selectivity using the columns. Terefore, different RP columns, including more polar Agilent Method Development system. Te results show that the analy- materials, were systematically screened using the Agilent 1290 Infin- sis could be shortened to 7 min without any matrix influences. ity Method Development system. Great selectivity differences can be seen between the unpolar phases like Eclipse Plus C18, Eclipse Agilent Technologies, Inc. Plus Phenyl-Hexyl, Poroshell EC-C18, and Stable Bond C18. Only 5301 Stevens Creek Blvd., Santa Clara, CA 95051 the Phenyl-Hexyl column shows separation of more than four of the tel: (877) 424-4536, fax: (408) 345-8474 explosives. Te polar RP-phases are more useful. In particular, the Website: www.agilent.com THE APPLICATION NOTEBOOK – JUNE 2011 EnvironmentalMaster 21

Direct Determination of Endothall in Water Samples by IC–MS

Leo (Jinyuan) Wang and William C. Schnute, Dionex Corporation

ndothall is a widely used herbicide for both terrestrial and aquatic weeds. Exposure to endothall in excess of E 8.0e4 mz IS: glutarate-d SIM_01: 137 the maximum contamination level (MCL) can cause illness. 6 Endothall is regulated by the U.S. Environmental Protection Agency (EPA) at 100 ppb in drinking water, and by the Cali- Intensity [counts] fornia EPA at 0.58 mg/L, or 580 ppb, as the Public Health 1.0e4 SIM_02: 185 mz Goal. Current analytical methods described in EPA method Endothall 548.1 for the quantitation of endothall in water samples involve time-consuming sample preparation and derivatization followed by a 20-min analysis by GC–MS or GC–FID. Intensity [counts]

fiis study describes the direct analysis of trace levels of endo- 35 - ECD_1 Br+NO3 thall in water samples by ion chromatography mass spectrometry H PO- NO- 2 4 2 - HSO4 (IC–MS). Water samples were directly injected for analysis and F- Cl- chromatographic separation was reduced to 10 min. fie MSQ Response [µS] Plus™ Mass Spectrometer was operated in selected ion monitoring 0 1 2 3 4 5 6 7 8 9 10 (SIM) mode, allowing minimum sample cleanup and ensuring Minutes sensitive (low ppb) and selective quantiTcation. Isotope labeled Figure 1: IC-MS of 20 ppb endothall spiked in a seven anions glutaric acid (Glutarate–d6) was used as the internal standard to matrix: 0.2–1.5 ppm. ensure quantitation accuracy. Results and Conclusion Experimental Conditions As seen in Figure 1, endothall was retained and separated from System: Dionex ICS-5000 RFIC System seven commonly seen anions within 10 min, and was detected Column: IonPac® AS16 and AG16 (2 mm) with great sensitivity and selectivity using SIM acquisition. fiis Temp.: 30 °C method features direct analysis without sample pretreatment and Flow Rate: 400 µL/min signiTcantly reduces run time relative to GC methods, thus im- Mobile Phase: Hydroxide gradient generated from EG-II proving throughput. Suflcient sensitivity was achieved in this KOH cartridge (15 mM hydroxide after 4 study to allow the routine quantiTcation of endothall below the min equilibration; ramp to 80 mM from 5 to lowest regulated level (100 ppb by U.S. EPA standards). 6 min and held for 3 min; return to initial condition in 0.5 min) IonPac is a registered trademark of Dionex Corporation. MSQ Plus is a trade- Detection: Suppressed conductivity and MSQ Plus mark of fiermo Fisher ScientiTc. mass spectrometer Ionization Source: Electrospray ionization (ESI) Probe Temp.: 500 °C Needle Voltage: 3000 V Solvent: 200 µL/min acetonitrile delivered by an AXP-MS pump Matrix Diversion: Eluent to MS: 4.2 – 6 min Acquisition Mode: Selected ion monitoring (SIM) Analyte: SIM (m/z) Endothall: 185 at 50V Dionex Corporation

IS (glutarate-d6): 137 at 35V 1228 Titan Way, P.O. Box 3603, Sunnyvale, CA 94088 tel. (408) 737-0700, fax (408) 730-9403 Website: www.dionex.com 22 MasterEnvironmental THE APPLICATION NOTEBOOK – JUNE 2011

Separation of Quaternary Ammonium Compounds Using a Bonded Polymeric Zwitterionic Stationary Phase

Patrik Appelblad, Merck SeQuant AB

Using a bonded polymeric zwitterionic stationary phase, in HILIC mode, polar quaternary herbicides like Diquat can easily be retained, with good peak symmetry. The impact of ionic strength and organic content in the mobile phase is discussed.

iquat (6,7-Dihydrodipyrido[1,2-a:2’,1’-c]pyrazinediium di- Dbromide) is an extremely polar compound, Figure 1, and commonly used as a nonselective contact herbicide. Diquat is a doubly charged cationic analyte. Different separation techniques have been used over time like gas chromatography, electrophoresis, and reversed phase (RP) liquid chromatography. Te most common procedure used to be RP-chromatography with addition of ion-pairing (IP) reagents. Most RP-columns are silica based, where presence of residual silanol groups imposes secondary interactions, causing peak tailing. IP reagents are added to the mobile phase to enhance the retention, Figure 1: Separation of diquat on a ZIC-pHILIC column; 150 × to mask unwanted secondary interactions, and to improve the analyte 4.6 mm, 5 µm. Mobile phase consisted of acetonitrile/250 mM of ammonium acetate pH 3.7 (X:Y, v/v). peak shape. Presence of IP reagents in the mobile phase decreases the sensitivity, especially when combined with mass spectrometric (MS) silanol interactions with the model compound. After initial scouting detection. IP reagents often show an effect of ion-suppression, decreas- experiments, a mobile phase with acetonitrile and ammonium acetate ing the quantity of ions that reaches the MS. To aid better detection, (40:60; v/v) was selected. Te pH was set at 3.7, and the impact of alternative separation techniques are therefore sought. ionic strength was evaluated. Due to the electrostatic interactions be- An attractive alternative is hydrophilic interaction liquid chroma- tween diquat and the stationary phase, the retention was strong at lower tography (HILIC), where a polar stationary phase is used in combina- ionic strength, and as the salt concentration was raised, narrower, more tion with aqueous–organic mobile phases. In HILIC mode, the elution symmetrical peaks were attained along with overall shorter retention. order is usually the opposite of reversed-phase chromatography, offer- To further optimize the separation, the proportion of organic solvent ing an increased retention of polar and hydrophilic solutes. Tis means was varied (see Figure 1). Te retention factor for diquat increased from that IP reagents are not needed and coupling to MS detection is eased. 1.5 to 7 when the acetonitrile content was raised from 40 to 70 volume Tis application note focuses on the retention of diquat on a polymeric percent in the mobile phase. Te peak symmetry was very good for all bonded zwitterionic HILIC stationary phase, the effect of ionic strength compositions; hence it was easy to modulate the retention and to obtain and percentage of the organic content in the mobile phase is discussed. useful experimental conditions for any detection mode.

Experimental Conditions Conclusion Column: ZIC®-pHILIC 150 × 4.6 mm, 5 µm Separation of extremely polar compounds like diquat is straightfor- Mobile phase compositions are given in each ﬁgure legend. ward using a bonded polymeric zwitterionic stationary phase. Altering Flow rate: 1.0 mL/min ionic strength and organic content in the mobile phase, the retention Temperature: 40 ∘C of diquat can easily be modulated and to allow for sensitive detection. Detection: UV at 313 nm Sample: Diquat, 50 ppm diluted in acetonitrile/water (1:1; v/v) Reference Injection volume: 5 µL (1) Dr. Norikazu Nagae, ChromaNik Technologies Inc., Japan.

Results EMD Chemicals, Inc., An afﬁliate of Merck KGaA A polymer-based HILIC column, with a bonded zwitterionic sta- 480 South Democrat Road, Gibbstown, NJ 08027 tionary phase was chosen for the separation of diquat, to alleviate tel. (800) 222-0342; Website: www.emdchemicals.com THE APPLICATION NOTEBOOK – JUNE 2011 EnvironmentalMaster 23

Table I: Results

Mean Spiked Congener Recovery % Rec STD Dev. ug/mL ug/L BZ #1 1 .911 91.1% 0.13152 BZ #5 1 .916 91.6% 0.16226 Automated One Step Solid BZ #18 1 .903 90.3% 0.02089 Phase Extraction and BZ #31 1 .905 90.5% 0.02668 BZ #44 1 .898 89.8% 0.02869 Concentration of PCBs in Water BZ #52 1 .897 89.7% 0.02518 Phil Bassignani, FMS, Inc. BZ #66 1 .907 90.7% 0.03209 BZ #87 1 .913 91.3% 0.03405 CBs are a group of synthetic organic chemicals that contain 209 BZ #101 1 .905 90.5% 0.032893 Pindividual compounds (known as congeners) with varying harm- BZ #110 1 .916 91.6% 0.03250 ful eﬁects. PCBs enter the environment as mixtures containing a variety BZ #138 1 .900 90.0% 0.03206 of individual components. PCBs have been widely used as coolants and lubricants in transformers, capacitors, and other electrical equipment. BZ #141 1 1.005 106.0% 0.03216 Te manufacture of PCBs stopped due to evidence that PCBs build up BZ #151 1 .895 89.5% 0.03165 in the environment and cause harmful eﬁects. Due their extreme toxic- BZ #153 1 .907 90.7% 0.31756 ity, durability, and wide industrial use, PCBs have found their way into BZ #170 1 .896 89.6% 0.03248 drinking water supplies. Te PowerPrep SPE and PowerVap Direct to BZ #180 1 .915 91.5% 0.03151 Vial Concentration system speeds up the sample prep process for the BZ #183 1 .924 92.4% 0.03215 analysis of PCBs by combining these sample prep steps into one. BZ #187 1 .914 91.4% 0.03139

Instrumentation • FMS, Inc. Power Prep SPE (solid phase extraction) system Procedure • Waters Oasis 1 g HLB SPE cartridge Five 1 L water samples were each spiked with a mixture of 19 • FMS, Inc. PowerVap Concentrator individual PCB congeners at 1 ug/mL each. Samples were also • Termo Fisher Scientiﬁc Polaris Q GCMS spiked with a 1 ug/mL tetrachloro-m-xylene solution as an extraction surrogate. Using the Power Prep SPE system, samples Method Summary were then loaded on prewet Oasis HLB cartridges using vacuum PowerPrep SPE to draw the samples across the cartridge. Sample bottles were 1. Condition Cartridge: 10 mL MeOH automatically rinsed. Cartridges were dried using a nitrogen

2. Condition Cartridge: 10 mL H2O stream blown across the cartridge to remove all remaining water. 3. Load Sample: 15 min Once dried, the cartridges were eluted with 20 mLs of DCM 4. Rinse bottle: 5 s directly to the FMS PowerVap Concentrator into direct to vial 5. Load rinse: 1 min tubes. Extracts were blown down to a 1 mL ﬂnal volume using 6. Dry Cartridge: 30 min the FMS direct to vial concentrator tubes. Extracts were then 7. Elute Sample: 20 mL DCM transferred for GC–MS analysis. PowerVap Concentrator 1. Pre-heat temp: 55 °C Conclusions 2. Pre-heat time: 30 min Te FMS PowerPrep SPE and PowerVap perform Automated 3. Heat in Sensor mode: 65 °C One Step and Direct to Vial Concentration. Combining sample 4. Nitrogen Pressure: 15 PSI prep processes is shown to eﬀciently extract PCB samples at a high rate of speed producing excellent recoveries and reproducibility for water samples. Te combination of the FMS PowerPrep SPE system and the Waters Oasis HLB cartridge demonstrates RT: 8.22 . 22.62 21.46 NL: 100 8.76E5 consistent, reproducible, reliable high throughput automated TIC MS 90 MSA1394 80 sample extraction of PCBs.

70 PCB-209 (ISTD) 60 50 PCB-87 PCB-110 PCB-66 PCB-52 PCB-101

40 PCB-187 PCB-44 PCB-153 PCB-141 PCB-31 15.44 PCB-151 PCB-183 PCB-170 PCB-18 30 13.78 15.03 PCB-180 FMS, Inc. PCB-138 PCB-5 TCMX 16.31 12.11 13.11 17.26 17.87 19.26 PCB-1 PCB-206 Relative Abundance 20 8.88 11.22 10.21 580 Pleasant Street Watertown, MA 02472 10 20.95 8.69 9.09 10.10 10.71 11.73 12.91 14.56 18.31 19.45 20.07 22.37 0 tel. (617) 393-2396, fax: (617) 393-0194 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Time (min) Email: [email protected] Figure 1: Results of PCB congeners in sample extract. Website: www.fmsenvironmental.com 24 MasterEnvironmental THE APPLICATION NOTEBOOK – JUNE 2011

Advantages of Multi-Pesticide Screening by GC–MS

Kory Kelly, Phenomenex Inc.

Due to the increasing number of toxic pesticides being used throughout the world, a robust method for App lD 16184 61 positively identifying low levels of multiple classes of 80-81 84 pesticides is required. A Zebron™ ZB-MultiResidue™ 21 24 25 36,37 41-43

column is used with GC–MS to identify over 100 com- 44-45

64-66 39 mon pesticides. 71-74

63 69 79 75 here is an increasing amount of trade in agricultural products, 85

33 resulting in a greater need to regulate those exports/imports 77,78 T 20 23 46-47 82 38 76 35 50 26 60 for toxic chemicals like pesticides. Many countries importing 17 86 62 67 87 52 13-15 59 40 4 22 49 58 68 70 foods are striving for testing protocols that account for the varia- 27 57 31 19 94,95 29 30 32 tion of pesticides being used as well as diﬁerent regulations of 48 83 34 22 24 28 51 12 18 other exporting countries. To this end, a widely accepted method 96 14 16 18 105 89 108 98 104 of pesticide residue identiTcation and quantitation relies on a 6,7 8 55 103 107 90 97 99 110,111 3 5 56 102 100 112 screening approach using GC–MS. 1 10 54 91 92, 109 16 53 88 93 101 106 2 9 11

Multiresidue screening analyses are typically extracted using 8 10 12 14 16 18 20 22 24 26 28 30 32 34 QuEChERS, which do not fractionate residues into separate classes. fle large number of resulting compounds must be sepa- Figure 1: Multiresidue pesticide screen. For peak identities, please contact Phenomenex. rated in one chromatographic run. A GC–MS is used to separate and positively identify compounds as well as give low detection improved separation of previously clustered peaks. Fewer coelu- limits. tions are advantageous for sensitive SIM methods because fewer For this work, a column speciTcally designed for multiple pes- peaks are included in the SIM window. flis results in easier iden- ticide detection is used, the ZB-MultiResidue-1. fle column was tiTcation, greater signal to noise, and lower detection limits. developed using a new stationary phase unlike any commercially fle column is also MS certiTed, providing low bleed in sensi- available columns today. fle column is also MS certiTed making tive MS detectors. flis lowers noise levels for later-eluting com- it ideal for use with GC–MS for multiresidue pesticide methods. pounds like permethrins, making detection and quantitation easier. Experimental Conditions A Zebron ZB-MultiResidue column of 30 m × 0.25 mm ID Conclusions × 0.25 µm dimensions (Phenomenex, Torrance, California) was A pesticide multiresidue screening method is presented using a new used in an Agilent 6890 with 5973 MSD (Palo Alto, California). gas chromatographic column (Zebron ZB-MultiResidue) coupled fle splitless injection of 1.0 µL of 1 ppm analytes was made with with mass spectrometry. flis method provides for improved ana- constant ffow helium at 0.9 mL/min. fle oven program was 80 lyte separation which can result in easier SIM method development °C for 0.5 min to 150 °C @ 10 °C/min to 240 °C @ 4 °C/ min to as well as improved quantitation. In addition, this MS certiTed col- 320 °C @ 15 °C/min for 3 min. umn provides low bleed to reduce instrument maintenance and provide better detection limits for later-eluting compounds. Results Additional methods are available that include retention of fle chromatogram for a multi-pesticide screen is presented in more than 300 difierent pesticides. For further information on Figure 1. flis chromatogram contains 112 of the most com- multiresidue methods, please contact Phenomenex. monly detected pesticides, including chlorinated, nitrogen, and phosphorous-containing pesticides classes. Phenomenex Inc. Typical chromatograms using 5% phenyl phases can have 411 Madrid Avenue, Torrance, CA 90501 clusters of coeluting peaks in the center of the chromatogram. tel. (310) 212-0555, fax (310) 328-7768 fle unique selectivity of the ZB-MultiResidue-1 column ofiers Website: www.phenomenex.com THE APPLICATION NOTEBOOK – JUNE 2011 EnvironmentalMaster 25

Reducing Cycle Time for Analysis of 1,4-Dioxane Using an Automated Purge and Trap Sample Prep System

Nathan Valentine, Teledyne Tekmar

With advances in instrumentation and without regulatory method constraints, 1,4-dioxane can be detected at the part-per-billion (ppb) level, despite its poor purge efficiency. This application will manipulate purge and trap, as well as GC–MS parameters, to create a more efficient method to reduce cycle times. This study will utilize a Teledyne Tekmar Atomx Automated Sample Prep System in conjunction with an Agilent 7890/5975 GC–MS. A linear calibration and method detection limits (MDLs) for 1,4-dioxane will be established using this method. Figure 1: Extracted ion chromatogram of a 10 ppb 1,4-dioxane standard. ,4-dioxane is commonly analyzed in water using purge and 1trap concentration to prepare the sample for evaluation by Results and Conclusions GC–MS. Because of the poor purge eTciency of the analyte, Relative response factors were evaluated for linearity and normal purge and trap methods must be modified to detect %RSD, which were 0.9998 and 5.12, respectively. fle MDL 1,4-dioxane at the part-per-billion level. flis method also utilizes was also established at 0.74 ppb for 1,4-dioxane by analyzing an Agilent 7890A/5975C with Triple Axis Detector in Selected seven replicates at a concentration of 5 ppb. An extracted ion Ion Monitoring (SIM) mode to provide the increased sensitivity chromatogram for 10 ppb 1,4-dioxane standard can be found in required by this low-level analysis. Figure 1. For this study, an Atomx Automated Sample Prep System was flroughput is a major factor in lab eTciency, and therefore, used in conjunction with a GC–MS system. flis “all-in-one” profitability. Automation and optimized purge and trap methods autosampler allows for complete automation of sample preparation can greatly reduce sample preparation and overall cycle times. flis for the analysis of soil and liquid samples for purge and trap. study demonstrates the capabilities of the Teledyne Tekmar Atomx flrough the features the Atomx provides, such as the 80-position Automated Sample Prep System coupled with an Agilent 7890/5975 sample tray and standard addition vessels, eTciency and throughput GC–MS to detect 1,4-dioxane at the ppb level. By optimizing the can be greatly increased, leading to cost and time savings. purge parameters, without compromising sensitivity, throughput Utilizing an Agilent 7890/5975 GC–MS system, a linear calibration can be greatly increased, leading to cost and time savings. was performed and percent relative standard deviation (%RSD) and MDLs were determined for 1,4-dioxane, using a 5-mL purge volume.

Experimental Conditions fle Atomx Automated Sample Prep System was coupled to an Agilent 7890A/5975 GC–MS with Triple Axis Detector for analysis. Teledyne Tekmar’s proprietary #9 trap was the analytical trap utilized. fle GC was configured with an Agilent DB-624 20 m × 0.18 mm × 1.0 µm column. A 50-ppm working calibration standard was prepared in methanol. Calibration standards were prepared in 50-mL volumetric ffasks filled to volume with de-ionized water over a range of 1–500 ppb. Samples were transferred to headspace-free 40-mL vials for analysis. fle internal standard (IS) was prepared in methanol at Teledyne Tekmar a 50 ppm concentration and added to each sample, bringing the 4736 Socialville Foster Rd., Mason, OH 45040 final concentration of 50 ppb. Agilent Chemstation software was tel. (800) 874-2004, fax (531) 229-7050 used to process the calibration data. Website: www teledynetekmar.com 26 MasterEnvironmental THE APPLICATION NOTEBOOK – JUNE 2011

Method 525.2 Update: Analyte Recoveries with New CCL3 Compounds

Xiaoyan Wang, Michael Telepchak, Jeffery Hackett, and Don Shelly, UCT, LLC

EPA is updating method 525.2. The CCL3 includes 104 j) Quantitatively transfer extract to auto sampler vial then bring analytes. UCT has evaluated the recoveries of these to 1 mL. compounds, including the current list, using proposed k) Add internal standards. Sample is ready for analysis. changes to the method with the C18-based UCT ENVIRO-CLEAN® 83 mL Extraction Cartridge ECUNI525. 4) Instrumentation • GC–MS: Agilent 6890N GC with Chemstation software cou- Experimental pled with 5975C MSD & 7683 auto sampler 1) Sample Preparation • GC capillary column: Restek Rxi-5sil MS 30 m × 0.25 mm × a) Weigh 0.1 g L-ascorbic acid, 0.35 g Na3EDTA and 9.4 g potas- 0.25 µm sium monobasic citrate into a 1-L amber bottle. • Injector: 1 µL splitless injection at 250 °C, 1 min split delay b) Fill with sample water. Do not ﬁush out preservation reagents. • Liner: 4 mm splitless gooseneck, 4 mm ID × 6.5 mm OD × c) Shake until salts dissolve. 78.5 mm (UCT GCLGN4MM) d) Spike with surrogates and analyte standards. Mix well. • Oven temperature program: Initial 55 °C, hold 1 min, ramp 10 °C/min to 200 °C, ramp 7 °C/min to 320 °C, and hold for 0.36 2) Sample Extraction min. Total run time is 33 min. Begin data acquisition at 5 min. a) Assemble the extraction system. Attach adaptors, 525 cartridg- • Carrier Gas: He 1.2 mL/min es, and bottle holders to 6-station manifold. • MSD condition: Aux temperature: 280 °C, MS Source: 230 °C, b) Wash the bottle holder and cartridge with 5 mL 1:1 MS Quad: 150 °C • EtOAc:MeCl2 (ethyl acetate: methylene chloride). Full scan: 45–500 amu c) Draw half through, soak for 1 min then draw through remaining solvent. Dry under vacuum for 2 min. Conclusion d) Condition the cartridge with 10 mL MeOH. Soak for 1 min Te UCT C18 ENVIRO-CLEAN® 83 mL Extraction Cartridge then draw most MeOH through, leaving a layer on the frit. ECUNI525 has been shown to provide excellent recoveries for all e) Add 10 mL of reagent water to the cartridge. Draw most of the Method 525 compounds as well as the EPA CCL3 analytes. water through the cartridge, leaving a layer on the frit. f) Do NOT let the cartridge go dry, otherwise re-condition. Table I: Representative analyte recoveries data generated from spiked ground water g) Place sample bottle on the bottle holder. h) Turn on vacuum, adjust the ﬁow to a fast drop-wise fashion. Analyte Recovery% Analyte Recovery% i) After extracting sample, rinse the sample bottle by adding 10 mL 3-BHA 102.0 Nitrofen 111.0 reagent water. Pass rinse through cartridge. BHT 95.5 Nonachlor, trans 105.0 j) Dry cartridge under full vacuum for 10 min. Captan 97.3 Oxyuorfen 91.8 DEET 101.0 Pentachlorophenol 98.3 3) Sample Elution DIMP 89.6 Phenanthrene 106.0 2,4 a) Insert a 40-mL glass vial into the manifold. 95.6 Phorate 95.5 Dinitrotoluene b) Elute C18 cartridge with 5 mL EtOAc. Endosulfan II 98.5 Phosphamidon 100.0 c) Draw half, soak 1 min then draw remaining through. Ethion 95.8 Profenofos 98.4 d) Repeat with 5 mL of MeCl2. Fluridone 111.0 Tebuconazole 94.4 e) Add 10 g sodium sulfate (Na SO ) to cartridge. Rinse bottle 2 4 HCH, δ 102.0 Vinclozoline 98.6 with 5 mL EtOAc then pour into cartridge. f) Repeat using 5 mL MeCl2. UCT, LLC 2731 Bartram Road, Bristol, PA 19007 g) Remove extract. Dry through a 10-g bed of Na2SO4. tel. (800) 385-3153 h) Rinse the Na2SO4 using both solvents. E-mail: [email protected] i) Concentrate extract to 0.7 mL under a gentle N2 stream at 40 °C. Do NOT dry to ≤ 0.5 mL. Website: www.unitedchem.com THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 27

Screening for Pesticide Residues in Food and Identiﬁcation with Highest Conﬁdence Using High Resolution and Accurate Mass LC–MS-MS André Schreiber and David Cox, AB Sciex

his note describes the use of the AB Sciex TripleTOF™ 5600 TLC–MS-MS system and data processing using the XIC Man- ager software to screen extracts of fruit and vegetable samples for pesticide residues. Detected compounds were automatically iden- tiﬁed based on retention times, accurate mass, isotopic pattern, and MS-MS library searching.

