the application - informa...

SUPPLEMENT TO

THE

APPLICATION NOTEBOOK

February 2015

www.chromatographyonline.com

TH

E A

PP

LIC

AT

ION

NO

TE

BO

OK

A

SU

PP

LE

ME

NT

TO

LC

GC

NO

RT

H A

ME

RIC

A

Fe

bru

ary 2

01

5

ES560786_LCGCAN0215_CV1.pgs 01.28.2015 22:57 ADV blackyellowmagentacyan

A New Industry-Standard Integrated HPLC

www.ssi.shimadzu.com/iseries

ES560448_LCGCAN0215_CVTP1_FP.pgs 01.28.2015 19:37 ADV blackyellowmagentacyan

Shimadzu Scientific Instruments Inc. 7102 Riverwood Dr.

Columbia, MD 21046, USA • (800) 477-1227

www.ssi.shimadzu.com/iseries

— Realization of Advanced Laboratory

• ICM (Interactive Communication Mode) provides maximum freedom of operation and monitoring− Starting an analysis or performing routine operations such as purging from the i-Series interface maximizes ease-of-use

− Remote monitoring using smart devices means you are never out-of-touch with your analysis progress

• Robust reliability and stability− Dual temp-control with TC-Optics for PDA and UV-Vis detectors results in superior stability regardless of room temperature fl uctuations

− Excellent linearity and reproducibility due to superior injection volume precision down to less than 1ul

− Optional video camera provides easy needle alignment verifi cation

— Achieving Easier Operation

• Unifi ed graphical user interface between system and workstation

• Color touch screen panel with intuitive software design

• Easy-to-see status indicator for quick determination of Ready, Pre-treatment, Run, and Error

— Smart Features Increase Work Effi ciency

• Migrate existing methods from either Shimadzu or non-Shimadzu systems

• Automatic ECO mode after non-user interaction reduces power consumption

A New Industry-Standard

Integrated HPLC

ES560449_LCGCAN0215_CVTP2_FP.pgs 01.28.2015 19:37 ADV blackyellowmagentacyan

SUPPLEMENT TO

THE

APPLICATION NOTEBOOK

February 2015


TH

E A

PP

LIC

AT

ION

NO

TE

BO

OK

A

SU

PP

LE

ME

NT

TO

LC

GC

NO

RT

H A

ME

RIC

A

Fe

bru

ary 2

01

5

ES560786_LCGCAN0215_CV1.pgs 01.28.2015 22:57 ADV blackyellowmagentacyan

Think of these instruments as your Tools for Macromolecular Characterization

A grocery scale is your go-to tool for weighing apples and oranges—you wouldn’t consider buying produce in a grocery store without

one! To determine the molar masses of your proteins or biopolymers in solution, the essential lab tool is a DAWN® or a miniDAWN™ Multi-Angle Light Scat-tering (MALS) detector connected to your favorite GPC/SEC. In fact, Wyatt Technology provides an en-tire biophysical characterization toolbox covering the macromolecular and nanoparticle essentials: molar mass, size, charge and interactions.

The toolbox includes the high-throughput DynaPro® Plate Reader, for making thousands and thousands of unattended Dynamic Light Scattering (DLS) measurements to assess size, aggregation and stability of your samples.

Wyatt’s Möbius® instruments are the most sensi-tive on earth for gently —yet quickly—determining the zeta potential and charge on your molecules or QDQRSDUWLFOHV��LQ�IRUPXODWLRQ�EX�HU��'LG�ZH�PHQWLRQ�

that it can be automated via an autosampler? And our unique Calypso® Composition-Gradi-

ent system enables you to investigate interactions in solution without labeling or immobilization, to deter-

PLQH�ELQGLQJ�D�QLW\�DQG�absolute complex stoichi-ometry of self- and hetero-associating biomolecules.

Our customers undergo intensive days of train-ing at Light Scattering University® in Santa Barbara, that earns them a Masters in Light Scattering diplo-ma. All work and no play? After a long day’s work we spoil our customers with gourmet dinners and award-winning accommodations.

Still not convinced? Search our on-line bibliog-raphy consisting of thousands of published refer-ences that rely on our tools (more than 10,000 at last count). An expert sample-analysis laboratory is at your disposal, too, in order to see which measure-ments are best.

So what are you waiting for? Open the box and select your next essential biophysical characterization tool! Find us at www.wyatt.com.

09

8

1

7

65

4

3

2

Molar Mass

Charge

SizeInteractions

SEC - MALS • FFF - MALS • CG - MALS • DLS • MP - PALS

Molar Mass • Size • Charge • Interactions

ES561702_LCGCAN0215_CV2_FP.pgs 01.29.2015 21:22 ADV blackyellowmagentacyan

Commercial environmental laboratories analyze

large numbers of samples per day using approved

methods, and produce data that needs to be

technically sound and legally defensible. The

Shimadzu GCMS-QP2010 Series helps meet

this need. Whether you’re analyzing VOCs, semi-

volatiles, pesticides, PAHs, or PCBs, these powerful,

high-quality instruments offer rapid analysis times,

long-term calibration stability, and greater uptime/

usability with easier system maintenance.

Shimadzu’s GCMS-QP2010 Series features:■ High-speed oven cooling for shorter cycle times

■ Front access to the ion source for easier, faster routine maintenance

■ Install two columns into one GCMS, and run two methods (such as 8260 and 8270) without venting to change the column

■ Automatic Adjustment of Retention Time (AART) after routine maintenance, with >99.9% accuracy and no need to adjust flow rates

■ Easy sTop eliminates the need to vent the MS during routine injection port maintenance, saving up to three hours

■ Simultaneous Scan/SIM for qualitative and quantitative data in a single run

■ High-speed scanning produces library-searchable mass spectra even on the narrowest GC peaks

Order consumables and accessories on-line at http://store.shimadzu.comShimadzu Scientific Instruments Inc., 7102 Riverwood Dr., Columbia, MD 21046, USA

Learn more about Shimadzu’s GCMS-QP2010 SE and GCMS-QP2010 Ultra.

Call (800) 477-1227 or visit us online at

www.ssi.shimadzu.com/QP2010

With Dependable, Long-term Performance, Shimadzu GCMS Systems Provide Ultimate Value

Reliability, Productivity, and Maximum Return on Investment

ES561701_LCGCAN0215_003_FP.pgs 01.29.2015 21:22 ADV blackyellowmagentacyan

THE APPLICATION

NOTEBOOK

Medical/Biological

9 Amino Acid Isomer Separation Using the

Intrada Amino Acid Column from Imtakt

Robert Puryear, Piotr Macech and Itaru Yazawa, Imtakt USA

10 High Throughput Sub-4 Min Separation of

Antibodies Using Size Exclusion Chromatography

Justin Steve and Atis Chakrabarti, PhD, Tosoh Bioscience LLC

12 Quantitative Analysis of Natural

Cannabinoids Using LC–MS-MS

Jonathan Edwardsen, Shimadzu Scientific Instruments

13 Quantitative Analysis of EtG and EtS in Urine Using

Ion Exchange SPE and an Aqueous C18 HPLC Column

Jody Searfoss, UCT, LLC

14 Molecular Weight Determination of Low-Molecular-

Weight Heparins: SEC-MALS Versus SEC-UV-RI

Wyatt Technology Corporation

Chiral

15 Detection of Ibuprofen Using a Circular

Dichroism Detector and Chiral Column

JASCO

4 THE APPLICATION NOTEBOOK – FEBRUARY 2015

TABLE OF CONTENTS

ES561391_LCGCAN0215_004.pgs 01.29.2015 15:46 ADV blackyellowmagentacyan

Environmental

16 Automated Extraction of Organochlorine

Pesticides with In-Line Florisil Cleanup

FMS

17 Using Nitrogen as an Alternate Purge Gas for

Analysis of Volatile Organic Compounds (VOCs)

Cynthia Elmore, OI Analytical

18 Volatile Organic Compounds in Wastewater:

Performance of US EPA Method 624 on Modern

GC–MS and Purge and Trap Instrumentation

Brahm Prakash, Laura Chambers, and William Lipps,

Shimadzu Scientific Instruments

Food and Beverage

19 Analysis of Diacetyl and Other Vicinal

Diketones (VDKs) in Alcoholic Beverages

Andrew James, Ellutia

20 Simultaneous Analysis of Vitamins B1, B2, B3,

and B6 in Protein Powders and Supplements

Maria Ofitserova and Sareeta Nerkar, Pickering Laboratories, Inc.

22 Separation of Lactulose and Epilactose

Melissa Turcotte* and Satoko Sakai†, *Showa Denko

America, Inc. and †Showa Denko K.K.

23 Analysis of (E)-2-Nonenal in Beer with the

Atomx and GC–MS SIM Detection

Roger Bardsley, Teledyne Tekmar

THE APPLICATION NOTEBOOK – FEBRUARY 2015 5

TABLE OF CONTENTS


Industrial

24 AFFF-MALS-RI for Determining the Mass and Size

Distributions of Amylose and Amylopectins in Starch

Wyatt Technologyl

Pharmaceutical/Drug Discovery

25 Separation of Macrocyclic Lactones

Diamond Analytics

26 Accurate Pain Management Analysis in Under

5 Min on Raptor™ Biphenyl Superficially

Porous Particle LC Columns

Sharon Lupo, Ty Kahler, and Paul Connolly, Restek Corporation

28 Rapid Separation of Basic Drug Compounds

on pH-Stable Hamilton PRP™-C18

Derek Jensen and Mark Carrier, Hamilton Company

General

29 Fast and Robust Analysis of Drugs of Abuse Using

Direct Exposure Probe on the Pegasus HT

Jonathan D. Byer and Joe Binkley, Life Science and

Chemical Analysis Centre, LECO Corporation

30 A Better Way to Profile Your Ion Beam in

Instrument Design for More Accurate Analysis

Photonis, USA

31 Ruggedness of YMC Cyano-High Strength Stationary

Phase Under Acidic and Basic Extremes

Jeffrey A. Kakaley and Ernest J. Sobkow, YMC America Inc.


TABLE OF CONTENTS


Articles

32 In Memoriam: Professor Georges A. Guiochon

Mark R. Schure and Lois Ann Beaver (Guiochon)

34 Quantification of Individual Mass Transfer

Phenomena in Liquid Chromatography for Further

Improvement of Column Efficiency

Fabrice Gritti

Departments

47 Call for Application Notes

Cover Photography: Getty Images


TABLE OF CONTENTS



®

Publishing & Sales

485F US Highway One South, Suite 210, Iselin, NJ 08830

tel. (732) 596-0276 fax (732) 647-1235

Science Group Publisher Michael J. Tessalone [email protected]

Associate Publisher Edward Fantuzzi [email protected]

East Coast Sales Manager Stephanie Shaffer [email protected]

Account Executive Lizzy Thomas [email protected]

EditorialEditorial Director Laura Bush

[email protected] Editor Megan L’Heureux

[email protected] Technical Editor Stephen A. Brown

[email protected] Editor Cindy Delonas

[email protected] Director Dan Ward

[email protected]

Marketing Manager Anne [email protected]

Classified/Recruitment Sales Representative Tod [email protected]

Direct List Rental Sales Tamara [email protected]

Permissions Maureen [email protected]

Reprint Services 877-652-5295 ext. 121/ [email protected] US, UK, direct dial: 281-419-5725. Ext. 121

Production Manager Jesse [email protected]

Audience Development Manager Wendy [email protected]

Assistant Audience Development Manager Gail [email protected]

Chief Executive Officer Joe Loggia

Executive Vice-President, Life Sciences Tom Ehardt

Executive Vice-President Georgiann DeCenzo

Executive Vice-President Chris DeMoulin

Executive Vice-President, Business Systems Rebecca Evangelou

Executive Vice-President, Human Resources Julie Molleston

Executive Vice-President, Strategy & Business Development Mike Alic

Sr Vice-President Tracy Harris

Vice-President, General Manager

Pharm/Science Group Dave Esola

Vice-President, Legal Michael Bernstein

Vice-President, Media Operations Francis Heid

Vice-President, Treasurer & Controller Adele Hartwick

UBM Advanstar (www.ubmadvanstar.com) is a leading worldwide media company providing integrated marketing solutions for the Fashion, Life Sciences and Powersports industries. UBM Advanstar serves business professionals and consumers in these industries with its portfolio of 91 events, 67 publications and directories, 150 electronic publications and Web sites, as well as educational and direct marketing products and services. Market leading brands and a commit-ment to delivering innovative, quality products and services enables UBM Advanstar to “Connect Our Customers With Theirs.” UBM Advanstar has approximately 1000 employees and currently

operates from multiple offices in North America and Europe.

©2015 Advanstar Communications Inc. All rights reserved. No part of this publica-

tion may be reproduced or transmitted in any form or by any means, electronic or

mechanical including by photocopy, recording, or information storage and retriev-

al without permission in writing from the publisher. Authorization to photocopy

items for internal/educational or personal use, or the internal/educational or per-

sonal use of specif c clients is granted by Advanstar Communications Inc. for librar-

ies and other users registered with the Copyright Clearance Center, 222 Rosewood

Dr. Danvers, MA 01923, 978-750-8400 fax 978-646-8700 or visit http://www.copy-

right.com online. For uses beyond those listed above, please direct your written

request to Permission Dept. fax 440-756-5255 or email: [email protected].

UBM Advanstar provides certain customer contact data (such as customer’s name,

addresses, phone numbers, and e-mail addresses) to third parties who wish to pro-

mote relevant products, services, and other opportunities that may be of interest

to you. If you do not want UBM Advanstar. to make your contact information avail-

able to third parties for marketing purposes, simply call toll-free 866-529-2922 be-

tween the hours of 7:30 a.m. and 5 p.m. CST and a customer service representative

will assist you in removing your name from UBM Advanstar’s lists. Outside the U.S.,

please phone 218-740-6477.

LCGC North America does not verify any claims or other information appearing

in any of the advertisements contained in the publication, and cannot take re-

sponsibility for any losses or other damages incurred by readers in reliance of

such content.

To subscribe, call toll-free 888-527-7008. Outside the U.S. call 218-740-6477.

MANUSCRIPTS: For manuscript preparation guidelines, see

chromatographyonline.com/AuthorInfo, or call The Editor, (732) 596-0276. LCGC

welcomes unsolicited articles, manuscripts, photographs, illustrations, and other

materials but cannot be held responsible for their safekeeping or return. Every

precaution is taken to ensure accuracy, but LCGC cannot accept responsibility for

the accuracy of information supplied herein or for any opinion expressed.

SUBSCRIPTIONS: For subscription and circulation information: LCGC, P.O. Box

6168, Duluth, MN 55806-6168, or call (888) 527-7008 (7:00 a.m.–6:00 p.m. central

time). International customers should call +1-218-740-6477. Delivery of LCGC

outside the United States is 14 days after printing. For single and back issues,

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

Western Europe and Asia and Australia, respectively.)

CHANGE OF ADDRESS: Send change of address to LCGC, P.O. Box 6168, Duluth,

MN 55806-6168; alternately, send change via e-mail to [email protected]

or go to the following URLs:

• Print: https://advanstar.replycentral.com/Default.aspx?PID=469

• Digital: https://advanstar.replycentral.com/?PID=469&V=DIGI

Allow four to six weeks for change. PUBLICATIONS MAIL AGREEMENT No.

40612608. Return all undeliverable Canadian addresses to: IMEX Global Solu-

tions, P.O. Box 25542, London, ON, N6C 6B2, CANADA. Canadian GST number:

R-124213133RT001.

DIRECT MAIL LIST RENTAL: Contact Tamara Phillips, tel. (440) 891-2773, e-mail

[email protected].

REPRINTS: Reprints of all articles in this issue and past issues of this publication

are available (500 minimum). Call 877-652-5295 ext. 121 or e-mail bkolb@wrights-

media.com. Outside US, UK, direct dial: 281-419-5725. Ext. 121

MARKETING DEVELOPMENT/CLASSIFIED: Contact Tod McCloskey, tel. (440) 891-

2739, fax (440) 826-2865.

RECRUITMENT: Contact Tod McCloskey, tel. (440) 891-2739, fax (440) 826-2865.

INTERNATIONAL LICENSING: Contact Maureen Cannon, tel. (440) 891-2742, fax

(440) 891-2650, or e-mail [email protected].



MEDICAL/BIOLOGICAL

Imtakt USA1104 NW Overton St., Portland, OR 97209

tel. (888) 456-HPLC, (215) 665-8902, fax (501)646-3497

Website: www.imtaktusa.com

The Intrada Amino Acid column has been shown to be a viable

solution for the analysis of amino acids by LC–MS without de-

rivatization. This current work addresses the limitation of this

detection method, where isomers and isobaric compounds may

not have distinguishable m/z ratios.

Separation of amino acids isomers by LC–MS faces multiple challenges.

First, these compounds may be isobaric (same m/z ratio), constraining

the application of a mass spectrometer (MS). Secondly, they may have

very similar molecular structure making them diff cult to resolve chro-

matographically using current HPLC column chemistries.

Here we show that the unique stationary phase of the Intrada Ami-

no Acid column can address these limitations, providing an excellent

solution for the analysis of amino acid isomers by LC–MS. Specif cally,

this note presents data for leucine and threonine isomers.

Experimental Conditions

A 5 µL sample at 1 mM concentration in 0.1 N HCl was injected

onto Intrada Amino Acid column. Eluent was analyzed using single

quadrupole MS in ESI, positive mode. Further experimental details

are provided as inserts in respective chromatograms.

Results and Discussion

Despite the fact that leucine isomers, Figure 1, are structurally simi-

lar, the Intrada Amino Acid column separates these compounds

well. The diff culty with this separation is that they differ primarily

only in the alkyl portion of their structure, specif cally, the methyl

group geometry. This challenge is exacerbated by the fact that the

Amino Acid Isomer Separation Using the Intrada Amino Acid Column from ImtaktRobert Puryear, Piotr Macech and Itaru Yazawa, Imtakt USA

Figure 1: LC–MS separations of leucine isomers.

Figure 2: LC–MS separation of threonine isomers. Intrada Amino Acid, 250 × 3 mm.

Intrada Amino Acid column has a normal phase mixed-mode sta-

tionary phase, which conceptually should struggle with separating

analytes based solely on their hydrophobic portions. Despite this,

the Intrada Amino Acid column is able to separate these isomers,

further highlighting its extraordinary resolving power.