Experimental Conditions Fruit and vegetable samples were extracted using a QuEChERS (quick, easy, cheap, effective, rugged, and safe) procedure (1,2). Ex- tracts were diluted 5× using the aqueous mobile phase to optimize chromatographic peak shape and minimize possible ion suppression effects. UHPLC separation was achieved using a Shimadzu UFLCXR system with a Restek Ultra Aqueous C18 3 μm (100 × 2.1 mm) column and a 15 min gradient of water and methanol with ammonium formate buffer. Te AB Sciex TripleTOF™ 5600 system equipped with an electrospray ionization (ESI) source was Figure 1: Imidacloprid, metalaxyl, spirotetramat, and cyprodinil used. Full scan TOF-MS spectra were acquired over a mass range were identified in a Chinese broccoli sample using retention of 100–1000 Da with an accumulation time of 100 ms. TOF-MS- time, accurate mass information, and MS-MS library searching MS spectra were automatically acquired throughout the chromato- (purity > 80%). graphic run using standardized collision energy settings of CE = 35V and CES = 15 V. Summary Results and Discussion Te AB Sciex TripleTOF™ 5600 LC–MS-MS system provides a A total number of 289 targeted pesticides were analyzed using high high sensitivity platform to screen for and identify pesticide residues resolution and accurate mass LC–MS-MS. LC–MS-MS data were in fruit and vegetable extracts. Te high resolution and accurate processed using the XIC Manager software which calculates the mass data can be automatically mined by the XIC Manager software expected molecular ion based on molecular formula, isotope, and to provide high confidence compound identification by combining adduct provided. Extracted ion chromatograms (XIC) of each pes- retention times, accurate mass MS, and MS-MS data. ticide were automatically generated using an extraction window of ± 10 mDa around the expected retention time and compared to a References specified intensity threshold. (1) M. Anastassiades et al., J. AOAC Int. 86, 412–431 (2003). Te AB Sciex TripleTOF™ 5600 LC–MS-MS system routinely (2) EN 15662:2007 (2007). operates with a resolution (full width at half height) of ~20000 at (3) A. Schreiber et al., 46th Florida Pesticide Residue Workshop (2009) St. Pete m/z 200 Da to ~40000 at m/z 1000 Da and with a mass accuracy of Beach, FL. ~1 ppm (0.1 to 1 mDa, depending on m/z). Different fruit and vegetable samples obtained from a supermar- ket were analyzed and processed. An established MS-MS library was used for pesticide identification (3). Figure 1 shows an example of identification of imidacloprid, metalaxyl, apirotetramat, and cypro- AB Sciex dinil in a Chinese broccoli sample based on retention times, accurate 110 Marsh Drive, Foster City, CA 94404 mass, isotopic pattern, and MS-MS library searching with a purity tel. (877) 740-2129, fax (650) 627-2803 > 80%. Website: www.absciex.com 28 Food and Beverage THE APPLICATION NOTEBOOK – JUNE 2011

Simultaneous Quantitation of 2- and 4-Methylimidazole in Carbonated Drinks

Leo (Jinyuan) Wang, Xiaodong Liu, Christopher Pohl, and William Schnute, Dionex Corporation

he chemicals 2- and 4-methylimidazole (2-MI and 4-MI) are Tby-products produced during the manufacturing of caramel coloring ingredients used to darken food products such as carbon- Direct Analysis of a Diluted Dark Soda (10X) A Diluted Dark Soda (10X) + 100ppb Spike 100 100 NL: 8.53E3 4-MI NL: 8.88E4 ated beverages and soy sauce. ﬁese two chemicals were revealed C-SRM 2-MI Q-SRM as probable carcinogens in studies performed by the National Toxicology Program (NTP) and other researchers (1,2). In ad-

Relative Abundance Relative SRM:83 42 Abundance Relative 4-MI dition, 4-MI is listed as a carcinogen by the California Oﬀce of C-SRM

Environmental Health Hazard Assessment (OEHHA) in Janu- 0 0 100 100 ary 2011 with a calculated No SigniTcant Risk Level (NSRL) of NL: 4.55E4 4-MI NL: 1.19E5 Q-SRM 4-MI 16 μg per person per day (3). ﬁus, the quantitation of these two Q-SRM carcinogens in foods is of particular importance for food safety

and human health. Abundance Relative Abundance Relative 2-MI Conventional methods for identiTcation of 2-MI and 4-MI SRM: 83 56 C-SRM 0 0 in caramel color include gas chromatography methods which 0 2 4 6 8 10 0 2 4 6 8 10 Minutes Minutes involve labor intensive procedures (4). Liquid chromatographic methods have also been reported for 4-MI analysis (5). Figure 1: SRM chromatograms of real samples. The chemical This study demonstrates simultaneous quantitation of 2- 4-MI is quantied at 396 ppb in original dark carbonated drink. and 4-methylimidazoles in carbonated drinks by ultrahigh performance liquid chromatography tandem mass spectrom- dilution with positive detection of 4-MI. fiis diluted sample was etry (UHPLC MS-MS). re-analyzed after spiking with 100 ppb of each 2-MI and 4-MI, with the result showing in Figure 1 (right). Experimental fie analyses were performed on a Dionex UltiMate® 3000 Rapid UltiMate and Acclaim are registered trademarks and Trinity is a trademark of Separation Liquid Chromatography UHPLC system. Separation Dionex Corporation. TSQ Quantum Access is a trademark of fiermo Fisher was achieved on an Acclaim® Trinity™ P1 mixed-mode column ScientiTc. (2.1 × 50 mm) with isocratic elution at 0.5 mL/min. fie mobile phase consisted of 10% methanol in pH 5.7 ammonium acetate References buffier (total buffier concentration at 5 mM). Column temperature (1) http://ntp.niehs.nih.gov/files/535_Web_Final.pdf (Accessed May, 2011). was maintained at 15 °C. A fiermo ScientiTc TSQ Quantum (2) http://ntp.niehs.nih.gov/files/516final_web.pdf (Accessed May, 2011). Access™ with heated electrospray ionization source (HESI) was (3) http://oehha.ca.gov/prop65/law/pdf_zip/010711NSRLrisk4EI.pdf (Accessed operated in selected reaction monitoring (SRM) mode with the May, 2011). settings listed with the chromatogram. Vaporizer temperature was (4) R.A. Wilks, M.W. Johnson, A.J. Shingler, J. Agric. Food Chem. 25(3), 1077 set at 350 °C and capillary temperature was set at 200 °C. Sheath (1977). gas and auxiliary gas were both set at 60 arbitrary units. (5) C. Moretten, G. Crétier, H. Nigay, J. Rocca, J. Agric. Food Chem. (2011) (published online March 7, 2011, available at http://pubs.acs.org/doi/ Results full/10.1021/jf104464f, accessed March, 2011). As shown in Figure 1 (right), two target analytes were baseline separated on the Trinity P1 column. Both analytes were observed with the same two SRM transitions with diffierent responses; the SRM with stronger response was used for quantitation (Q-SRM) and the other was used for conTrmation (C-SRM). fie ratio of Dionex Corporation the two SRM responses was used as additional conTrmation to 1228 Titan Way, P.O. Box 3603, Sunnyvale, CA 94088 avoid possible false positives. Figure 1 (left) also demonstrates the tel. (408) 737-0700, fax (408) 730-9403 direct analysis of a diluted dark carbonated drink after ten-fold Website: www.dionex.com THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 29

Fast Determination of Anthocyanins to Authenticate Pomegranate Juice

Pranathi R. Perati, Brian De Borba, and Jeffrey S. Rohrer, Dionex Corporation

nthocyanins are a sub-class of naturally electron deficient Experimental Aand powerful antioxidants called ffavonoids that are respon- A Dionex UltiMate® 3000 Rapid Separation LC System was sible for the red, orange, and blue coloration in fruits and ffow- used for this study. For individual components of the system and ers. Pomegranate juice (PJ) is a major anthocyanin source and, preparation of solutions, please refer to Dionex Application Note therefore, very popular in health drinks. But, because of the short 264 (1). supply of pomegranates, adulteration of PJ is widespread. Tis study describes a sensitive, fast, and accurate method to deter- Results and Discussion mine anthocyanins in commercially available PJ with a simple di- Figure 1a shows the separation of six signature anthocyanins in lution and separation on a Dionex Acclaim® C18 RSLC column 100% PJ. Te sample was diluted 1:5 to prevent overloading. followed by detection at 540 nm. Te six signature anthocyanins Grape juice is one of several juices used to adulterate PJ. Tere- in PJ: delphinidin 3,5-diglucoside (Dp3,5), cyanidin 3,5-diglu- fore, a 50:50 mixture of grape and 100% PJ were used in this coside (Cy3,5), delphinidin 3-glucoside (Dp3), pelargonidin study to simulate an adulterated sample. Te chromatogram in 3,5-diglucoside (Pg3,5), cyanidin 3-glucoside (Cy3), and pelar- Figure 1a shows a separation of six anthocyanins characteristic gonidin 3-glucoside (Pg3) are separated in <8 min. to PJ ranging in concentrations from 2.1 to 73 μg/mL. Te simulated adulterated sample in Figure 1b shows the six signature anthocyanins and several late eluting peaks that are not characteristic of PJ. Te figure illustrates how the method could be used to 6.0 A identify adulterated samples. Adulterated juices often show un- 2 5 characteristic concentration ratios of signature anthocyanins and additional peaks not present in pure PJ. 1 To evaluate accuracy, the juices were spiked with known amounts mAU 3 of the six anthocyanins. Recoveries ranged from 98.0% to 108.3%, suggesting good method accuracy. 4 6 Acclaim and UltiMate are registered trademarks of Dionex Corporation. -1.0 3.5 B 5 References 2 3 (1) Dionex Corporation. Fast Analysis of Anthocyanins in Pomegranate Juice; 1 Application Note 264, LPN 2647: Sunnyvale, CA. mAU

4 6

-1.0 0 2 4 6 8 10 Minutes

Figure 1: Column: Acclaim 120, C18, 2.2 μm, Analytical 2.1 × 150 mm. Eluent: A: 9% CH3CN, 10% Formic Acid, B: 36% CH3CN, 10% Formic Acid. Gradient: 0.0–0.9 min: 100% A, 0.9–8.0 min: 28.5% B, Hold for 2 min at 28.5% B. Temp: 30 °C, Flow-rate: 0.475 mL/min. Detection: Vis 540 nm, Injection Volume: 0.5 µL. Separation of anthocyanins in (a) pomegranate juice sample showing Peak 1: Dp3,5, Peak 2: Cy3,5, Peak 3: Dp3, Peak 4: Pg3,5, Peak 5: Cy3, Peak 6: Pg3 at Dionex Corporation concentrations of 65.5, 123, 27.5, 8.00, 121, and 4.45 μg/mL, respec- 1228 Titan Way, P.O. Box 3603, Sunnyvale, CA 94088 tively; (b) simulated adulterated PJ showing Peak 1: Dp3,5, Peak 2: Cy3,5, Peak 3: Dp3, Peak 4: Pg3,5, Peak 5: Cy3, Peak 6: Pg3 at concen- tel. (408) 737-0700, fax (408) 730-9403 trations of 33.5, 61.0, 26.0, 5.15, 73.0 and 2.10 μg/mL, respectively. Website: www.dionex.com 30 Food and Beverage THE APPLICATION NOTEBOOK – JUNE 2011

Fast Determination of Catechins in Tea

Pranathi R. Perati, Brian De Borba, and Jeffrey S. Rohrer, Dionex Corporation

ea, one of the most consumed non-alcoholic drinks world- ffe most abundant catechins in tea include catechin (C), epicat- Twide, contains catechins that are powerful antioxidants echin (EC), epigallocatechin (EGC), epicatechin gallate (ECG), gal- thought to provide many health benefits, including reduction locatechin (GC), gallocatechin gallate (GCG), and epigallocatechin of cholesterol and obesity, and protection against cardiovascular gallate (EGCG). ffe method described here uses a high-resolution disease and cancer. ffe four major varieties of teas (white, green, Dionex Acclaim® C18 RSLC column to separate and an absorbance oolong, and black) are derived from the same Camellia sinensis of 280 nm to detect and quantify catechins in green and black teas plant, but as they are prepared by diTerent processing methods, in less than 20 min per injection. their catechin concentrations can vary. ffus, it is important to establish a simple and reliable analytical method to determine cat- Experimental echin concentrations in diTerent tea products. A Dionex UltiMate® 3000 Rapid Separation LC (RSLC) system was used; for system components and preparation of solutions, please refer to Dionex Application Note 275 (1).

Results and Discussion 10 A 6 Figure 1a shows the catechins present in a commercially available green tea. ﬀe most abundant catechin in this green tea is EGCG, 3 which is about 50% of the total catechin content, at 135.4 mg/g. 8 Figure 1b shows an analysis of black tea, where, unlike green mAU 1 2 4 5 7 tea, the EGC concentration is higher than the EGCG content and the total catechin content is 63.3 mg/g, nearly 50% less than that of the green tea. ﬀe caTeine levels are also higher in black tea as 6 compared to green tea. Both the lower total catechin concentration and the higher caTeine concentrations are consistent with the in- 15 B 3 creased processing of black tea. To evaluate method accuracy, the teas were spiked with known amounts of the seven catechins. Recoveries ranged from 90.2% to 102.3%, suggesting method accuracy. An accurate analysis of these complex samples requires less than 20 min mAU per injection.

6 8 1 UltiMate and Acclaim are registered trademarks of Dionex Corporation. 2 4 5 7 -2 References 0 2.5 5 7.5 10 12.5 15 Minutes (1) Dionex Corporation. Determination of Catechins in Tea; Application Note 275, LPN 2799: Sunnyvale, CA. Figure 1: Column: Acclaim 120, C18, 2.2 μm, analytical 2.1 × 150 mm. Eluent: A: 0.1% TFA, 5% CH3CN, B: 0.1% TFA in CH3CN; Gra- dient: 0.0–1.2 min: 100% A, 1.2–15.5 min: 28.5% B, hold for 1.5 min at 28.5% B; Temp: 25 °C, Flow-rate: 0.45 mL/min; Detection: UV 280 nm, Injection Volume: 1.0 µL. Separation of catechins in (a) Green tea and (b) Black tea with Peak 1: Gallic Acid, Peak 2: EGC, Peak 3: Caffeine, Peak 4: C, Peak 5: EC, Peak 6: EGCG, Peak 7: GCG, and Peak 8: ECG. Gallic acid and caffeine concentrations were not calculated in either of the teas. Concentration of EGC, Dionex Corporation C, EC, EGCG, GCG, and ECG were as follows: Green tea: 45.3, 3.45, 1228 Titan Way, P.O. Box 3603, Sunnyvale, CA 94088 6.14, 64.0, 6.74, and 9.81 mg/g; Black tea: 27.8, 4.35, 2.25, 12.3, tel. (408) 737-0700, fax (408) 730-9403 9.20, and 7.47 mg/g. Website: www.dionex.com THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 31

Sensitive Determination of Hydroxymethyl Furfural in Honey, Syrup, and Fructose Solution

Lipika Basumallick, Deanna Hurum, and Jeff Rohrer, Dionex Corporation

ydroxymethylfurfural (HMF), an organic compound con- Htaining aldehyde and alcohol functional groups, is naturally 275 98 found in very low amounts in sugar-containing foods. It is also 3 4 2 produced during the heat treatment of foods from the dehydra- nC

tion of sugars such as glucose and fructose (1). Its presence is A D used as an indicator for spoilage, excessive heat-treatment, and 60 4 5 76 adulteration. Many countries have imposed restrictions on the Minutes maximum levels of HMF in food and beverages because of the 1 carcinogenic potential of similar compounds. 5 Commonly used HMF determination methods are based on nC spectral absorbance at 284 nm; however, direct absorbance measurements could have interferences from other compounds present in complex matrices. Here, we present a high-performance anion- 2 exchange chromatography with pulsed amperometric detection A (HPAE-PAD) method for HMF determination of HMF. ﬁis meth- B od exhibits broad linearity, low detection limits, high precisions, and C

good recovery of HMF, indicating that PAD is an appropriate detec- 50 tion technique for HMF. 0 2 4 6 8 10 12 14 16 18 20 22 24 25 Minutes

Equipment Figure 1: HMF in (A) thermally stressed honey, (B) fructose solu- A Dionex ICS-3000 system was used in this study. HMF (10 μL) tion, (C) pancake syrup, and (D) HMF standard. Peaks (1) Glycerol was separated on a CarboPac® PA1 column set, with electrolyti- (2) HMF (3) Glucose (4) Fructose (5) Sucrose. cally generated 50 mM hydroxide at 0.5 mL/min, and detected by PAD. ﬁe CarboPac PA1, a high capacity rugged column, de- were 0.5%, which can be attributed to consistent generation of high livered high resolution for HMF in a variety of matrices. Samples purity hydroxide by the eluent generator. Accuracy of the method were prepared according to procedures described in Dionex Ap- was veriﬀed by determining recoveries of HMF in spiked honey over plication Note 270 (2). three days (average recovery 103%).

Results Conclusion Figure 1(A) shows HMF (in a thermally stressed honey sample) de- fie described HPAE-PAD based method is accurate, reliable, re- tected at 4.8 min without interference from the other sugars. Figure quires no manual eluent preparation, and because of good sensitivity 1(C) shows the separation of HMF from other degradation products and consistent response, can be adapted for routine HMF analysis. in thermally stressed pancake syrup. HMF is a product of thermal degradation of fructose, the main constituent of pancake syrup. References Complex matrices, such as syrup, may have later eluting peaks (at (1) H.D. Belitz, W. Grosch, 1999 Food Chemistry. Berlin: Springer-verlag. ~55 min in pancake syrup; not shown), which could interfere with (2) Dionex Corporation, Application Note 270, 2011, LPN 2677. subsequent injections if a shorter run time is used. fiis method was also applied for HMF detection in fructose. (HMF, present as an impurity in fructose, must be quantiffed and meet FCC and USP requirements before use as a food substance or in infusion Tuids.) Figure 1(B) shows the separation of HMF from other thermal degra- Dionex Corporation dation products in fructose. fie described method was linear in the 1228 Titan Way, P.O. Box 3603, Sunnyvale, CA 94088 range of 0.1–50 μg/mL. fie LOD and LOQ were 0.04 μg/mL and tel. (408) 737-0700, fax (408) 730-9403 0.10 μg/mL, respectively. Retention time and peak area precisions Website: www.dionex.com 32 Food and Beverage THE APPLICATION NOTEBOOK – JUNE 2011

Cleanup of Baby Food Samples Using Gel Permeation Chromatography (GPC)

Elizabeth Badgett, Laura Chambers, and Gary Engelhart, OI Analytical

he Food Safety Modernization Act, which was recently signed Tinto law, is the country’s ﬁrst major food legislation since

1938, and increases the FDA’s authority over the U.S. food sup- TIC: GB+SPK-BEFORE GPC-001.D\data.ms TIC: GB+SPK-GPC 17-37-002.D\data.ms 1.05e+07 ply. ﬀese new regulations require that foods, including baby 1e+07 food, be tested for pesticide residues. 9500000 9000000 ﬀis application note illustrates the use of the AutoPrep 2000 8500000 8000000 GPC system to remove matrix interferences from baby food sample 7500000 7000000 extracts prior to pesticide analysis by GC–MS. 6500000

6000000

5500000 Experimental Conditions 5000000 4500000 Eight baby foods make up 96% of all sales from the three main 4000000 3500000 food manufacturers (1). Two vegetables, green beans and sweet po- 3000000 2500000 tatoes, and two fruits, pears and peaches, were used in this study. 2000000 A 100-g aliquot of each sample was extracted by the Luke proce- 1500000 1000000 dure (2), and divided into two aliquots. One aliquot was exchanged 5000000 0 into 1.5 mL hexane for immediate GC–MS analysis. ﬀe second 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 aliquot was exchanged into 10 mL of solvent for GPC cleanup. ﬀe Figure 1: GC–MS chromatogram of green bean extract before GPC column was packed with Envirobeads S-X3 swollen with 1:1 (black) and after (red) GPC cleanup. cyclohexane:ethyl acetate and calibrated according to U.S. EPA Meth- od 3640 to determine collection time for the pesticide residue fraction.

79.1 Scan 725 (11.062min): GB+SPK-GPC 17-37-002D\data.ms

Results 9000 Comparison of chromatograms from all four baby food extracts before 8000 7000 GPC cleanup and after GPC clean-up show signiﬁcant removal of ma- 6000 5000 44.0 trix interferences. Overlaid chromatograms of the green bean “Extract 4000 3000 55.1 207.1 262.9 Before GPC” and “Extract After GPC” are shown in Figure 1. 2000 67.1 108.1 276.9 95.1 149.0 236.9 1000 121.1 135.1 193.0 173.0 345.0 379.3 161.0 219.0 251.0 315.1 Figure 2 is the expanded total ion chromatogram (TIC) of the 288.9 300.7 326.9 357.2 369.3 396.4 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 79.0 green bean “Extract After GPC” and NIST library search illustrating #164256: Dieldrin

9000 positive identiﬁcation of dieldrin. 8000 For complete results of this study, refer to OI Analytical Applica- 7000 6000 tion Note # 3679 (3). 5000 4000

3000

2000 108.0 Conclusions 39.0 53.0 237.0 1000 27.0 66.0 97.0 143.0 173.0 193.0 209.0 121.0 132.0 251.0 309.0 0 ffe AutoPrep 2000 GPC system eTectively removed matrix inter- 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 ferences from baby food extracts prior to GC–MS analysis, facilitat- Figure 2: Expanded TIC and pesticide identification. ing pesticide peak identification and quantitation.

References (1) Pesticides in Baby Food, Environmental Working Group, http://www.ewg. org/book/export/html/7575. (2) Method I for Non-fatty Foods. Pesticide Analytical Manual, Vol. I: Multi- OI Analytical residue Methods, U.S. FDA, College Park, MD, 1994; Section 302. P.O. Box 9010, College Station, TX 77842 (3) OI Analytical Application Note # 3679, “Clean-up of Baby Food Samples tel. (800) 653-1711, fax (979) 690-0440 Using Gel Permeation Chromatography.” Email: [email protected], Website: www.oico.com THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 33

Analysis of Pesticides in Citrus Oil Using PTV Backush with GC–MS-MS Triple Quadrupole for High Sample Throughput

Charles Lyle, Eric Phillips, and Hans-Joachim Huebschmann, Thermo Fisher Scientiﬁc Inc.

International regulations on maximum residue levels analyzed and the areas were used to demonstrate stability of the (MRLs) of pesticides in food cover hundreds of indi- final method, including backffush. Te relative standard devia- vidual contaminants at the 10 ppb or below range. tions for the areas ranged from 0.69% to 3.68%. The analysis of citrus oil for pesticide contamination holds specific challenges. Conclusion Te use of PTV with backffush allowed the development of a esticides that are used on citrus crops are concentrated along method with very little sample preparation. Backffush of the PTV Pwith the matrix in this oil. Tis matrix is known for its high kept all of the high-boiling-point contamination from reaching boiling point components. Te system of the Termo Scientific any part of the analytical system, reducing maintenance. It also TRACE GC Ultra and TSQ Quantum XLS GC–MS-MS allows provided an increased sample throughput by decreasing time for the analysis of more than 40 pesticides in citrus oil with mini- needed for sample or system cleanup. Te timed-SRM function mal sample preparation. of the TSQ Quantum XLS provided the most efficient use of the instrument time, only monitoring the expected SRMs when Experimental Conditions needed. Te timed-SRM function provided the precision needed Te samples were diluted 20-to-1 using hexane. Te samples for for the low level analysis of pesticides in citrus oil. method development and calibration curve were made by spiking the list of pesticides into a known clean sample. Te TRACE GC Ultra™ PTV injector with backffush was the in- 14000000 2-Phenylphenol 6000000 -Lindane jector chosen due to the heavy contamination potential from the cit- R2 = 0.9998 12000000 2 5000000 R = 0.9999 10000000 rus oil matrix. During the method development process it was found RT: 6.40 RT: 7.73 AA: 744520 4000000 AA: 299977 100 100 8000000 90 90 80 80

Area 3000000 70 that there was a signiﬁcant amount of high boiling point matrix. Area 70 60 6000000 60 50 50 40 40 2000000 30 Relative Abundance 4000000 30 Relative Abundance All selected reaction monitoring (SRM) transitions were opti- 20 20 10 6.49 7.58 7.67 7.83 7.86 7.99 10 6.54 6.61 6.62 7.56 6.67 1000000 0 2000000 0 7.67.57.4 8.8.07.97.87.7 6.4 6.5 6.6 6.7 Time (min) mized for best collision energy and full conversion from precursor Time (min) 0 0 0 50 100 150 200 0 50 100 150 200 to product ion for monitoring. Each analyte had at least one con- ppb ppb

firming ion. Te information for these transitions as provided by Calibration cure for 2-phenylphenol with the Calibration cure for -lindane (BHC) with the the Termo Scientific Pesticide Analyzer Reference. Te function quantitation peak at 10 pph quantitation peak at 10 pph of timed-SRM was used to allow the instrument to automatically 25000000 4-4’-DDT lprodione R2 = 1.0000 25000000 R2 = 0.9998 determine the most efficient use of its time. 20000000

RT: 15.44 2000000 RT: 16.70 AA: 1062242 AA: 109848 100 100 Te timed-SRM function of the TSQ Quantum XLS™ system 15000000 90 90 80 80 70 1500000 70 Area 60 Area 60 10000000 50 50 provides more points across the peaks with better sensitivity and 40 40 1000000 30

30 Relative Abundance Relative Abundance 20 20 16.75 15.50 10 15.38 10 16.56 16.62 16.86 5000000 15.13 15.24 15.62 15.64 16.52 16.92 16.97 precision. Te method used for the analysis was a 1 μL injection of 0 500000 0 15.2 15.4 15.6 15.8 16.4 16.6 16.8 17.0 Time (min) Time (min) 0 0 the citrus oil, diluted 20:1 in hexane. Te TRACE GC Ultra PTV 0 50 100 150 200 0 50 100 150 200 backffush time was set to the point just after the last compound of ppb ppb Calibration cure for 4,4’-DDT with the Calibration cure for for iprodione with the interest was loaded onto the analytical column. All samples used in quantitation peak at 10 pph quantitation peak at 10 pph the method development and calibration curve were in the citrus oil. Te calibration curve ranged from 10 to 200 ppb. Figure 1: Calibration curves for 2-phenylphenol with the quantitation peak at 10 ppb; α-lindane (BHC) with the quantitation peak at 10 ppb; 4,4’-DDT with the quantitation peak at 10 ppb; Results and iprodione with the quantitation peak at 10 ppb. In the final developed method, the last peak of the more than 40 pesticides eluted in less than 20 min. Te extracted standards showed better than 0.995 r2 values for the correlation coefficient. Thermo Fisher Scientific Inc. Te peaks shown with the curves show the low point on the cali- 2215 Grand Avenue Parkway, Austin, TX bration curve, at 10 ppb, extracted from the citrus oil matrix. Six tel. (800) 532-4752, fax (561) 688-8731 matrix samples were spiked at 25 ppb level. Tese samples were Website: www.thermoscientific.com 34 Food and Beverage THE APPLICATION NOTEBOOK – JUNE 2011

Glyphosate Analysis in Soy Beans, Corn, and Sunﬁower Seeds by HPLC with Post-Column Derivatization and Fluorescence Detection

Maria Otserova, Rebecca Smith, and Michael Pickering, Pickering Laboratories, Inc.

lyphosate is a broad-spectrum herbicide widely used around is 125 mL. Blend at high speed for 3–5 min and centrifuge for Gthe world. Monitoring of glyphosate in crops and water is man- 10 min. Transfer 20 mL of the aqueous extract into a centrifuge dated in many countries. We describe a sensitive and robust HPLC tube and add 15 mL of methylene chloride. Shake for 2–3 min method for analysis of glyphosate in soy beans, corn, and sunfiower and centrifuge for 10 min. Transfer 4.5 mL of aqueous layer seeds. Tis method utilizes a simpliffed sample preparation procedure to another centrifuge tube and add 0.5 mL of acidic modiffer that has proven to be effiective even for challenging matrices. solution. Shake and centrifuge for 10 min. Filter through a 0.45 μm fflter. Method Analytical Conditions Matrix-Specic Modications Column: Cation-exchange column, K+ form, P/N Matrix with high 1) Water; 2) Protein; 3) Fat Content: 1954150 1) For samples that absorb large amounts of water, reduce Guard Column: Cation-exchange GARD™ Column test portion to 12.5 g while keeping water volume the Protection System or Cation-exchange guard same. column P/N 1953020 2) For samples with high protein content, add 100 μL of concen- Column Temperature: 55 °C trated HCl to 20 mL of crude extract. Shake and centrifuge Flow Rate: 0.4 mL/min for 10 min. Mobile Phase: K200, RG019 3) For samples with high fat content, do the methylene chloride Injection Volume: 100 μL partitioning twice.

Post-column Conditions SPE Cleanup Post-column System: Pinnacle PCX or Vector PCX Remove the top cap ffrst, then the bottom cap of the SPE col- Heated Reactor Volume: 0.5 mL umns and place them into the manifold. Drain the solution to Temperature: 36 °C the top of the resin bed. Transfer 1 mL of extract into the column Ambient reactor: 0.1 mL and elute to the top of the resin bed. Add 0.7 mL of the elu- Reagent 1: 100 μL of 5% NaOCl (Bleach) in tion solution and discard the efluent. Repeat with a second 0.7 950 mL of GA116 Diluent mL portion of the elution solution and discard the efluent. Elute Reagent 2: 100 mg of OPA and 2 g of Tio- glyphosate with 12 mL of the elution solution and collect the ef- fiuor in 950 mL of GA104 Diluent fiuent in a round bottom fiask. Evaporate to dryness at 40 °C us- Reagent Flow rate: 0.3 mL/min each reagent ing a rotary evaporator. Dissolve the residue in 2.0 mL of a solu- λ λ Detection: EX 330 nm, EM 465 nm tion of 10% RESTORE™ in water (use 1.5 mL for dry samples), fflter through a 0.45 μm syringe fflter and inject onto the HPLC Supplies for Sample Preparation column. Extracts can be stored refrigerated for up to seven days Methylene Chloride, HPLC Grade before the evaporation step.

Acidic Modiﬀer Solution (16 g KH2PO4, 160 mL of water, 40 ml of Methanol, 13.4 mL of conc. HCl) Elution Solution (160 mL of water, 40 mL of Methanol, 2.7 mL of HCl) RESTORE™ SPE sample clean-up cartridges P/N 1705-0001

Sample Preparation Extraction To 25 g of homogenized sample, add enough water (after esti- mating moisture content) such that the total volume of water THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 35

Table I: HPLC Gradient Table II: Recoveries for Glyphosate Time (min) K200 (%) RG019 (%) Spike level Soy Beans Corn Sunﬁower seeds

0 100 0 0.2 ng/g 109% 102 % 70%

15 100 0 0.1 ng/g 90% 93% 82%

15.1 0 100 0.05 ng/g 93% 93% 71%

17 0 100

17.1 100 0

25 100 0

Glyphosate

0 2 4 6 8 10 min

Figure 1: Chromatogram of soy beans sample spiked with Glyphosate at 0.1 ppb level.

Glyphosate

0 2 4 6 8 10 min

Figure 2: Chromatogram of corn sample spiked with Glyphosate at 0.1 ppb level.

Glyphosate

0 2 4 6 8 10 min Pickering Laboratories 1280 Space Park Way, Mountain View, CA 94043 Figure 3: Chromatogram of sunﬁower seeds sample spiked with tel. (800) 654-3330, fax (408) 694-6700 Glyphosate at 0.1 ppb level. Website: www.pickeringlabs.com 36 Food and Beverage THE APPLICATION NOTEBOOK – JUNE 2011

Simultaneous and Direct Analysis of Biogenic Amines in Food by LC–MS-MS Using Hydrophilic Chromatography Seiji Ito and Fumiya Nakata, Tosoh Corporation

iogenic amines are present in protein-rich foods such as Bfish, meat, and milk. ffeir concentration increases during food decomposition. ffe reasons to monitor biogenic amines in food products are to determine that the food is suitable for Extraction consumption and to establish appropriate storage conditions. Sample 5 g Analysis methods for biogenic amines include HPLC with Tuo- rescence detection — using derivatization with dansyl chloride Cut into small pieces or o-phthalaldehyde. ffe drawbacks to these methods are a 40- min analysis time, sample pretreatment that requires Tuores- 20% trichloroacetic acid, 5 mL cence derivatization, and liquid phase extraction. In this study, we investigated a simple, highly sensitive, and direct analytical Homogenize method that does not require derivatization and uses a TSKgel Amide-80 column under HILIC conditions followed by MS-MS Adjust to 100 mL with Ultrapure water detection (Figure 1). Room temperature, 30 min Experimental conditions Filtration using membrane lter LC System : Agilent 1200SL Series (pore size: 0.45 µm, cellulose acetate) Column: TSKgel Amide-80, 3 μm, 2.0 mm ID × 15 cm Mobile phase: A: 30 mmol/L ammonium formate in H2O, Extraction pH 4.0 B: ACN Gradient: 0 min (90%B), 12 min (40%B), 14 min (40%B), 16 min (90%B) Dilution Flow rate: 0.2 mL/min Extraction 1 mL Temperature: 50 °C Injection vol.: 2 μL

Adjust to 10 mL with acetonitrile MS: QTRAP® (AB SCIEX) Ion source: ESI Polarity: Positive Mode: MRM Precursor ion/ Product ion: Spermidine (Spd): 146.3/72.1 Putrescine (Put): 89.1/72.1 Cadaverine (Cad): 103.1/86.1 LC–MS-MS Histamine (His): 112.0/95.0 Tyramine (Tyr): 138.0/121.0 Tryptamine (Trp): 161.0/115.0 Figure 1: Sample pretreatment procedure of LC–MS-MS method.

Results and Discussion shown), this LC–MS-MS method can be completed in half the Figure 2 and Table I detail the results of the analysis of six bioanalysis time (20 min versus 40 min). The limits of quantita- genic amines. Excellent separation of the analytes and linearity tion for the analytes were 0.001–0.02 mg in a 100 g sample, an of the calibration curves were obtained. When compared to a improvement of 5–1300 times compared with the fluorescence fluorescence method using a reversed phase column (data not method. THE APPLICATION NOTEBOOK – JUNE 2011 Food and Beverage 37

A study was conducted to compare the LC–MS-MS method to the fiuorescence method in the evaluation of tuna samples under differing storage conditions. Table II lists the quantitative values 60,000 of biogenic amines in tuna preserved for two days at freezing and Spd: 50µg/L 50,000 room temperatures. Good correlation was obtained between the two Put: 250µg/L methods. 40,000 Cad: 250µg/L 30,000 Conclusions His: 50µg/L A new LC–MS-MS method for the analysis of biogenic amines 20,000 was investigated by Tosoh scientists. Six biogenic amines were ana- Intensity, cps Tyr : 20µg/L 10,000 lyzed in 15 min under HILIC conditions using a TSKgel Amide-80 Trp : 50µg/L column without the need for a complex and time-consuming de- 0 rivatization procedure. Limits of quantitation of the analytes were improved 5–1300 times using this new method compared with the 0 5 10 15 20 conventional fiuorescence method. Since good correlation between Retention time (minutes) the two methods was obtained in a comparative study of quantita- Figure 2: Chromatograms of biogenic amines. tive values, the LC–MS-MS method is a proven alternative method that is both simple and sensitive for the analysis of biogenic amines in food products.

Table I: Comparison of limit of detection and quantitation values (LC–MS-MS versus uorescence method) RSD LOQ Calibration curve LOD LOQ (n = 5) (Fluorescence method) Analytes (mg/100 g; in Range (μg/L) r2 (at 10 μg/L) (μg/L) (μg/L) (mg/100 g; in sh) sh)

Spd 1.0–500 0.998 2.1 0.10 0.20 0.004 0.36 Put 5.0–500 0.995 1.7 0.70 2.10 0.040 0.08 Cad 1.0–500 0.992 1.5 0.30 1.00 0.020 0.10 His 1.0–500 0.993 0.8 0.10 0.20 0.004 2.20 Tyr 1.0–500 0.999 0.6 0.01 0.05 0.001 1.40 Trp 1.0–500 0.996 0.9 0.03 0.10 0.002 0.18

Table II: Comparison of quantitative values of biogenic amines in tuna sample (mg/100 g)

Sample Spd Put Cad His Tyr Trp

MS method n.d. n.d. n.d. 1.6 0.1 0.2 tuna1) (RSD (%): n = 5) — — — (1.3) (1.8) (1.4) Fluorescence method N.D. N.D. N.D. N.D. N.D. N.D. MS method 0.6 3.6 11.7 77.4 5.0 0.20 tuna2) (RSD (%): n = 5) (2.1) (1.2) (1.1) (0.8) (0.9) (1.1) Fluorescence method N.D. 2.6 14.3 68.4 6.5 N.D. 1) preserved for 2 days under freezing; 2) preserved for 2 days under room temperature

Tosoh Bioscience and TSKgel are registered trademarks of Tosoh Corporation. QTrap is a registered trademark of Applied Biosystems/MDS SCIEX Instruments MDS Inc.

Tosoh Bioscience LLC 3604 Horizon Drive, Suite 100, King of Prussia, PA 19406 tel. (484) 805-1219, fax (610) 272-3028 Website: www.tosohbioscience.com 38 Industrial THE APPLICATION NOTEBOOK – JUNE 2011

Simultaneous Determination of Mineral Acids, Fluoride, and Silicate in Etching Baths by Ion Chromatography with Dual Detection

German Bogenschütz, Thomas Kolb, Beni Galliker, Andrea Wille, and Alfred Steinbach, Metrohm International Headquarters

The presented ion chromatographic method is used baths (also known as texturing baths) before being spiked with for the simultaneous determination of HF, HNO3, foreign atoms (P, B). Te etching solutions consist of various H2SO4, short-chain organic acids and H2SiF6 in acidic acids, which act as an oxidizing agent (HNO3), complexing texturing baths that are used in the wet chemical agent (HF), stabilizer and wetting agent (CH3COOH), or buf- etching process of solar cell production. Fluoride, ni- fers (H3PO4, CH3COOH) and determine the surface structure trate, sulfate, and acetate are determined by conduc- and thus the efficiency of the solar cells. Te replenishment of tivity detection after chemical suppression, while the components used up in the etching process extends the bath life silicon present in the form of hexafluorosilicate is de- and saves costs, though it does require knowledge of the exact tected spectrophotometrically as molybdosilicic acid composition of the bath, especially the concentration of silicon after derivatization in the same analysis. The analyti- and hexafluorosilicate. By using titration and ion chromatogra- cal results are validated by titration. phy (IC), it is possible to determine the key components quickly and precisely. nergy production from renewable sources such as biomass, Tis article describes an ion chromatographic method that Ebiogas, biofuels, water, wind, and solar power is becoming separates all relevant components in the bath on an anion- increasingly important in our energy-hungry society. Particular exchange column and identifies them by dual detection in a single interest is given to solar energy, which by human criteria is inex- run. After suppressed conductivity detection of the acid anions, haustible. Solar cells used in photovoltaic units convert the radia- the undissociated silicic acid reacts in a post-column reaction tion energy in sunlight directly into electric energy. (PCR) to form molybdosilicic acid, which is determined spectro- Solar cells are manufactured from ultrapure mono- or poly- photometrically at 410 nm. Te concentrations of fluoride and crystalline silicon wafers whose surface is treated in acid etching hexafluorosilicate are determined by way of a simple stoichiomet- ric calculation that is performed by the chromatography software.

Separation column: Metrosep A Supp 15 - 250/4.0 Column temperature: 45 ˚C 12 Eluent: 3.5 mmol/L Na2CO3 3.0 mmol/L NaHCO

uoride 3 11 Flow-rate: 0.7 mL/min Sample volume: 1.5 µL 10

Conductivity (µS/cm) 4

2 acetate nitrate 1

0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Time (min)

Figure 2: Conductivity chromatogram of a simulated etching bath with 25 mg/L fluoride, 20 mg/L acetate, and 10 mg/L nitrate. The Figure 1: 850 Professional IC Anion – MCS and 858 Professional undissociated orthosilicic acid is not recorded in the conductivity Sample Processor. detector. THE APPLICATION NOTEBOOK – JUNE 2011 Industrial 39

(a) Separation column: Metrosep A Supp 15 - 250/4.0 Column temperature: 45 ˚C

1.0 silicate 30 Eluent: 3.5 mmol/L Na2CO3

3.0 mmol/L NaHCO3 Flow-rate: 0.7 mL/min Sample volume: 1.5 µL sulphate 0.5 25 PCR reagent: 200 mmol/L HNO3

20 mmol/L Na2MoO4

Flow-ratePCR reagent: 0.25 mL/min 0.0 Wavelength: 410 nm 20

-0.5 15 uoride

Intensity (mV) -1.0 10 Conductivity (µS/cm) nitrate -1.5 5

-2.0 0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Time (min) Time (min) (b) Figure 3: UV–vis chromatogram of a 10 mg/L silicic acid standard. -0.0 Silicic acid is derivatized to molybdosilicic acid which is then detected -0.2 spectrophotometrically. -0.4 silicate -0.6 Instruments and Reagents -0.8 a) Instrument setup -1.0 • 850 Professional IC Anion – MCS with post-column reactor -1.2 • 858 Professional Sample Processor Intensity (mV) • Lambda 1010 UV/VIS Detector -1.4 • 771 IC Compact Interface -1.6 • MagIC Net chromatography software -1.8 b) Reagents and eluent -2.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Te standard solutions were prepared with CertiPUR standards Time (min) from Merck (SiO in NaOH; solutions of NaF and NaNO in ul- 2 3 trapure water) and the TraceCERT standard from Fluka (acetate Figure 4: (a) Conductivity and (b) UV–vis chromatograms of solution). All the standard and eluent solutions were prepared with etching bath sample 1 with 1:2000 dilution. The chromatographic ultrapure water with a speciﬁc resistance of more than 18 MΩ·cm. parameters are the same as those of the prior chromatograms. Etching bath samples were provided by a solar cell manufacturer from Germany. During the etching process, the concentrations of HF and

Etching of Silicon HNO3 in the etching bath decrease and the concentrations of wa- In the wet chemical etching of silicon surfaces, nitric acid is used ter and hexaTuorosilicate increase. To ensure a constant etching to oxidize silicon to form silicon dioxide which is further etched by rate and surface properties, the etching bath can be regenerated hydroTuoric acid. a few times by subsequent replenishment of spent acids. ﬃe in-

creasing concentration of H2SiF6, however, limits the number of + + + + 3 Si 4 HNO3 18 HF —> 3 H2SiF6 4 NO 8 H2O possible recycling cycles. ﬃis requires semi-continuous monitoring

Table 1: Comparison of the concentrations of a few selected bath components determined by ion chromatography and titration

Etching Bath Sample 1c Etching Bath Sample 2 Etching Bath Sample 3 Etching Bath Sample 4

a b a b a b a b Si HF HNO3 Si HF HNO3 Si HF HNO3 Si HF HNO3 IC (g/L) 3.3 22.4 216.6 34.8 47.2 248.4 17.6 98.9 504.4 19.3 94.8 516.8

Titration 3.7 26.4 224.3 28.1 48.4 255.9 17.6 86.2 476.1 18.1 80.7 478.1 (g/L)

RSDIC (%) 2.2 1.2 0.3 1.8 5.6 0.6 2.0 2.0 0.4 2.2 3.1 0.7 RSD Titration 1.2 3.3 0.3 0.5 2.4 1.1 0.4 1.3 0.6 0.2 1.8 1.1 (%)

a Calculated from uoride and hexauorosilicate concentrations. b Determination via nitrate concentration. c In etching bath sample 1, also 651 g/L sulfuric acid were detected. 40 Industrial THE APPLICATION NOTEBOOK – JUNE 2011 of the bath components, which can be achieved conveniently by Conclusion automated ion chromatography. Ion chromatography with dual detection allows determining the concentration of all relevant constituents of texturing baths for Dual Detection solar cell production in less than 30 min. Te acids consumed in Te acid anions present in the etching bath — mainly ﬂuoride and the texturing process can subsequently be replenished in a tar- nitrate, though sometimes also sulfate and acetate — are separated geted way. Tis extends the life of the etching baths, guarantees under alkaline elution conditions and determined by conductivity clean and reproducible wafer surfaces, cuts costs, and protects the detection (Figure 2). In the alkaline eluent, the hexaﬂuorosilicate is environment. converted into orthosilicic acid which is undissociated and therefore “invisible” in the conductivity detector. References (1) A. Henssge and J. Acker, Talanta 73, 220–226 (2007). + + Talanta 72, Na2SiF6 4 NaOH —> Si(OH)4 6 NaF (2) J. Acker and A. Henssge, 1540–1545 (2007). (3) M. Zimmer et al., 22nd European Photovoltaic Solar Energy Conference Te undissociated orthosilicic acid is determined by way of and Exhibition, Milan, Italy (2007). post-column reaction with an acid molybdate solution and subse- (4) G. Bogenschütz, C. Wilde, and C. Hack, Pittcon 2009 (http://www. quent UV-vis detection at 410 nm (Figure 3). metrohm.com/com/Applications, search for 8.000.6041EN).

+ 2− + + + H4SiO4 12 MoO4 24 H —> H4[Si(Mo3O10)4] 12 H2O

2− Te injection of SiF6 produces a fluoride peak in the conductivity detector and a silicate peak in the UV–vis detector. Te mass balance derived from the respective peak areas conffirms that 2− the concentration of SiF6 results stoichiometrically from the determined fluoride and silicate concentrations, provided there are no other sources of fluoride or silicate present. Tus the concentration of free HF can be calculated as the difierence between the total fluoride concentration and the fluoride concentration from the hexafluorosilicate:

− − [HF] = [F ]total − [F ]hexaﬂuorosilicate

Analysis of the Texturing Baths and Validation After being diluted at ratios between 1:1000 and 1:5000, four samples from diﬁerent texturing baths are analysed for their constituents by using the above IC method with dual detection. Figure 4 shows the chromatograms of etching bath sample 1 obtained by (a) conductivity and (b) UV–vis detection. Table I provides an overview of the concentrations of the relevant bath components determined by ion chromatography with Metrohm International Headquarters dual detection. For comparison, the concentrations obtained by Ionenstrasse, CH-9101 Herisau, Switzerland titration are also shown. Te potentiometric determination of tel. +41 71 353 85 04 E-mail: [email protected] acid concentrations and H2SiF6 content was carried out using aqueous acid–base titration with 1 mol/L NaOH solution. Website: www.metrohm.com THE APPLICATION NOTEBOOK – JUNE 2011 Industrial 41

Determination of Phthalate Esters in Child Care Products and Children’s Toys by Gas Chromatography–Mass Spectrometry (GC–MS) Richard Whitney, Shimadzu Scientic Instruments

he Consumer Product Safety Improvement Act of 2008 (CP- TSIA) requires testing of child care products and toys for selected phthalate esters by GC–MS. Te CPSC test method specifies GC– MS analysis in the SIM mode to monitor for low-intensity ions that are unique to speciflc phthalate esters, but full-scan mass spectra are valuable in qualitative identiflcation. Operation of the mass spectrometer in the FASST Scan/SIM mode allows concurrent acquisition of full-scan and SIM mass spectral data, to provide improved qualitative identiflcation while still maintaining optimum sensitivity. Two of the regulated compounds are mixtures of isomers, and quantitation ions for these compounds are of very low relative inten- Figure 1: Chromatogram of phthalate esters on RXI-5MS column. sity. So trace detection for these two substances is more challenging than for the other regulated compounds. Terephthalate esters (non-regulated isomers of phthalate esters) are frequently present at high concentration in real-world samples and interfere with identiflcation and quantitation of several of the regulated phthalate esters. Analysis on an alternate column provides chromatographic resolution of interferences from the regulated phthalate esters.