Threonine isomers separation is shown in Figure 2. Overall, the peaks

are well-resolved (elution times of 11, 15, 16, and 18 min), with only

a minimal coelution for peaks 2 and 3 observed. This separation pro-

vides an additional example of the resolving power the Intrada Amino

Acid column in that the only difference between L-threonine and allo-L-

threonine is in their three-dimensional shape, indicating the presence of

steric selectivity in addition to the other modes indicated.

Conclusion

In summary, the Intrada Amino Acid column has been shown to provide

excellent separation power for underivatized amino acid isomers. Re-

solved compounds shown here include ones with distinctive m/z ratios,

isomers containing minimal aliphatic differences, as well as only steric

differences. These results are of particular importance as isomers are

extremely challenging to separate using more conventional approaches,

yet, the Intrada Amino Acid column shows adequate chromatographic

resolution of both positional isomers and stereoisomers. Therefore, this

evidence validates the

use of this column for the

separation of a wide range

of amino acids when used

in concert with mass spec-

trometry detection.



MEDICAL/BIOLOGICAL

High Throughput Sub-4 Min Separation of Antibodies Using Size Exclusion ChromatographyJustin Steve and Atis Chakrabarti, PhD, Tosoh Bioscience LLC

Gel f ltration chromatography (GFC) is a powerful analytical tool in the

separation of antibodies. Traditionally, GFC columns with a dimension

of 7.8 mm i.d. × 30 cm are used for analytical purposes. The longer

column dimension, however, leads to longer run times and sample

dilution, as well as substantial solvent waste. Alternatively, a column of

smaller dimension provides high throughput separation with shorter

run times, high resolution, and minimal solvent waste when used

on a conventional HPLC system. This application note demonstrates

the use of a 4.6 mm i.d. × 15 cm TSKgel® SuperSW mAb HTP SEC

column for the highly reproducible separation of antibodies in less

than 3.5 min by using a moderate f ow rate of 0.75 mL/min. The

TSKgel SuperSW mAb HTP column provides excellent stability for

high speed, sub-4 min separations of monoclonal antibodies.

Materials and Methods

Column: TSKgel SuperSW mAb HTP, 4 μm,

4.6 mm i.d. × 15 cm

Instrument: Agilent 1100 HPLC system

Mobile phas e: 100 mmol/L phosphate/100 mmol/L sulfate buffer,

pH 6.7 + 0.05% NaN3

Flow rate: 0.75 mL/min

Detection: UV @ 280 nm

Temperature: ambient

Injection vol.: 5 μL

Samples: PABA, 0.01 mg/mL

mAb 01, 4.6 mg/mL

mAb 02, 4.6 mg/mL

human IgG, 4.6 mg/mL

mouse IgG, 4.6 mg/mL


Figure 1 shows the separation of four different monoclonal

antibodies in less than 3 min using the TSKgel SuperSW mAb HTP

column at a f ow rate of 0.75 mL/min. High resolution separation of

the monomer, dimer, and fragment peaks of the mouse IgG sample

are clearly shown under these conditions. The separation of these

IgG-based proteins within 3 min using the TSKgel SuperSW mAb

HTP column corresponds to a 3.75-fold decrease in analysis time

relative to conventional (7.8 mm i.d. × 30 cm) SEC columns.

Sustained pressure from operating at elevated f ow rates can lead

to voids within the column, generating poor peak shapes and drifting

retention time. As shown in Figure 2, 540 consecutive injections of mAb

02 and PABA separated on the TSKgel SuperSW mAb HTP column Figure 1: Separation of monoclonal antibodies on TSKgel SuperSW mAb HTP column under high f ow conditions.

0

500

1000

1500

2000

0 0.5 1 1.5 2 2.5 3 3.5 4

De

tec

tor

resp

on

se (

mA

U)

Mouse IgG

Human IgG

mAb 02

mAb 01

Retention time (minutes)

Figure 2: Retention time reproducibility of the TSKgel SuperSW mAb HTP column under high f ow conditions.

Figure 3: Resolution stability of the TSKgel SuperSW mAb HTP col-umn under high f ow conditions.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 100 200 300 400 500 600

Re

solu

tio

n

Number of Injections

Rs (monomer/dimer)



MEDICAL/BIOLOGICAL

at 0.75 mL/min show good reproducibility with no discernible drift in

retention. Additionally, no signif cant loss in resolution between the mAb

monomer and dimer was observed on the TSKgel SuperSW mAb HTP

column operated at 0.75 mL/min, yielding a %RSD < 3 (Figure 3).

Conclusions

Separations of IgG-based proteins using the TSKgel SuperSW mAb HTP

column at 0.75 mL/min yielded highly reproducible results with high

resolution and moderate back pressure within 3 min. This corresponds

to a 3.75-fold decrease in analysis time relative to traditional SEC

methodology. Additionally, due to the smaller dimension of the TSKgel

SuperSW mAb HTP column, minimal solvent waste is observed even at

increased f ow rates, making this a cost effective and “green” method

for protein separations when compared to that of traditional 7.8 mm

i.d. × 30 cm SEC columns. The TSKgel SuperSW mAb HTP column

operated at 0.75 mL/min for 540 injections of monoclonal antibody

showed no drift in retention and good reproducibility. These results

demonstrate that the TSKgel SuperSW mAb HTP, 4 μm, 4.6 mm i.d.

× 15 cm column clearly has a competitive advantage in fast assay

and high throughput analysis of antibodies using a conventional HPLC

system.

Tosoh Bioscience and TSKgel are registered trademarks of Tosoh Corporation.

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



MEDICAL/BIOLOGICAL

Quantitative analysis of natural cannabinoids was conduct-

ed using the LCMS-8050 triple quadrupole mass spectrom-

eter. A lower limit of quantitation (LLOQ) of 1–4 ng/mL was

achieved depending on the specif c cannabinoid. This meth-

od showed certain medicinal oils or tinctures available over

the internet contained naturally occurring cannabinoids.

The analysis of natural cannabinoids is necessary not only because

of potential medical uses for these compounds, but also in the

regulation and quality control testing of products containing these

compounds. To ensure the authenticity, quality, and amount of each

cannabinoid contained in the product, an LC–MS-MS method was

developed using the Shimadzu LCMS-8050 triple quadrupole mass

spectrometer.


After diluting in methanol neat standards of the naturally occurring

cannabinoids, f ow injection analysis was used to optimize source, CID

conditions, and product ion selection. Optimized LC conditions were

developed empirically and a 3-min gradient method was developed

Quantitative Analysis of Natural Cannabinoids Using LC–MS-MSJonathan Edwardsen, Shimadzu Scientif c Instruments

Shimadzu Scientif c Instruments, Inc.7102 Riverwood Drive, Columbia, MD 21046

tel. (800) 477-1227

Website: www.ssi.shimadzu.com

4000000

3500000

3000000

2500000

2000000

1500000

1000000

500000

0

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1

Time min

40000000

45000000

35000000

30000000

25000000

20000000

15000000

10000000

5000000

0

(a)

(b)

CB

G

CB

DA

CB

DA

CB

DC

BD

CB

G

CB

NC

BN

CB

CC

BC

TH

CA

TH

CA

∆9

TH

C∆

9 T

HC

Figure 1: (a) Chromatogram of seven cannabinoids at 100 ng/mL in solvent. (b) Chromatogram of a commercially available tincture containing seven cannabinoids.

Table I: Quantitative results for each cannabinoid at the limit of quantitation and the concentration of the commercial tincture.

Quantitative Results at LLOQ (n=6)

Compound LOD (ng/mL) %RSD %Accuracy S/N Weighting Commercial Tincture

CBN 1 4.516099 99.998± 4.2% 58.96 1/C² 0.016% ± 0.001%

THCA 1 7.023558 99.998 ± 9.1% 21.14 1/C 0.452% ± 0.018%

CBDA 1 6.671582 100.001 ± 5.7% 70.42 1/C² 0.019% ± 0.001%

∆9 THC 1 6.414479 99.997 ± 6.3% 85.89 1/C² 0.370% ± 0.021%

CBG 1 3.666911 100.000 ± 3.7% 2397.6 1/C² 0.018% ± 0.0004%

CBD 1 7.770838 100.123 ± 6.8% 107.4 1/C² 0.006% ± 0.001%

CBC 2.5 8.193242 100.006± 5.7% 70.64 1/C 0.029% ± 0.006%

using a Restek column. Using solvent standards, calibration curves

were created and various medical tinctures were then analyzed.


Cannabinoid optimization identif ed one quantif er and two qualif er

ions for each naturally occurring cannabinoid. The two ions were

selected based on ion intensity and repeatability across multiple

collision energies. The precursor ions selected were the [M+H]+

for all of the cannabinoid compounds. Following MRM optimization

and development of chromatographic conditions, a standard curve

was generated for each cannabinoid with n = 6.

The lower limits of quantitation (LLOQ) were established for each

cannabinoid at 1 ng/mL except for CBC, which was 4 ng/m. The min-

imum signal-to-noise ratio for all of the cannabinoids was determined

to be greater than or equal to 20:1. Calibration curve weighting of

either 1/Concentration (1/C) or 1/Concentration2 (1/C2) was applied.

The chromatographic method that was developed yielded baseline

separation of six of the seven cannabinoids with CBG (m/z 317.25)

and CBD (m/z 314.95) co-eluting. Even though these peaks co-elut-

ed, the LCMS-8050 was able to identify and quantify them without

complete baseline separation. All seven of the cannabinoids were

detected in the commercially available tincture purchased online

(Figure 1). The concentrations for each cannabinoid are presented

in Table I. There was no measurable carryover in the blank injected

immediately after the highest level standard.

Conclusion

This work demonstrates a rapid method for the detection of natu-

rally occurring cannabinoids by using the Shimadzu LCMS-8050.

All seven cannabinoids were detected at levels as low as 1 ng/mL (1

pg on column) with a S/N of at least 20:1. This method is useful for

quantitating cannabinoids in raw or commercial products.



MEDICAL/BIOLOGICAL

Ethyl glucuronide (EtG) and ethyl sulfate (EtS) are conjugated etha-

nol metabolites formed in low amounts in the body following alcohol

consumption. EtG and EtS are excreted in urine for a prolonged time

(EtG up to 80 h and EtS up to 24 h after ingestion). This makes them

valuable, sensitive alcohol biomarkers for complied abstinence.

This SPE method uses a strong anion exchange (QAX) column to

extract the acidic EtG and EtS from urine utilizing a single elution for

both EtG and EtS that plays to the chemical nature of both analytes.

Sample Pretreatment

1. To 0.5 mL of urine sample containing deuterated analogues of

EtG/EtS add 4.5 mL of D.I. H2O

2. Vortex for 30 s

SPE Method

1. Precondition SPE column with 5 mL of MeOH followed by 5 mL

of D.I. H2O.

2. Apply sample to SPE column.

3. Wash SPE column with 5 mL of ACN followed by 5 mL of MEOH.

4. Dry column (10 min at full vacuum or pressure).

5. Elute EtG/EtS with 5 mL of 2% HCl in ACN (collect eluate at

1–2 mL/min).

6. Evaporate to dryness at < 50 °C.

7. Reconstitute sample in 100 μL of D.I. H2O.

UCT, LLC 2731 Bartram Road, Bristol, PA 19007

tel: (800) 385-3153; Email: [email protected]

Website: www.unitedchem.com

Quantitative Analysis of EtG and EtS in Urine Using Ion Exchange SPE and an Aqueous C18 HPLC ColumnJody Searfoss, UCT, LLC

Table III: Gradient Program

Mobile Phase A: 0.1% formic acid in water

Mobile Phase B: 0.1% in methanol

Time (min) %A %B

0 100 0

2.5 100 0

4.0 5 95

6.0 5 95

6.1 100 0

11.0 100 0

Table IV: Results

CompoundRT

(min)

100 ng/mL 500 ng/mL

Result % RSD% (n=5) Result % RSD% (n=5)

EtG 1.31 86.2 4.8 97.9 10.9

EtS 1.69 95.1 6.5 84.9 5.9

Overall mean 90.7 5.65 91.4 8.4

Table I: Materials

CUQAX156 Clean-Up® 6 mL SPE cartridge with 500 mg QAX

SLAQ100ID21-3UMSelectra® Aqueous C18 HPLC column,

100 × 2.1 mm, 3 μm

SLAQGDC20-3UMSelectra® Aqueous C18 guard column,

10 × 2.0 mm, 3 μm

Table II: Instrumentation

System: AB Sciex API 4000 QTrap MS/MS with Agilent 1200 Binary Pump SL

Column: UCT Selectra® Aqueous C18 100 × 2.1 mm, 3 μm

Guard Column: UCT Selectra® Aqueous C18 10 × 2.0 mm, 3 μm

Column Temperature: 50 °C

Column Flow Rate: 0.3 mL/min

Injection Volume: 10 µL

Conclusion

Excellent recoveries were achieved with EtG at 97.9% and EtS at 84.9%.

The extraction eff ciency was evaluated by fortifying samples at two vary-

ing concentrations (100 ng/mL and 500 ng/mL). RSD values were less

than 11% (n = 5 at each concentration). Matrix-matched calibration

curves were used for quantif cation with R2 values ranging from 0.9983

to 0.9998 over the entire concentration range (50–1500 ng/mL). The

limits of detection and quantif cation for this method were determined

to be 25 ng/mL and 50 ng/mL, respectively for EtG and EtS.



MEDICAL/BIOLOGICAL

Molecular Weight Determination of Low-Molecular-Weight Heparins: SEC-MALS Versus SEC-UV-RI Wyatt Technology Corporation

Low-molecular-weight heparins (LMWHs) are obtained by fraction-

ation or depolymerization of natural heparins. They are def ned as

having a mass-average molecular weight of less than 8000 and for

which at least 60% of the total weight has a molecular mass less

than 8000.

Size-exclusion chromatography (SEC) has been the most com-

mon way of measuring the molecular weight and molecular weight

distributions of LMWHs by using the two most common detection

technologies: ultraviolet (UV) coupled with refractive index (RI)

detection. However, these detectors embody a relative method in

order to determine molecular weights, requiring calibration stan-

dards. A newer, absolute method involves the use of multi-angle

light scattering (MALS), which does not require any standards.

The European Pharmacopeia (EP) monograph for LMWH speci-

f es the use of the UV–RI detection method and provides a known

calibration standard. Many laboratories around the world have

adopted this method.

We previously developed an SEC–MALS method and found

it to be very suitable for the analysis of LMWHs. We have

recently adopted the UV-RI method described in the EP mono-

graph and compared the molecular weight results generated for

LMWH using each detection type. The adopted method uses an

Agilent LC-1200 series HPLC, 0.2 M sodium sulfate pH 5.0 mo-

bile phase, Tosoh TSK-gel G2000 SWxl column with Tosoh TSK-gel

Guard SWxl, Waters 2487 dual wavelength UV detector, and Wyatt

Optilab rEX refractive index detector. For MALS analysis, the UV

detector was replaced with a Wyatt miniDAWN TREOS detector; all

other methods aspects remained the same.

The results indicated that both detection types are suitable

and acceptable for the analysis of LMWHs. The molecular weight

and distribution results generated using each detection type are

comparable. This indicates that a SEC–MALS method could be

adopted in place of the SEC–UV-RI method currently required by

the EP monograph, and that it would result in less time because it

obviates the need for calibration standards.

This note was graciously submitted by Lin Rao and John Beirne of Scien-

tif c Protein Laboratories LLC.

Wyatt Technology6300 Hollister Avenue, Santa Barbara, CA 93117

tel. +1 (805) 681-9009, fax +1 (805) 0123

Website: www.wyatt.com Figure 1: Examples of UV and RI traces for an LMWH sample.

Figure 2: Examples of LS and RI traces for an LMWH sample.

LS dRI UV

Define Peaks: LMWH Sample

0.8

0.6

0.4Rel

ativ

e sc

ale

0.2

0.0

5.0 10.0

Time (min)

15.0 20.0 25.0 30.0 35.0

Define Peaks: LMWH Sample

1.0

0.5

0.0

Rel

ativ

e sc

ale

5.0 10.0

Time (min)

15.0 20.0 25.0 30.0 35.0

LS dRI



CHIRAL

Ibuprofen is a non-steroidal anti-inf ammatory drug and widely used

around the world for pain relief and fever reduction. In ibuprofen,

only the S enantiomer has active antipyretic characteristics, but

within the human body an enzymatic reaction can convert the R

enantiomer into the S enantiomer.

The CD detector is well known for high accuracy and selectivity

for chiral analysis on compounds which have circular dichroism.

Both CD and UV signal are simultaneously measured with the ability

to run spectral scanning, which is extremely useful information for

chiral chemicals. Not only can the CD detector help differentiate be-

tween two enantiomers, but it can also distinguish between achiral

impurities and enantiomers of interest.

Experimental

Equipment

Pump: PU-4185

Pump option: DG unit

Autosampler: AS-4050

Column oven: CO-4060

Detector: CD-4095

Structure

Detection of Ibuprofen Using a Circular Dichroism Detector and Chiral Column JASCO

JASCO28600 Mary’s Court, Easton, MD 21601

tel. (800) 822-1220

Website: www.jascoinc.com

Figure 1: Chromatogram of caffeine and racemic ibuprofen standards.Top: CD detection, Bottom: UV detection, 1: caffeine, 2: R-(-)-ibuprofen, 3: S-(+)-ibuprofen.

Figure 2: Spectral measurement. Top: CD, Bottom: UV.

Conditions

Column: CHIRALPAK AD-RH (4.6 mm i.d. × 150 mm L,

5 µm)

Eluent A: 0.1% phosphoric acid aqueous solution

Eluent B: Acetonitrile

Composition: Eluent A / B (60 / 40)

Flow rate: 0.5 mL/min

Column temp.: 25 ºC

Wavelength: 230 nm

Response: 1.5 s

Scan speed: 10 nm/s

Injection volume: 10 mL

Standard sample: Mixture of 50 µg/mL caffeine and 200 µg/mL

racemic form of ibuprofen in eluent A / B (60 / 40)

Results

Figure 1 shows the chromatograms of caffeine and racemic ibuprofen

standards. In the UV chromatogram, both samples are detected but

only ibuprofen can be detected in CD chromatogram as the caffeine

has no chirality. Figure 2 shows the CD and UV spectra with stopped-

f ow technique.



ENVIRONMENTAL

Automated Extraction Of Organochlorine Pesticides with In-Line Florisil Cleanup FMS

Organochlorine pesticides are among the more notorious organic pol-

lutants. Gaining wide spread attention into the latter part of the 20th

century, and ultimately leading to their ban, OCPs remain with us as

a legacy contaminant in the environment. In water samples, OCPs

are monitored by the U.S. Environmental Protection Agency by vari-

ous methods including methods EPA 508, EPA 608, and EPA 8081.