Results Analyses were conducted using a Shimadzu GCMS-QP2010S op- - erated in the FASST (Scan/SIM) mode. Two chromatographic col Figure 2: Chromatogram of phthalate esters on RXI-17 SilMS column. umns (0.25 mm × 30 m × 0.25 µm) were employed: RXI-5MS and RXI-17Sil MS (Restek Corporation). Te following instrument respectively. Under these conditions, di-n-octyl terephthalate conditions were employed for the analyses: (DOTP) coelutes with di-n-octyl phthalate. A chromatogram of the regulated phthalate esters on an RXI- GC Conditions 17SilMS column is shown in Figure 2. Under these conditions, • Injector Temp: 330 °C DOTP is well resolved from the regulated phthalate esters. Te • Injection Mode: Splitless (1 min) chromatographic separation of the phthalate esters is also very • Carrier: He; Constant Linear Velocity (36 cm/s) good on the RXI-17SilMS column, but elution of butyl benzyl • Column Temperature Program: 50 °C (hold 1 min), program phthalate is much later on this column. 30 °C/min to 280 °C, then 15 °C/min to 330 °C (hold 3 min) Conclusion MS Conditions In summary, qualitative identiﬂcation is improved and excellent sen- • Interface Temp: 330 °C sitivity is maintained for this method using FASST Scan/SIM, rela- • Ion Source Temp: 255 °C tive to SIM (only) data acquisition. One common chromatographic • Data Acquisition: FASST Scan/SIM Mode; Scan 50–500 0.15 interference on the method-speciﬂed chromatographic column was s, SIM (group) 0.10 s resolved from the target analytes using an alternate column.

A chromatogram of the regulated phthalate esters on an Shimadzu Scientic Instruments RXI-5MS column is shown in Figure 1. Te traces for diisono- 7102 Riverwood Drive, Columbia, MD 21046 nyl phthalate and diisodecyl phthalate on the chromatogram are tel. (800) 477-1227, fax (410) 381-1222 mass chromatograms for the quantitation ions m/z 293 and 307, Website: www.ssi.shimadzu.com 42 Pharmaceutical THE APPLICATION NOTEBOOK – JUNE 2011

Analysis of Pharmaceuticals in Whole Blood by Poroshell 120, Using a Modiﬁed Mini-QuEChERS Approach for Sample Preparation

Joan Stevens, Agilent Technologies, Inc.

etermination of pharmaceuticals in biological matrices is commonly employed in ADME (DMPK), clinical, and D x103 +ESI MRM Frag=112.0V [email protected] (278.2000 -> 117.0000) HQC_6.d 5.1 1 1 5 forensic analysis. Te main techniques used for drug monitor- 4.9 4.8 4.7 4.6 ing and analysis are immunoassays, LC, and GC methods. Mass 4.5 4.4 4.3 4.2 spectral chromatographic methods are the ﬁrst choice for many 4.1 4 3.9 3.8 applications based on their ﬂexibility, selectivity, sensitivity, quali- 3.7 3.6 3.5

3.4 Diltiazem tative, and quantitative capabilities. Analysis of pharmaceuti- 3.3 Tramadol 3.2 3.1 3 Lidocaine cals in biological samples requires sample preparation that can 2.9 Biperiden 2.8 2.7 2.6 range from simple protein precipitation (PPT) to more complex 2.5 2.4 2.3 2.2 solid-phase extraction (SPE). Tere is a need in classic sample 2.1 2 1.9 1.8 1.7 preparation for an approach to determine multi-classes of phar- 1.6 1.5 1.4 1.3 maceuticals in biological samples. Polymeric or mixed-mode SPE 1.2 1.1 1 0.9 sorbents which can isolate acidic, neutral, and basic drugs by hy- 0.8 0.7 Nortriptyline (ls) 0.6 Amitriptyline

0.5 Chlorpromazine drophobic or ion-exchange interactions have addressed this void 0.4 0.3 Naloxone Lorazepam

0.2 Oxazepam but there is always room for sample preparation techniques that 0.1 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.91 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.93 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.94 4.1 4.2 4.3 4.4 4.5 4.6 are rapid and inexpensive to implement. Counts vs. Acquistion Time (min) In 2003, Anastassiades and colleagues. reported a method for Figure 1: LC–MS-MS chromatograms of 10 ng/mL spiked whole the analysis of multiresidue pesticides in foods known as QuECh- blood sample after mini-QuEChERS extraction; AOAC (NaAc) and d-SPE (PSA). ERS; which equates to a quick, easy, cheap, eﬀective, rugged, and safe sample preparation approach (1). Te authors reported outstanding recoveries for a wide range of pesticide classes (1). Since 50, and 100 ng/mL of the component mix, add 20 uL of IS its inception there have been many reported articles employing stock solution (nortriptyline), two ceramic homogenizers then QuEChERS for the analysis of a wide range of compounds in- vortex. Then 2 mL of acetonitrile solutions (with or without cluding, but not speciﬁc to: antibiotics (2), toxins (3), contami- acid) was added and vortexed. A premixed amount (see Table nants (4), and pharmaceuticals (5). I) of the extraction salts is added and vigorously shaken, centrifuged at 5000 rpm for 5 min. One milliliter of the extract is Experimental transferred into a d-SPE tube (2 mL centrifuge tube) contain-

In this application note we describe an extension of the work ing 50 mg of PSA and 100 mg of MgSO4 for matrix clean-up; presented by Plössl and colleagues in 2006 (5), for the determi- vortex for 1 min and centrifuge at 18,000 rpm for 3 min. nation of pharmaceuticals in whole blood resulting in a modi- A 200 µL aliquot of the extract is transferred into a LC vial fied mini-QuEChERS procedure with LC–MS-MS analysis. containing 800 uL of water, vortexed and analyzed. A matrix Te experiments used whole blood from human donors and matched calibration curve from 10–250 ng/mL was employed evaluated both nonbuffered and buffered extraction salts of the to determine recovery. QuEChERS method, namely the Original, AOAC 2007.01 and EN 15662. Modifications to the acetonitrile (extraction sol- Results vent) used in the first step (extraction/partitioning) were also The experiments showed that the use of ACN (0.4% FA) as evaluated (Table I). Te experiments were performed using nine the extraction solvent offered a better lysed sample versus the different pharmaceuticals (lidocaine, tramadol, amitriptyline, other extraction solvents where the sample became a solid biperidene, oxazepam, lorazepam, chlorpromazine, diltiazem, mass. The AOAC buffered salts yielded the cleanest extract, and naloxone) which offer a broad range of hydrophobicity and visually and was chosen to be used with the d-SPE containing K dissociation constants, p a. 50 mg PSA, 150 mg MgSO4 for the extraction of the phar- The general procedure is as follows. A 1 mL aliquot of whole maceuticals in whole blood. Typical chromatogram shown in blood was added to a centrifuge tube and spiked with appro- Figure 1 is the analysis of a 10 ng/mL whole blood sample after priate volume from a concentrated stock mixture to yield 25, mini-QuEChERS procedure. THE APPLICATION NOTEBOOK – JUNE 2011 Pharmaceutical 43

Table I

Sample, 1 mL Extraction Solvent Extraction Salts, mg d-SPE Observation

WB ACN none none Sample: solid mass WB ACN, 1 % AA none none Sample: solid mass WB ACN, 0.4% FA none none Sample: loose particles WB ACN, 0.4% FA Non-buffered, 500 none Dark extract WB ACN, 0.4% FA AOAC, 500 none Clear extract WB ACN, 0.4% FA EN, 650 none Dark extract WB ACN, 0.4%FA Non-buffered, 500 50 mg PSA, 150 mg MgSO4 Clear extract WB ACN, 0.4%FA AOAC, 500 50 mg PSA, 150 mg MgSO4 Clear extract WB ACN, 0.4%FA EN, 650 25 mg PSA, 150 mg MgSO4 Clear extract WB ACN, 0.4%FA EN, 650 50 mg PSA, 150 mg MgSO4 Clear extract WB= whole blood; AA= acetic acid; FA= formic acid

Table II: Recovery and reproducibility

Compound 50 ng/mL Spiked 100 ng/mL Spiked

Recovery RSD Recovery RSD

Lidocaine 98.7 15.7 100 11.8 Tramadol 105 3.0 104 8.2 Amitriptyline 104 2.1 104 8.2 Biperidene 97 4.5 99 8.2 Oxazepam 77.0 9.2 78 8.6 Lorazepam 81.9 6.8 81.8 8.6 Chlorpromazine 110 10.3 105 6.3 Diltiazem 88.1 2.7 91.7 8.3 Naloxone 80.6 9.0 75.5 7.7

Method Parameters from whole blood oﬁers an alternative sample preparation tech- Flow Rate: 0.4 mL/min nique that is easily implemented into laboratories. Poroshell 120 Mobile Phase: A: 5 mM Ammonium acetate, pH 5; 20:80 is an excellent column choice for this analysis, in part because it MeOH:Water; has standard 2 µm frits and is more forgiving for more complex B: 5 mM Ammonium Acetate in ACN samples relative to a sub-2 µm column. It has mass transfer such Column: Poroshell 120 EC-18, 2.7 µm, 2.1 × 100 mm that it acts very much like a sub-2 µm particle LC column, but Injection: 10 µL doesn’t have the back pressure. Te eﬂcient mass transfer equates Instrument: Agilent LC 1200, 6460 LC–MS-MS, positive mode, with faster analysis time and higher throughput, with optimum GT (°C): 300 resolution. GF (l/min): 7 Nebulizer (psi): 40 References SGT (°C): 400 (1) M. Anastassiades, S.J. Lehotay, D. Štajnbaher, and F.J. Schenk, J. SFG (l/min): 12 AOAC Int. 86, 4121 (2003). Capillary (V): 3500 (2) G. Stubbings and T. Bigwood, Anal. Chim. Acta 637, 68 (2009). NV (V): 500 (3) R.R. Rasmussen, I.M.L.D. Storm, P.ZH. Rasmussen, J. Smedsgaard, Gradient: 20 to 75% B over 5.5 min and K.F. Nielsen, Anal. Bioanal. Chem. 397, 765 (2010). Te mini-QuEChERS approach has shown to be a viable sam- (4) D. Smith and K. Lynam, GC/µECD analysis and confirmation of ple preparation technique for the analysis of pharmaceuticals in PCBs in fish tissue with Agilent J&W DB-35ms and DB-XLB GC biologicals like whole blood as shown with average recoveries > columns Agilent Technologies, 5990-6236EN, (2010). 90% and 7% RSDs for 50–100 ng/mL, in Table II. (5) F. Plössl, M. Giera, and F. Bracher, J. Chrom. A 1135, 19 (2006).

Conclusion Te mini-QuEChERS sample preparation is a simple, easy, and Agilent Technologies, Inc. cost eﬁective approach, requiring minimal sample preparation 2850 Centerville Road, Wilmington, DE 19808 expertise, solvent, equipment, and is a green technology. Te tel. (800) 227-9770, fax (302) 633-8901 mini-QuEChERS approach for the extraction of pharmaceuticals Website: www.agilent.com 44 Pharmaceutical THE APPLICATION NOTEBOOK – JUNE 2011

LC Analysis of Aminoglycoside Antibiotics Kanamycin and Amikacin

Lipika Basumallick, Deanna Hurum, and Jeff Rohrer, Dionex Corporation

anamycin and amikacin are aminoglycoside antibiotics used to treat serious bacterial infections. Amikacin is used for in- K 108 fections resistant to other aminoglycosides because it is less suscep- 1 tible to enzymatic reactions. PuriTed kanamycin from Streptomyces kanamyceticus is mainly kanamycin A, from which amikacin is synthesized by acylation of the kanamycin A amino group with L-(-)-γ-amino-α-hydroxybutyric acid (L-HABA). Kanamycin A 2 and L-HABA are, therefore, expected impurities in amikacin synthesis. nC fiese antibiotics must meet speciTed purity criteria before clini- A cal use. fie United States Pharmacopeia (USP) monographs for the assay of kanamycin and amikacin use high-performance, anion- B exchange chromatography with pulsed amperometric detection C (HPAE-PAD) (1,2). Here we evaluate the USP assays using dispos- 18 able gold on polytetrafluoroethylene (PTFE) working electrodes 0 2.5 5 7.5 10 with a 4-potential waveform suitable for use with these electrodes. Minutes Disposable electrodes are more convenient than conventional elec- Figure 1: Typical chromatograms of: (A) resolution solution (ka- trodes because they are easier to install and do not require time- namycin 0.008 mg/mL and amikacin 0.02 mg/mL), (B) commercial consuming electrode polishing. Compared to other disposable gold kanamycin A sulfate sample, (C) commercial amikacin sample. electrodes, the gold on PTFE electrodes have longer lifetimes and Peaks: 1. kanamycin (8 µg/mL), 2. amikacin (20 µg/mL). can operate at the hydroxide concentration used in the USP assays. peak at ~21 min is similar to a late-eluting peak reported for de- Experimental graded streptomycin (3), which could interfere with quantiTcation A Dionex ICS-3000 system and Chromeleon® Chromatography if a shorter run time is used. Under basic conditions, amikacin loses Data System software were used in this study. Kanamycin and its acetylated group resulting in a kanamycin-like molecule. amikacin (20 µL) were separated using a CarboPac® MA1 (USP L47) column set with 115 mM sodium hydroxide at 0.5 mL/min. Conclusion fiis method matches and exceeds USP requirements, achieves Results good sensitivity, and has high sample throughput. Additionally, Figure 1 shows the <10 min separation of kanamycin and ami- using disposable electrodes provide shorter equilibrium time and kacin. Peak resolution was >4, (exceeding the USP requirement better electrode-to-electrode reproducibility. of 3); asymmetry for both antibiotics was 1.1 (USP requirement of <2). Retention time RSDs were 0.16 for kanamycin and 0.07 References for amikacin for nine replicate injections (USP requirement of (1) United States Pharmacopeia (USP), Kanamycin Sulfate, USP34-NF29, 3244. <0.3). Intra- and between-day peak area precisions (RSDs) were (2) United States Pharmacopeia (USP), Amikacin Sulfate, USP34-NF29, 1846. 0.99 and 1.3 for kanamycin and 1.2 and 2.3 for amikacin. fiese (3) Dionex Corporation, Application Note 181; Sunnyvale, CA. values suggest that this method effectively assays these antibiotics without column regeneration, and that using a disposable elec- CarboPac, Chromeleon, and UltiMate are registered trademarks of Dionex trode with its associated waveform meets the current USP assay Corporation. requirements. To demonstrate method capability for stability assays, these anti- Dionex Corporation biotics were studied after exposure to elevated temperatures under 1228 Titan Way, P.O. Box 3603, Sunnyvale CA 94088 acidic or basic conditions. For both antibiotics, most of the degra- tel. (408) 737-0700, fax (408) 730-9403 dation products eluted within 10 min. An unidentiTed late-eluting Website: www.dionex.com THE APPLICATION NOTEBOOK – JUNE 2011 Pharmaceutical 45

Enantiomeric Separation of Proton Pump Inhibitors Using Polysaccharide-Based Chiral Stationary Phases in Reversed-Phase HPLC Conditions Liming Peng, Michael McCoy, Jeff Layne, and Kari Carlson*, Phenomenex, Inc.

The chiral analysis and purication of proton pump inhibitors (PPIs) has become a popular topic as more of (a) these drugs fall out of patent protection each year. Due 19718 Lansoprazole/Dexlansoprazole on Lux Cellulose-4 90:10:01 MeOH:20mMAmmbi:DEA to their wide range of enantiomeric selectivity, a set of 1 polysaccharide-based chiral stationary phases (CSPs) was screened to identify methods for the successful 2 enantioseparation of four benzimidazoles; rabeprazole, lansoprazole, omeprazole, and pantoprazole. = 1.4

Experimental 0 2 4 6 8 10 min

(b) A screening protocol was performed on Tve diﬁerent polysaccha- 19721 Omeprazole/Esomeprazole on Lux Cellulose-2 80:20:01 AcN:20mMAmmbi:DEA ride-based chiral stationary phases under reversed phase condi- 1 tions using Lux® 5 µm columns (Table I). To reduce solvent usage, 150 × 4.6 mm columns were used for the initial screen. Once 2 the best chiral stationary phase was identiTed for each sample, the mobile phase conditions were further optimized on 250 mm = 1.54 length columns of the same particle size and diameter to get increased resolution. Additional optimization of the chromato- 0 2 4 6 8 min graphic conditions with respect to retention, enantioseparation, (c) 19716 Rabeprazole on Lux cellulose-4 and resolution was achieved by variation of the mobile phase con- 80:20:01 MeOH:20mMAmmbi:DEA stituents at room temperature. 1

Results and Conclusions 2

ﬂe HPLC analysis of the four benzimidazoles allows for fast = 1.70 and accurate identiTcation of their enantiomers. We have shown analytical techniques in reversed phase conditions conducive to 0 2 4 6 8 10 12 14 16 18 min

MS detection for each. Preparative techniques can be explored (d)

19717 Pantoprazole on Lux Cellulose-2 using these analytical techniques by Trst optimizing mobile 60:40:01: AcN:20mMAmmbi:DEA phase conditions for best resolution and selectivity followed by 1 loading tests. 2

Table I: Lux® chiral stationary phases = 1.21 Name Selector

Amylose-2 Amylose tris(5-chloro-2-methylphenylcarbamate) 0 2 4 6 8 10 12 14 Cellulose-1 Cellulose tris(3, 5-dimethylphenylcarbamate) Figure 1: Figures 1a through 1d show the racemic mixtures of Cellulose-2 Cellulose tris(3-chloro-4-methylphenylcarbamate) lansoprazole, omeprazole, rabeprazole, and pantoprazole. In Figure 1a, the single enantiomer, dexlansoprazole, is labeled as Cellulose-3 Cellulose tris(4-methylbenzoate) peak 2. In Figure 1b, the single enantiomer, esomeprazole, is la- Cellulose-4 Cellulose tris(4-chloro-3-methylphenylcarbamate) beled as peak 1.

Phenomenex, Inc. 411 Madrid Avenue, Torrance, CA 90501 tel. (310) 212-0555, fax (310)328-7768 Website: www.phenomenex.com 46 Polymer THE APPLICATION NOTEBOOK – JUNE 2011

Chemical Composition Analysis of Polyoleﬁns by Multiple Detection GPC-IR5

Wallace W. Yau1, Alberto Ortín2, and Pilar del Hierro2, 1Polymer Char Scientiﬁc Consultant and 2Polymer Char

olyolefin (PO) is the largest volume industrial polymer in Pthe world for making a wide range of commercial products that touch nearly every aspect of our daily lives, such as auto- mobile parts, pipes, packaging films, household bottles, baby diapers, and so on. Polyolefin as a group includes high-density and low-density polyethylene (HDPE, LDPE), polypropylene (PP), EP rubber, and linear low density (LLDPE) copolymers of ethylene with alpha olefins (propylene, 1-butene, 1-hexene, 1-octene). Tough chemically simple, being made up of only carbon and hydrogen atoms, PO products derive their wide range of end-use properties from their semicrystalline structure. Te ability to incorporate co-monomers in PE to create short chain branches (SCB) makes it possible to control the polymer crys- tallinity and crystalline morphology, and thus control the rigid- ity and flexibility of PO products. Te controlling factor also depends on the SCB variations across the molar mass distribu- Figure 1: Molar mass distribution curves of the three samples. tion (MMD). In dilute solution, the effect of SCB causes a reduction in polymer coil size with an increase of methyl groups off the backbone. Such structural changes can be studied by high temperature size exclusion chromatography (SEC, also known as gel permeation chromatography, or GPC), equipped with a multiwavelength infrared detector (IR5 MCT) (1), a light scattering (LS) detector, and a viscosity detector in the triple-detector GPC configuration (TD-GPC) (2). Tese techniques allow the analysis of SCB variation across MMD as it is illustrated by using the homopolymer PP and PE samples described in Table I together with their blend.

Experimental Conditions • Eluent: TCB (stabilized with 300 ppm BHT). × × • Columns: 3 PLOlexis, 13 µm, 300 7.5 mm. Figure 2: Analysis of the test blend sample. IR5 MCT CH2 and CH3 • Flow rate: 1.0 mL/min. absorbance channels versus Log M (solid lines); ethylene % variation across MMD, predicted versus IR5 MCT measured (top part). • Injection volume: 200 µL. • Sample concentration: 2 mg/mL. • Temperature: dissolution 160 °C; columns: 145 °C; detectors Results and Discussion 160 °C; MALLS 150 °C. 1. Ethylene-propylene Composition Analysis by GPC-IR5 MCT: • Chromatographic system: GPC-IR by Polymer Char. Te MMD obtained by GPC-LS for the three samples in Table • Detectors: IR5 MCT infrared detector by Polymer Char. I are shown in Figure 1, where the IR5 MCT detector total CH 4-capillary viscometer by Polymer Char. channel is used to record the sample concentration across the Multiple angle laser light scattering HELEOS™ GPC elution curve. 8+ by Wyatt Technology. For studying chemical composition variation across MMD, • Data handling: GPC One by Polymer Char. the combination of detector signals from the IR5 MCT methylene THE APPLICATION NOTEBOOK – JUNE 2011 Polymer 47

Table I: Description of test samples used in this study M M M M Sample w (g/mol) n (g/mol) w/ n IV (dL/g) Broad PP 530,000 141,200 3.75 2.44 Broad HDPE 51,900 20,600 2.52 1.01 Blend (50/50) 291,000 36,000 8.09 1.73

Both light scattering and viscosity detector signals are more responsive to higher MM molecules. This is the rea- son that the LS and viscosity curves of the blend sample are skewed more to the higher MM region than the IR concentration curve shown in Figure 3. The corresponding MH plots for the three samples are shown in Figure 4. The MH line of the PP sample is seen to have a shift to higher MM from that of the PE sample. This is because of the extra methyl group off the chain backbone Figure 3: Triple detector signal overlay of the test blend sample. that leads to higher MM and lower IV for the PP sample. Te MH plot of the blend sample in Figure 4 is seen to go through a transition from the PE line to the PP line as the MM increases. ﬁe experimental data agrees with the predicted curve very well. ﬁis agreement supports the use of TD-GPC for studying polymer blends and block copolymers of EP, PE, EO (ethylene/octene), and so on.

Conclusions ﬁe micro-structural diTerence of Polyethylene and Polypropylene as well as the compositional variation in EP can be detected by GPC- IR5 MCT and TD-GPC, as it has been validated by using a test sample of known composition created by a 50-50 PE-PP blend. Tis result of SCB variation across MMD provides the additional polymer structural information that complements perfectly the CRYS- TAF (crystallization analysis fractionation), TREF (temperature rising elution fractionation), and CEF (crystallization elution frac- Figure 4: Mark-Houwink plot of homopolymer PP, HDPE, and tionation) results which give SCB distribution across the crystalliza- their blend. tion temperatures (3–5) and to the cross-fractionation technique (6).

(CH2) and methyl (CH3) channels, shown in the bottom part References of Figure 2, is used. Since the PP part in the blend sample is of (1) J. Montesinos, R. Tarín, A. Ortín, and B. Monrabal. ICPC 2006 Houston. higher MM than the PE part, a higher methyl signal is seen in (2) A. M. Striegel et al., Modern Size-Exclusion Liquid Chromatography (Wiley, the higher MM end of the sample, as expected. New York, 2009). Te trend of changing ethylene percent, seen in Figure 2, can (3) W. W. Yau and D. Gillespie, Polymer 42, 8947–8958 (2001). be predicted a priori from the co-adding of the original PE and (4) B. Monrabal, in Encyclopedia of Analytical Chemistry, R. A. Meyers, Ed. PP GPC elution curves. ﬁis ethylene to propylene transition (John Wiley & Sons Ltd., 2000). also can be determined from the diTerence between the CH2 and (5) B. Monrabal, L. Romero, N. Mayo, and J. Sancho-Tello, Macromolecular 282, CH3 signals of the IR5 MCT detector (1). ﬁe validity of this Symposia 14–24 (2009). analytical approach is supported by the agreement seen in the (6) A. Ortin, B. Monrabal, and J. Sancho-Tello, Macromolecular Symposia 257, Figure between the predicted and the IR5 MCT results. 13–28 (2007).

2. Ethylene-propylene Composition Analysis by TD-GPC: In the triple detector GPC, the ratio of LS signal over sample concentration at each GPC elution volume gives a measure of the polymer weight-average MM (Mw). Similarly, the ratio of the viscosity detector signal over sample concentration gives the polymer intrinsic viscosity (IV). The Polymer Char plot of Log IV versus Log Mw by LS is called the Mark- Gustave Eiffel 8, Paterna, Valencia, 46980, Spain Houwink (MH) plot, which is an often-used tool to reveal tel. +34 96 131 8120, fax +34 96 131 8122 polymer branching and conformational structures (3). E-mail: [email protected], Website: www.polymerchar.com 48 General THE APPLICATION NOTEBOOK – JUNE 2011

Isolation of Benzylideneacetophenone from a Crude Reaction Mixture

A. Talamona, BUCHI Corporation

enzylideneacetophenone forms the central core for a variety of Bimportant biological compounds, known collectively as chal- cones. Many of these compounds have significant medicinal value as they demonstrate antibacterial, antifungal, antitumor, and anti- inflammatory properties. Tey are synthesized by an aldol condensa- tion between a benzaldehyde and an acetophenone in the presence of NaOH as catalyst. In this application note, benzylideneacetophenone is isolated from a crude reaction mixture using flash chromatographic technique, including consideration for scale-up.