The use of solid phase extraction (SPE) can rapidly increase both

extraction eff ciency and reduce lab solvent usage. By implement-

ing the FMS TurboTrace® ABN system with its dual cartridge func-

tionality, an in-line f orisil cartridge can be added to perform auto-

mated extract clean-up during the elution step.

Instrumentation

• FMS, Inc. TurboTrace® ABN SPE system

• FMS, Inc. SuperVap® 12 Concentrator

• 55 mL SuperVap Tubes vials

• Agilent 7890 GC with µECD detector

Consumables

• FMS Inc. 1 g C18 SPE cartridges

• FMS Inc. Florisil SPE cartridges

• Acetone, pesticide grade or equivalent

• n-hexane, pesticide grade or equivalent

• HPLC grade water

• 6 N HCL

• Anhydrous sodium sulfate, ACS grade or equivalent

SPE Procedure

1. Condition Florisil/Na2SO

4 SPE cartridge with 20 mL 10% acetone

in hexane

2. Condition C18 SPE cartridge with 10 mL acetone

3. Condition C18 SPE cartridges with 20 mL H2O

4. Load samples across C18 SPE cartridges via vacuum

5. Dry C18 cartridge with N2 for 10 min

6. Automatic solvent rinse of sample bottle with 10% acetone in hexane

7. Bottle rinse loaded across both cartridges and collected in SuperVap

concentrator

8. Additional 10 mL elution solvent passed C18 cartridges

9. 10 mL elution solvent passed across Florisil cartridge and nitrogen

purged to SuperVap

Total elution solvent: ~40 mL

SuperVap

1. Preheat temperature: 55 °C

2. Evap mode: 10 psi nitrogen with sensor

FMS, Inc.580 Pleasant Street, Watertown, MA 02472

tel. (617) 393-2396, fax (617) 393-0194

Website: www.fms-inc.com

Results

Mean

Analyte Rec RSD

α-BHC 105.9% 2.2%

γ-BHC (lindane) 105.1% 2.0%

ß-BHC 98.8% 2.1%

δ-BHC 110.8% 2.1%

Heptachlor 100.3% 9.8%

Aldrin 76.4% 6.9%

Heptachlor epoxide 101.3% 1.9%

α-Chlordane 92.1% 3.1%

Endosulfan I 94.9% 2.0%

α-Chlordane 99.7% 2.0%

4,4’-DDE 93.9% 2.6%

Dieldrin 102.0% 2.1%

Endrin 110.5% 2.7%

4,4’-DDD 102.9% 3.7%

Endosulfan II 101.8% 2.7%

Endrin aldehyde 84.1% 5.8%

4,4’-DDT 103.9% 4.1%

Endosulfan sulfate 103.5% 3.0%

Methoxychlor 100.1% 6.3%

Endrin ketone 99.8% 4.5%

Surrogate

TCMX 72.3% 5.7%

PCB-209 78.7% 11.2%

Results of IPR study performed on Turbo Trace ABN system with in-line

Florisil

ConclusionsExtraction of the water samples performed on the TurboTrace ABN

system enabled a fully automated extraction process using its dual

cartridge capabilities. Samples were loaded across the C18 cartridges,

then eluted through the Florisil cartridges in succession. The addition of

Na2SO

4 to the clean-up cartridge enabled residual water removal from

the extract without a separate, manual step. Analytical results yielded

spiked recoveries of Or-

ganochlorine pesticides

to all be between 70–

130%, with low RSDs.


Figure 1: Chromatogram of an LFB extraction at .5 µg/L.


ENVIRONMENTAL

Volatile organic compounds can be concentrated on an

OI Analytical 4660 Purge and Trap using nitrogen as a cost-

effective alternative to helium as a purge gas.

Most United States Environmental Protection Agency (US EPA)

methods for analysis of volatile organic compounds (VOCs) call for

extraction of the analytes by purging with helium for 11 min at 40

mL/min, making purge and trap (P&T) one of the largest consumers

of helium in a laboratory. Many labs are seeking less expensive op-

tions and have turned to nitrogen as a viable alternative. Compared

to helium, nitrogen is abundantly available, inert, safe to use, and

currently a third of the price of helium, resulting in signif cant cost

savings.


This study is based on one of the most widely used methods for

volatiles analysis by GC–MS; US EPA Method 8260B. Instrumenta-

tion for this study included an OI Analytical Eclipse 4660 Purge &

Trap Sample Concentrator, an OI Analytical 4100 Sample Proces-

sor, and an Agilent 7890A/5975C GC–MS. The nitrogen molecule

has a higher heat capacity than helium and is able to withdraw

more thermal energy from the solution through which it is purged;

thus, it is necessary to increase the thermal energy present in the

purged volume. For that reason, several changes were made to the

standard conditions including slowing purge ý ow to 35–37 mL/min,

increasing purge temperature to 55 °C, and installing a 0.6 mm

draw-out plate into the mass spectrometer.

Using Nitrogen as an Alternate Purge Gas for Analysis of Volatile Organic Compounds (VOCs)Cynthia Elmore, OI Analytical

OI Analytical151 Graham Road, P.O. Box 9010, College Station, TX

tel. (979) 690-1711, (800) 653-1711 (US and Canada)

Website: www.oico.com

Table I: Nitrogen purge calibration (range 0.5–200 ppb)

Analyte Compound Waters RF 4100 %RSD Soils RF 4100 %RSD

6 Chloromethane 0.726 9.07 0.670 10.39

7 Vinyl Chloride 0.739 8.08 0.706 10.85

12 1,1-Dichloroethene 0.514 9.06 0.471 10.57

17 1,1-Dichloroethane 1.124 5.31 1.133 7.95

23 Chloroform 1.047 4.79 0.993 7.80

46 Toluene 0.793 7.27 0.785 11.70

47 1,2-Dichloropropane 0.406 3.44 0.376 2.57

53 Chlorobenzene 0.857 3.86 0.825 6.43

54 Ethylbenzene 1.535 10.10 1.329 9.03

56 Bromoform 0.272 5.37 0.240 4.49

58 1,1,2,2-Tetrachloroethane 0.560 10.83 0.473 11.80

A 10-point calibration was prepared with concentrations ranging

from 0.5 ppb to 200 ppb. Standards were purged for 11 min with zero-

grade nitrogen. Following purge, the standards were desorbed for 0.5

min onto an Agilent DB-VRX column. Response factors were calculated

using the internal standard approach. Response factors and %RSD for

all compounds were checked in accordance with Method 8260B.

Results

A total of 72 target compounds were analyzed. A representative por-

tion of the analytes is presented in Table I, which includes 8260B

System Performance Check Compounds (SPCCs) and Calibration

Check Compounds (CCCs). Method 8260B criteria were met (1).

Conclusions

Nitrogen is a viable alternative purge gas for the analysis of volatiles

by Method 8260B, which can result in considerable cost savings

for the laboratory.

References

(1) US EPA Method 8260B, December 1996, Revision 2, “Volatile Organic

Compounds by Gas Chromatography/Mass Spectrometry” (GC-MS), “Test

Methods for Evaluating Solid Waste,” Volume 1 B, Chapter 4, Section 4.3.2:

Laboratory Manual Physical/Chemical Methods, SW-846, December 1996.



ENVIRONMENTAL

US EPA wastewater method 624 Purgeables was originally

published as 40 CFR Part 136 Appendix A in the mid 1980s.

While instrumentation has advanced, there have been no

signif cant updates to method 624. This paper describes the

performance of EPA 624 on modern GC–MS instrumentation.

The US EPA is currently in the process of developing a revision to US

EPA Method 624 (1). A study was conducted to evaluate operating con-

ditions for the existing EPA Method 624 component list using updated

technology and advanced GC–MS instrumentation.

Experimental

A Shimadzu GCMS-QP2010 SE was conf gured with a Restek capillary

column designed specif cally for volatiles analysis. The GC was oper-

ated using constant linear velocity mode. A narrow i.d. inlet liner was

used to minimize band broadening during transfer to the purge and

trap, while still allowing high split injections. Data were acquired in full

scan mode. The EST Evolution and Centurion Water/Soil Autosampler

were used for extraction, concentration, and sample introduction.

Results

Each day, prior to running samples and at intervals no longer than

12 h an aliquot of BFB was purged and the spectra compared to

EPA Method 624 tuning criteria. BFB spectra met all criteria for all

samples evaluated and the GC–MS did not require retuning for the

entire project (2½ months).

A series of nine calibration standards across the range of

0.5–200 µg/L were prepared. Three internal standards were held

Shimadzu Scientif c Instruments, Inc.7102 Riverwood Drive, Columbia, MD 21046

tel. (800) 477-1227

Website: www.ssi.shimadzu.com

Volatile Organic Compounds in Wastewater: Performance of US EPA Method 624 on Modern GC–MS and Purge and Trap InstrumentationBrahm Prakash, Laura Chambers, and Wil-liam Lipps, Shimadzu Scientif c Instruments

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

1000000

0 5 10 15 20 25

Are

a C

ounts

Run Number

Internal Standard Stability Over 12 Hours

Bromochloromethane 2-Bromo-1-chloropropane 1,4-Dichlorobutane

RSD = 2%

RSD = 4%

RSD = 4%

constant at 30 µg/L and three surrogates were added at 10 µg/L. A

total ion chromatogram (TIC) at 10 µg/L is shown in Figure 1 along

with an expanded view of the early eluting light gases.

Continuing calibration veriå cation (CCV) standards were analyzed

as speciå ed in Method 624 and % recoveries were determined. All

recoveries were between within 80–105%. Method detection limits

(MDL) (2) were determined using replicates of a 0.5 µg/L standard.

All MDLs easily exceeded the criteria in US EPA Method 624.

Eight replicate aliquots each of 1.0 µg/L and 20 µg/L were analyzed

for precision and accuracy. Recoveries were all between 80% and

120% and %RSD were all ≤15%, exceeding method criteria. Also,

surrogate recovery and internal standard responses were monitored.

Surrogate recoveries fell within 80–120% for the entire study and

internal standard responses remained stable at ≤4% (Figure 2).

Conclusion

Modern GC–MS and purge and trap instrumentation far exceed ex-

isting criteria for US EPA Method 624. MDLs can easily be met and

a high level of precision and accuracy can be expected across the

entire calibration range.

References

(1) Appendix A to Part 136, Methods for Organic Chemical Analysis of Munici-

pal and Industrial Wastewater, Method 624 – Purgeables.

(2) Definition and Procedure for the Determination of the Method Detection

Limit, Fed. Regist. 49 (209), Appendix B to Part 136 (1984).


Figure 1: Total ion chromatogram of a 10 µg/L standard.

Figure 2: Internal standard response over a 12 h period.


FOOD & BEVERAGE

Analysis of Diacetyl and Other Vicinal Diketones (VDKs) in Alcoholic BeveragesAndrew James, Ellutia

Vicinal diketones (VDKs) are naturally produced compounds in the

fermentation process. The monitoring of VDK levels is of great impor-

tance as the concentration of these compounds can greatly alter the

f avor of a beverage. VDKs produce a butter-like f avor in the beverage

so they are undesirable in lighter beers that want a clean crisp taste,

but are actually wanted at higher levels in wines to give a smoother

taste and feel. The most commonly monitored VDKs are 2,3-butane-

dione also known as diacetyl and 2,3-pentanedione also known as

acetyl propionyl. VDKs are typically found in the 0–100 ppb range in

beers, and can be anywhere from 100 to 5000 ppb in wine.

The rise in popularity of craft beer has seen a dramatic rise in the

number of smaller independent breweries, many of which cannot

justify they cost and space of large expensive GC systems. A meth-

od has been developed using the Ellutia 200 series gas chromato-

graph and the Ellutia EL-2100 Head-Space autosampler to provide

an offering that can address the analysis at a lower entry price.

Experimental-Instrument Conditions

The EL-2100 Head-Space autosampler was coupled with a 200 se-

ries GC f tted with an ECD. The column used for the analysis was a

1.5 m glass packed column, 2 mm I.D packed with 80/20 Carbo-

pack BAW / 5% Carbowax 20M. The GC method was as per Table I.

A number of standards of different strengths ranging from 1000 ppb

down to 5 ppb of 2,3-butanedione, 2,3-pentanedione, and 2,3-hexane-

dione where made up in a 5% ethanol/water mix. These were then used

to perform a calibration for the analysis of real samples.

Sample Preparation

Before any analysis can take place the sample must f rst be de-

gassed, the simplest way to achieve this is to repeatedly pass

Ellutia Inc.660 Riverland Dr., Suite D, Charleston, SC 29412

tel. (843) 259-2307

Website: www.ellutia.com

380

360

340

320

300

2 4 6 8 10 12 14

Time (min)

2,3

-bu

tan

ed

ion

e 9

.99

pp

b

2,3

-Pe

nta

ne

dio

ne

2.4

9 p

pb

Vo

lta

ge

(m

v)

550

500

450

2,3

-bu

tan

ed

ion

e

2,3

-Pe

nta

ne

dio

ne

2,3

-he

xa

ne

dio

ne

Vo

lta

ge

(m

v)

400

350

300

2 4 6 8 10 12 14

Time (min)

the sample between

two beakers until the

sample stops foaming.

This is to help prevent

pressure forming when

the sample is heated in

the headspace vial and

from the CO2 eluting

during the chromatog-

raphy. Once degassed

2 mL of the sample is

then placed into 20 mL headspace vial that is then purged with

nitrogen. This is to help prevent oxygen in the air interfering with

the chromatography.

Results

The system was able to analyze the compounds of interest down to

the required low ppb levels. Figure 1 shows a chromatogram of the

25 ppb standard mix used in the calibration. Figure 2 shows an ac-

tual sample of a pale ale that contained just under 10 ppb diacetyl.

Conclusions

The Ellutia 200 series GC and EL-2100 autosampler can provide a

simple lower cost solution to VDK analysis allowing breweries with

smaller budgets the ability to perform this useful analysis.

Table I: GC method

Injector mode Split

Injector temp. 160 ˚C

Split f ow 6 mL/min

Column pressure 12 psi

Column oven Isothermal 135 ˚C

Detector temperature 200 ˚C

Detector current 10


Figure 2: Sample of a pale ale showing 10 ppb diacetyl.

Figure 1: 25 ppb VDK standard.


FOOD & BEVERAGE

B vitamins are a group of water soluble vitamins that play an important

role in cell metabolism. This group consists of a number of compounds

including thiamine (vitamin B1), ribof avin (vitamin B2), niacin and

nicotinamide (vitamin B3), and pyridoxine and pyridoxal (vitamin B6). B

vitamins are found in plant and animal food sources, such as legumes,

nuts, green leafy vegetables, red meat, and poultry. Many commercial

food products are fortif ed with vitamin B complex, and people could

take multivitamins supplements to help f ght vitamin B def ciencies.

Pickering Laboratories offers a method for simultaneous

determination of vitamins B1, B2, B3, and B6 in supplements and

protein powders. The method uses chemical and photochemical

post-column derivatization with f uorescence detection that increases

sensitivity and selectivity of analysis. Photochemical derivatization

required for niacin and nicotinamide and chemical derivatization is

needed for thiamine. Vitamins B2 and B6 have natural f uorescence.

Method

Calibration

• Calibration ranges: thiamine: 0.03–10 µg/mL; niacin: 0.125–10

µg/mL; nicotinamide: 0.3–100 µg/mL; riboý avin: 0.03–10 µg/mL;

pyridoxal: 0.125–10 µg/mL; pyridoxine: 0.125–10 µg/mL.

• To make riboý avin stock solution, dissolve 20 mg of riboý avin in

5 mL of 0.1 M NaOH and immediately add 50 mL of 0.1 M HCl

and make up the solution with DI water to 500 mL. Make work-

ing standards by diluting the stock solution with 0.01 M HCl. Store

protected from light.

Simultaneous Analysis of Vitamins B1, B2, B3, and B6 in Protein Powders and SupplementsMaria Of tserova and Sareeta Nerkar, Pickering Laboratories, Inc.

Pyridoxine

Thiamine

Ribofavin

Nicotinamide

min0 5 10 15 20 25 30

Niacin

Pyridoxal

Pyridoxine

Thiamine

Ribofavin

Nicotinamide

min0 5 10 15 20 25 30

Pyridoxine

Thiamine

Riboflavin

Nicotinamide

min0 5 10 15 20 25 30

Figure 1: Chromatogram of mixed B vitamins standard solution. Niacin, riboflavin, pyridoxine, pyridoxal, and thiamine – 1 ug/mL; nicotinamide – 10 ug/mL.

Table I: HPLC gradient

Time, min Buffer, % Acetonitrile, %

0 100 0

8 100 0

15 90 10

22 90 10

30 40 60

35 40 60

35.1 100 0

45 100 0

Table II: Pinnacle PCX

pump program

Time, minFlow rate,

mL/min

0 0

16.9 0

17 0.5

23 0.5

23.1 0

35 0

Run time: 35 min

Equilibration time: 10 min

• Stock and working standards for thiamine, niacin, nicotinamide,

pyridoxal, and pyridoxine are made in 0.01 M HCl and stored

protected from light.

Instrument Set Up

Connect the instruments in the following order: HPLC pump – HPLC

autosampler – analytical column – Pinnacle PCX post-column

derivatization instrument – UVETM photochemical reactor – f uorescence

detector.


Figure 2: Chromatogram of NIST multivitamin tablets sample.

Figure 3: Chromatogram of soy protein shake powder sample.


FOOD & BEVERAGE

Sample Preparation

For protein powders: To 0.5 g of samples add 50 mL of extraction

buffer (0.1N NaOH adjusted to pH 2 with phosphoric acid).

Homogenize using hand held homogenizer for 30 s and heat on

a water bath at 100 ºC for 30 min. Cool the solution down, f lter

through 0.45 µm nylon å lter, and inject. Protect from light.

For multivitamins supplements tablets: Blend at least 10 tablets

to a f ne powder and mix the entire sample thoroughly. Weigh 250

mg of sample and add 90 mL of DI water acidif ed to pH 2.6 with

0.1 N HCl. Stir using magnetic stirring plate for 2 h, protecting from

light. Make the volume up to 100 mL with acidif ed water. Filter the

sample through 0.45 µm nylon å lter and inject. Protect from light.