TLC of the Crude Reaction Mixture n Figure 2: Separation 1. TLC on silica gel Si60F254, developed in -hexane/ethyl acetate 19:1, detection UV 254nm.

Sepacore Conﬁguration Cartridge 12 × 150 mm, prepacked with silica gel 60, 40–63 μm 2 Pump modules C-605 Fraction collector C-660 Control-Unit C-620 with SepacoreControl software UV Photometer C-635

Separation Conditions Eluent: see below Flow rate: 10 mL/min Figure 3: Separation 2. 100 mg crude mixture, dissolved in n-hexane and some drops of toluene (solubility of the sample in pure n-hexane too low) Loading: approx. 0.4 mL Separation 2 Eluent: n-hexane with 1%, 2% and 3% ethyl acetate, step gradi- 1. Test runs ent. Each step was initiated after a component was completely Separation 1 eluted (at the end of a peak). Eluent: n-hexane with 2% ethyl acetate, isocratic TLC Check n TLC on silica gel Si60F254, developed in -hexane/ethyl acetate 19:1, detection: UV 254 nm 1 = Benzaldehyde 2 = Benzylideneacetophenone 3= Benzophenone Recovery Fraction 7: 65 mg 2. Scale-up Calculation of the scale-up factor Cartridge 12 × 150 mm: cross sectional area = 1.13 cm2 Cartridge 40 × 150 mm: cross sectional area = 12.56 cm2 Figure 1: TLC of crude mixture. Scale-up-factor = 12.56 cm2 / 1.13 cm2 = 11.1 ≈ 10 THE APPLICATION NOTEBOOK – JUNE 2011 General 49

Figure 4: TLC check of Separation 2.

Figure 5: Scale-up separation.

Figure 6: TLC check of scale-up.

Separation conditions Eluent: n-hexane with 1%, 2% and 3% ethyl acetate, step gradient Sample: 1 g crude mixture, dissolved in n-hexane/toluene 7:3 (solubility of the sample in n-hexane too low, toluene is eluted as front peak) Injection volume: 2 mL TLC-Check n TLC on silica gel Si60F254, developed in -hexane/ethyl acetate 19:1, detection: UV 254 nm Recovery Fraction 13–16: 512 mg crystalline product, mp 57 °C

BUCHI Corporation 19 Lukens Drive, Suite 400, New Castle, DE 19720 tel. (302) 652-3000, fax (302) 652-8777 Website: www.mybuchi.com 50 General THE APPLICATION NOTEBOOK – JUNE 2011

Characterization of an Unknown Cannabinomimetic Compound in an Herbal “Incense” Sample by Gas Chromatography–High Resolution Time-of-Flight Mass Spectrometry Joe Binkley, LECO Corporation

ynthetic cannabis products have garnered a great deal of me- Sdia attention recently. Countless products are being sold at smoke shops, convenience stores, and online sites. fiese products are labeled “not for human consumption,” but have been reported to have eflects similar to cannabis when smoked. fiese reports have prompted testing in crime laboratories across the country. Some of the products tested have been conTrmed to contain synthetic cannabinoids. fie identiTcation of these compounds by GC–MS can be challenging, as most of them are not present in commercially available mass spectral libraries. A particular product that received a great deal of media attention in the midwest- ern United States during 2010 was “Mr. Smiley.” fiis product was reported to contain natural substances such as Mullein and Damiana leaf. It was also reported that when smoked, this product exhibited a “THC-like” high. Shortly before this product was Figure 1: GC–HRT TIC and extracted ion chromatogram for the analyte identied as 1-pentyl-3-(1-naphthoyl)indole or JWH-018. pulled from store shelves, some was purchased for analysis to de- Notice a mass accuracy of 0.2 ppm was achieved. termine which compounds may be present. fie sample was initially analyzed by GC–TOFMS, but the most abundant analyte detected was not present in the commercially avail- Flight Path: High Resolution (R = 25,000 FWHM) able NIST mass spectral database. Acquisition rate: 20 spectra/s fiis application note shows the use of LECO’s Pegasus GC– Source temperature: 250 °C HRT high resolution time-of-ffight mass spectrometer to provide an elemental composition based on accurate mass data, which Results ultimately lead to identiTcation of the unknown compound. fie accurate mass data provided by the Pegasus GC-HRT was

consistent with the formula C24H23NO. A literature search lead to Sample identiTcation of this formula as 1-pentyl-3-(1-naphthoyl)indole, or A 1-gram sample of “Mr. Smiley” was weighed into a 20-mL vial. A JWH-018, one of the most common synthetic cannabinoids used in 10-mL aliquot of ethyl acetate was added and the sample was soni- “herbal incense” products. cated for 10 min. Conclusions Experimental fiis application shows how the high resolving power and mass fie “Mr. Smiley” extract was analyzed on a LECO Pegasus GC– accuracy of the Pegasus GC–HRT facilitate identiTcation of sub- HRT system using the conditions shown below. stances absent from commercially available mass spectral libraries. fiis ability will be required as new abused substances are continu- GC: Agilent 7890 ally developed. Injection: 1 μL, split 50:1 at 275 °C Column: Rtx-5, 10 m x 0.18 mm x 0.2 μm Carrier gas: helium at 1.0 mL/min Oven: 60 °C to 330 °C @ 50 °C/min, hold 10 min Transfer Line: 300 °C LECO Corporation MS: LECO Pegasus GC–HRT 3000 Lakeview Avenue, St. Joseph, MI 49085 Acquisition Delay: 120 s tel. (269) 985-5496, fax (269) 982-8977 Saved mass range: 40–550 m/z Website: www.leco.com THE APPLICATION NOTEBOOK – JUNE 2011 General 51

Rapid Analysis of Amphetamines in Biological Samples

Michael Rummel, Matthew Trass, and Jeff Layne, Phenomenex, Inc.

This study demonstrates a faster method of analysis when coupling a simpliﬁed and effective SPE procedure 7,8 Intensity, cps 5,6 with a core-shell HPLC column and LC–MS-MS detection. 9,10 1.6e4

1.4e4 3 4 ecently, the Substance Abuse and Mental Health Services Adminis- 1.2e4 tration (SAMHSA) has lowered the amphetamine drug class cutoﬂ R 1.3e4

level and has added MDMA, MDA, and MDEA to the panel. As a re- 8000 sult, there will undoubtedly be an increase in positive result conforma- 1 6000 2

tional testing, placing additional stress on the toxicology laboratory. We 4000

demonstrate how a specialized SPE sorbent and method, together with a 2000

high eTciency core-shell HPLC column and LC–MS-MS can decrease 0 analysis time and increase sample throughput. 0.5 1 1.5 2 2.5 3.5 3.5 4 4.5 min Figure 1: Rapid analysis of amphetamines by LC–MS-MS using a Experimental Conditions Kinetex XB-C18 2.6 μm, 50 × 2.1 mm. Urine samples were pretreated and cleaned up using Strata™-X-Drug B solid phase extraction tubes. (It is important to note that neither Results a conditioning or equilibration step is required using this specialized ffe Strata-X-Drug B SPE procedure consists of only 5 steps, a load, SPE sorbent and method.) After diluting samples by a factor of 20 to two washes, dry, and an elution step, compared to the 9 step pro- bring the concentration into a suitable range for analysis, samples were cedure called for by the traditional SPE method. ffe new Strata-X- injected onto the LC–MS-MS. Drug B resulted in a 7 min time savings and 11 mL solvent savings. Analyses were performed using an HP 1100 LC system (Agilent Figure 1 demonstrates the ability of the Kinetex 2.6 μm XB-C18 Technologies, Palo Alto, California) with an upper pressure limit of columns to rapidly screen amphetamines while providing eTcient 400 bar, equipped with an API 3000™ LC–MS-MS detector. conformational results. Complete analysis is completed in less than 3 All analytes were present at a concentration of 10 ng/mL each and min with an average peak width of 0.147. are listed in order of elution. LC analysis was completed at operating pressures less than 400 Column: Kinetex® 2.6 μm XB-C18 100 Å bar and may therefore be used on any HPLC system without the Dimensions: 50 x 2.1 mm need for specialized ultra-high pressure equipment. Mobile Phase: A: 5 mM Ammonium formate with 0.1 % Formic acid Conclusion B: Methanol with 0.1 % Formic acid A rapid cleanup and analysis for amphetamines was developed us- Gradient: Time (min) %B ing Strata-X-Drug B SPE and a Kinetex XB-C18 HPLC column 0.00 10 which can dramatically improve the eTciency of toxicology labora- 1.00 70 tories while simultaneously reducing cost due to solvent consump- 3.00 70 tion. ffese time and solvent savings can be multiplied to provide Flow Rate: 0.4 mL/min substantial savings when running multiple samples in a high capacity Detection: API 3000™ MS/MS, ESI negative (ESI-) laboratory. Sample: 1. D11-Amphetamine; 2. Amphetamine; 3. D14-Metham- phetamine; 4. Methamphetamine; 5. D5-MDA; 6. MDA; 7. D5- MDMA; 8. MDMA; 9. D5-MDEA; 10. MDEA

Request a copy of Technical Note TN-1096 for complete Phenomenex, Inc. method details. 411Madrid Avenue, Torrance, CA 90501 tel. (310) 212-0555, fax (310) 328-7768 Website: www.phenomenex.com 52 General THE APPLICATION NOTEBOOK – JUNE 2011

Improved Analysis of Preservatives in Cosmetics Using a Unique C18 Core-Shell Phase

Terrell Matthews, Zeshan Aqeel, and Jeff Layne, Phenomenex, Inc.

This study evaluates the performance of a C18 core- shell phase that incorporates a C18 ligand with iso- 19498 ® butyl side chains. The Kinetex XB-C18 HPLC/ UHPLC mAU

14 column delivers a fast and effective separation of sev- 200 eral common preservatives in cosmetics. 12 8 6 15 5 4 9 10 13 11 reservatives prevent product deterioration and deter any 100 possible health risks microorganisms may cause to the 1 2 3 P 7 consumer; however, a disadvantage of using such agents is that they may cause adverse efiects, such as allergic responses and 0 irritation. flere is a need today for a fast, eTcient, and selec- 0 2 4 6 8 10 12 14 16 min tive HPLC method to screen for such common and dangerous Time (min) preservatives. Figure 1: High performance separation of 15 preservatives on Kinetex 2.6 µm XB-C18. Experimental Conditions for a gradient separation and high peak capacity values indicate in- Analyses performed using an HP 1100 LC system (Agilent Tech- creased analyte resolution over a given analysis time. nologies, Palo Alto, California) with an upper pressure limit of fle core-shell particle morphology allows for faster mass trans- 400 bar, equipped with a UV detector. fer of analytes into and out of the stationary phase as compared to Column: Kinetex 2.6 μm XB-C18 100 Å fully-porous silica particles. In addition, the very narrow particle Dimensions: 100 × 4.6 mm size distribution inherent in core-shell silica particles, as compared Mobile Phase: A: Water with 0.1 % TFA to fully-porous particles, results in less band broadening. B: Acetonitrile with 0.1 % TFA Kinetex 2.6 μm XB-C18 was able to separate the chlorinated com- Gradient: (85:15) A/B for 20 min, then to (15:85) A/B pounds triclosan and triclocarban, which can be challenging. flis is Flow Rate: 1.5 mL/min likely due to the unique XB-C18 selectivity. fle Kinetex XB-C18 Column Temperature: 30 °C chemistry contains protective di-isobutyl side chains that shield the sil- Detection: UV @ 214 mm (ambient) ica surface. In addition, the surface is endcapped with trimethylsilane. Injection Concentration: 50 μg/mL Analysis of preservatives in cosmetics was accomplished at Sample: 1. Benzyl alcohol; 2. Phenoxyethanol; 3. Sorbic acid; 4. Benzoic an operating pressure under 400 bar and may therefore be used acid; 5. Methyl paraben; 6. p-Anisic acid; 7. Dehydroacetic acid; 8. Sali- on conventional HPLC systems without the need for specialized cylic acid; 9. Ethyl paraben; 10. Isopropyl paraben; 11. Propyl paraben; ultra-high pressure equipment. 12. Isobutyl paraben; 13. Butyl paraben; 14. Triclosan; 15. Triclocarban Conclusion Results An ultra-high performance liquid chromatography method has Cosmetic products can only use a limited number of preservatives been developed for the simultaneous determination of 15 preser- selected from a positive list, Annex VI of the Cosmetics Direc- vatives in cosmetics. fle method was developed to achieve the tive, which also deffnes preservative maximum permitted levels best balance of analysis time and separation. and areas of use. fle esters of parahydroxybenzoic acid (paraben), fle Kinetex XB-C18, 2.6 μm column provided high peak ca- methyl paraben, ethyl paraben, propyl paraben, and butyl para- pacity and the unique selectivity of the XB-C18 was well-suited ben are standard substances among the preservatives list. for the application. Figure 1 illustrates the ability of the Kinetex XB-C18, 2.6 μm core-shell column to rapidly screen and separate all 15 compounds. Phenomenex, Inc. In this separation, the Kinetex XB-C18 column ofiered a peak ca- 411Madrid Avenue, Torrance, CA 90501 pacity of 445, which was higher than any other column evaluated tel. (310) 212-0555, fax (310) 328-7768 in the experiment. Peak capacity is the best measure of performance Website: www.phenomenex.com THE APPLICATION NOTEBOOK – JUNE 2011 General 53

Rapid and Streamlined Screening of Barbiturates

Matthew Trass, Seyed Sadjadi, Jeff Layne, Sky Countryman, Michael Rummel, and Erica Pike, Phenomenex, Inc.

arbiturates are a class of antidepressants whose abuse and ad- Table I: Solid phase extraction: Strata-X-Drug N 100 mg/6 mL Bdiction by recreational users has become a widespread prob- (part no. 8B-S129-ECH) lem (www.nlm.nih.gov). In our work we strived to streamline the Condition/ Equilibrate NOT REQUIRED barbiturate screening process to provide a fast, cost-efiective, and Dilute 2 mL of spiked urine with 2 mL 100 mM Sodium acetate buffer Load reproducible method from start to flnish for forensic labs who are (spiked with I.S. at 300 ng/mL). Load involved in high-throughput processing. diluted sample onto SPE sorbent Wash 1 2 mL 0.1 N HCl Experimental Conditions Wash 2 2x 2 mL Methanol/0.1 N HCl (30:70) Phenobarbital, butalbital, pentobarbital, amobarbital, and secobarbital Dry 10 minutes at 10 in. of Hg 2 mL Ethyl acetate/Isopropanol were spiked into urine samples at 40, 100, and 125% of cutofi level (300 Elute (85:15) ng/mL). Te spiked urine samples were then subjected to a pretreatment Dry down To dryness at 50 °C followed by SPE on a 100 mg/6 mL Strata™-X-Drug N tube as specifled Reconstitute 1 mL of 10 % Acetonitrile in Table I. After extraction, the barbiturates were analyzed by LC–MS- MS using a Kinetex® 2.6 μm C18 100 × 2.1 mm core-shell HPLC/UH-

PLC column with the MS operating in APCI negative mode (Figure 1). 3,4 2.6e6 2.5e6 6 2.4e6 7 2.3e6 5 Results and Discussion 2.2e6 2.1e6 1.0e6 1,2 While developing an extraction method for the barbiturate spiked urine 1.9e6 1.8e6 samples it was discovered that conditioning the Strata™-X-Drug N SPE 1.7e6 1.6e6 1.5e6 sorbent was not necessary as the elimination of this step did not afiect 1.4e6 1.3e6 recovery. By skipping this step we were able to save both solvent and 1.2e6 1.1e6 time which could drastically improve the effciency of a high throughput cps intensity, 1.0e6 9.0e5 8.0e5 lab. Te more effcient SPE extraction also provided high recoveries and 7.0e5 6.0e5 excellent reproducibility which is noted in Table II. 5.0e5 4.0e5 3.0e5 Downstream LC–MS-MS analysis on the Kinetex® core-shell 2.0e5 1.0e5 HPLC/UHPLC column also provided signiflcant beneflts as we were 0.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 min able to successfully separate pentobarbital and amobarbital, which difier Figure 1: LC–MS-MS analysis of barbiturates. from each other by the placement of a methyl group which has historically made it diffcult to resolve the two compounds. It is thought that the separation on the Kinetex column was made possible by the high Table II: Relative recovery, RSD, and linearity of barbiturates RSD % Analyte Relative Recovery (%) Linearity peak capacity and effciency of the Kinetex core-shell particle. Core-shell (N = 3) particles contain a 1.9 μm solid core surrounded by a 0.35 μm porous Phenobarbital 100.6 0.5 0.994 layer of silica. Tis allows the particle to perform like a sub-2 μm col- Butalbital 99.37 1.2 0.998 umn in terms of resolution, effciency, and speed without the limiting Pentobarbital 99.28 1.4 0.989 backpressures that are associated with sub-2 μm particles making it easy Amobarbital 98.83 1.6 0.991 to adopt this technology without having to invest in a UHPLC system. Secobarbital 99.40 1.5 0.990

Conclusion μm C18 core-shell HPLC/UHPLC columns provides separation and By pairing a streamlined SPE extraction with an effcient LC–MS-MS resolution of flve barbiturates. analysis, it is possible for any forensic lab to improve their barbiturate screening. Strata-X-Drug N SPE sorbent provided both time and solvent Phenomenex, Inc. savings which are multiplied when screening numerous samples at once, 411Madrid Avenue, Torrance, CA 90501 making it faster and more profltable for forensic labs to screen barbiturate tel. (310) 212-0555, fax (310) 328-7768 samples. After cleanup, a sensitive LC–MS-MS method on Kinetex 2.6 Website: www.phenomenex.com 54 General THE APPLICATION NOTEBOOK – JUNE 2011

Resistive Glass Inlet Tubes Increase Ion Throughput

Paula Holmes, Ph.D., and Bruce N. Laprade, Photonis USA

Resistive glass tubes and plates are designed to guide ions by generating a uniform electric ﬁeld. PHOTONIS resistive glass products are composed of a proprietary lead silicate glass that has been specially processed to create a resistive layer at the surface. The resistiv- ity can be varied over several orders of magnitude to suit the speciﬁc application.

esistive glass is manufactured by using a hydrogen firing pro- Rcess to create an integral semi-conductive layer on the surface. flis reduced lead silicate layer is typically several hundred Figure 1: PHOTONIS’ multi-capillary resistive glass inlet tubes angstroms thick. Resistive glass can be formed into plates, tubes, offer six individual channels in a standard footprint. cylinders, sheets, washers, or other shapes. fle products are resistant to scratches from light to moderate abrasions, and can easily vides containment for counter-ffow gas, eliminating the need for be cleaned ultrasonically with water, acetone, methanol, or IPA an additional enclosure. without degrading the performance. Results Application An increase in ion transfer eTciency by factor of 100 has been One application of resistive glass is capillary inlet tubes for at- reported from using PHOTONIS single capillary inlet tubes. mospheric pressure ionization sources. Single capillary inlet tubes An increase in ion transmission of up to 10× using multicap- made from resistive glass significantly improve ion transfer eT- illary tubes when compared to single capillary inlet tubes has been ciency when compared to conventional quartz inlet tubes. Volt- achieved by a leading mass spectrometer manufacturer, dramati- age applied across nickel-chromium electrodes at each end of the cally enhancing instrument sensitivity. Resistive glass multicapil- inlet tube creates an electric field that preferentially attracts either lary tubes therefore provide an increase in ion transfer eTciency positive or negative ions. Polarity switching can also be accom- of up to 1000× when compared to conventional quartz tubes. plished more quickly than with conventional inlet tubes. A demonstrated improvement in ion transmission is also real- fle properties of resistive glass help prevent ions from collid- ized with the use of single-piece construction resistive glass IMS ing with the tube walls and with each other, reducing ion loss and drift tubes when compared to traditional multipiece lens and ring resulting in a more eTcient sample transfer by forcing more ions assemblies. into the mass spectrometer. Resistive glass reffectron tubes provided equal or better perfor- PHOTONIS has also developed a multicapillary resistive glass mance in an orthogonal TOF system. flis comparison showed inlet tube. A proprietary multibore extrusion process creates a cir- superior resolution, indicating better energy focusing, while spec- cular array of six individual channels in the same footprint as a tra between the two were nearly identical. single capillary inlet tube (See Figure 1). Overall, resistive glass tubes offier benefits to a variety of mass Multicapillary resistive glass inlet tubes provide increased sen- spectrometer applications, many of which can be realized by re- sitivity by further improving ion transmission when compared to placing an existing tube with one made from resistive glass. single capillary inlet tubes. Tubes made from resistive glass can also be used in other mass spectrometry applications, such as for drift tubes, collision cells, ion mirrors, voltage dividers, or reffectron lenses. Another application of resistive glass is for use in ion mobility Photonis USA spectrometry drift tubes. Resistive glass drift tubes operate on the 660 Main Street, Sturbridge, MA 01566 same principle as capillary inlet tubes, and demonstrate a similar Tel. (508) 347-4000, fax: (508) 347-3849 improvement in ion transmission. fle solid tube body also pro- www.photonis.com THE APPLICATION NOTEBOOK – JUNE 2011 General 55

Selection of Optical Fiber for Chromatographic Detectors and Remote Sensing Applications

Joe Macomber and Rick Timmerman, Polymicro Technologies

Optical ﬁbers are routinely used in liquid chromatographic

detectors as a means of simplifying optical designs. 100% Selection of the appropriate ﬁber is an important factor FVP-UVMI FDP in achieving optimal system performance. 80%

ptical fiber has been used for many years in chromatographic ap- 60% FVP-UVM Oplications which employ UV-Vis spectroscopy for sample detection. Fiber has allowed advances such as remote sensing, where in the 40% FVP “detection cell” no longer has to reside inside the detector itself, with Transmission 1m dissolution sampling probes being a prime example. Important factors 20% to consider when selecting a fiber are core size, –OH content, cladding 0% thickness, potential bending radius, and optical attenuation. Of special 0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 interest is the phenomenon of UV solarization, which occurs when the Time fiber absorbs high intensity radiation below 240 nm and bonds within Figure 1: Effect of UV radiation on 214 nm transmission. the glass structure are broken. fle resulting “color centers” can exhibit strong absorbance which results in a marked increase in attenuation of diameter fiber (i.e., 400 μm core or larger). Note that once the the fiber (1). Alternately, many researchers would prefer to describe this hydrogen diTuses out it behaves like FVP-UVM fiber. as a decrease in light transmission. • FDP fiber utilizes proprietary materials and processing steps, providing excellent solarization resistance without the concerns related to hy- Optical Fiber Attenuation in Deep UV drogen diTusion. It has excellent lifetime and superior long term sta- flere are three key performance attributes to consider when compar- bility with only minimal damage. It is the fiber of choice for most deep ing optical fiber types for use in the deep UV: 1) fle initial attenua- UV applications, especially when smaller diameter fiber is required. tion of the fiber prior to significant exposure to UV radiation; 2) fle additional attenuation that appears with exposure to UV radiation Fiber Comparison (eventually saturating or stabilizing); and 3) fle stability of the atten- Figure 1 contains a comparison of 214 nm transmission (relative to the uation during periods of nonexposure (commonly called “recovery”). initial) for various fiber types when exposed to Deuterium lamp radiation. fle amount of recovery and redamage that takes place in repeated fle degradation can be seen, in most cases, to stabilize at a level beyond on/oT cycles is generally a small fraction of the original degradation. which no further damage occurs. Additional data on longer term exposure, and stability in on/oT cycles has also been collected. Fiber Types Available As most chromatographic analysis is done in the UV-Vis spectral range, Conclusion designers typically use a high –OH fiber. flere are four fiber options: flis note discusses solarization of optical fiber and provides recommenda- • FVP is the traditional high –OH fiber series and is excellent for applications for selecting the most appropriate fiber type for UV-Vis applications. tions at or above 240 nm. Additional attenuation is significant at lower wavelengths due to rapid solarization. References • FVP-UVM offers moderate solarization resistance below 240 nm. Even (1) J. Zhou, J. Shannon, and J. Clarkin, Biophotonic Int., 42–44 (Jan 2008). though it experiences solarization damage during initial exposure, ad- (2) K.-F. Klein, R. Kaminski, S. Hüttel, J. Kirchhof, S. Grimm, and G. Nelson, ditional attenuation is minimal providing stable performance thereafter. SPIE-Proc. Vol. 3262C (BiOS 98), 150–160 (1998). • FVP-UVMI is a hydrogen loaded fiber series. It provides outstanding solarization resistance to below 200 nm, but perfor- Polymicro Technologies, mance decreases dramatically once the hydrogen diTuses out A subsidiary of Molex Incorporated of the fiber. Lifetime is dependent upon hydrogen content and 18019 North 25th Avenue, Phoenix, AZ 85023 therefore both the fiber diameter and operating temperature tel. (602) 375-4100, fax (602) 375-4110 should be considered (2). flis fiber type is thus limited to larger Website: www.polymicro.com 56 General THE APPLICATION NOTEBOOK – JUNE 2011

Advantages of Shodex™ NH2P Series, Polymer-Based Amino HILIC Column, Over Silica-Based Amino HILIC Columns

Kanna Ito, Shodex/Showa Denko America, Inc.

ydrophilic interaction chromatography (HILIC) has become Ha popularly used separation mode for carbohydrate HPLC method. fiere are wide variations in HILIC stationary phase. fie base material can be silica or polymer, and they are modifled with diTerent types of polar functionalities such as amide, amino, diol, and cyano. Shodex™ Asahipak NH2P series column is fllled with a polyvinyl alcohol based gel, modifled with polyamine. In this article, we compare the performance of a NH2P series column and a silica-based amino HILIC column.