Analytical Conditions

Analytical column: Thermo Hypersil, Aquasil C18

(4.6 × 150 mm)

Column temperature: 40 ºC

Flow rate: 1 mL/min

Mobile phase: Solvent A: Dissolve 4.77 g of potassium

phosphate monobasic in 900 mL of DI

water, adjust pH to 5.9 with KOH.

Bring volume to 1 L with DI water.

Solvent B: Acetonitrile. See Table I

for gradient conditions.

Injection volume: 20 µL Pickering Laboratories, Inc.1280 Space Park Way, Mountain View, CA 94043

tel. (800) 654-3330, (650) 694-6700

Website: www.pickeringlabs.com

Thiamine

Riboflavin

Nicotinamide

min0 5 10 15 20 25 30

Figure 4: Chromatogram of NIST chocolate protein drink mix sample.

Table III: FLD program

Time, min Excitation, nm Emission, nm

0 322 400

17.9 322 400

18 370 440

25 370 440

25.1 350 470

35 350 470

Table IV: Analysis of NIST samples

Found in

Protein

Powder

RSD, %; n=3

NIST value

for Protein

Powder

Found in

Vitamins

Tablets

RSD; n=3

NIST Value

for Vitamins

Tablets

Nicotinamide 282 µg/g 4.5% 258 µg/g 15288 µg/g 0.8% 13907 µg/g

Thiamine 12.6 µg/g 9.0% 15.8 µg/g 1145 µg/g 0.6% 1045 µg/g

Ribof avin 25.9 µg/g 11.2% 26.9 µg/g 1204 µg/g 1.8% 1302 µg/g

Post-Column Conditions

Post-column derivatization system: Pinnacle PCX

Reactor volume: 0.5 mL

Reactor temperature: 30 ºC

Reagent: Dissolve 10 g of sodium

hydroxide in 500 mL of

DI water, add 2 g of

sodium sulf te, mix till

fully dissolved

Reagent ý ow rate: Initial ý ow rate 0 mL/min.

See Table II for pump

program

Detection: FLD, see Table III for

detector program UVETM

photochemical reactor



FOOD & BEVERAGE

Separation of Lactulose and Epilactose Melissa Turcotte* and Satoko Sakai†, *Showa Denko America, Inc., †Showa Denko K.K.

Lactulose, a semi-synthetic disaccharide, is a useful indicator for

heat induced modif cations in milk. Lactulose can be derived from

lactose by alkaline isomerization or enzymatic transgalactosylation

of fructose. Once in solution, lactulose is less stable in than lactose

in an alkaline environment and can degrade into galactose, taga-

tose, and low molecular weight acids.

US Pharmacopeia describes the separation of lactulose together

with lactose and galactose, as well as traces of other related sugars.

The USP method describes HPLC separation is achieved on an L8

amino propyl functionalized silica support column. Herein, we de-

scribe an improved separation for lactulose, epilactose, and lactose

on Shodex VG-50 4E, an amino HILIC column.

Shodex introduces the VG-50 4E column packed with a durable

polymer based packing material modif ed with chemically stable

tertiary amino functional groups. Shodex VG-50 4E is suitable for

saccharide analysis and provides a fast reliable method for the de-

termination of lactose and lactulose.


The analysis of three disaccharides is accomplished with Shodex

VG-50 4E (4.6 mm ID × 250 mm, 5 μm), a polymer-based amino

HILIC column. Column temperature was 40 ºC and f ow rate was

1.0 mL/min for the VG-50 4E analysis. Eluent conditions are 75/20/5

acetonitrile/methanol/water. Injection volume of 5 μL of 5 mg/mL of

each sugar was used for the experiment. The HPLC system was

coupled with RI detector.

Results

Lactulose, lactose, and epilactose were analyzed successfully by

HPLC and RI detection with VG-50 4E (Figure 1). Shodex VG-50 4E

demonstrates baseline separation of the three sugars with an elution

volume of less than 15 min with the elution order of lactulose, then

epilactose, and f nally lactose. The USP monograph describes the

elution order as f rst epilactose, lactulose, and then lactose with

resolution between lactulose and lactose not less than 1.5 and

resolution between lactulose and epilactose not less than 0.9. The

Shodex analysis demonstrates resolution between lactulose and

epilactose as 2.29 and resolution between epilactose and lactose

as 1.98.

Conclusions

Shodex VG-50 4E, a polymer-based hydrophilic interaction (HILIC)

chromatography column is suitable for the analysis of lactulose, epi-

lactose, and lactose.

Shodex™/Showa Denko America, Inc.420 Lexington Avenue Suite 2335A, New York, NY 10170

tel. (212) 370-0033 x109

Website: www.shodex.net

TM

Figure 1: Separation of lactulose and epilactose. Column: Shodex VG-50 4E; column temperature: 40 ºC; injection volume: 5 µL; eluent: CH3CN/CH3OH/H2O = 75/20/5; f ow rate: 1.0 mL/min; detector: Shodex RI. Sample: 1. lactulose, 2. epilactose, 3. lactose.



FOOD & BEVERAGE

Numerous compounds con-

tribute to changes in beer

f avor as it becomes stale.

One of these compounds,

(E)-2-nonenal, has been in-

vestigated as a major source

of the papery/cardboard

f avor that develops in aged

beer (1).

This application note demonstrates the purge and trap capability

of the Teledyne Tekmar Atomx to adequately separate this com-

pound from beer using the soil extraction feature. A GC–MS system

in a selected ion monitoring (SIM) detects and quantif es (E)-2-non-

enal levels in f ve beer samples.

Teledyne Tekmar4736 Socialville Foster Drive, Mason, OH 45040

tel. (513) 229-7000

Website: www.tekmar.com

Analysis of (E)-2-Nonenal in Beer with the Atomx and GCÐMS SIM DetectionRoger Bardsley, Teledyne Tekmar

Figure 1: Comparison of the SIM total ion chromatogram of a beer sample (red, lower) to its spiked sample (blue, upper).

Table I: Modified soil value of the default method for the

Atomx automated VOC sample prep system

Purge Variable Value Purge Variable Value

Valve oven temp 200 °C Condensate purge temp 70 °C

Transfer line temp 210 °C Dry purge time 0.50 min

Sample vial temp 50 °C Dry purge f ow 50 mL/min

Soil valve temp 120 °C Dry purge temp 70 °C

Standby f ow 25 mL/min Desorb variable Value

Purge ready

temp70 °C Sweep needle time 0.50 min

Condensate ready

temp70 °C Desorb preheat temp 220 °C

Purge mix speed Fast Desorb temp 225 °C

Purge time 10.00 min Bake variable Value

Purge f ow 25 mL/min Bake time 8.00 min

Purge temp 70 °C Bake f ow 200 mL/min

Bake temp 230 °C

Table II: Method values for the GC–MS system

GC Settings

ColumnRtx®-VMS, 20 m × 0.18 mm i.d. × 1 m, 0.9 mL/min

constant f ow, helium

Inlet/transfer line 230 °C, split ratio 60:1, transfer line 230 °C

Oven program95 °C, 3.5 °C/min to 145 °C, 15 °C/min to 240 °C,

3 min f nal hold, 21.7 min run

MS Settings

Scan35.0 m/z to 200 m/z, gain Factor 1, ATune, 1.4 min

solvent delay

SIM55 m/z, 70 m/z, 83 m/z, 96 m/z, 111 m/z, dwell

time 100

Temperatures Source 230 °C, Quad 150 °C

Method Conditions

Trap 1A Part No. 12-0083-503

Standards and Samples

Standard: (E)-2-Nonenal, Aldrich, 255653, 0.27 to 5.4 ppb in 5%

(v/v) ethanol solution

Samples: 5 Lots of a commercially available pale lager beer, 10 mL

degassed sample with 6 g NaCl

Results

References

(1) R. Scherer, R. Wagner, C.H. Kowal-

ski, and H.T. Godoy, “(E)-2-Nonenal

Determination in Brazilian Beers

using Headspace Solid-Phase Micro-

extraction and Gas Chromatographic

Coupled Mass spectrometry” (HS-

SPME-GC-MS) Cinc. Tecnol Aliment.,

Campinas, 30(Supl.1), 161–165, Maio

(2010) [Online] http://www.scielo.

br/scielo.php?script=sci_arttext&pi

d=S0101-20612010000500024.

Table III: Concentration

of (E)-2-nonenal in five

lots of a commercial

pale lager

Sample Lot ppb

A 0.180

B 0.463

C 0.261

D 0.184

E 0.082



INDUSTRIAL

Starch is used for a variety of industrial and nutritional purposes. Its

functional properties are inf uenced by the ratio and molar masses

of its macromolecular constituents, which vary with source, crop

year, and climate. Starch contains large homopolymers of amylose

(AMY) and amylopectin (AMP).

Linear AMY consists of long chains of (1⇒4)-α-D-glucose link-

ages, while the higher molar mass AMP is a branched structure con-

taining a mixture of (1⇒4)-α- and (1⇒6)-α-D-glucose linked resi-

dues. The goal of this work was to apply AF4-MALS-RI to separate

AMY and AMP in order to calculate a mass ratio, to determine the

molar mass distributions, the average molecular weights (Mw), and

the mean-square radius (Rz) of the AMP component. We applied the

technique to starches with AMY:AMP ratios covering a wide range.

An Eclipse AF4 system (Wyatt Technology) was equipped with a

short (18 cm) channel, a 350 μm spacer, and a regenerated cel-

lulose (10 kDa cutoff) membrane. Detection was accomplished with

DAWN Multi Angle Light Scattering (MALS) and Optilab RI detectors

(both instruments Wyatt Technology). The channel f ow was main-

tained at 1.0 mL/min and the cross-f ow was varied linearly from 1.0

to 0.1 mL/min for 10 min, then abruptly switched to 0.0 mL/min.

Integration of RI peak areas enabled calculation of the AMY:AMP

ratios, in excellent agreement with the nominal values. The values

for Mw and R

z fall within the generally accepted limits found in the

literature. Conformational plots for the AMP component verify its

branched nature.

This note graciously submitted by Rick White and Eija Chiaramonte, Global Analytical Sciences—

Personal Health, The Procter & Gamble Company, Mason, OH.

DAWN®, miniDAWN®, ASTRA®, Optilab® and the Wyatt Technology logo are registered trade-

marks of Wyatt Technology Corporation. ©2013 Wyatt Technology Corporation 4/4/13

AFFF-MALS-RI for Determining the Mass and Size Distributions of Amylose and Amylopectins in Starch Wyatt Technology

Wyatt Technology6300 Hollister Avenue, Santa Barbara, CA 93117

tel. +1 (805) 681-9009, fax +1 (805) 681-0123

Website: www.wyatt.com

Molar mass (g/mol)

rms

rad

ius

(nm

)

Hi-Maize (0.39) Hylon VII (0.40) Hylon V (0.41) Melojel (0.39) Amioca (0.41)

1.0x107 1.0x1091.0x108

100.0

Hi-Maize

1.0x109

1.0x108

1.0x106

1.0x104

1000.0

100.0

10.0

10.0 15.0 20.0

1.0

0.1

0.01

1.0x105

1.0x107

Hylon VII Hylon V Melojel Amioca

Amylose

Amylopectin

Mo

lar

mass

(g

/mo

l)

Volume (mL)

Sample

AMY:AMP AMY:AMP Amylose Amylopectin

(nominal) (calc) Mw (g mol-1) Mw (g mol-1) Rz (nm) Mw / Mn

Hi-Maize 75:25 74:26 227,900 109,200,000 145 2.4

Hylon VII 75:25 77:23 204,000 94,340,000 158 1.9

Hylon V 55.45 56.44 367,200 251,100,000 245 2.5

Melojel 25:75 26:74 419,700 418,500,000 316 2.9

Amioca 1:99 1:99 –– 321,000,000 261 2.1

Figure 2: Conformation plot (log Rz versus log Mw) for the amylopectin component of f ve starches (slopes 0.39–0.41 indicative of branching).

Figure 1: AMY:AMP ratio: AF4-RI fractograms with molar mass distributions overlayed. (Cross-f ow (Vx) = 1.0 to 0.1 mL/min in 10 min, then Vx = 0.0 mL/min.)


AF4-MALS-RI results for f ve native starches of varying


PHARMA/DRUG DISCOVERY

Avermectins are a series of 16-membered macrocyclic lactone derivatives that are used extensively in animal and crop protec-tion. They occur in nature and can be produced as fermentation by-products by the micro-organism, streptomyces avermitilis. Ex-amples of well-known avermectins and derived products include ivermectin, eprinomectin, selamectin, doramectin, and abamec-tin. All the avermectins have high anthelmintic and insecticidal properties even at low dose levels and residues of these veterinary drug components reach the environment through manufacturing and animal waste and may potentially affect terrestrial and aquatic life forms.

Several crop protection companies have strong interest in the synthesis, production, and analysis of these compounds. Therefore, there exists a need to develop a fast and eff cient analytical method capable of determining avermectin residues on animals, in food, and in the environment. The HPLC method presented here is fast and eff cient and can be used for residual analysis as well as for quality control of avermectins in drug formulations. The peaks pro-duced are sharp resulting in high sensitivity.

The mixture can be baseline separated using gradient elution with an MS-compatible mobile phase containing acetonitrile, methanol, and water. What is even more interesting is that the four avermectin products considered and up to 20 separate degradation products could be identif ed.

HPLC Conditions

Column name: FLARE C18 MMColumn dimensions: 4.6 × 150 mm (15698-14-2, 3.6 μm, 120 Å) HPLC system: Agilent 1200Inj. vol. and conc.: 5.0 μL, ~0.2 mg/mL of each major analyte in MeOHDetection: UV at 244 nmFlow rate: 1.0 mL/minSolvents: A: 50 mL ACN, 100 mL MeOH, 850 mL H2O; B: 600 ml ACN, 200 mL MeOH, 200 mL H2OGradient: 0.00 min, 10% B; 10.00 min, 90% B; 10.01 min, 10% B; 18.00 min, 10% BTemperature: 35 °C

The FLARE C18 MM column is manufactured with 3.6 µm diamond core-shell particles that have 120 Å pores. The particle geometry and tight particle size distribution contribute to high packing den-sity and column eff ciency with reduced plate height, h, of ~2. The columns come in various lengths and diameters and are compatible

Diamond Analytics11260 S 1600 W., Orem, UT

tel. (801) 235-9001, fax (801) 235-9141

Website: www.diamond-analytics.com

Separation of Macrocyclic Lactones Diamond Analytics

with 100% aqueous to 100% organic solvents. They are pH and temperature stable and available to ship worldwide.


Figure 1: Structures of selected macrocyclic lactones.

Figure 2: Separation of macrocyclic lactones using FLARE C18 MM column.

Figure 3: Exploded view showing peaks between 8 and 12 min.



Accurate Pain Management Analysis in Under 5 Min on Raptor™ Biphenyl Superf cially Porous Particle LC ColumnsSharon Lupo, Ty Kahler, and Paul Connolly, Restek Corporation

Pain management LC analyses can be diff cult to optimize

due to the limited selectivity of C18 and phenyl-hexyl phas-

es. In contrast, the selectivity of Raptor™ Biphenyl superf -

cially porous particle (SPP) LC columns provides complete

resolution of isobaric pain medications with a total cycle

time of 5 min.

Accurate, reliable analysis of pain medications is a key component

in monitoring appropriate medical use and preventing drug diver-

sion and abuse. As the demand for fast, multicomponent methods

grows, LC–MS-MS methods are increasingly desired for pain man-

agement and therapeutic drug monitoring due to the low detection

limits that can be achieved with this highly sensitive and selective

technique. However, despite the selectivity offered by mass spec-

trometry, hydrophilic matrix components can still interfere with ear-

ly-eluting drug compounds resulting in ion suppression. In addition,

isobaric pairs must be chromatographically separated for positive

identif cation. The need for highly selective and accurate methods

makes LC column selection critical.

While C18 and phenyl-hexyl phases are frequently used for bio-

analytical LC–MS-MS applications, Restek’s Biphenyl phase offers

better aromatic retention and selectivity for pharmaceutical and

drug-like compounds, giving it a signif cant advantage over other

phases for the analysis of pain management medications or other

drugs of abuse. The Biphenyl phase, originally developed a decade

ago by Restek, has recently been combined with Raptor™ SPP

(“core-shell”) silica particles to allow for faster separations without

the need for expensive UHPLC instrumentation. Here, we demon-

strate the fast, selective separation of commonly tested pain drugs

that can be achieved using the new Raptor™ SPP Biphenyl LC

column.


A standard containing multiple pain management drugs was pre-

pared in blank human urine and diluted with mobile phase as fol-

Figure 1: Baseline resolution of isobaric pain management drugs in sub-5-min runs on the Raptor™ Biphenyl column.

Table I: Mobile phase gradient

Time (min) Flow (mL/min) %A %B

0.00 0.6 90 10

1.50 0.6 55 45

2.50 0.6 0 100

3.70 0.6 0 100

3.71 0.6 90 10

5.00 0.6 90 10

lows, urine:mobile phase A:mobile phase B (17:76:7). The f nal

concentration for all analytes was 10 ng/mL except for lorazepam,

which was 100 ng/mL. Samples were then analyzed by LC–MS-

MS using an AB SCIEX API 4000™ MS-MS in ESI+ mode. Chro-

matographic conditions, retention times, and mass transitions are

presented here and in Tables I and II:

Column: Raptor™ Biphenyl, 50 mm × 3.0 mm i.d. × 2.7 µm

Sample: Fortif ed urine

Inj. vol.: 10 μL

Inj. temp.: 30 °C

Mobile phase A: Water + 0.1% formic acid

Mobile phase B: Methanol + 0.1% formic acid

Results

As shown in Figure 1, 18 commonly tested pain management

drugs were analyzed with the last compound eluting in less

than 3.5 min, giving a total cycle time of 5 min on Restek’s

Raptor™ SPP Biphenyl LC column. Analyte retention times

are presented in Table II. Important isobaric pairs (morphine/

hydromorphone and codeine/hydrocodone) were completely

resolved and eluted as symmetrical peaks, allowing accurate

identification and integration. In addition, early-eluting com-

pounds such as morphine, oxymorphone, and hydromorphone

are separated from hydrophilic matrix interferences, resulting

in decreased ion-suppression and increased sensitivity. Similar

analyses on C18 and phenyl-hexyl columns often exhibit poor

peak shape and resolution (for example, peak tailing between




closely eluting isobars), which makes identification and accu-

rate quantification more difficult.

Conclusions

Complete separation of critical pain management drug analytes

from hydrophilic matrix components and isobaric interferences was

achieved using the new Raptor™ SPP Biphenyl LC column in less

than 5 min. The fast, complete separations produced in this method

allow accurate quantif cation of pain management drugs and sup-

port increased sample throughput and improved lab productivity.

To learn more, visit www.restek.com/raptor

Restek Corporation110 Benner Circle, Bellefonte, PA 16823

tel. (800) 356-1688, fax (814) 353-1309

Website: www.restek.com/raptor

Table II: Analyte retention times and transitions

Peaks tR (min) Precursor Ion Product Ion 1 Product Ion 2

Morphine* 1.34 286.2 152.3 165.3

Oxymorphone 1.40 302.1 227.3 198.2

Hydromorphone* 1.52 286.1 185.3 128.2

Amphetamine 1.62 136.0 91.3 119.2

Methamphetamine 1.84 150.0 91.2 119.3

Codeine* 1.91 300.2 165.4 153.2

Oxycodone 2.02 316.1 241.3 256.4

Hydrocodone* 2.06 300.1 199.3 128.3

Norbuprenorphine 2.59 414.1 83.4 101.0

Meprobamate 2.61 219.0 158.4 97.2

Fentanyl 2.70 337.2 188.4 105.2

Buprenorphine 2.70 468.3 396.4 414.5

Flurazepam 2.73 388.2 315.2 288.3

Sufentanil 2.77 387.2 238.5 111.3

Methadone 2.86 310.2 265.3 105.3

Carisoprodol 2.87 261.2 176.3 158.1

Lorazepam 3.03 321.0 275.4 303.1

Diazepam 3.31 285.1 193.2 153.9

*An extracted ion chromatogram (XIC) of these isobars is presented in the inset of Figure 1.




Mobile phase pH is a powerful tool in methods development,

particularly for separation of neutral forms of amines or other

organic bases under alkaline conditions. In this study a generic,

5-min linear gradient was used to separate six basic drug com-

pounds on a short (50 mm) PRP-C18 column.

More than 70% of all pharmaceutical drug compounds are cationic

solutes that carry a formal positive charge below pH 7. Separation

of these and other organic bases has historically been problematic.

Ionization has a dominating effect in reversed phase chromatogra-

phy that tends to dictate retention. Consequently, the elution win-

dow for a sample of ionized amines is narrow. The task is further

complicated by secondary interactions that occur between positively

charged solutes and residual silanols on the column stationary phase.

These secondary mechanisms of retention are the principle source for

anomalous chromatographic activity, such as poor peak shape, shifts

in retention times and loss of eff ciency that progressively worsen over

the life of the column.

The PRP-C18 is a new column designed for high-eff ciency reversed

phase separations under any mobile phase conditions. The stationary

phase for the PRP-C18 is devoid of free silanols, does not strip, bleed,

or dissolve at any pH, and therefore can be expected to perform reliably

and reproducibly throughout the extended life of the column, regardless

of mobile phase conditions. Use of alkaline mobile phase (pH > 11) per-

mits separation of basic solutes in their neutral forms. This broadens the

window for elution, whereby subtle structural nuances among chemi-

cally similar compounds can be exploited to effect resolution.

Although some recent C18 columns boast stability in alkaline pH,

all silica-based supports experience measurable degradation at pH

> 6, where column life is still considerably shorter than if used under

more favorable conditions. On the other hand, the PRP-C18 stands

up to prolonged exposure to concentrations as high as 1 M NaOH

and H2SO

4, with no measurable decrease in performance.


Column: PRP-C18, 4.1 × 50 mm, 5 μm

Instrumentation: Agilent 1100 quaternary pump with UV detector

Standards: nicotine, metropolol, quinine, doxylamine, dexmetho-

rphan, amitriptyline

Mobile phase A : 30 mM diethylamine

Mobile phase B: A + 95% ACN, 5% H2O

Gradient: 10 to 100% B in 5 min

Flow rate: 2 mL/min

Temperature: Ambient

Injection volume: 10 µL

Detection: UV at 265 nm

Results and Conclusion

In modern drug discovery science where analytical HPLC can be a bot-

tleneck, the trend is to streamline production through the use of shorter

columns with smaller particles operated at elevated f ow rates. The f ex-

ibility to employ a high pH mobile phase is another valuable tool that

permits separation of basic solutes in their neutral forms. Oftentimes

this greatly simplif es the process of methods development. In this study,

separation of a set of structurally diverse pharmaceutical compounds is

achieved on a short (50 mm) PRP-C18 column using a generic 5 min

linear gradient.

Rapid Separation of Basic Drug Compoundson pH-Stable Hamilton PRP™-C18Derek Jensen and Mark Carrier, Hamilton Company

Hamilton Company4970 Energy Way, Reno, NV 89502

tel. (800) 648-5950, fax (775) 858-3026

Website: www.hamiltoncompany.com

0 2

2

3

4

1

5

6

4

Time (min)

6 8 10


Figure 1: Rapid separation of six basic drug compounds on a 50 mm PRP-C18.


GENERAL

Figure 1: DEP analysis was used for the fast and reliable identif cation of the unknown components in MTN-767, a recreational synthetic cannabinoid.


Direct exposure probe (DEP) coupled to the Pegasus HT,

time-of-f ight mass spectrometer, was used for the identif -

cation of synthetic cannabinoids in MTN-series compounds.

Crime laboratories are responsible for the timely identif cation of

seized drugs used as evidence in criminal investigations. The objec-

tive was to identify drugs of abuse using DEP instead of traditional

LC or GC for time-saving and ease-of-use.


Sample: MTN-XXX dissolved in methanol

(1 mg/mL)

Direct exposure probe: SIS (Ringoes, NJ)

DEP conditions: 0 A to 1.5 A at 1.0 A/min

Source temperature 200 °C

Mass range m/z 35–1000

Acquisition rate 3 spectra/s

Fast and Robust Analysis of Drugs of Abuse Using Direct Exposure Probe on the Pegasus HTJonathan D. Byer and Joe Binkley, Life Science and Chemical Analysis Centre, LECO Corporation

LECO Corporation

3000 Lakeview Avenue, St. Joseph, MI 49085

tel. (269) 985-5496, fax (269) 982-8977

Website: www.leco.com

Results

The MTN-series of compounds are allegedly blends of JWH-series

compounds, but have not been studied as extensively. Very rapidly,

it was determined that MTN-767 is most likely a blend of JWH-018

and JWH-081 (Figure 1). True Signal Deconvolution® produced two

mass spectra that were compared to the NIST, Wiley, and Cayman

library databases and yielded hits with similarity scores greater than

800 out of 1000. This result, as well as the analysis of other MTN

compounds, conf rmed that MTN-series compounds are blends of

JWH compounds, and was accomplished within a few min.

Advantages of DEP experiments on LECO’s Pegasus HT:

1. Open source design - robust and low maintenance

2. Time-of-f ight MS - no mass spectral skewing

3. True Signal Deconvolution® - identify more components

4. Probe installation does not interfere with GC or GC×GC


GENERAL

Photonis has combined two of our superior detection and analysis

technologies into a new digital ion beam prof ling unit. The new

Ion Beam Prof ler combines a large microchannel plate (up to 120

mm) in a complete assembly with the Photonis Nocturn, a high-

resolution digital CMOS low-light camera which is immune to sud-

den light damage. Together these highly sensitive imaging compo-

nents offer a detailed image of your instrument’s ion beam, allowing

system designers to identify and correct any areas of ion loss that

had previously been overlooked. This new diagnostic tool allows the

instrument to be designed to maximize the amount of ions collected

for a superior analysis.

Tracing the Path

Ion optics modeling software is often used to design and predict the

ion path within a mass spectrometer. The conventional method for

aligning an ion beam consists of scanning the beam over a Faraday

cup or electron multiplier, integrating the current, and f nding the

settings which produce the highest signal. However, ion trajectories

can be inf uenced by many factors which are not considered in

the model. Eff ciently transporting ions from the source through the

mass f lter is critical for maximizing instrument sensitivity.

The new Photonis Ion Beam Prof ler can visualize the location of

any charged particle (ion, electron, UV, photon, or soft X-ray), en-

abling the instrument designer to ensure all available signal ions are

collected. The unit can capture images from the phosphor screen

at rates up to 100 frames per second and store them on a PC for

collaboration and comparison testing. A strobe trigger is available

to synchronize the camera to a specif c event. The unit can also be

used as a real-time tuning device, allowing an instrument designer

to receive immediate feedback on the focus, alignment, and trajec-

tory of the ion beam, and can also show any obstructions in the

pathway, such as a loose wire (Figure 1).

Connectivity Options

The Nocturn camera provided with the Ion Beam Prof ler can be

ordered with a variety of connectivity options for simple video analy-

sis, including GigE Vision®, USB3 GenICam®, USB, CameraLink®

compatible, or analog NTSC or PAL. Some models are also offered

with a color video sensor, which can provide color video at higher

light levels and monochrome images at light levels less than 100

mlx to eliminate color noise from the image.

Versatile Applications

The new Ion Beam Prof ling unit can be used in wide range of ap-

plications, including ion optic model verif cation, imaging TOF, VUV

spectroscopy, and high energy physics as well as ion beam prof ling.

The unit can be custom-f t with a specif c microchannel plate and

electro-optic housing to f t your unique application.

Photonis USA660 Main Street, Sturbridge, MA 01566

tel. (508) 347-4000

Website: www.photonis.com

A Better Way to Prof le Your Ion Beam in Instrument Design for More Accurate AnalysisPhotonis, USA

Figure 1: Actual images taken of the ion beam during ion optimiza-tion process. Top left shows a well-focused beam; top right shows a poorly focused beam. Bottom left shows a defocused beam while bot-tom right demonstrates a close up of a region of interest.



GENERAL

Silica-based cyano stationary phases have historically exhibited

poor stability and short column lifetime at pH extremes. This appli-

cation note details an investigation of the stability of YMC’s Cyano-

HG (high strength) 10 μm silica-based stationary phase, which was

performed at a customer’s request. The columns were stressed un-

der acidic (0.1% trif uoroacetic acid) and basic (10 mM ammonium

bicarbonate, pH = 8.2) conditions, and evaluated for retention time,

USP tailing, and USP plate count using two probes (naphthalene

and methyl benzoate) and the following method:

Ruggedness of YMC Cyano-High Strength Stationary Phase Under Acidic and Basic ExtremesJeffrey A. Kakaley and Ernest J. Sobkow, YMC America Inc.

YMC America, Inc.941 Marcon Blvd., Suite 201, Allentown, PA 18109

tel. (610) 266-8650

Website: www.ymcamerica.com

Figure 2: Cyano-HG stability — acidic conditions.

An example chromatogram for the column performance test can

be seen below:

The f rst column was run under the basic test conditions, exhibit-

ing a modest loss in retention time for both analytes (0.087 min for

methyl benzoate and 0.265 min for naphthalene). Tailing factor re-

mained at 1.0 for both analytes and plate count was also very stable

for both over the entire 50 h span.

The second column was tested in the same manner except under

the acidic conditions. This column exhibited essentially no loss in

retention time (0.010 min) for methyl benzoate and only a slight de-

crease in retention time for naphthalene (0.065 min). Tailing factor

remained unchanged (1.2 for methyl benzoate and 1.1 for naph-

thalene) and plate count was very stable under these conditions,

similar to what was seen during the basic stress runs.

Overall, 10 μm YMC-Pack Cyano-HG stationary phase proved to be

very rugged under both stress conditions and is suitable for use in

preparative separations when used under the test conditions.


Figure 1: Cyano-HG stability — basic conditions.


In Memoriam: Professor Georges A. GuiochonOn October 21, 2014, University of Tennessee Distinguished Scientist Georges Andre Guiochon succumbed to neuromuscular failure caused by post-polio syndrome. Here, we pay tribute to his remarkable career and life.

Mark R. Schure and Lois Ann Beaver (Guiochon)

Georges Andre Guiochon was enthusiastically in-terested in broad areas of separation science from preparative to analytical chromatography. He

probed deeply and elucidated fundamental and practi-cal principles based on studies ranging from isotherms in preparative chromatography, mechanisms of liquid chromatography (LC), and gas chromatography (GC) to two-dimensional (2D) and three-dimensional (3D) LC. He developed the theory of nonlinear chromatography and applied it to GC, LC, and supercritical f luid chromatogra-phy (SFC). Georges’s last years were spent in the pursuit of understanding SFC and core–shell particle performance.

Early Career in FranceGeorges was born in Nantes, France, on September 6, 1931. He attended Ecole Polytechnique in Paris where he received the degree of Ingenieur after having spent a year hospitalized with polio that rendered him unable to walk. His remark-able determination prevailed, and he walked and went on to receive a PhD in chemistry from the University of Paris. At the request of the French military, his work on the cause of explosions involving ammonium nitrate remained confiden-tial for decades although his counsel on the safe handling of that compound is posted in factories where it is produced. It was only in 1995, after U.S. officials investigating the bomb-ing of a federal building in Oklahoma went to the Oak Ridge National Laboratory to seek his advice, that Georges felt free to share his knowledge. Later, after an explosion in a manu-facturing facility in Toulouse, France, Georges was called upon as an expert witness in the court trial and the appeal.

Georges’s career in France was outstanding and yielded many awards and public accolades, including two Tswett medals (Advances in Chromatography and the Academy of Sciences of USSR), the Stephen Dal Nogare Award, the AJP Martin Medal, a Silver Medal from the Centre National de la Recherche Scientifique, and a JSPS Fellowship for Research in Japan. Initially, his work focused on the fundamentals and application of GC, but later he became involved in the development of high performance liquid chromatography (HPLC). Georges moved into the elevated position of Direc-tor of Chemistry and Physics at the Ecole Polytechnique as well as that of Professor at the University of Paris. Aside from

his publications and publicly known activities, Georges was again drawn into confidential work involving locating the sites in southern France where heroin was being produced by reacting opium and acetic anhydride. He applied his analyti-cal chemistry skills to assist officials in thwarting the French Connection, a subject he was loath to reveal.

FamilyDuring this time, Georges and his wife Claudine raised three daughters. The oldest, Anne, became a physician with a doctorate in molecular biology. She is a leader at a medi-cal university. Alice, the second daughter, majored in com-puter science and now supervises a large team of experts. The youngest, Odile, is an economist overseeing a staff in an insurance company. Along with their successful careers, his daughters and their husbands produced seven grandchildren. Georges’s second wife, Lois Ann Beaver, who worked at the Food and Drug Administration (FDA), is extremely happy to be close to such a vibrant family.

Work and Achievements in the U.S.In 1984, Georges accepted an invitation to be a Professor of Chemistry at Georgetown University in Washington, DC. Three years later, he agreed to become a Distinguished Scientist with simultaneous positions at the University of Tennessee in Knoxville and at Oak Ridge National Labora-tory. Georges was pleased with the computer access and the instrumentation he acquired. Many students sought to study under him. This period was especially productive, leading to many innovations and basic studies in chromatography and separation science.

The recognition earned by Georges during this period is extraordinary, and his awards are too numerous to list. We note two American Chemical Society awards: the ACS Award in Chromatography and the ACS Award in Separa-tion Science. Other prestigious awards include the LCGCLifetime Achievement Award, the Horvath Medal, the Ist-ván Halász Award, the von Humboldt Award, and more. Georges received honorary doctorates from the Technical University in Budapest, Hungary, in 1982; the University of Pardubice, in the Czech Republic, in 1999; Ramon Llull University in Barcelona, Spain, in 2002; the University

ES560998_LCGCAN0215_032.pgs 01.29.2015 00:38 ADV blackyellowmagenta


of Ferrara, in Italy, in 2003; and the University of Science and Technology Laoning, in China, in 2010. He was inducted into both the Catalonian and the Spanish Academies of Science in 2011. He was a permanent member of the HPLC conference series commit-tee and in 1985 became associated with the PREP Symposium, which is still a vibrant conference covering developments in preparative and non-linear chromatography.

Georges’s research culminated in the publication of more than 1200 papers. Given this impressive achieve-ment, Professor Peter Carr of the Uni-versity of Minnesota suggested that the number of research papers pub-lished should be standardized in units of “the Georges” (G), with 1 G = 1000 papers. If you published 120 papers you would have published 0.12 G or 120 milliGeorges of papers.

Georges published seven books: AGuide to the HPLC Literature, Vol. 1(1966–1979), Vol. 2 (1980–1981) and Vol. 3 (1983) all with coauthors J.-L. Excoffier, H. Colin, and A.M. Krstu-lovic; Gas Chromatography in Inor-ganics and Organometallics with C. Pommier in 1971; Quantitative Gas Chromatography with C.L. Guillemin in 1988; Fundamentals of Preparative and Nonlinear Chromatography with S. Golshan-Shirazi and A.M. Katti in 1994 (an updated edition, published

in February of 2006 also includes A. Felinger); and Modeling for Prepara-tive Chromatography with B.C. Lin in 2003. The book on preparative and nonlinear chromatography is consid-ered the definitive work in this area and is affectionately known as “Bob” (the Big Orange Book).

Georges taught more than 100 graduate students and post-docs and is considered one of the major lead-ers in chromatography research of the past 50 years. Many of his students are employed by industrial, academic, and national laboratories throughout the world in fields as diverse as the phar-maceutical industry, chromatography and separation science research, chro-matographic equipment, and business.

His presence in the field of chroma-tography was evidenced by numerous invitations to speak at scientific confer-ences. Georges was immersed in chro-matography like few others have been, could be, or will be. He was a brilliant researcher who mentored hundreds of people, not only in the university envi-ronment but also in the context of the PREP and HPLC meetings. At these meetings Georges had an entourage of people, including friends, fellow researchers, students, and those who had technical questions for him. When asked, Georges would say his motiva-tions were “understanding phenomena, solving problems and training people.” This was evident right to the end.

Fond Memories of GeorgesIn addition to being a great scientist, Georges was a unique individual. His love of king crab is legendary. When in Chicago, Georges and Lois would seek accommodations across the street from Joe’s Stone Crab Restaurant, so that he could eat nearly every meal there. Georges’s knowledge of great restaurants follows in the tradition of the French. If you found him in a res-taurant, you knew it was a good choice.

Georges loved his family, and he and Lois spent considerable time with them exploring different countries and showing them the United States. His knowledge of history was remark-able. Georges followed the entire Lewis and Clark trail and especially studied

geologic phenomena caused by ice age flooding of Lake Missoula.