Experimental Conditions A NH2P-50 4E column and a silica-based amino column, obtained from another company, were compared for their analytical performances. Mobile phase consisted of a mixture of acetonitrile and water. A conventional HPLC system was used with a Shodex™ Figure 1: Comparison of saccharides analysis by a NH2P-50 4 E and a silica-based amino column. Column dimensions; 4.6 × 250 RI detector. mm, Eluent; CH3CN/H2O = 75/25, Flow rate; 1.0 mL/min, Temp.; 25 °C, Detector; Shodex™ RI. Results When acetonitrile concentration was increased, NH2P-50 4E HPLC with a corona charged aerosol detection or an evaporative provided stronger retention and a higher theoretical plate number light scattering detector systems. for all saccharides and sugar-alcohols. In contrast, the silica-based amino column showed nonconsistent changes in the theoretical Conclusions plate number and sensitivity to some saccharides. fiis was more Advantages of polymer-based amino column, Shodex™ NH2P se- obvious for galactose measurement that showed considerably low- ries, over conventional silica-based amino columns are: (i) no ad- er sensitivity with a silica-based amino column (Figure 1). sorption of particular saccharides or sugar-alcohols, (ii) tolerances In addition to saccharide separations, NH2P columns are suit- towards wide range of working conditions, and (iii) small bleed- able for sugar-alcohol and saccharides separations. Separation of ing. Using those advantages, the possible applications of NH2P some sugar-alcohol and saccharide pairs (for example, sorbitol series include analysis of carbohydrates in food stuT, and in bio- and glucose) is diffcult by a silica-based HILIC column, whereas medical areas such as analysis of glycoproteins and glycolipids. a NH2P column provides a base-line separation of those two. Another advantage of polymer-based NH2P columns is the inherent chemical stability, solving the deterioration problem over time that conventional silica-based columns have. One proof of high chemical tolerance is its wide pH working capability (2–13). fiis allows a large selection of usable eluents including diTerent cleaning solvents. fie performance tests under very extreme conditions (over 60 h of 0.1 N H2SO4 and over 160 h of 0.005 N NaOH at pH 11.4) showed no considerable deterioration of the column. fius, it is also suitable for the method that requires an alkaline condition, such as separation Shodex/Showa Denko America, Inc. of pyridylamino-derivatives. 420 Lexington Avenue Suite 2850, New York, NY, 10170 Moreover, very small “bleeding” of particles has been reported tel. (212) 370-0033, fax (212) 370-4566 in NH2P series column. Consequently, this reduces the back- Website: www.shodex.net ground noise level. fiis is especially beneflcial when coupling the Email: [email protected] THE APPLICATION NOTEBOOK – JUNE 2011 General 57

Fast and Accurate LC–MS Analysis of Vitamin D Metabolites Using Ascentis® Express F5 HPLC Columns

Craig R. Aurand and David S. Bell, Supelco/Sigma-Aldrich

itamin D deﬁciency has become a topic of interest in recent Vpublications (1–3). Vitamin D is present in two forms, Vitamin D3 and Vitamin D2. D3 is produced after ultraviolet light-stimulated conversion of 7-dehydrocholesterol in the skin (3). Vitamin D2 is derived from plant sources. Both D2 and D3 are metabolized in the liver to form 25-hydroxyvitamin D2 (25-OH D2) and 25-hydroxyvitamin D3 (25-OH D3), respectively. In addition, biologically inactive 3-epi analogs of 25-OH D2 and 25-OH D3 have been reported, especially in young children (3). ﬂe levels of the 25-hydroxy metabolites are routinely measured for diagnostic assessment of vitamin D related diseases; however, recent studies have indicated that separation from the inactive 3-epi analogs may provide more accurate information for treatment and prevention. Analytical methods that can accurately quantitate both of the 25-hydroxyvitamin D analytes in the presence of 3-epi analogs may become essential for diagnosis and monitoring of patients with vitamin D disorders.

LC–MS analysis of 25-OH D2 and 25-OH D3 is classically performed using C18 stationary phases. Under such conditions, the 3-epi analogs are not resolved and thus are included in the overall reported Figure 1: Fast, LC–MS analysis of vitamin D metabolites using value. Recently, Phinney and colleagues, reported the use of a cyano Ascentis Express F5. Column: Ascentis Express F5, 10 cm × 2.1 column for the eTective separation of the 25-OH and the 3-epi forms mm I.D., 2.7 µm (53569-U), Mobile phase: 25% 5mM ammonium for use in reference measurement procedures (1). Although eTective, formate water, 75% 5mM ammonium formate methanol, Flow rate: 0.4 mL/min., Temp: 40 °C, Inj. vol.: 1 μL, UV detection: 265 the conditions necessitate a run time of better than 40 min limiting its nm, MS detection: m/z 100-1000, Peaks: 1. 25-Hydroxyvitamin D3 utility for routine high-throughput analyses. (2.57 min), 2. 3-epi-25-Hydroxyvitamin D3 (2.76 min), 3. 25-hy- As an outcome of some recent application development eTorts, droxyvitamin D2, (2.77 min). Top: Separation of 25-Hydroxyvi- tamin D and 3-epi-25-Hydroxyvitamin D . Bottom: Analysis of it was observed that Ascentis Express F5 stationary phase provided 3 3 25-Hydroxyvitamin D3 and 25-Hydroxyvitamin D2. increased selectivity toward 25-OH D3 and the corresponding 3-epi analog relative to reported methods. flis report provides a of the closely related 25-OH D3 and 3-epi-25-OH D3 as com- brief synopsis of continuing eTorts to assess the potential impact of pared to methods reported in the literature. Initial eTorts to show this additional selectivity on routine clinical vitamin D diagnostics. selectivity in a fast, LC–MS system provides promising evidence for implementation in real-world situations. Discussion fle initial conditions were adopted for fast LC–MS methodology. References 82, Figure 1 shows some preliminary results indicating that 25-OH D3 (1) S.S .-C. Tai, M. Bedner, and K.W. Phinney, Analytical Chemistry and 3-epi-25-OH D3 can be rapidly resolved. 25-OH D2 and 3-epi- 1942–1948 (2010). 25-OH D3 coelute under these high throughput conditions, how- (2) T. Higashi, S. Homma, H. Iwata, and K. Shimada, Journal of Pharma- ever they are easily resolved by mass response. fle methodology thus ceutical and Biomedical Analysis 29, 947–955 (2002). enables quantification of all three components in one analysis. (3) T. Higashi, K. Shimada, and T. Toyo’oka, Journal of Chromatography B 878, 1654–1661 (2010). Conclusions Separation of the biologically inactive 3-epi analog may serve to Supelco/Sigma-Aldrich provide improved data in support of vitamin D related clinical di- 595 North Harrison Road, Bellefonte, PA 16823 agnostics and treatment. fle pentaffuorophenyl stationary phase tel. (800) 359-3041, fax (800) 359-3044 has been shown to provide superior selectivity for the separation Website: www.sigma-aldrich.com 58 THE APPLICATION NOTEBOOK – JUNE 2011

Aggregated Singletons for Automated Purification Workflow

Aggregated singletons for automated purification (ASAP) is a purification workflow that provides singleton sample purification, registration, and delivery to the materials management department as 30 mM dimethyl sulfoxide solutions for biological screening. The singleton samples submitted are aggregated in a mini-array of 10–12 samples and then are analyzed using an automated purification process. The steps in the process include pre-quality control (QC), preparative chromatography, solvent evaporation, reformat dilution and duplication, final QC, and final registration. The turnaround time from samples received to delivery to materials management is two or three business days. The final QC data are uploaded to a database and are available to chemists via electronic laboratory notebook software. The final purity, weight recovery, and registration information is available in a research database and an e-mail notification of completion is sent to the chemist. ASAP enables a high purification success rate, increases the likelihood of running mini-arrays that generate 5–10 analogs in a final step rather than one or two with the same 2–3 day turnaround time, and provides expert-level service and technology.

he aggregated singletons for auto- solutions for biological screening. The Tmated purification (ASAP) work- workflow for compound synthesis to flow was introduced at Pfizer biological screening is shown in Figure 1. (Groton, Connecticut) in April 2009. The turnaround time from samples Singletons are unique, individual final received for purification to delivery to compounds synthesized for in-vitro the materials management department testing. The main driver and objective is two to three business days. Centraliz- behind this initiative was to save synthe- ing this activity in the purification group sis costs and time. Typically a majority allows greater time for higher value tasks of the singletons are made in 100–300 to be completed by practicing chemists; mg amounts. Datasets across various the purification scientists can provide projects show that only ~20% of the expert-level service and technology, and singletons survive primary and second- also more opportunity for harmoniza- ary screenings and hence ~80% of the tion with screening (consistent and high compounds are made unnecessarily on quality samples delivered for biological Bhagyashree A. Khunte and the large scale. Pharmaceutical chem- assays). Laurence Philippe ists can save considerable time and cost by making smaller (30–50 mg) batches. Experimental Analytical Chemistry, Pfizer Global The chemists can submit synthesized Following are the steps for sample sub- R&D, Eastern Point Road, Groton, compounds (30–50 mg) to ASAP to mission to ASAP: The samples submit- CT 06340 be purified, registered, and delivered to ted are final products (compounds going Direct correspondence to the materials management department for biological screening — no intermedi- [email protected] as 30 mM dimethyl sulfoxide (DMSO) ates). The crude weight of the compound THE APPLICATION NOTEBOOK – JUNE 2011 59 is 10–50 mg. Some type of an initial sample workup (for example, liquid–liquid extraction, solid-phase extraction, Synthesis and so forth) is highly recommended to remove metals or reagents used during Crude registration the synthesis. The crude sample is fully dissolved in 900 µL of DMSO, filtered, Compound submittal for screening and placed in a 2D bar-coded, matrix capped tube. The crude sample is then Autopuriﬁcation submission registered as an “in-production singleton” sample via Chemistry e-Notebook Sample aggregation electronic laboratory notebook software (CambridgeSoft, Cambridge, Massa- Autopuriﬁcation chusetts) using the software’s Global Registration tool, which has been cus- Pure registration tomized for Pfizer. The biology assay– screen panel is assigned and the sample is submitted for purification using Pfizer’s TekCel storage Auto-Purification submission website. The sample is then dropped off in the Local order management assigned submission dry boxes in a Sin- gleton Matrix-Box container. Screening The ASAP process is outlined in Fig- Figure 1: Flow chart showing the steps in the process from compound synthesis to ure 2. A total of 10–12 singletons are screening. aggregated in a plate format and associated with a bar code using a 2D Matrix tube reader (Thermo Fisher Scientific, Waltham, Massachusetts). An analytical plate of fixed concentration (1 mg/ Singleton submission mL) is created using a Tecan Freedom Evo liquid handling system (Män- nedorf, Switzerland). The samples in the analytical plate are then analyzed Duplication using reversed-phase liquid chroma- Liquid Handler tography–mass spectrometry (LC–MS) and evaporative light scattering detec- Pre-Prep Analysis LCMS/UV/ELSD tion (ELSD) techniques. The analytical system comprised a Waters 2795 Alli- Prep LCMS/Analog ance high performance liquid chromatography (HPLC) system and ZQ mass Evaporation Evaporator spectrometer (Milford, Massachusetts) and a Polymer Labs 2100 ELSD sys- Weigh Weigh Station tem (Agilent Technologies, Santa Clara, California). The initial purity of the Fraction Selection sample can range from 5% to 80%. The pre-quality control (QC) gradient meth- Reformat Duplication Liquid Handler ods are 5 min in length with 2-mL/min flow rates. Different column chemis- Quality Control Analysis LCMS/UV/ELSD tries, modifiers, and gradient conditions are used to develop methods best suited for purification. Even though the samples are aggregated in a miniplate REG and analyzed in a high-throughput mode, a method is developed for each individual sample. The best method and conditions are then transferred to the Solpl preparative purification step. A Waters Auto-Purification Fractionlynx system Figure 2: Outline of the ASAP process. 60 THE APPLICATION NOTEBOOK – JUNE 2011

is used for purification. Fraction collection is mass triggered. The prepara- (a) tive methods are gradients run over 10 MS ES+:TIC 100 1.32E7 100 (6), 1.59 240.2 5.48E5 (11),28%,(16),2294D%,2.99 min. The collected fractions are evapo- 50 (2),23%,1.12 rated using Genevac Mega evaporators 0 0.0 0.0 1.0 2.0 3.0 4.0 (Gardiner, New York). The dry fraction 75 MS ES+ :240.081 1.2E6 (6),100%,1.59 tubes are then weighed on Tecan weigh- ing stations.

0.0 The dried, purified material is dis-

50 DAD: 215 solved in DMSO to make 30 mM solu- 100 2.84E5 (4),48%,1.46 tions. A maximum of 900 µL of the 30 50 (7),17%, (12),17%,2.311.62 mM DMSO solution is transferred to a 0 0.0 0.0 1.0 2.0 3.0 4.0 25 479.3 plate. A daughter analytical plate is also PL-EL 2100 100 4.28E5 241.2 (3),55%,1.44(11),26%,2.28 created by transferring a 5-µL aliquot

90 (15),10%,2.95 480.3 and adding enough DMSO to result

318.2 80 3.43E5 0 0.0 1.0 2.0 3.0 4.0 in a 0.5-mg/mL concentration. Excess 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 30 mM DMSO solution is transferred (b) to bar-coded 2-mL vials. Tecan liquid DAD:215 100 37 256768 handling systems are used for dispens- 22 50 1 18 0 8 ing the DMSO and transfer of samples 0 0 0 1 2 0 0 0 0 0 0 0.0 to the plates and vials. The samples in NaN 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 MS ES+ :240.081 the vials are evaporated using a Genevac 100 94 807316 HT-12 evaporator in a two-step evapora- 50 5 0 0 0 tion process. 0 0.0 NaN 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 The final QC is performed using MS ES+ :TIC 100 66 8.0493 the HPLC–MS–ELSD system or

50 7 12 11 Waters Acquity UPLC/SQD/PL 2100 4 0 0.0 ELSD units, and the samples are reg- NaN 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 100 255.2 istered if they meet the purity crite- 50 ria (>80% purity by UV at 215 nm, 0 > > 220.0 240.0 260.0 280.0 300.0 320.0 340.0 85 % purity by ELSD, and 50% mass spectral purity). Orthogonal QC (c) methods are selected to ensure the

DAD:215 100 100 224076 final purity of the samples. The solu-

50 bilized plate (maximum up to 900 µL

0 0.0 of 30 mM DMSO stock) is sent to the 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 MS ES+ :240.081 100 100 708460 materials management department for 50 assay preparations. The compound is

0 0.0 then released for the requested screen- 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 MS ES+ :TIC 100 100 922749 ing. The dry compounds are regis- 50 tered and shipped to the Pfizer Neat 0 0.0 Store. The pre-QC and final QC 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 PL-ELS 2100 100 100 100096 data are uploaded in Pfizer’s Global 50 Analytical database as a PDF and the

0 0.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 file also gets linked to the chemist’s 100 240.2 e-Notebook submission page. The 50 0 recovery amount, purity, and reg- 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 istration data are available through Figure 3: Singleton compound analyzed using ASAP: (a) pre-QC analysis, (b) prepara- Pfizer’s Research database. An e-mail tive run, and (c) final QC analysis. (a) 50 mm X 4.6 mm, 5-µm Waters XBridge C18; notification is sent to the chemists to mobile phase A: 0.03% ammonium hydroxide in water (v/v); mobile phase B: 0.03% inform them that the samples have ammonium hydroxide in acetonitrile (v/v); gradient: 5–95% B (linear) in 4.0 min, hold been delivered to the materials man- at 95% B to 5.0 min; flow rate: 2.0 mL/min. (b) Column: 100 mm X 19 mm, 5-µm Wa- ters XBridge C18; mobile phase A: 0.03% ammonium hydroxide in water (v/v); mobile agement department. A results spread- phase B: 0.03% ammonium hydroxide in acetonitrile (v/v); gradient: 5–50% B (linear) sheet is attached that provides detailed in 8.0 min, 50–100% B (linear) to 8.5 min, hold at 100% B to 10.0 min; flow rate: 25 information regarding the samples, mL/min. (c) Column: 50 mm X 4.6 mm, 5-µm Waters Atlantis dC18; mobile phase A: including the QC gradient conditions, 0.05% trifluoroacetic acid in water (v/v); mobile phase B: 0.05% trifluoroacetic acid in acetonitrile (v/v); gradient: 5–95% B (linear) in 4.0 min, hold at 95% B to 5.0 min; mass observed, retention time, purity flow rate: 2.0mL/min. data, and total recovery after purification. THE APPLICATION NOTEBOOK – JUNE 2011 61

The entire process and data flow are handled through customized soft- (a) Peak 4 ware. The software has been specifi- MS ES+ :TIC (13), 2.55 100 1.9E8 100 370.1 9.98E6 13%, 2.11 (22) 11% 2.47 (9) (21)(27) 10% 10% 1.43 0. cally designed for automated batch- 50 singleton and library purification 0 0.0 0.0 1.0 2.0 3.0 4.0 workflow. The software enables a user 75 MS ES+ :370.099 100 3.27E7 (13), 92%, 2.55 to handle multiple plates with accu- 50 rate data flow. It also provides struc- 0 0.0 0.0 1.0 2.0 3.0 4.0 tures, chemical properties, CLogD 50

100 DAD: 215 (12), 18%, 2.4 2.2E6 (9), 18%, 2.11 data, and acid–base labile prediction 50 372.0 (22), 40%, 3.46 information for the compounds asso- 0 0.0 0.0 1.0 2.0 3.0 4.0 25 ciated in the plate. This information

PL-ELS 2100 100 3.96E5 helps minimize the time required for (12),(13), 23%, 20%, 2.41 2.56 648.1 50 739.1 method development. (23), 22%, 3.5 741.1 0 0.0 0 0.0 1.0 2.0 3.0 4.0 0.0 200.0 400.0 600.0 800.0 Results and Discussion The ASAP process was introduced in (b) Pfizer R&D (Groton) in April 2009. DAD: TIC 100 5.3040 A major contributor to the success of 75 10 2 3 0 ASAP is the strong partnership between 50 0 2 33 21 11 7 2 the company’s Analytical Chemistry 25 0 3 0 1 11 1 0 0 0 000 00 000 0 0 00 0 0 0.0 and Medicinal Chemistry lines. A point 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 MS ES+ :370.099 of contact (POC) was established for 100 34 1.0084 63 75 each project using ASAP to facilitate

50 communication between the two lines.

25 21 1 Training sessions were conducted for 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 each project joining ASAP. There was MS ES+ :TIC 100 20 15 5.0287 a gradual ramp-up of projects added to 5 75 9 ASAP to ensure continuous progress of 50 8 3 43 2 3 3 3 implementation. 25 4

0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Examples of Samples 100 369.9

50 Purified by ASAP 0 Figures 3 and 4 show pre-QC, prep, and 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 final QC data for samples analyzed by ASAP. Figure 3 is an example of a sam- (c)

MS ES+ :TIC (2), 2.3 ple that was submitted to ASAP with an 100 3.08E7 100 370.1 6.49E0 (2), 83%, 2.31

50 initial purity of ~15%. The target mass (1), 17%, 2.21

0 0.0 was isolated and enough material was 0.0 1.0 2.0 3.0 4.0

MS ES+ :370.099 75 delivered to the materials management 100 1.97E7 (2), 97%, 2.3

50 department for screening. Figure 4 is

0 0.0 an example of a complex sample. Under 0.0 1.0 2.0 3.0 4.0 50 DAD: 215 acidic conditions peak 4 in Figure 4a was 100 1.02E6 (2), 95%, 2.3

50 372.1 coeluted with the target peak. However,

0 0.0 after a method was developed in basic 0.0 1.0 2.0 3.0 4.0 25

PL-ELS 2100 conditions, separation was obtained and 100 1.91E5 (2), 100%, 2.3

50 the purified compound easily passed the 371.1

0 0.0 448.1 0 final QC (Figure 4c). 0.0 1.0 2.0 3.0 4.0 0.0 200.0 400.0 600.0 800.0

Figure 4: Singleton compound analyzed using ASAP: (a) pre-QC analysis, (b) pre- ASAP Workflow parative run, and (c) final QC analysis. (a) Column: 50 mm X 4.6 mm, 5-µm Waters Deliverables and Advantages XBridge C18; mobile phase A: 0.03% ammonium hydroxide in water (v/v); mobile phase B: 0.03% ammonium hydroxide in acetonitrile (v/v); gradient: 10–95% B (linear) • A high quality of purified sample is in 4.0 min, hold at 95% B to 5.0 min; flow rate: 2.0 mL/min. (b) Column: 100 mm X 19 delivered to the materials manage- mm, 5-µm Waters XBridge C18; mobile phase A: 0.03% ammonium hydroxide in wa- ment department with a 2–3 day turn- ter (v/v); mobile phase B: 0.03% ammonium hydroxide in acetonitrile (v/v); gradient: around time. 10–100% B (linear) in 8.5 min, hold at 100% B to 10.0 min; flow rate: 25 mL/min. (c) 50 mm X 4.6 mm, 5-µm Waters Atlantis dC18; mobile phase A: 0.05% trifluoroacetic acid • The process provides a high level of in water (v/v); mobile phase B: 0.05% trifluoroacetic acid in acetonitrile (v/v); gradi- purification expertise and technology. ent: 5–95% B in 4.0 min, hold at 95% B to 5.0 min; flow rate: 2.0 mL/min. ASAP can purify compounds that for 62 THE APPLICATION NOTEBOOK – JUNE 2011

chemists find difficult to purify using • The ASAP process saves time and lets Acknowledgments conventional techniques. chemists focus on high-value activities The authors would like to thank David • The process provides a high success rate such as making compounds for pre- Price, Rose Gonzales, and Mark Noe (>90% samples are registered). liminary toxicity studies. for sponsoring ASAP and providing • Postpurification analysis: The final resources and staffing. purity is assessed by MS–UV–ELSD Conclusions and QC data are uploaded into the To date, more than 5700 singletons Bhagyashree A. Khunte is a Global Analytical database. have been submitted and >90% of Principal Scientist and Laurence Philippe is an Associate Research • Final purity information (purity and the samples have been successfully Fellow in the Analytical Chemistry weight recovery) is available in the purified. ASAP is currently support- department at Pfizer Global R&D research database. ing approximately 25 projects across (Groton, Connecticut). Direct • An e-mail notification is sent to different therapeutic areas. Although correspondence to Bhagyashree. the chemist. Data can be accessed the samples are purified in an auto- [email protected]. through electronic laboratory note- mated purif ication mode, the versatil- book software. ity of the technology allows purifica- • The process increases the likelihood of tion of diverse molecules with respect For more information about this topic, running mini-arrays to generate 5–10 to compound solubility, polarity, please visit analogs in a final step rather than one and molecular weight. The next www.chromatographyonline.com or two, with the same 2–3 day turn- steps involve enhancing purification around time. technologies such as adding normal- • The ASAP process allows projects to phase chromatography, supercriti- choose synthetic routes that diverge cal fluid chromatography, and on- at a late stage, which is the most column solvent exchange techniques

effective way to prepare large sets of to increase the scope of available Visit ChromAcademy on LCGC’s Homepage analogs. chemical space. www.chromacademy.com

on the web

UPCOMING EdUCatIONal lytical chemistry, a discussion of solvents Absolute Characterization of Proteins WEbCast and solvent selection from the green per- and Biopolymers: Combining SEC spective will follow. Finally, means to with Multi-Angle Light Scattering Green Analytical Chemistry: evaluate the green attributes of analytical (MALS) Detection A Convergence of Thought Processes methods will be discussed. Tuesday, July 12, 2011 Reagent-Free Ion Chromatography: 11:00 AM EDT RECENt WEbCasts NOW Generate High Purity Eluents and Reagents for Environmental and Analytical chemistry involves the knowl- aVaIlablE ON dEMaNd Pharmaceutical Applications edge of chemistry and chemical measure- Practical Operation of ments to solve a problem. Green chemis- UHPLC Columns Modern Size Exclusion Chromatog- try, as defined by Anastas and Warner, raphy for Biopolymer and Synthetic is “the use of a set of principles to elimi- Faster GC...or Fastest Polymer Characterization nate the use or generation of hazardous Expanding Your Lab Capabilities and substances...” In other words, analytical Be on the Safe Side: Combined Quan- Productivity with the Latest Genera- chemistry and green chemistry funda- titative Multi-Target Screening and tion LC–MS mentally are ways to think about the Spectral Confirmation of Pesticide practice of chemistry. So, green analyti- Residues in Food and Environmental Addressing Chemical Diversity and cal chemistry is a convergence of these Samples by LC–MS-MS Expanding Analytical Capability Using thought processes. Following an intro- Atmospheric Pressure GC (APGC) duction to the 12 principles of green The Challenge to Find the Right Needle More Sensitive Analysis of Hexavalent chemistry, an overview of green ana- in the Haystack: Using Multi-tier LC– Chromium in Water and Soil Extracts lytical techniques, including sampling, MS-MS Methodologies to Implement separations, and spectroscopy, will be Non-targeted Screening for Residues presented. Since solvent use is among the of Pesticides and their Metabolites in Register or watch on demand at largest environmental impacts in ana- Water and Food Samples www.chromatographyonline.com/webseminar THE APPLICATION NOTEBOOK – JUNE 2011 63