Although attending conferences was not always easy for Georges, he did so regularly, because he enjoyed inter-acting with his scientific colleagues. Often he would be pleased with the talks he heard, but occasionally he would show disdain. When Georges challenged a presenter it was because he was concerned that a young and developing scientist might be misled.

He remarked that István Halász, a famous Hungarian chromatographer and the PhD advisor to Csaba Hor-váth, maintained that very high pres-sures were unnecessary for extremely high performance separations. When core–shell particles were shown to demonstrate this, Georges was pleased. He had a deep understand-ing of how chromatography functions and explained matters in clear, under-standable terms. Georges’s work is widely referenced not only by chroma-tographers, but also by applied phys-ics and f luid mechanics researchers, chemical engineers, and biochemists interested in the mechanics of separa-tions. He was prolific in the subject matters, angles, and approaches he took to separation science.

Georges was unrelentingly gener-ous with his time in mentoring sepa-ration scientists who demonstrated great potential. He would often praise their accomplishments and felt he had a responsibility to usher them into the chromatography community.

With Georges’s passing, we have lost a great person and a great scientist. We can be comforted, however, by know-ing that Georges’s legacy will continue to inspire future advances.

Mark R. Schure is chief technology

officer with Kroungold Analytical in Blue

Bell, Pennsylvania. Lois Ann Beaver

(Guiochon) is retired from the Food

and Drug Administration. Direct correspon-

dence to: [email protected] ◾

2015 Events Honoring GuiochonVarious events in honor of Georges Guiochon are planned for 2015. These include• The Chromatography Community

Mixer on March 10 at Pittcon 2015

in New Orleans

• The “Celebration of a Great Scien-

tist” at the University of Tennessee

in Knoxville on March 14 (for more

information contact: [email protected])

• Sessions at HPLC 2015 in Geneva,

Switzerland, June 21–25

• A session honoring Georges’s life-

time achievements in chromatogra-

phy at the 2015 PREP meeting in

Philadelphia, July 26–29.

For more information on this topic, please visit our homepage at:


ES560999_LCGCAN0215_033.pgs 01.29.2015 00:38 ADV blackyellowmagenta


Quantification of Individual Mass Transfer Phenomena in Liquid Chromatography for Further Improvement of Column EfficiencyThis article reports on the physical phenomena that control column efficien-cy and on experimental protocols designed to accurately measure their con-tributions to band broadening of analytes during their passage from the injec-tion to the detection device. The results of these protocols are analyzed, allow-ing for the accurate determination of the complete mass transfer mecha-nism in different separation modes and providing solutions and future direc-tions to further improve the efficiency of liquid chromatography columns.

Fabrice Gritti

The successful separation of complex mixtures and critical pairs of similar compounds by liquid chro-matography (LC) ultimately depends on how many

plates the chromatographic column can generate. This need for higher plate numbers has continuously pushed column manufacturers forward to produce finer particles from about 100 μm in the early days of LC (1,2) down to nearly 1 μm today thanks to the advent of ultrahigh-pressure liquid chromatography (UHPLC) in the mid-2000s (3,4). From a practical viewpoint, LC practitioners often wonder what causes their peak widths to become unexpectedly large under certain experimental condi-tions and, therefore, they search for some rationale or, at least, reasonable explanations for their observations. The scientific principles and theories of band broadening and mass transfer in LC columns have continuously been revisited over the last 50 years. They were recently re-viewed, refined, and adjusted based on the accumulation of a large amount of accurate experimental data and vali-dated computer-simulated data (5). The lack of and need for a quantitative measurement of the different physical phenomena involved in the overall band broadening of a sample molecule as it progresses from the injection loop through the column and to the detection cell unit were addressed. The combination of a series of experimental protocols necessary to unravel the complete mass trans-fer mechanism in a chromatographic column was then proposed (6).

In the 1960s, Giddings established the most general and comprehensive theory of band broadening in chromato-graphic columns (7). It is still valid today. The finite width of an eluted peak results from the combination of several in-dependent sources of sample dispersion. They include longi-tudinal diffusion (the B/u term in the van Deemter equation), eddy dispersion caused by mobile-phase velocity inequali-ties (the A(u) term), and solid–liquid mass transfer resistance

due to the finite rate of diffusion across the stationary phase and to slow adsorption–desorption (the Cu term). To those three main height equivalent to a theoretical plate (HETP) terms, one can add extracolumn band broadening, which has become increasingly important with the new generation of sub-2-μm particles and narrow-bore columns operated in UHPLC (8,9), and the friction of the mobile phase against the packed bed (generating heat) and its expansion during the inlet-to-outlet decompression (consuming heat), which both cause additional peak band broadening under nonadiabatic conditions (5).

The first purpose of this work is to redefine all the potential sources of band broadening inside and outside the chromatographic column and to discuss when, in practice, each of them is dominant in LC. The second goal is to il lustrate how the B/u , A(u), and Cu HETP terms of the van Deemter equation can be accurately measured from a series of well designed protocols. These quantitative results are used to decipher the complete mass transfer mechanism in reversed-phase LC C18 columns packed with 2.5-μm fully porous particles and 2.6-μm superficially porous particles, in hydrophilic in-teraction chromatography (HILIC) columns packed with 3.5-μm fully porous particles, in chiral reversed-phase columns packed with 5-μm fully porous particles coated with cellulose polymer, and in reversed-phase LC C18 silica-based monolithic columns of the first and second generation. The final goal is to reveal from these experi-mental investigations the key directions to follow for the improvement of the efficiency of the future generation of LC columns.

Column Efficiency: A Dimensionless ApproachThe efficiency of a chromatographic column, which is de-fined as the ratio of its length (L) to its plate height (H), is a function of a very large number of physical parameters.



These are the linear velocity of the mo-bile phase (u, cm/s), the particle size or the characteristic size of the separation medium (dp, cm), the diffusion coeffi-cient of the analyte in the bulk mobile phase (Dm, cm2/s), the diffusivity of the analyte in the stationary phase (Dp

= Ω Dm, cm2/s), the retention factor (k′), the external porosity of the chromato-graphic bed (ε

e), the bed aspect ratio or

the ratio of the inner diameter of the column to the particle size (dc/dp), and the ratio of the column length to its diameter (L/dc). A simpler, easy-to-use relationship between the column effi-

ciency N and these experimental vari-ables is needed. This can be achieved from a dimensionless approach that is familiar in chemical engineering (7). The column efficiency N is first nor-malized to the column length (plate height) and to the particle size to define the reduced plate height (h):

h = L/Ndp [1]

The linear velocity u is reduced to the particle diameter and the bulk diffu-sion coefficient of the analyte. This de-fines the reduced velocity (ν) as

ν = udp/Dm [2]

These two dimensionless variables are related through a general expres-sion, the dimensionless or reduced van Deemter equation (10):

h = B/ν + A(ν) + Cν [3]

The kinetic performance of the column is now a function of only three vari-ables (B, A, and C), which vary little provided that the practitioner operates with a certain retention mode (such as HILIC, reversed-phase LC, chiral LC, or ion-exchange chromatography) and uses a similar column format (short or wide internal diameter columns, capil-lary columns, particulate, or monolithic columns). B accounts for the longitudi-nal diffusion along the packed bed im-mersed in the eluent; A accounts for all sources of eddies in the column; and Caccounts for the finite rate of diffusion across the stationary phase and the po-tentially slow adsorption–desorption process.

The advantage of equation 3 is straightforward: Chromatographers can easily anticipate, with a good approxi-mation, the efficiency of any given column under various experimental conditions. For instance, the left graph in Figure 1 illustrates the expected change in the reduced plate height h

when transferring a method from high performance liquid chromatography (HPLC) to UHPLC (large to small par-ticle size, low to high pressures). For the maximum allowable pressure (400 bar in HPLC and 1200 bar in UHPLC), the reduced velocities are smaller in UHPLC than those in HPLC because of the diminution of both the particle size (× 1/3) and the linear velocity (the column permeability decreases by about one order of magnitude because it is proportional to the square of the particle diameter). Therefore, at maxi-mum pressure, the column efficiency of UHPLC columns is essentially con-trolled by the longitudinal diffusion B/ν and the eddy dispersion A(ν) HETP terms in the van Deemter equation, and in conventional HPLC the column effi-ciency is governed by the A(ν) and Cν

HETP terms. The right graph in Figure 1

1000p = 0.1p = 0.2p = 0.3p = 0.4p = 0.5p = 0.6p = 0.7p = 0.8p = 0.9

100

Syst

em

vari

an

ce (

µL

2)

10

Retention factor (k’)

System 1

System 2

System 31

0.1

0 2 4 6 8 10

Figure 2: Plot of the minimum system variance versus the retention factor k′ required to observe at least a certain fraction p of the maximum intrinsic plate number (Nmax = 39,000 plates for a 100 mm × 2.1 mm column packed with 1.6-mm core–shell particles).

6

10

8

6

4

2

00 20

*

40 60 0 100 200 300 400 500ν ν

P = 120 bar

P = 1200 bar

P = 600 bar

P = 60 bar

P = 60 bar P = 600 bar

P = 40 bar

P = 400 bar

Small molecule, Dm = 10-5 cm2

dp = 5 µm, L = 15 cm

Protein BSA : Dm = 7 10-7 cm2/s

Small molecule : Dm = 10-5 cm2/s

dp = 1.5 µm, L = 10 cm

L = 10 cm, dp = 3.5 µm

4

h =

H/d

p

h =

H/d

p

2

0

Figure 1: Impact of the experimental chromatographic conditions (conventional HPLC versus UHPLC, left graph; analysis of small and large molecules, right graph) on the ex-pected reduced plate height and column eff ciency. The universal reduced van Deem-ter plot (B, A, and C coeff cients) was obtained for a conventional 150 mm × 4.6 mm C18 reversed-phase LC column and a retention factor of about 3.



shows the shifts in reduced velocity and plate height when using the same column, but for different applications (small to large molecules, same linear velocity). When analyzing large biomolecules (~10 kDa) instead of small molecules (<500 Da), the bulk diffusion coefficient typically decreases by one order of magnitude. The reduced velocity increases then by a factor of 10 and the column efficiency essentially becomes depen-dent on the slow rate of diffusion across the stationary phase: it is mostly controlled by the Cν HETP term.

The Different Sources of Band Broadening in LCThe different physical phenomena causing the chromato-graphic band to broaden are redefined and listed below.

Extracolumn Band Broadening

Band broadening does not only take place inside the column. To some extent, the chromatographic system itself also con-tributes to sample dispersion after the passage of the injected sample volume along the injection loop, the needle seat cap-illary (in most HPLC instruments), the injection valve, the inlet and outlet capillaries, and the detection cell (including its electronic components [11]). About 15 years ago, ana-lysts did not have to worry much about extracolumn band broadening in LC because the column dimensions were large enough and the column efficiency still moderate when 250 mm × 4.6 mm and 5 μm were the standard column format and particle size, respectively. Indeed, the volume variance of the chromatographic peak because of the sole chromato-graphic column is given by equation 4(8):

σ2

v, column=

V 2

N1+k′

02

[4]

According to equation 4, the system contribution to the total peak variance observed by the analyst becomes in-creasingly important when small diameter columns (small hold-up volume V0) are packed with fine particles (large efficiency N) or when small retention factors k′ are ap-plied. In 2014, with the emergence of sub-2-μm core–shell particles packed in 100 mm × 2.1 mm columns, the intrin-sic efficiencies of LC columns can now reach values close to 450,000 plates/m (dp = 1.3 μm [4]). Unfortunately, this exceptionally high level of efficiency cannot be achieved or observed by analysts because of the band broadening along the UHPLC instrument, even with the most advanced instruments. Figure 2 illustrates this statement by plotting the maximum system volume variance required to observe at least a certain fraction (p) of the maximum intrinsic plate number (dp = 1.6 μm, core-shell particles, Nintrinsic = 39,000 for a 100 mm × 2.1 mm column) versus the retention factor k′. For instance, if a chromatographer imposes p = 90%, this goal (Nobserved = 0.9 × 39,000 = 35,100) can only be satisfied with system 1, system 2, or system 3 if k′ is larger than 8, 5, and 2, respectively. If p = 50%, the k′ values should only be larger than about 2 and 1 for systems 1 and 2, while the goal set for Nobserved = 0.5 × 39,000 = 19,500 will always be satisfied with system 3 irrespective of the retention factor.

The importance of instrumentation for fast liquid chro-matography in pharmaceutical applications was recently documented (69).

Longitudinal Diffusion

Longitudinal diffusion is caused by the inescapable relax-ation of the axial concentration gradients during the resi-dence time of the sample zone in the chromatographic col-umn. More quantitatively, this source of band broadening is directly dependent on the effective diffusion coefficient (Deff) of the analyte along the heterogeneous bed made of a solid porous material, the internal eluent (its average composi-tion may differ from that of the bulk mobile phase), and the external mobile phase. The coefficient B in the reduced van Deemter equation is given by equation 5 (5):

B = 2 1+k1

D

Def

m[5]

where k1 is the zone retention factor.Longitudinal diffusion is dominant for low reduced veloci-

ties (small linear velocity u, large diffusion coefficient Dm, and small particle diameter dp) and is dependent on the sample dif-fusivity inside the stationary phase (dp

= Ω Dm), which affects the value of Deff (12). This is particularly true in reversed-phase LC because of so-called “surface diffusion” in the stationary phase (13–15) unlike in HILIC (16) or chiral chromatography (17) where the intraparticle diffusivity is the smallest and is mostly controlled by the sole pore diffusion (18).

Solid–Liquid Mass Transfer Resistance Because of

a Finite Rate of Diffusion Across the Stationary Phase

This source of band dispersion is because of the velocity difference between molecules trapped inside the stationary phase (zero velocity) and those of other molecules, which simultaneously progress forward along the column in the interstitial mobile phase at an average linear velocity (u). As-suming spherical particles and a radial symmetry for the diffusion process across the porous particles, the general expression of the reduced mass transfer coefficient (C

ρ) for

core–shell particles (ρ is the core-to-particle diameter ratio; ρ = 0 for fully porous particles) is given by (19):

Cp =

30

1 1+2ρ+3ρ 2−ρ 3−5ρ 4

1−εe

1+k1

1

Ω

2εe k

1

1+ρ+ρ 2 2[6]

The contribution of the solid–liquid mass transfer resistance to the overall peak width is important for the largest reduced velocities (ν) (fast average linear velocity u, large particle di-ameter dp, small diffusion coefficient Dm), the lowest reduced intraparticle diffusivities (Ω), and the largest zone retention factors k1.

Eddy Dispersion

Band broadening because of eddy dispersion is related to all sources of velocity biases in the interstitial eluent. Addition-ally, these velocity biases combine with the border effects associated with the nonideal distribution and collection of



the sample zone at the inlet and outlet of the column, respectively. Eddy dis-persion is divided into two categories: the short range and the long range (7).

Short-Range Eddy DispersionShort-range eddy dispersion involves all the velocity biases that take place over distances smaller than a few par-ticle diameters. The left graph in Figure 3 shows the square cross-section area of the bulk region of a packed bed (the particles are represented in white) with the intensity (from dark blue for the lowest local velocities to brown for the largest ones) of the local eluent veloc-ity along the direction normal to this surface. Significant differences in fluid velocity are observed between those at the surface of the particle (zero velocity, dark blue) and those in the center of the interstitial volumes between adjacent

particles (maximum velocity, brown). This represents the so-called tran-schannel eddy dispersion (7). Smaller differences in the mobile phase velocity are also observed over distances cov-ering one to a few particle diameters because of the random structure of a packed bed. This accounts for the so-called short-range interchannel eddy dispersion (7). These two HETP terms are additive and each is quantitatively described by the Giddings’s coupling theory of eddy dispersion. Their sum is written as follows (7):

Ashort-range

(υ )=1+(ω

1υ ) -1

2λ1

+1+(ω

2υ ) -1

2λ2

[7]

The two pairs of parameters λ1 and ω1

(trans-channel eddy dispersion) and λ2

and ω2 (short-range interchannel eddy dispersion) can either be guessed (as

Giddings did in his book on dynamics of chromatography [7]) or measured from the reconstruction of the actual three-dimensional (3D) bulk structures of packed beds (with nonporous particles) after numerically solving the Navier-Stokes equations for the determination of the complete flow field u(x,y,z) along these bulk structures and calculating the band dispersion (H) from a standard advection-diffusion model. Examples are available in the literature for col-umns packed with core–shell particles (20) and for silica-based monolithic col-umns (21). They permit the correction and adjustment of Giddings’s original guesses for the four parameters in equa-tion 7. In real columns, the impact of the short-range eddy dispersion on column efficiency is small compared to that of the long-range eddy dispersion, which is defined in the next section.

Long-Range Eddy DispersionThe presence of a cylindrical wall sur-rounding the packed bed (the stainless steel column tube) induces a radial het-erogeneity of the bed structure across the column diameter. This is illustrated in the central plot of Figure 3 for a 50-μm i.d. capillary column packed with 1.7-μm particles. A peripheral and a cen-tral region can be clearly distinguished. The thickness of the wall region is about five times the particle size (dp). The aver-age interstitial velocity in this peripheral volume is close to 10% larger than the average velocity in the bulk central re-gion of the capillary column. This long-range eddy dispersion (also called trans-column eddy dispersion [7]) is combined with the nonidealities of the distribution and collection of the sample zone at the inlet and outlet of the column, respec-tively. The right graph in Figure 3 shows that the sample molecules do not actu-ally enter the packed bed as a perfectly f lat rectangular plug, but rather as a bowl (see the trajectories along the blue lines). Regarding the sample collection through the central aperture of the out-let endfitting, it takes a slightly longer time for a molecule located in the wall and corner region of the column to be eluted than for those present in its cen-tral zone. This is illustrated by the red streamlines in the right graph of Figure

Hydrodynamic Random bed Center-to-wall

Long-range eddies and border effects

Large impact on effciencyWeak impact on effciency

Short-range eddies

Inlet–outlet sample distribution and collection

bed heterogeneitystructure

4

50 0

0

11

22

10

15

20

25

30

+

3

2

Radial velocityprofle

u(r)/u

1

Flo

w in

ten

sity

Rad

ial d

ista

nce

/dp

boundary layer

0.09

(a) (b)

90

1.0 1.1 1.2 1.3

80

70

70

C18

C1 Air 1.0

60

60

50

50

40

40

30

30

20

20

10

100

0

Twall

= 299 K0.06

Ab

sorb

an

ce a

t 294 n

m (

mA

U)

Resp

on

se (

mV

)

0.03

0.00

Inlet elution volume (cm3) Time (s)

Figure 3: Visualization of short-range and long-range eddy dispersion in a packed column. Adapted with permission from reference 67.

Figure 4: Visualization of peak distortion in (a) UHPLC and (b) SFC. Adapted with per-mission from references 21 and 22.



3. The long-range eddy dispersion ac-counts for at least 50% and up to 90% of the total plate height of small molecules at high reduced velocities (5).

Adsorption–Desorption Kinetics

When the number of adsorption–de-sorption steps (from the internal eluent to the surface of the mesopores for ad-sorption and vice versa for desorption) during the migration of the analyte along the column becomes too small, both the plate height and the peak skew-ness (asymmetry) can reach excessively large values. This result was predicted a long time ago by the molecular or sto-chastic theory of chromatography (2), and it is usually observed for large mol-ecules involving slow conformational changes in the stationary phase (27) and in chiral chromatography (17). The number of adsorption steps is governed by the rate constant of adsorption kads

(unit Hz), which depends on four inde-

pendent properties of the adsorption system, as shown in equation 8 and ex-plained below:

kads ∝ exp 1− f

refection

T

MS

p RT

Ea

− [8]

• kads is proportional to the average molecular speed of the analyte in the bulk phase in contact with the solid adsorbent.

• High temperatures and low-molecu-lar-weight compounds contribute to increase the number of adsorption steps. It is also proportional to the specific surface area per volume unit of the adsorbent (Sp).

• Porous particles with a larger sur-face-to-volume ratio are preferred. However, the adsorption process is also characterized by the activation energy for adsorption because only a fraction of the analytes with an energy equal to or larger than Ea can be adsorbed. Again, high tempera-

tures are definitely recommended to increase the intensity of kads.

• Finally, even though some analyte molecules have the required energy for adsorption (E > Ea), a fraction of them ( freflection) will not be adsorbed because they simply bounce back to the internal eluent after collision with the solid adsorbent. Therefore, this stresses the importance and need for a high surface coverage of the bonded ligand in chiral separations.Assuming a first-order adsorption–

desorption kinetics (Langmuir adsorp-tion), the contribution (Cads) of the adsorption–desorption process to the overall coefficient C = Cp + Cads in the reduced van Deemter equation 3 is given by equation 9 (28):

Cads =

εe

1−εe

21+k

1

2

k1

1+kp

2kp

1

1−εp

1

1−ρ 3

Dm

kads

d 2p

[9]

where kp is the retention factor for the particle volume only (no external eluent) (28).

Recent investigations have shown that when kads is larger than 104 Hz, the adsorption–desorption kinetics has no impact on the column efficiency and on the peak resolution in LC (29). That is the case for most reversed-phase LC and HILIC applications based on the analy-sis of small molecules. This is no longer the case in chiral chromatography for the analysis of large biomolecules.

Friction Expansion

of the Mobile Phase

Under certain operating conditions, such as high pressures in UHPLC, large linear velocities in HPLC, an external thermal environment with a thermo-stated column wall, nonadiabatic ther-mal conditions in general, and expan-sible eluents used in supercritical fluid chromatography (SFC), the friction of the mobile phase against the packed bed and its decompression can be accompa-nied by either a significant production (when friction dominates) or adsorption (when decompression dominates) of heat in the packed bed volume. Overall, the heat power generated per unit length of the column is given by equation 10 (22):

1

400

300

200

100

0

tp = 15 min

tp = 120 min

tp = 480 min

-1.2 -0.6 0.0 0 1x104 2x104 3x104

Time (min) Parking time (s)

0.6 1.2

(a) (b)

C/C

o

Peak v

ari

an

ce (

s2)

0

Second generationx

y

z

Figure 5: Illustration of the peak parking method applied for thiourea on an end-capped C1 5-μm column (150 mm × 4.6 mm) using 75:25 (v/v) methanol–water as the mobile phase. Flow rate: 0.5 mL/min. (a) Eluted peaks for different peak parking times (0, 5, 10, 15, 30, 60, 120, 180, 300, and 480 min). (b) Corresponding plot of the time peak variance versus the arresting time. Adapted with permission from reference 30.

Figure 6: Visualization of the bulk, homogeneous, and random structures of packed (left) and silica-based monolithic (right) columns from confocal laser scanning micros-copy and image analysis. Adapted with permission from reference 42.



Pfriction / expansion

=F

v1+αT L

ΔP [10]

where α (<0) is the thermal expan-sion coefficient of the eluent, T is the

temperature, Fv is the f low rate, and ΔP is the pressure drop along the column.

In UHPLC, the average product αTis around –1/3 (23) (1 + αT is positive)

so that a significant amount of heat can be produced inside the column for the largest product of the f low rate by the pressure drop. A first fraction of this heat is dissipated out of the column by convection (at the column outlet) and by conduction (at the column wall). The remaining fraction is dissipated by conduction, which produces radial temperature gradients and, therefore, radial gradients of the local migration velocity for retained compounds across the column diameter. The intensity of this fraction depends on the thermal environment of the column (adiabatic versus nonadiabatic). It usually repre-sents about 25% of the heat produced by friction when the column is kept horizontal under still-air conditions (68). Severe band distortions are then observed, as shown in the left graph of Figure 4 when the wall temperature is uniformly maintained (24). Analyte molecules in the hot central region of the column migrate faster than those in the cooler wall region. Inversely, in SFC the pressure drops are relatively small but the product of the thermal expansion of pure carbon dioxide by the temperature is around –10 (25) (1 + αT is negative). Heat is now absorbed by the column and the temperature at the center of the column is cooler than the temperature of its peripheral region. Overall, the effect on column efficiency is the same as in UHPLC and the peak shape becomes severely affected (see Figure 4b).

Measurement of the Various Sources of Band Broadening in LCIn this section, a series of experi-mental protocols are proposed for the experimental determination of the reduced longitudinal coefficient (B) (12,30), including: the solid–liq-uid mass transfer resistance coeffi-cient (Cp) because of finite diffusion across the stationary phase (5,6), the short-range (31) and long-range eddy dispersion A(ν) terms (6), and the solid–liquid mass transfer resistance coefficient (Cads) because of the slow adsorption–desorption process (17). Before performing these measure-ments, the diffusion coefficient of the

6

4

2

(a) (b)

0 010 1020 20

Reduced velocity ν Reduced velocity ν

B/ν + Cp ν + A(ν )

Experimental h data

B/ν + Cp ν + A(ν )

Experimental h data

h

6

4

2

h

30 3040

Figure 7: Comparison of the total HETP measured on a chiral column (red symbols) and the sum of the longitudinal diffusion (B/ν), the solid–liquid mass transfer resis-tance because of a f nite diffusivity across the stationary phase (Cν), the short-range eddy dispersion (all three measured on the chiral column), and the long-range eddy dispersion (measured on a reference arbitrary achiral column). The analytes were (a) the achiral compound tri-tert-butylbenzene and (b) the chiral compound (R,S) trans-stilbene oxide. Adapted with permission from reference 17.

60

40

20

0

Hap

pare

nt (

µm

)

0.0 0.2 0.4

1/(1+k’)2

0.6

System 1

System 2

System 3

System 4

σ2 v,ex = 41.9 µL2

σ2 v,ex = 11.1 µL2

σ2 v,ex = 3.0 µL2

σ2 v,ex = 0.5 µL2

Figure 8: Plots of the apparent HETPs of homologous compounds versus the recip-rocal of (1 + k′)2 for four different UHPLC instruments. Column: 100 mm × 2.1 mm packed with 1.6-μm prototype reversed-phase LC C18 core–shell particles. Eluent: 75:25 (v/v) acetonitrile–water. Temperature = 24 °C. Note that the system 2 has been modif ed with 140-μm i.d. connectors. It is not representative of the commercial stan-dard instrument. Adapted with permission from reference 46.



analyte in the bulk phase should first be accurately measured.

The Bulk Diffusion Coefficient Dm

All of the effective diffusion coef-f icients involved in the chromato-graphic process (such as the effective diffusivity [Deff] along the heteroge-neous packed bed made of particles immersed in the liquid phase and the intraparticle diffusivity [Dp]) are scaled to the bulk diffusion coeffi-cient (Dm). The measurement of Dm

is also necessary for the determina-tion of the reduced velocity (ν). Two methods are available: the peak park-ing method and the capillary method.

The Peak Parking MethodThis method consists of parking the sample zone inside a column packed with nonporous particles at the very

same axial position for a series of in-creasing arresting times (tp) and in re-cording the time peak variance (σt

2) of the eluted band (32). The best slope of the expected linear plot of σt

2 versus tp

provides an excellent estimate (within 5%) of the diffusion coefficient (Dm) as long as the analyte is not retained on the surface of the nonporous particles and that a reference standard is avail-able for which the diffusion coefficient Dm,ref is known accurately. This refer-ence compound permits the measure-ment of the external obstruction factor (γe) of the column used in this proto-col. If u is the linear velocity applied for the elution of the sample zone, the expressions of γe and Dm are then di-rectly given by equations 11 and 12 (5):

γe =

2Dm, ref

1

Δtp

Δσ2

t, refu 2 [11]

Dm=

2γe

1

Δtp

Δσ2

tu 2 [12]

The Capillary MethodIf the studied compound is adsorbed onto the nonporous particles, then the capillary method can be a suitable alter-native solution (33,34). It is valid if the following five conditions are satisfied:

• The radial equilibration of the sam-ple concentration is effective along a tube of length (L) and inner radius (rtube), operated at the average linear velocity u. This condition is achieved when (35)

<< 3.82D

m

r 2tube

2u

L

[13]

• The contribution of axial molecular diffusion to the total spatial vari-ance is negligible. This condition is achieved if (36)

<<148D2

m

r 2

tube u 2

[14]

• The extracapillary time peak vari-ance is negligible compared to the observed time variance.

• There is no secondary flow circulation in the coiled tube. The product of the Dean number by the Schmidt number should be smaller than 100 (37).

• The diffusion coefficient of the stud-ied compound is validated from the value Dm,ref measured for a standard reference compound for which the true diffusion coefficient is accu-rately and independently known from another technique. Using the capillary method, only one

injection is necessary. The diffusion coefficient is then obtained from the retention time (tR) and the time peak variance (σt

2) of the eluted peak:

Dm =

24

1

σ2

t

tR

rtube

2 [15]

The Reduced

Longitudinal Coefficient B

The reduced B coefficient in the re-duced plate height equation is mea-sured from the peak parking method described above using the column of interest. Figure 5 shows the typi-cal and expected observations when increasing the peak parking time (tp):

Reduced velocity ν

Re

du

ced

pla

te h

eig

ht h 15

10

10 15 20

5

50

15(b)(a)

10

5

00

Reduced velocity ν

10 15 2050

Eddy dispersionLongitudinal diffusionSolid–liquid mass transfer

Eddy dispersionLongitudinal diffusionSolid–liquid mass transfer

Re

du

ced

pla

te h

eig

ht h

1.7 µm

2.6 µm

Pro

bab

ilit

y d

en

sity 6

3 1.7-µm fully porous, ε = 0.460

1.7-µm fully porous, ε = 0.366

2.6-µm core–shell, ε = 0.460

2.6-µm core–shell, ε = 0.366

Monosized, ε = 0.460

Monosized, ε = 0.366

2.5

2

1.5

1

0.5

100 101 102 103

5

4

32

1

01.5 2.5 3.52 3

1.5 2.5 3.52 3

Pro

bab

ilit

y d

en

sity 6

5

4

32

1

0

Particle diameter (µm)

εe = 0.460

εe = 0.366

h =

H/d

A

� = UdA/DM

Figure 9: Contributions of the longitudinal diffusion (green), eddy dispersion (red), and solid–liquid mass transfer resistance (blue) terms to the overall reduced plate height of 100 mm × 4.6 mm columns packed with (a) 2.5-μm fully porous particles and (b) 2.6-μm core–shell particles. Analyte: naphthalene. Mobile phase: 65:35 (v/v) acetonitrile–water. Temperature = 297 K. Adapted with permission from reference 47.

Figure 10: Effect of the RSD of the particle size distribution on the short-range eddy dispersion for dense (bed porosity = 0.366) and loose (bed porosity = 0.46) randomly packed beds. Adapted with permission from reference 48.



the peak width becomes larger and the time peak variance increases linearly with increasing tp. The most relevant experimental information includes the best slope of the linear plot shown in the Figure 5b, the zone retention factor k1 that can be measured from a single injection for tp = 0, and the applied con-stant linear velocity u during the peak parking experiments. B is then given by equation 16 (6):

B =

Dm

1

Δtp

Δσ2

t

1+k1

u2

[16]

The Reduced Mass Transfer Coefficient Cp

The determination of this mass trans-fer coefficient is relatively approximate and challenging compared to that of the coefficient B. First, it is based on the assumption that the symmetry of diffusion across the spherical particles is strictly radial (so that the concentra-tion is uniform at the external surface area of the particles). The general ex-pression of Cp for core–shell particles is given by equation 6 (19). Secondly, the

parameter Ω or the ratio of the intra-particle diffusivity Dp to the bulk dif-fusion coefficient Dm in equation 6 is unknown and must be estimated. This can be done by combining the previ-ous peak parking results for the deter-mination of the B coefficient and the selection of a validated mathematical model of effective diffusion along the chromatographic bed (38). By simple inversion of the mathematical formula of the best model of effective diffu-sion, the parameter Ω can be extracted and Cp can be estimated directly from equation 6. See reference 5 for the list of available models of effective diffusion in chromatographic beds.

The Reduced Short-Range Eddy Dispersion ASR(ν)No chromatographic methods are cur-rently available for the direct measure-ment of the short-range eddy dispersion in real chromatographic columns term because columns have an inlet, an outlet, and a wall boundary. These boundaries induce an additional source of eddy dis-persion (the so-called long-range eddy dis-

persion) as previously described and illus-trated in the middle and right graphs of Figure 3. Therefore, original and differ-ent approaches were recently developed based on either a computer-based (39,40) or an image-based (41–43) reconstruc-tion of the actual bulk random structure of packed beds. The computer-based approach is based on the Jodrey-Tory algorithm with independent adjustment of the bed porosity (from random dense to random loose packings) and degree of heterogeneity (from long-range uniform to long-range heterogeneous distribution of spherical particles). The image-based approach involves confocal laser scan-ning microscopy (CLSM) in combina-tion with a sophisticated treatment of the image data (41). It has been success-fully applied to the direct visualization of the homogeneous, random structure of both particulate and monolithic columns as shown in Figure 6. After the precise 3D structure of the bulk region of the chromatographic bed is known, the flow field and the band broadening of analytes along these structures can be calculated and used to measure the best coefficients λ1, ω1, λ2, and ω2 in equation 7.

The Reduced Long-Range Eddy Dispersion ALR(ν)The long-range eddy dispersion HETP term, which results from the border and wall effects on band broadening in relatively short or wide columns, is accessible experimentally provided that the friction-expansion HETP term pre-viously described can be neglected and the adsorption–desorption kinetics is fast enough (kads > 104 Hz). ALR(ν) is then directly measured after subtrac-tion of the B/ν, Cpν, and ASR(ν) HETP terms from the total reduced HETP, h, obtained from the numerical integra-tion of the whole peak for the deter-mination of the first (μ1) and second central (μ2′) moments (44):

ALR (υ )= C

p υ −A

SR (υ )L

μ1

− −2

μ2

υ

B' [17]

The Apparent Adsorption Rate Constant kads

In a recent attempt to identify and quan-tify the different sources of band broad-ening in a cellulose-based chiral column, the adsorption rate constant (kads) was

40

HETP (

µm

)

20

0

0.0 0.1 0.2

Superfcial linear velocity (cm/s)

Skeleton–throughpore mass transfer

Eddy diffusion

Longitudinal diffusion

Total HETP

0.3 0.4

Figure 11: Contributions of the longitudinal diffusion (red), eddy dispersion (green), and solid–liquid mass transfer resistance (blue) terms to the overall reduced plate height of 100 mm × 4.6 mm silica-based reversed-phase LC monolithic column of the f rst generation. Analyte: naphthalene. Mobile phase: 55:45 (v/v) acetonitrile–water. Temperature = 297 K. Adapted with permission from reference 53.



estimated by assuming a homogeneous adsorption process (only one adsorp-tion rate constant will be measured) and a first-order kinetics (the analyti-cal solution of the general rate model of chromatography is available after La-place transform) (17). Accordingly, the expression of the corresponding reduced HETP term, Cadsν, is given by equation 9. The method was based on the use of an achiral compound and an achiral col-umn for the estimation of the long-range

eddy dispersion ALR(ν) of the chiral col-umn. The achiral column was not ran-domly picked: It should have the same dimensions (length and inner diameter) as those of the chiral column, should be packed with particles having the same size as the particles in the chiral col-umn, and should be slurry-packed by the same column manufacturer (same column endfitting, same column tube, comparable slurry-packing process). Therefore, the long-range eddy disper-

sion is likely to be similar on both the chiral and achiral columns. The ad-vantage of the achiral column is that its long-range eddy dispersion HETP can be directly measured from equation 16 since the adsorption–desorption process is fast enough and does not contribute to the total reduced HETP. This choice for the achiral column was validated from the injection of the achiral com-pound onto the chiral column and by comparing its total HETP to the sum of B/ν, Cpν, ASR(ν) (all three of these HETP terms were measured on the chiral col-umn as mentioned above), and ALR(ν) (the only HETP term measured on the achiral column). Figure 7 compares these two reduced HETP plots for tri-tert-butylbenzene (nonchiral compound, Figure 7a) and (R,S) trans-stilbene oxide (chiral compound, Figure 7b). The ob-served difference between the red and black symbols represents the contribu-tion of the adsorption–desorption kinet-ics HETP term to the total plate height. The term kads can finally be estimated from equation 9 and the best slope of the plot of this h difference versus the reduced velocity (ν).