A New Path to High-Resolution HPLC–TOF- MS — Survey, Targeted, and Trace Analysis Applications of TOF-MS in the Analysis of Complex Biochemical Matrices

High performance time-of-flight mass spectrometry (TOF-MS) is applied to the analysis and characterization of complex biological samples. High mass accuracy, high resolution, and accurate relative isotope abundance are all applied to the determination of analytes covering a range of concentrations in complex matrices including plasma and urine. Qualitative and quantitative evaluations are provided and include demonstrations of the impact of high- performance MS on sensitivity and selectivity. The ability to leverage high-performance MS in conjunction with a broad dynamic range in rapid ultrahigh-pressure liquid chromatography (UHPLC) analyses to identify unknowns and to propose putative metabolic biomarkers is demonstrated. The impact of speed of analysis and selectivity to the depth of coverage and accuracy of the analyses are discussed.

emands on modern ana- The utility and value of high-res- D lytical chemistry are driven olution, high mass accuracy MS has by enhanced information been emphasized in numerous recently content, faster analyses, faster data reported applications of the technol- processing, and higher throughput. ogy (4–9). High-resolution MS has Tools in separation science continue been the domain of Fourier-transform to increase the speed and information (FT) MS and magnetic-sector instru- content of traditional analyses as dem- ments for decades. The paradigm is onstrated by the growing popularity of shifting away from just high-reso- ultrahigh-pressure liquid chromatog- lution MS to high-performance MS raphy (UHPLC) (1), fast gas chroma- that includes mass resolving power, tography (GC) (2), and GC×GC (3). mass accuracy, isotope abundance, This creates demands for faster data and acquisition speed in its consid- acquisition and higher fidelity data, erations. New developments in high- which in mass spectrometry (MS) performance mass spectrometers have translate to mass accuracy and resolu- shown significant impact. Time-of- tion. To address these demands, the flight (TOF) instruments with higher mass resolution, mass accuracy, and resolution and enhanced performance data acquisition speeds of mass spec- are a significant portion of the efforts. trometry have been pushed to improve Recent advances in TOF mass spec- instrument capabilities. A historical trometers have enabled these instru- Jeffrey S. Patrick, Kevin Siek, overview of mass analyzers is provided ments to provide high resolution and Joe Binkley, Viatcheslav Artaev, in Table I with the representative per- mass accuracy with speed, simplicity, and Michael Mason formance attributes provided. and convenience. The measurement of 64 THE APPLICATION NOTEBOOK – JUNE 2011

Table I: Overview of high-performance mass spectrometers Isotopic Platform Resolving Power Mass Accuracy Acquisition Rate Mass Range Abundance

Magnetic sector >50,000 <100 ppb <10 spectra/s Excellent Limited

Limited by Fourier transform <10 ppb (trade-off with Trade-off with other Poor (space duration, ﬁeld (includes both magnetic >10 6 acquisition rate and attributes including charge) strength, and and electrostatic traps) magnet size) resolving power experiment

Good Time of ﬂight (includes 1–5 ppm (lower values Typically trade- Limited by pulse >10 4 (new systems (limited by TOF and Q-TOF conﬁgu- achievable with special off with resolving frequency and approaching 105) dynamic rations) consideration) power analyzer attributes range)

on the order of a few meters. Other efforts have attempted to extend the Gridless Mirror flight path using cyclotron and tor- Periodic lon Lenses roidal analyzers (13). Duty cycle and effective mass range are typical com- promises to high levels of mass measurement performance. With these advances, TOF has achieved a position in the realm of mass analyzers in which it provides sensitivity, mass Gridless Mirror accuracy and resolution that surpass some magnetic sector systems and encroach on the analyses historically reserved for FTMS systems. The prin- Detector ciple advantage of a TOF system is the absence of scanning. This provides a lon Source mass measurement system well suited to analyte surveys rich in information content and high in duty cycle. These advances have opened the world of high resolution accurate mass analy- Figure 1: Depiction of multireflecting TOF or folded flight path design. sis, historically the domain of FTMS and magnetic sectors, to TOF mass analyzers with the added benefit of mass-to-charge ratio (m/z) in a TOF orthogonal acceleration (10), high- simplicity and speed. The signifi- analyzer is described by the following speed electronics, delayed extraction cant advances of TOF mass analyzers equation: (10), and pulsed ionization (11), which toward high-performance mass spec- permit the most effective utilization trometers continue in the approach t = D(m/2z)1/2 of the flight path available. In spite discussed below. of these advances, TOF mass analyz- in which t is time, D is the flight ers are highly sensitive to the initial The New Technology distance, m is the mass, and z is the conditions of the ions as they enter the In one of the most recent advances charge on the ion. This provides the mass analyzer. This includes distribu- in TOF-MS, Verentchikov and col- fundamental relationship that t is pro- tions in velocity, time, and space dur- leagues (14,15) have introduced the portional to D, and in turn, resolution ing ion generation and at entry to the multireflecting time-of-flight ana- (defined as [m/δm] or [t/2*δt]) is also a mass analyzer. To provide an extended lyzer that uses a Folded Flight Path function of D and t. This has been one flight path and address initial ion dis- (FFP). This is depicted in Figure 1, of the challenges to TOF — to achieve persion, electrostatic time-focusing which shows how the ions effectively long flight paths in reasonable physical elements such as reflectrons (12) have pass between each of two planar ion space and provide high performance. A been implemented, improving resolu- mirrors and through an Einzel lens number of key advances have occurred tion significantly. array. A major challenge in the multi- over the past two decades that have However, these approaches are lim- reflecting systems is poor transmission improved the performance of TOF ited by dispersions and higher order because of defocusing of ion trajectories. mass analyzers and have included aberrations, as well as flight lengths In the FFP configuration, transmission THE APPLICATION NOTEBOOK – JUNE 2011 65

This design is applied to the analysis of analytes in complex matrices in the Ultrahigh Resolution High Resolution Nominal Resolution form of UHPLC coupled to atmospheric pressure ionization with TOF-MS. The matrices include plasma and urine, and the analytes include compounds of

A A A pharmaceutical interest and naturally D D D occurring biological analytes associated with disease states. The applications discussed include the identification and relative quantitation of metabolites in plasma from three strains of Zucker rats, an example of the study of animal 706 706.2 706.4 706.6 706.8 707 707.2 707.4 707.6 707.8 708 706 706.2 706.4 706.6 706.8 707 707.2 707.4 707.6 707.8 708 690 700 710 720 730 740 m/z m/z m/z models of disease (16), and the analysis R = 100,000 R = 50,000 R = 2500 for metabolites of common cold medica- m/z m/z 4:1 mass range = 50–2500 = 50–2500 tions in human urine.

Figure 2: The three modes of operation of the MRT/HRT system. The FFP design Experimental facilitates multiple flight distances for ions, and as such, the potential for multiple levels of resolving power. Ions pass between electrostatic mirrors and are refocused The following represent the basic con- by central lenses for the length of the analyzer. Two full passes across the analyzer ditions used in the analyses described provide a 40-m flight path (ultrahigh resolution) and 100,000 resolving power, one in the results and discussion which fol- pass through the entire analyzer provides a 20-m flight path (high resolution) and low. Chromatographic separation was 50,000 resolving power, and a partial pass through the analyzer provides a 2-m path (nominal resolution) and 2500 resolving power. (Note: A = accelerator and D = achieved using an Agilent Technolo- detector.) gies (Santa Clara, California) 1290 UHPLC system with mobile phases consisting of water with 0.1% (v/v) formic acid (A) and acetonitrile with 0.1% (a) (b) (v/v) formic acid (B). Mobile phase was 7 2.5 delivered at 0.1–0.2 mL/min and with Fatty:Lean Fatty:Lean gradients covering 0–80% B between 6 Fatty:Obese 2 Fatty:Obese 3 and 30 min. Injection volumes var- 5 ied between 1 and 5 μL. The column Obese:Lean Obese:Lean 1.5 4 used was a 50 mm × 1 mm Hypersil Gold C18 AQ (Thermo Fisher Scien- 3 1 tific, West Palm Beach, Florida). MS Response ratio 2 Response ratio was achieved on a LECO Citius LC- 0.5 HRT system equipped with a LECO 1 electrospray ionization (ESI) source 0 0 (LECO Corporation, St. Joseph, Michigan) and operated in high reso- Urate Urate ADMA ADMA Tyrosine Tyrosine Cytidine Cytidine lution mode (R = 50,000 [FWHM]). Hippurate Hippurate Tryptophan Tryptophan Kynurenine Kynurenine Pantothenate Pantothenate Phenylalanine Phenylalanine Data were processed using Chroma- Leucine/lsoleucine Leucine/lsoleucine 3-OH-anthranilate 3-OH-anthranilate TOF-HRT software (LECO). Acquisi- tion rates of 2–40 spectra/s were used Analyte Analyte both with and without in-source collision-induced dissociation. Calibration Figure 3: Relative response ratios for select analytes in plasma from Zucker lean, was achieved using external calibration fatty, and obese rats. (a) Full-scale plot; (b) zoomed plot. with Agilent Tune Mix. Rat plasma samples were obtained is improved by using nonlinear electro- 2-, 20-, and 40- m lengths with opera- from Bioreclamation (Hicksville, New static fields in the gridless mirrors. tions in the same planar flight “tube.” York) and were from lean, fatty, and The ions are constantly refocused as These operational options are depicted obese Zucker animals. The plasma they traverse the flight path and cre- in Figure 2. The number of passes was from terminal bleeds with the rats ate the extended flight path needed to or reflections determines the effec- being 7–9 weeks of age. The plasma enhance the resolution. The gridless tive pathlength, and by inference, the samples were deproteinated using design minimizes ion loss during the available resolving power. The perfor- Microcon (Billerica, Massachusetts) flight. The mass analyzer offers three mance capabilities of this system are centrifugal devices with a 5K cutoff. effective flight paths of approximately provided in Table I. After protein removal, samples were 66 THE APPLICATION NOTEBOOK – JUNE 2011

and Lean:Obese ratios relative to the Fatty:Obese ratio. These include leucine/isoleucine, hippurate, dimethyl- arginine, kynurenine, and urate. The nearly sixfold change in leucine/isoleucine is the most striking and largest proportional positive change for the “fatty” phenotype, while the change in kynurenine is the highest magnitude change negatively correlated with the “fatty” state. Both kynurenine and leucine/isoleucine have been previously associated with diabetes and obesity (19–23). At the beginning of this study, standards were not available for confirma- Figure 4: Extracted ion chromatogram of 232.1554 ± 0.01 from selected rat samples. tion of identity. As such, experiments including the fragmentation of “precursor” ions before the mass analyzer Table II: Mass accuracy in the assignment of identities to doxylamine metabolites (so-called in source CID) with accurate mass measurement of fragment m/z m/z Assignment Expected Observed Relative Error (ppm) ions and the use of relative isotope Didesmethyldoxylamine 243.149190 243.14925 0.2 abundance were used to clarify or support identification. One analyte Desmethyldoxylamine 257.164840 257.16497 0.5 identified in these studies was butyryl Doxylamine 271.180490 271.18049 0.0 carnitine. In this case, an ion at m/z = 232.1554 was observed to change in Doxylamine + O 287.175404 287.17543 – 0.1 intensity between lean and fatty/obese states. The formula search for this m/z diluted 5× into aqueous 0.2% hep- of the data was achieved by means provided two formulas within 5 ppm m/z tafluorobutyric acid before UHPLC of two mechanisms. In one, a list of of the target : C11H21N1O4 (0.77 analysis. established analytes was used to search ppm error) and C12H17N5 (1.3 ppm Urine was obtained from a healthy the data for these analytes using an error). To provide additional informa- male volunteer before and after a sin- accurate mass targeted approach. In tion on the structures of the analyte, gle dose of cold medicine containing these instances, the exact m/z value the fragment ions were extracted into 6.5 mg doxylamine, 15 mg dextro- was searched against the existing a separate data channel and evalu- methorphan, and unspecified amounts data to within 0.001 m/z unit. In the ated. Figure 4 shows an extracted ion of polyethyleneglycol (PEG) and other other, the accurate m/z values obtained chromatogram for both low (lean) and excipients. To remove salt and protein from the analyses were used to deter- high (obese/fatty) samples along with from each sample, a 0.15-mL aliquot mine molecular formulas, and the the fragment ion spectrum from a lean was diluted with 1.05 mL methanol formulas or the accurate m/z values sample. The two prominent fragment and centrifuged. A 1.0-mL aliquot of were searched in ChemSpider (17) or ions at m/z 173.082 and 85.03 match the clear supernatant was evaporated, KEGG (18) or other similar metabo- the loss of trimethylamine from 232 reconstituted with 0.025 mL methanol lite databases to facilitate the identi- and the loss of butyric acid group and 0.10 mL water with 0.1% formic fication of the analyte in question. and a trimethylamino group from acid, and analyzed by UHPLC–TOF- In both of these instances, the ratio 232, respectively. These are consistent MS under high resolution conditions. of area responses in the samples from with the identification of 232.1554 All other chemicals used were obtained the different phenotypes were then as butyryl carnitine. To provide yet from Fisher Scientific (Fairlawn, New averaged and compared to each other another piece of supporting evidence, Jersey) or Sigma-Aldrich (St. Louis, for differences as three ratios. Spe- the relative isotope abundance was Missouri). cifically, the Lean:Fatty, Lean:Obese, generated for the two formula above and Fatty:Obese response ratios were and these were compared to what was Results and Discussion generated, and the averages for each observed in high (fatty/obese) and low Rat Plasma Metabolomics: Plasma phenotype are provided graphically in (lean) intensity samples. These find- samples from lean (n = 10), fatty (n Figure 3 for representative analytes. ings are provided in Table II. For the = 9) and obese (n = 10) Zucker rats The plot shows several selected or monoisotopic peak and first isotope were analyzed by LC–MS using high- readily identified analytes that show the agreement with the theoretical dis- performance TOF-MS. The analysis positive changes in the Lean:Fatty tribution is better than ±5% (relative) THE APPLICATION NOTEBOOK – JUNE 2011 67

analytes simultaneously, the linearity of response across several orders 180 of magnitude is necessary. To explore C H NO C H N 160 11 21 4 12 17 5 this and the limit of detection of the accurate mass system, a standard curve 140 of butyryl carnitine was created that 120 covered the range of 0.16 through 100 >500 nM. The plots of response versus

80 concentration are shown in Figure 6 (Figure 6a: 0.16–500 nM; Figure 6b: 60 0.16–20 nM) and demonstrate that 40 Percent theoretical Percent the dynamic range achievable here is 20 clearly in excess of the 500 nM point shown (2500 nM not shown). This is 0 12 12 13 13 13 13 equivalent to a linear dynamic range C C 1* C 1* C 2* C 2* C greater than 3000-fold (15,000-fold

lsotope examined at 2500 nM). The experiment has its lowest detectable level at 0.16 nM or Figure 5: Comparison of relative isotope abundance measurements for the analyte approximately 200 fg of butyrylcarni- at m/z 232.1554. The percent of the theoretical abundance is provided versus the tine on-column. This translates to the expected values for each option and is shown as Option 1/Option 2. The options are the two highest hits in the formula match. Isotope abundance was calculated using ability to examine either dramatic dif- the ChromaTOF-HRT tool. Signals were considered from each of five samples, which ferences in concentrations of a single covered a range of response intensities, and all three phenotypes. analyte or a range of analytes at sub- stantially different concentrations in the same analysis. Analysis and Characterization of Drug Metabolites in Urine: One of 3500000 65000 the most complex, but also most readily available, biological fluids is urine. 3000000 55000 It is difficult to standardize and is 2500000 affected by diet, physiological state, 45000 disease, and other hard-to-control fac- 2000000 tors (24). With all of this said, urine is 35000 still able to indicate the health of indi- 1500000

Area response Area viduals and to provide insights into Area response Area 25000 what the body has been exposed to, 1000000 as it does give up its metabolites easily 15000 through renal clearance mechanisms. 500000 5000 Here, urine from a healthy donor was 0 analyzed by UHPLC with high-per-

0 200 400 600 -5000 0 10 20 30 formance detection. A minimal level of cleanup was performed with the Analyte concentration (nM) Analyte concentration (nM) objective of providing a global survey for metabolites of doxylamine and Figure 6: Standard curve of butyryl carnitine from single injections: (a) 0.16–500 nM, (b) zoomed region of 0.16–20 nM. dextromethorphan. Figure 7 shows a representative set of chromatograms for actives and metabolites detected in all cases versus Option 1 (butyryl rate mass precursor analysis, accurate in urine. The extraction window of carnitine). For the second isotope (2 mass fragment ion analysis, and high ±2 ppm shows the necessary stability 13C), the agreement is still better than integrity relative isotope evaluation is of the mass accuracy across the entire ±17% (relative) for four of the five a clear demonstration of the impact peak. This facilitates accurate inte- samples, with the fifth being of very that high-performance MS can have in gration of signal with high resolution low signal intensity. In all cases, the metabolomic analysis and the charac- and accuracy relative to other peaks. relative isotope abundance defines terization of analytes. Although there are known compounds which of the two formula options Of additional importance in metab- which should be present from the dos- best fits the data. This is most clearly olomic analysis is the dynamic range ing, urine is a complex matrix, and as seen in the first isotope data shown in of the instrument. To effectively inter- such, the identity of the analytes was Figure 5. This combination of accu- rogate both abundant and trace level confirmed by accurate mass analysis 68 THE APPLICATION NOTEBOOK – JUNE 2011

potential biomarkers for obesity and as metabolic byproducts of pharmaceuti- (a) cal agents. Having high mass accuracy,

800000 resolving power, and isotope measures available in both precursor and frag- 700000 ment ions make the confidence an

600000 analyst might have in the identifica-

243.149190+0.000486 (Parent Ions) tion of analytes extremely high. As 500000 257.164840+0.000514 (Parent Ions) 271.180490+0.000542 (Parent Ions) this can be achieved on femtogram 434.217329+0.000868 (Parent Ions) amounts of material using high acqui- 400000 258.185241+0.000516 (Parent Ions) 272.200891+0.001361 (Parent Ions) BPI sition rates with analysis times of a few 300000 minutes and provide high performance under all acquisition conditions, high- 200000 performance MS is both fast and

100000 of high information content. These findings clearly define the ability of 0 Times (s) 25 50 75 100 125 150 175 200 225 high-performance MS to provide a positive impact in the advancement of metabolomic and metabanomic analyses.

Acknowledgments The authors wish to thank Celine Aguer and Mary-Ellen Harper from the University of Ottawa (Ottawa, Canada), Oliver Fiehn from UC Davis (Davis, California), and Sean Adams from the Western Human Nutrition Research Center (WHNRC) at UC Davis for butyryl carnitine standards through work funded by the National Institute of Health (DK 078328, Bethesda, Maryland).

Figure 7: (a) Base peak signal and extracted ion chromatograms of active ingredients References and potential metabolites in urine. The signal for PEGs is clear between 50–100 s. (b) (1) M. Schwarz, LCGC North America 28(5), Precursor (low energy) and product (high energy) spectra from m/z 271 showing key 376–385 (2010); N. Wu, A. Clausen, fragment ions. The extraction window is at ±2 ppm for each analyte. L. Wright, K. Vogel, and F. Bernar- doni, Amer. Pharm. Review March/April (Table II), which showed less than 0.5 sured resolving power was in excess 2008. ppm mass errors for the proposed dox- of 40,000 at m/z 271, and even at m/z (2) F.L. Dorman, J.J. Whiting, J.W. Cochran, ylamine-related peaks. Similar to what 90, the resolving power was a robust J. Gardea-Torresdey, Anal. Chem. 82(12), was achieved with the butyryl carni- 35,000. 4775–4785 (2010); L. Mondello, P.Q. tine peak, relative isotope abundance Tranchida, A. Casilli, P. Dugo and G. for the doxylamine provided highly Conclusions Dugo, THE APPLICATIONS BOOK accurate agreement with the theoreti- The above applications of high-per- September 2004, 39–40 (2004); P. cal isotope abundance. This is shown formance MS to the analysis of physi- Magni, R. Facchetti, A. Cadoppi, F. in the top of Table III. In addition, ological analytes and metabolites in Pigozzo, and C. Brunelli, THE APPLI- the application of “in-source CID” complex biological matrices provide a CATIONS BOOK September 2004, 59 to generate fragment ions provided representative overview of the impact (2004); R. Murray, Anal. Chem. 77(1), both accurate mass fragment identi- that this type of experimentation can 5 A (2005). fication and accurate relative isotope have on analyte identification and (3) J. Harynuk and P.J. Marriott, Anal. abundance of the fragment ions. These quantitation. High mass accuracy, Chem. 78(6), 2028–2034 (2006); M. findings are shown in the bottom of mass resolving power, and relative Junge, S. Bieri, H. Huegel, and P.J. Mar- Table III. All but the lowest of signals isotope abundance prove to be invalu- riott, Anal. Chem. 79(12), 4448–4454 provide better than 5% relative dif- able tools for analyte identification in (2007). ference versus the theoretical isotope the experiments discussed and facili- (4) T. Kind and O. Fiehn, BMC Bioinformatics abundance. Furthermore, the mea- tated the identification of analytes as 8, 105–124 (2007). THE APPLICATION NOTEBOOK – JUNE 2011 69

Table III: Mass accuracy and relative isotope abundance for doxylamine using “in source CID.” Parent ion data are provided at the top of the table and product ion data at the bottom. Parent Ion Channel

Mass Resolving m/z Error Expected Relative Observed Relative Relative Isotopic Assignment Observed m/z Power (ppm) Abundance Abundance Accuracy

+ [C17H23N2O] 271.18049 49409 0.0 M+1 272.18398 49044 0.1942 0.1854 –4.5%

M+2 273.18671 41602 0.0180 0.0173 –3.8%

Product Ion Channel

Mass Resolving m/z Error Expected Relative Observed Relative Relative Isotopic Assignment Observed m/z Power (ppm) Abundance Abundance Accuracy

+ [C4H12NO] 90.09145 35232 1.2

+ [C13H12NO] 182.09672 45503 1.6 M+1 183.10023 43879 0.1456 0.1487 2.1%

M+2 184.10332 42877 0.0091 0.0097 5.8%

+ [C17H23N2O] 271.18105 48606 2.1 M+1 272.18446 49421 0.1942 0.1853 –4.6%

M+2 273.18732 49147 0.0180 0.0232 29.2%

(5) T. Kind and O. Fiehn, BMC Bioinformat- (17) ChemSpider, http://www.chemspider. ics 7, 234–243 (2006). com/FullSearch.aspx. on the web (6) M. Kellman, H. Muenster, P. Zomer, and (18) Kyoto Encyclopedia of Genes and H. Mol, J. Am. Soc. Mass Spectrom. 20, Genomes, http://www.genome.jp/kegg/ on demand 1464–1476 (2009). pathway.html. (7) J.M. Herniman, G.J. Langley, T.W.T. (19) D.K. Layman and D.A. Walker, J. Nutri- educational Webcast Bristow, and G. O’Connor, J. Am. Soc. tion 136, 319S–326S (2006). Faster GC...or Fastest? Mass Spectrom. 16, 1100–1108 (2005). (20) P. She, C. Van Horn, T. Reid, S.M. Hut- (8) R. Ketterlinus and I. Sanders, G.I.T. Lab- son, R.N. Cooney, and C.J. Lynch, Am. With the right combination of injection, oratory Journal, 7–8(12), 26–27 (2008). J. Physiol. Endocrin. Metab. 293, E1552– column, oven temperature and carrier gas (9) K. Yu, M. Xin, J. Castro-Perez, X. Fu, Y. E1563 (2007). flow, modern capillary gas chromatographs Chen, H. Guo, J. Shockor, and B. Mur- (21) X. Zhao, J. Fritsche, J. Wang, J. Chen, are capable of performing fast, high resolu- phy, Current Trends in Mass Spectrom. K. Rittig, P. Schmitt-Kopplin, A. Frit- tion separations. When optimizing a gas Oct. 2009, 36–43 (2009). sche, H-U. Haring, E.D. Schleicher, G. chromatographic method for speed, there (10) M.L. Vestal, P. Juhasz, and S.A. Martin Xu, and R. Lehmann, Metabolomics 6(3), are two possible approaches: faster-GC, Rapid Commun. Mass Spectrom. 9, 1044– 362–374 (2010). where an existing system or method is opti- 1050 (1995). (22) R. Brauer, A.B. Leichtle, G.M. Fielder, J. mized, or fastest-GC, in which a new sys- (11) R.J. Cotter, “Time of Flight Mass Spec- Thiery, U. Ceglarek, Metabolomics First tem, optimized for fast-GC is desired. This trometry: Instrumentation and Applica- Online, doi.10.1007/s11306-010-0256-1, webinar will introduce and discuss tech- tions in Biological Research”, Chapter March (2011). niques for making existing methods faster 7, 137–168, American Chemical Society (23) L. Akesson, J. Trygg, J.M. Fuller, R. and the technology, benefits and challenges (1997). Madsen, J. Gabrielsson, S. Bruce, H. for obtaining the fastest separations. (13) M. Toyoda, D. Okumura, M. Ishihara, Stenlund, T. Tupling, R. Pefley, T. Register or watch on demand at and I. Katakuse, J. Mass Spectrom. 3, Lundstedt, A. Lernmark, and T. Moritz, www.chromatographyonline.com/webseminar 1125–1142 (2003). Metabolomics First Online, doi.10.1007/ (14) A.N. Verentchikov, M.I. Yavor, Yu. I. s11306-010-0278-3, March (2011). Hasin, and M.A. Gavrik, Technical Phys- (24) L.G. Rasmussen, F. Savorani, T.M. ics 50(1), 73–81 (2005). Larsen, L.O. Dragsted, A. Astrip, and S.B. 7, Visit ChromAcademy on LCGC’s Homepage (15) A.N. Verentchikov, M.I. Yavor, Yu. I. Engelsen, Metabolomics 71–83 (2011). www.chromacademy.com Hasin, and M.A. Gavrik, Technical Phys- ics 50(1), 82–86 (2005). Jeffrey S. Patrick, Kevin Siek, (16) R.S. Plumb, K.A. Johnson, P. Rainville, Joe Binkley, Viatcheslav For more information on this topic, J.P. Shockcor, R. Williams, J.H. Granger, Artaev, and Michael Mason please visit and I.D. Wilson, Rapid Commun. Mass are with LECO Corporation, Separation www.chromatographyonline.com/majors Spectrom. 20, 2800–2806 (2006). Science Division, St. Joseph, Michigan. ◾ 70 THE APPLICATION NOTEBOOK – JUNE 2011

25-Hydroxyvitamin D2/D3 Analysis in Human Plasma Using LC–MS

One of the biggest debates in clinical medicine currently revolves around understanding the proper levels of vitamin D in patients’ plasma. Besides well-known diseases like rickets and osteomalacia, vitamin D deficiency has been linked to cancer and heart disease and is rapidly becoming the most widely used liquid chromatography–mass spectrometry (LC–MS)– based clinical test. Increases in testing frequency have required more rapid and cost-effective solutions for determining vitamin D levels in

plasma. An LC–MS method for vitamin D2/D3 was adapted for use with core-shell columns to achieve run times of less than 4 min. This method provides rapid, sensitive, rugged, and robust LC–MS-MS analysis of

vitamin D levels in patient serum (LOD of 1 and 2 ng/mL for 25-OH D3

and 25-OH D2, respectively, with CV of 4–7%) and speeds the diagnosis by hospital and clinical laboratories of potential vitamin D deficiencies.

itamin D is recognized as an ing has increased tremendously across a Vessential nutrient with its pri- variety of patient groups (1,2). mary physiological function Vitamin D is metabolized to 25-OH D being to increase intestinal absorption in the liver. Total vitamin D is best deter- of calcium and phosphate and to pro- mined by measuring total 25-hydroxyvi-

mote deposition of these minerals in tamin D (D2 and D3) in serum because newly formed bones. Deficiency and the half-life of 25-OH D is about three abnormal vitamin D levels result in weeks with serum concentrations of impaired bone mineralization and lead 10–50 ng/mL. Vitamin D supplementa- to bone softening diseases — rickets in tion in both food and tablets comes in

children and osteomalacia in adults. In both the D2 and D3 forms, making it addition, a large number of bone dis- imperative to measure 25-OH D2 and orders and mineral metabolism defects 25-OH D3. Although optimal serum have been associated with abnormal concentrations of total 25-OH D are vitamin D levels, including nephritic generally agreed to be ≥30 ng/mL, there syndrome, granulomatous diseases, is considerable discussion on the serum and hypocalcemia, as well as secondary concentration of 25-OH D considered hyperparathyroidism, which frequently to be inadequate for bone and over- complicates renal failure. Additional all health, but <20 ng/mL is generally well-known maladies, including car- regarded as deficient. Serum concentra- diovascular disease, cancer, and auto- tions >100 ng/mL are generally regarded immune disorders, recently have been as potentially toxic (3). Phil Koerner and Michael found to be influenced by vitamin D Vitamin D exhibits a high propensity McGinley deficiencies; as a result, vitamin D test- for inherent endogenous serum protein THE APPLICATION NOTEBOOK – JUNE 2011 71 binding and association. Vitamin D is Table I: MS-MS operating conditions typically not found free in serum sam- Instrument API 4000 w/TurboV Source 4000 QTRAP ples, thereby posing a challenge for sensitive and reproducible high performance Ionization APCI APCI liquid chromatography (HPLC) analysis Scan type MRM MRM without appropriate sample prepara- Polarity Positive Positive tion. Additionally, various serum sample Curtain cas (CUR) 10.00 15.00 matrix constituents are found to cause Nebulizer current 5.00 5.00 ion suppression that reduces accuracy Temperature (TEM) 450.00 450.00 and reproducibility during patient sam- Gas 1 (GS1) 75.00 55.00 ple analysis. This ion suppression can be Gas 2 (GS2) 0.00 0.00 especially troublesome with rapid LC– mass spectrometry (MS) methods where Collision gas (CAD) 6.00 High insufficient resolution can be observed Entrance potential (EP) 10.00 3.30 between matrix contaminants and ana- Interface heater (ihe) ON ON lyte peaks. A recent innovation in HPLC column of ethanol. The precipitation reagent at 8% B for 5 s, then 8–100% B in 200 technology has been the introduction of was prepared by adding 10 µL of inter- s (3.33 min). The column was washed core-shell silica particles. Unlike fully nal standard stock solution to 60 mL and re-equilibrated between injections porous silica media, core-shell media of 95:5 (v/v) acetonitrile–methanol in a and maintained at 35 °C. employ structured particle geometry 100-mL volumetric flask and diluting where a thin porous shell is grafted on to volume with 95:5 acetonitrile–meth- Results and Discussion a nonporous core particle. This particle anol. To analyze patient samples, 350 Analysis of 25-OH D2 and 25-OH D3 geometry has a reduced analyte diffu- µL of precipitating reagent containing from serum necessitates the use of sam- sion distance along with a tight particle internal standard was combined with ple preparation procedures to remove size distribution that results in column 100 µL of serum and vortexed for 30 s. potential matrix constituents, which efficiency on par with or better than sub- After inspecting the tube to ensure that will interfere with accurate and precise 2-µm media while maintaining column the sample was fully mixed, the sample determination of 25-OH D in serum back pressures more in line with tra- was centrifuged at 13,000 rpm for 15 and reduce HPLC column life. Vita- ditional HPLC columns. Higher per- min. Supernatant was transferred to an min D and the hydroxy D metabolites formance at lower back pressure using HPLC vial and 40 µL was injected onto are relatively more hydrophobic than core-shell HPLC columns offers the an LC–MS-MS system. virtually all endogenous and exog- promise of improving existing HPLC LC–MS-MS was performed using enous compounds that are typically methods without having to replace an HPLC system equipped with binary quantified in biological matrices. The existing HPLC instrumentation with pump, autosampler, and column oven propensity for vitamin D to inherently ultrahigh-pressure-compatible LC sys- interfaced with an Applied Biosystems associate with serum proteins reduces tems. This article discusses an example 4000 Qtrap mass spectrometer (AB the bioavailability of free vitamin D in of using core-shell columns to improve Sciex, Foster City, California) with a serum; therefore, it is necessary to dis- the resolution of a separation while Turbo V ion source or with an Applied rupt this association for improved assay reducing analysis run time. Biosystems API 4000 triple-quad- accuracy and precision. Protein precip- rupole mass spectrometer. An atmo- itation is the easiest means of disrupt- Experimental spheric pressure chemical ionization ing the serum protein association with Reagents and Chemicals (APCI) source was used and was run in hydrophobic analytes. In this method,

25-Hydroxyvitamin D3 a nd positive ion mode. The MS-MS operat- protein precipitation is performed in d 25-hydroxyvitamin D3- 6 were obtained ing conditions are listed in Table I. The 1.5-mL centrifuge tubes with the addi- from Medical Isotopes (Pelham, New HPLC column used for all analyses was tion of 100 µL of serum sample to 350

Hampshire) and 25-hydroxyvitamin D2 a 50 mm × 4.6 mm, 2.6-µm core-shell µL of 95:5 acetonitrile–methanol con- d was obtained from Sigma-Aldrich (St. Kinetex C18 column (Phenomenex, taining the 25-OH D3- 6 internal stan- Louis, Missouri). HPLC-grade water Torrance, California) operating at a dard. The low solubility of the endog- (Milli-Q, Millipore, Billerica, Mas- flow rate of 1 mL/min. This method enous proteins in acetonitrile results sachusetts) was used to prepare HPLC was compared against a methodology in their precipitation from the sample; mobile phase and for sample prepara- using a fully porous 5-µm C18 column. mixing and centrifugation cause the tion. Methanol and acetonitrile were A gradient method was employed using precipitated protein to form a pellet at obtained from Honeywell, Burdick & an aqueous mobile phase A of 0.05% the bottom of the centrifuge tube, and Jackson (Muskegon, Michigan). Stock formic acid in water and an organic the supernatant is then analyzed. An internal standard solution was prepared mobile phase B of 0.1% formic acid in alternative to the traditional protein by dissolving the contents of a 5-mg vial methanol with 5 mM ammonium ace- precipitation approach involves the use d of 25-hydroxyvitamin D3- 6 in 5.0 mL tate. The gradient was as follows: hold of a protein precipitation plate, such 72 THE APPLICATION NOTEBOOK – JUNE 2011

Table II: API 4000 triple-quadrupole MS system precipitation plate to eliminate the precipitated protein from the sample. MRM Pair Dwell Time (s) DP CE CXP Analyte (Q1/Q3) LC–MS was performed in multiple reaction monitoring (MRM) mode (see 25-OH D 395.3 / 209.3 200 66.0 20.0 6.0 2 Tables II and III for the MRM tran- 25-OH D 383.2 / 257.2 200 66.0 31.0 13.0 3 sitions monitored using the API 4000 25-OH D -d 389.3 / 263.3 200 82.0 30.0 15.0 3 6 and QTRAP MS systems, respectively). The use of MRM is important because

Table III: 4000 QTrap MS system 25-OH D2 and 25-OH D3 are not separated chromatographically; how- MRM Pair Dwell Time (s) DP CE CXP Analyte (Q1/Q3) ever, the unique parent–daughter ion 25-OH D 395.3 / 209.3 200 65.0 20.0 6.4 combination for each analyte allows for 2 specificity and accurate determination 25-OH D 383.2 / 257.2 200 70.0 34.0 16.8 3 of the concentration for each analyte in d 25-OH D3- 6 389.3 / 263.4 200 88.0 23.0 19.0 d the sample. 25-OH D3- 6 was used as an internal standard and signal intensity of each 25-OH D analyte rela-

4.5e4 tive to the internal standard was used 3.0e4 2.5e4 for determining the concentration of 2.0e4 1.5e4 25-OH D and 25-OH D in the sam- 1.0e4 2 3 Intensity (cps) 5000 0 ple. The calibration curves were lin- 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 Time (min) ear over the range from 0 to >500 ng/ 7000 6000 mL, with observed limits of detection 5000 4000 (LOD) of 1 and 2 ng/mL for 25-OH 3000 2000 D and 25-OH D , respectively. The Intensity (cps) 1000 3 2 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 reproducibility of this assay was very Time (min) good with CV of 4–7%. 2.5e4 2.0e4 Both the HPLC and UHPLC condi- 1.5e4 tions using a 50 mm × 2.0 mm, 5-µm 1.0e4

Intensity (cps) 6000 fully porous C18 column and the 2.6- 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 µm core-shell column, respectively, Time (min) allow for sufficient retention of 25-OH Figure 1: LC–MS-MS analysis of 25-OH D and D standards run on a fully porous 2 3 D2 and 25-OH D3, further minimizing 5-µm C18 column. Note the similar elution time for 25-OH D , 25-OH D -d internal 2 3 6 the potential for interference and ion- standard, and the 25-OH D peaks. 3 suppression from any weakly retained impurities (Figures 1 and 2). Using the 5-µm C18 column, elution of 25-OH 6.0e4 D and 25-OH D occurred in just 5.0e4 25-OH 2 3 4.0e4 D3-d6 (IS) under 5 min, with an overall chromato- 3.0e4

Intensity (cps) 2.0e4 graphic run time of 8 min, including 1.0e4 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 column re-equilibration. The separa- Time (min) tion using the 2.6-µm core-shell C18 1.3e4 1.2e4 25-OH column is similar; however, 25-OH D 1.0e4 2 D3 6000 and 25-OH D are eluted in less than 4 5000 3 4000 Intensity (cps) min. This allows the overall chromato- 2000 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 graphic run time, including column re- Time (min) equilibration, to be reduced to 6 min. 3.5e4 3.0e4 2.5e4 25-OH The shorter analysis time using the D3 2.0e4 core-shell column is a significant benefit 1.5e4 1.0e4 Intensity (cps) for laboratories analyzing a large num- 5000 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 ber of samples in a high-throughput Time (min) sample environment — the reduction Figure 2: UHPLC–MS-MS analysis of 25-OH D and D standards run on a Kinetex 2 3 in overall chromatographic run time 2.6-µm C18 column. Note the reduced retention time, improved peak shape, and increased peak height for the standards run on the Kinetex core-shell column. translates into a 25% increase in sample throughput and corresponding reduc- as Strata Impact (Phenomenex), which to the serum sample to facilitate pro- tion in solvent usage. In addition to the contains an oleophobic membrane fil- tein precipitation, the sample passes faster chromatographic separation, the ter. Following addition of acetonitrile through a 0.2-µm cut-off filter of the peak intensities are significantly larger THE APPLICATION NOTEBOOK – JUNE 2011 73

serum samples and speeds the diagnosis

4.5e4 of potential vitamin D deficiencies indic- 3.0e4 2.5e4 ative of specific disease states quickly and 2.0e4 1.5e4 with a high degree of precision. Protein 1.0e4 Intensity (cps) 5000 precipitation is used for sample prepara- 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 tion of patient serum samples, effectively Time (min) 7000 disrupting the serum protein association 6000 5000 with 25-OH D and 25-OH D and 4000 2 3 3000 providing sufficient sample cleanup 2000 Intensity (cps) 1000 before LC–MS-MS analysis. Protein 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 precipitation plates would be an effec- Time (min) 1.00e4 tive alternative for sample preparation 8000 6000 in a high-throughput clinical laboratory 4000 environment.

Intensity (cps) 2000 0 The core-shell column allows for faster 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3 5.4 Time (min) chromatographic analysis of patient samples and increased signal intensity Figure 3: LC–MS-MS analysis of 25-OH D2 and D3 standards spiked into plasma run on the fully porous 5-µm C18 column after protein precipitation. Note the interfering for improved sensitivity. The increase

D3 cholesterol peak that is eluted just after the 25-OH D3 peak. This interference can in chromatographic resolution from an greatly perturb the quantitation of the D3 peak. endogenous compound present in patient serum samples provides improved accuracy and reproducibility and a decrease in 0.44 6.0e4 sample reanalysis. This clinical example 25-OH 5.0e4 D3-d6 (IS) shows the advantages of using core-shell 4.0e4 3.0e4 columns with existing HPLC instru- 2.0e4

Intensity (cps) 1.0e4 mentation to realize an improvement 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 in throughput and performance. Other Time (min) 0.47 UHPLC columns require ultrahigh- 1.14e4 1.00e4 25-OH D2 pressure instrumentation, which can be 8000 6000 a difficult (and unnecessary) choice in 4000

Intensity (cps) 2000 the current economic climate. 0 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 Time (min) References 0.44 1.6e4 1.4e4 25-OH D3 (1) Dietary Supplement Fact Sheet: Vitamin D; 1.2e4 1.0e4 Office of Dietary Supplements, National 8000 6000 4000 Institutes of Health (http://ods.od.nih.gov/ Intensity (cps) 2000 0 factsheets/vitamind.asp). 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 Time (min) (2) M.F. Holick, American Journal of Clinical Nutrition 80(6), 1678–1688 (2004). Figure 4: LC–MS-MS analysis of 25-OH D2 and D3 standards spiked into plasma run on the core-shell Kinetex 2.6-µm C18 column after protein precipitation. Note that (3) Institute of Medicine, Food and Nutrition the interfering peak in Figure 3 with the fully porous column is not eluted near the Board. Dietary Reference Intakes: Calcium,

25-OH D3 peak, thus peak quantitation is not affected when the core-shell column is Phosphorus, Magnesium, Vitamin D, and used for this application. Fluoride. Washington, DC: National Acad- emy Press, 1997. on the core-shell column resulting in pound in patient samples was resolved improved sensitivity and lower quanti- from 25-OH D3 and was eluted out- Phil Koerner and Michael tation limits. side the quantitation window on the McGinley are with Phenomenex, The most important benefit is the core-shell column (Figure 4), resulting Torrance, California. ◾ improved efficiency and resolution pro- in increased analytical accuracy. This vided by the core-shell column. With improvement in throughput and resolu- the fully porous 5 µm C18 column, an tion was achieved at system back pres- Visit ChromAcademy on LCGC’s Homepage endogenous compound (D3 cholesterol) sures compatible with standard HPLC www.chromacademy.com present in patient samples was found to systems (400-bar back-pressure limit). be coeluted with the 25-OH D3 peak, impacting accurate quantitation and Conclusions For more information on this topic, increasing the need to reanalyze a large The analytical method presented above please visit percentage of patient samples (Figure 3). allows for the rapid and accurate deter- www.chromatographyonline.com The presence of the endogenous com- mination of vitamin D levels in patient 74 THE APPLICATION NOTEBOOK – JUNE 2011 THE APPLICATION NOTEBOOK Call for Application Notes

LCGC is planning to publish the next issue Format .TIF or .EPS files with a minimum resolution of Te Application Notebook special supple- • Title: short, specific, and clear of 300 dpi. Lines of chromatograms must be ment in September. Te publication will in- • Abstract: brief, one- or two-sentence heavy enough to remain legible after reduc- clude vendor application notes that describe abstract tion. Provide peak labels and identification. techniques and applications of all forms of • Introduction Provide figure captions as part of the text, chromatography and capillary electrophore- • Experimental Conditions each identified by its proper number and title. sis that are of immediate interest to users in • Results If you wish to submit a figure or chromato- industry, academia, and government. If your • Conclusions gram, please follow the format of the sample company is interested in participating in • References provided below. these special supplements, contact: • Two graphic elements: one is the company logo; the other may be a sample chromato- Tables Michael J. Tessalone, Group Publisher, gram, figure, or table Each table should be typed as part of the main (732) 346-3016 • Te company’s full mailing address, tele- text document. Refer to tables in the text by Edward Fantuzzi, Associate Publisher, phone number, fax number, and Internet roman numerals in consecutive order (Table (732) 346-3015 address I, etc.). Every table and each column within Stephanie Shaffer, East Coast Sales All text will be published in accordance with the table must have an appropriate heading. Manager, LCGC’s style to maintain uniformity through- Table number and title must be placed in a (508) 481-5885 out the book. It also will be checked for gram- continuous heading above the data presented. matical accuracy, although the content will If you wish to submit a table, please follow Application Note Preparation not be edited. Text should be sent in electronic the format of the sample provided below. It is important that each company’s mate- format, preferably using Microsoft Word. rial fit within the allotted space. Te edi- References tors cannot be responsible for substantial Figures Literature citations must be indicated by ara- editing or handling of application notes Refer to photographs, line drawings, and bic numerals in parentheses. List cited refer- that deviate from the following guidelines: graphs in the text using arabic numerals in ences at the end in the order of their appear- Each application note page should be no consecutive order (Figure 1, etc.). Company ance. Use the following format for references: more than 500 words in length and should logos, line drawings, graphs, and charts must (1) T.L. Einmann and C. Champaign, Science follow the following format. be professionally rendered and submitted as 387, 922–930 (1981).

Table I: Factor levels used in the designs Factor Nominal value Lower level (−1) Upper level (+1) Gradient profile 1 0 2 Column temperature (°C) 40 38 42 Buffer concentration 40 36 44 Mobile-phase buffer pH 5 4.8 5.2 Detection wavelength (nm) 446 441 451 Triethylamine (%) 0.23 0.21 0.25 Dimethylformamide 10 9.5 10.5 Figure 1: Chromatograms obtained using the conditions under which the ion sup- Te deadline for submitting application notes for the pression problem was originally discov- September issue of The Application Notebook is: ered. The ion suppression trace is shown on the bottom. Column: 75 mm × 4.6 mm July 26, 2011 ODS-3; mobile-phase A: 0.05% heptafluorobutyric acid in water; mobile-phase B: 0.05% heptafluorobutyric acid in aceto- Tis opportunity is limited to advertisers in LCGC North America. nitrile; gradient: 5–30% B in 4 min. Peaks: For more information, contact: = = 1 metabolite, 2 internal standard, Mike Tessalone at (732) 346-3016, Ed Fantuzzi at (732) 346-3015, 3 = parent drug. or Stephanie Shaffer at (508) 481-5885. Chromatography Online: A Trusted Resource for Separation Scientists

LCGC’s website, www.chromatographyonline.com, spans across the world providing original and fresh material across all key segments including food analysis, bioanalysis, environmental analysis, pharmaceutical analysis and petroleum and biofuels research.

Access the most comprehensive and essential resource for professionals in the chromatography industry providing up to date news, interactive web seminars, podcasts, archived articles and opinion surveys. www.chromatographyonline.com SpeedVisit • Simplicity us at HPLC • Solutions 2011 www.dionex.com/hplc2011 Run IC Samples Any Time 24/7 Speed • Simplicity • Solutions

ICS-5000 Capillary IC allows you to run your samples when you need to . . .

that’s IC on demand.

On Demand means: • No Start-Up Time = Instant Results • No Equilibration = Saves Time • Less Frequent Calibration = Faster Results

ICS-5000 Capillary IC delivers speed, simplicity, and solutions. Visit www.dionex.com/gettheproof

PIN 1070