The Extracolumn

Volume Variance σv,ex

2

The measurement of extracolumn band broadening should be performed without replacing the chromatographic column with a zero-dead-volume union connector. Indeed, the back pressure generated by the zero-dead-volume union connector is much smaller com-pared to that generated by the column. Because the pressure affects the diffu-sion coefficients (the eluent viscosity increases with increasing pressure), it also has an impact on the sample band broadening in the precolumn volume of the instrument (45). When the extracol-umn volume is much smaller than the hold-up column volume, the use of a se-ries of homologous compounds (such as n-alkybenzenes or n-alkylphenones) al-lows for the noninvasive measurement of the system band broadening (46). The plot of the apparent plate height as a function of the reciprocal of (1 + k′)2

is expected to be linear provided that the intrinsic plate height varies little with increasing retention factor:

205

4

3

2

1

First generation

Second generation

500,000 plates/m

625,000 plates/m

4.6-mm i.d. monolith frst generation

Nmax

= 50,000 plates/m

Nmax

= 170,000 plates/m

L = 10 cm

4.6-mm i.d. monolith second generation

Hed

dy (

µm

)

H (

µm

)

10

-12 µm

0.0 0.1 0.2 0.3 0 1 2 3

uav

(mm/s)

4 5 6Superfcial linear velocity (cm/s)

-0.5 µm

0

8

Red

uce

d B

co

eff

cien

t

Zone retention factor (k1)

6

Reversed-phase LC

HILIC

0 5 10 15 20

4

2

0

Figure 12: Diminution of the total eddy dispersion HETP (left) and comparison between the short-range eddy dispersion (right) and for the f rst and second generation of silica monolithic columns. Adapted with permission from references 53 and 52.

Figure 13: Comparison between the reduced longitudinal coeff cient B measured on a reversed-phase LC C18 (black) and HILIC (red) column. Reproduced with permission from reference 18.



H(k′) = H

intrinsic LV

0

σ2

v, ex

(1+k′)

1+ 2 2 [18]

Knowing the column length (L) and the hold-up column volume (V0), the slope of this representation pro-vides direct ly the system volume variance σv,ex

2. This experimental protocol was recent ly appl ied to a series of n-alkylphenones using four d i f ferent conf igurat ions of UHPLC instruments. Their extra-column volume variance increases from about 0.5 to 3, 10, and 40 μL2

(46). Figure 8 shows the correspond-ing plots of H versus (1 + k′)-2 for a 100 mm × 2.1 mm column packed with 1.6-μm core–shell particles at a constant f low rate of 0.4 mL/min. The experimental plots are quasilin-ear and their y-intercept is virtually the same and equal to the intrinsic plate height of the column. The sig-nificant difference in the observed slopes ref lects directly on the dif-ferent levels of band spreading along the four UHPLC systems reported in the legend of the graph.

Determining the Mass Transfer Mechanism in Different Columns and LC Retention ModesThe experimental protocols described in the previous sections are applied here for the determination of the com-plete mass transfer mechanism of four different classes of chromatographic columns: reversed-phase LC C18 col-umns packed with fully and superfi-cially porous particles, reversed-phase LC C18 silica-based monolithic col-umns, a HILIC column packed with fully porous particles, and a cellulosed-based chiral reversed-phase column packed with fully porous particles.

Reversed-Phase LC Particulate

Columns: Core–Shell Versus

Fully Porous Particles

Figure 9 shows the contributions of the B/ν, A(ν), and Cpν HETP terms to the total observed h values for two columns having the same dimension (100 mm × 4.6 mm). One (Figure 9a) is packed with 2.5-μm fully porous particles and the second (Figure 9b) is packed

with 2.6-μm core–shell particles (47). The most remarkable observation is that all three reduced HETP terms are smaller for core–shell particles. B is smaller because the impermeable solid core occupies about 25% of the column volume and sample diffusion is not al-lowed across the volume of these cores. This is an advantageous feature for the shell particles at the lowest reduced velocities and at those close to the op-timum speed (νopt = 8–10). Cp is also smaller because the average length of the diffusion path across the core–shell particles is shorter. However, for small retained molecules like naphthalene (k′

≈ 3, Dm ≈ 1.5 × 10-5 cm2/s) and 2.5-μm particles, the maximum reduced veloc-ity is about 18 and the largest gain in HETP is just 0.1 h unit. So, in practice, the decrease of the Cpν HETP term is virtually negligible at optimum speed. Regarding the total eddy dispersion HETP A(ν) term, it is about 0.5 h units smaller for the shell than for the fully porous particles at high velocities. The question remains whether this is be-cause of a diminution of the short- or the long-range eddy dispersion HETP. It is known that the particle size distri-bution (PSD) of core–shell particles is much tighter than that of conventional particles with a relative standard devia-tion (RSD) around 5% versus 15–25%. Figure 10 shows the calculated reduced HETPs along computer-based gener-ated bulk structures of actual 1.7-μm fully porous particles (wide PSD) and 2.6-μm core–shell particles (narrow PSD) (48). The calculations were also carried out for monosize particles (RSD = 0%). The results demonstrated that the RSD of the PSD had no impact whatsoever on ASR(ν) at a fixed bed po-rosity. In fact, ASR(ν) is mostly a func-tion of the bed porosity. Additionally, it is known experimentally that the external porosity of columns packed with core–shell particles is around 0.40 (49–51) versus only 0.36 for col-umns packed with conventional fully porous particles. Therefore, the short-range eddy dispersion is even larger for core–shell than for standard particles (48). In conclusion, the higher kinetic performance of the 2.6-μm core–shell particles shown in Figure 9 is because

4

2

6

0 10

Reduced velocity υ

HILIC, 3.5 µm

Reversed-phase LC, 5 µm

Chiral reversed phase, 5 µm

20

h

Figure 14: Comparison between the reduced longitudinal coeff cients B measured on a reversed-phase LC C18 column (black) and a HILIC column (red). Reproduced with permission from reference 18.



of a significant reduction of the long-range eddy dispersion HETP. This re-flects the higher degree of homogeneity of the packed bed (from the center to the inner wall of the column tube) of columns packed with core–shell parti-cles than that of columns packed with fully porous particles.

Reversed-Phase LC

Silica-Based Monolithic Columns:

First Versus Second Generation

Figure 11 shows the contribution of the same sources of band broadening as in Figure 9, but for a silica-based mono-lithic column (100 mm × 4.6 mm) of the first generation commercialized in the early 2000s. Remarkably, the minimum HETP is no smaller than 18 μm because of the large intensity of the eddy dispersion HETP term at and beyond the optimal velocity (52,53). The manufacturer of these monolithic columns prepared a second generation of silica-based monolithic columns (same dimension) with smaller mac-ropore and skeleton sizes (or domain size) and a much reduced impact of the A(ν) term on the observed column ef-ficiency (54,55). Figure 12 shows, from a quantitative viewpoint, that the total eddy dispersion HETP decreases by about 12 μm, which is considerable. The maximum column eff iciency improved then from about 50,000 plates/m (for the first generation in-strument) to 170,000 plates/m (for the second generation). However, the

reconstruction of the bulk structures of these two types of monolithic col-umns and the calculation of the short-range eddy dispersion (right graph in Figure 12) revealed a difference of only 0.5 μm (54), a value far smaller than the 12 μm observed. If the silica-based monolithic columns had no wall (in-finite diameter column) and no bor-der (infinitely long column), the first and second generation of silica-based monolithic columns would generate as many as 500,000 and 625,000 plates/m. This shows evidence that the perfor-mance of 4.6-mm i.d. silica monolithic columns is controlled by either the transcolumn velocity biases or the bor-der effects. Additionally, the analysis of the center-to-wall heterogeneity of the monolith structure by CLSM shows no significant default for the second generation column (54). In conclusion, further improvement of the efficiency of silica-based monolithic columns is linked to a better distribution and col-lection of the sample zone at the inlet and outlet of the column, respectively.

Mass Transfer Mechanism

in HILIC Columns

The mass transfer mechanism in HILIC columns differs from that in reversed-phase LC columns (same di-mension, same particle size) because the intraparticle diffusivity is much smaller across nonbonded porous silica than across the same (but C18-bonded) silica particles. This was unambigu-

ously revealed from the measurement of the reduced longitudinal diffusion coefficient B as shown in Figure 13 (18). For retained analytes, B is typi-cally twice as large in reversed-phase LC as in HILIC columns because of the concentration excess of the organic el-uent (acetonitrile), which forms a thick multilayer (three monolayers) onto the silica-C18 surface in reversed-phase LC (56), and the presence of a water-rich layer onto neat silica in HILIC (57). Therefore, in reversed-phase LC, the analytes are adsorbed and retained through weak dispersive interactions onto nonlocalized adsorption sites and they accumulate by a partition mecha-nism in a poorly viscous acetonitrile-rich eluent (the partition of the sample molecules between pure acetonitrile and the aqueous eluent is large). Their diffusion or mobility in the station-ary phase is then fast. In contrast, in HILIC, the mobility of the retained analytes is relatively small because they are either strongly bound to specific adsorption sites (zero mobility) or con-centrated (partition mechanism) in an aqueous layer that is three times more viscous (58,59). Consequently, the re-duced coefficient Cp is larger in HILIC than in reversed-phase LC (because of the smaller intensity of the parameter Ω in equation 6 for HILIC particles) and the optimum reduced velocity is usually observed around five in HILIC instead of 10 in reversed-phase LC retention mode. This explains why reversed-phase LC has been such a suc-cessful mode of retention in LC.

Mass Transfer Mechanism

in Chiral Columns

Chiral separations may involve an additional source of band broaden-ing compared to reversed-phase LC and HILIC retention modes. The ad-sorption–desorption process in chiral chromatography is usually slower than that in reversed-phase LC and HILIC columns. It can contribute significantly to the intensity of the overall C coef-ficient in the van Deemter plot. Figure 14 illustrates this statement by compar-ing the experimental reduced HETPs of reversed-phase LC C18, HILIC, and cellulosed-based chiral reversed-phase

6

1.0

Longitudinal diffusion

Short-range eddy dispersion

Long-range eddy dispersionSolid–liquid mass transfer resistance: D

p

Solid–liquid mass transfer resistance: kads

0.5

0.00 10 20 30

Experimental h data

Reduced velocity ν

hi/h

tota

l

4

2

0 10 20 30

B/ν + Cp ν + A(ν)

ν

h

Figure 15: Impact of a slow adsorption–desorption process (kads

) on the total reduced plate height of trans-stilbene oxide on a cellulose-based chiral stationary phase. Re-produced with permission from reference 17.



columns. The optimum reduced ve-locity is the smallest for the chiral reversed-phase column at about three. In the absence of activation energy (Ea) (no kinetic barrier for adsorption) and if all molecular collisions between the surface area of the adsorbent and the analyte lead to an adsorption event, then the adsorption rate constant (kads) would be as large as 109 Hz (60) and the adsorption–desorption process would have no impact whatsoever on column efficiency. In fact, the measurement of all the sources of band broadening in a chiral column packed with 5-μm cellulose-based particles revealed that kads was of the order of 103 Hz only (17). For such small values of kads, the kinetics of adsorption definitively lim-its the column performance, as shown in Figure 14. Figure 15 shows the con-tribution of the slow rate of adsorption–desorption to the total HETP for the enantiomer trans-stilbene oxide on a cellulose-based chiral column. Roughly, the reduced coefficient (Cads) represents more than 50% of the total reduced co-efficient C in the van Deemter plot and the band broadening because the slow adsorption rate of trans-stilbene oxide accounts for about 25% of the total plate height at a reduced velocity of 30.

Conclusions: Further Improvement of the Observed Column EfficiencyBased on the previous experimental protocols designed for the measure-ment of all sources of band broaden-ing in reversed-phase LC, HILIC, and chiral reversed-phase columns, the practitioner now has direct access to the quantitative contributions of lon-gitudinal diffusion, eddy dispersion (short- and long-range), mass transfer resistance because of a finite rate of dif-fusion across the particle, mass transfer resistance because of a slow adsorption–desorption process, and of the HPLC system band spreading to the overall observed peak width. The column or instrument manufacturers are finally able to pinpoint with confidence the intrinsic weaknesses of their products. They are becoming more inclined to invest time and financial resources to cope with the relevant problems, which

are limiting the column efficiency ob-served by their customers.

Pursuing the transition from conven-tional HPLC to faster UHPLC methods, smaller column dimensions and finer particles are increasingly invading ana-lytical laboratories. Irrespective of the retention mode applied (reversed-phase LC, HILIC, chiral), the efficiency of this new generation of chromatographic columns can be seriously limited by the band dispersion that takes place along the different parts (injection loop, valve, connecting tubes, detector, electronics) of the instrument. For instance, to get at least 80% of the intrinsic efficiency (Nintrinsic ≈ 22,500) of a 50 mm × 2.1 mm column packed with 1.3-µm core–shell particles and for a retention fac-tor of 1 (the column volume variance is only equal to 1.6 µL2), the system band spreading should not be larger than 0.4 µL2, which is a level not yet achieved even with the best UHPLC instruments currently available.

Regarding the intrinsic kinetic per-formance of a chromatographic column, the measurement of the eddy dispersion HETPs in reversed-phase LC, HILIC, and chiral reversed-phase columns has unambiguously revealed that it was governed by the long-range eddy disper-sion. In other words, the column hard-ware should be redesigned to minimize the effects of the center-to-wall f low heterogeneity, of the nonideal sample distribution at the column inlet, and of the asynchronous collection of sample zone at the column outlet. Active flow technology or parallel segmented out-let (and curtain) flow chromatography was recently developed to get rid of the molecules migrating along the periph-eral region of the bed (61,62). Only the molecules leaving the column through the central region of the outlet cross-section area are collected and detected. The efficiency of the shortest columns packed with sub-3-µm particles can then be multiplied by a factor close to two after elimination of the rear peak tailing and reduction of the long-range eddy dispersion (63,64). Neverthe-less, it is important to remember that this new column technology also has some limitations in terms of gain in column efficiency: It is not advanta-

geous for the longest columns (because of the radial equilibration across the whole bed diameter [64,65]), for the largest particle sizes (because of large transverse diffusion coefficients [66]), and for large molecules since the mass transfer mechanism is essentially gov-erned by the Cν and not by the A(ν) HETP term anymore.

In conclusion, the results of this series of experimental protocols stress the need for the integration of the injection device, separation medium, and detec-tion system into a single unit free from poor connections that can cause severe additional band broadening. Technologi-cal and scientific skills are then required in terms of miniaturization, micro- and nano-fluidics, and 3D printing to design and prepare tomorrow’s most efficient separations systems (67).

References(1) U.D. Neue, HPLC Columns, Theory, Tech-

nology, and Practice (Wiley-VCH, New York,

1997).

(2) C. Horvath, B. Preiss, and S. Lipsky, Anal.

Chem. 39, 1422–1428 (1967).

(3) J.R. Mazzeo, U.D. Neue, M. Kele, and R.S.

Plumb, Anal. Chem. 77, 460A–467A (2005).

(4) A.C. Sanchez, G. Friedlander, S. Fekete, J.

Anspach, D. Guillarme, M. Chitty, and T.

Farkas, J. Chromatogr. A 1311, 90–97 (2013).

(5) F. Gritti and G. Guiochon, J. Chromatogr. A

1221, 2–40 (2012).


1217, 5137–5151 (2010).

(7) J.C. Giddings, Dynamics of Chromatography

(Marcel Dekker, New York, 1965).

(8) F. Gritti, C. Sanchez, T. Farkas, and G. Gui-

ochon, J. Chromatogr. A 1217, 3000–3012

(2010).

(9) K.J. Fountain, U.D. Neue, E.S. Grumbach,

and D.M. Diehl, J. Chromatogr. A 1216,

5979–5988 (2009).

(10) J. van Deemter, F. Zuiderweg, and A. Klinken-

berg, Chem. Eng. Sci. 5, 271–289 (1956)

(11) J.C. Sternberg, Advances in Chromatography

(Marcel Dekker, New York, 1966).


1218, 3476–3488 (2011).

(13) F. Gritti and G. Guiochon, Anal. Chem. 78,

5329–5347 (2006).

(14) K. Miyabe and G. Guiochon, J. Chromatogr.

A 1217, 1713–1734 (2010).

(15) F. Gritti and G. Guiochon, AIChE J. 57,

346–358 (2011).




1302, 55–64 (2013).


1332, 35–45 (2014).


1297, 85–95 (2013).

(19) K. Kaczmarski and G. Guiochon, Anal.

Chem. 79, 4648–4656 (2007).

(20) S. Bruns and U. Tallarek, J. Chromatogr. A

1218, 1849–1860 (2011).

(21) K. Hormann, T. Müllner, S. Bruns, A. Hölt-

zel, and U. Tallarek, J. Chromatogr. A 1222,

46–58 (2012).

(22) H.-J. Lin and Cs. Horvath, Chem. Eng. Sci.

36, 47–55 (1981).

(23) M. Martin and G. Guiochon, J. Chromatogr.

A 1090, 16–38 (2005).


1216, 1353–1362 (2009).

(25) D. Poe and J. Schroden, J. Chromatogr. A

1216, 7915–7926 (2009).

(26) J.C. Calvin and H. Eyring, J. Phys. Chem.

1216, 416–421 (1955).

(27) B.C.S. To and A. Lenhoff, J. Chromatogr. A

1141, 191–205 (2007).

(28) K. Kaczmarski, J. Chromatogr. A 1218, 951–

958 (2011).


1348, 87–96 (2014).

(30) F. Gritti and G. Guiochon, Chem. Eng. Sci.

61, 7636–7650 (2006).

(31) D. Hlushkou, S. Bruns, and U. Tallarek, J.

Chromatogr. A 1217, 3674–3682 (2010).

(32) J.H. Knox and L. McLaren, Anal. Chem. 36,

1477–1482 (1964).

(33) J. Li and P.W. Carr, Anal. Chem. 69, 2530–

2536 (1997).

(34) J. Li and P.W. Carr, Anal. Chem. 69, 2550–

2553 (1997).

(35) G. Taylor, Proc. R. Soc. Lond. A 219, 186–

203 (1953).

(36) R. Aris, Proc. R. Soc. Lond. A 235, 67–77 (1956).

(37) L.A.M. Janssen, Chem. Eng. Sci. 31, 215–

218 (1976).

(38) F. Gritti and G. Guiochon, Chem. Eng. Sci.

66, 6168–6179 (2011).

(39) S. Khirevich, A. Höltzel, S. Ehlert, A.

Seidel-Morgenstern, and U. Tallarek, Anal.

Chem. 81, 4937–4945 (2009).

(40) S. Khirevich, A. Höltze, A. Seidel-Mor-

genstern, and U. Tallarek, Anal. Chem. 81,

7057–7066 (2009).

(41) S. Bruns, T. Müllner, M. Kollmann, J.

Schachtner, A. Höltzel, and U. Tallarek,

Anal. Chem. 82, 6569–6575 (2010).

(42) D. Hlushkou, S. Bruns, A. Höltzel, and U.

Tallarek, Anal. Chem. 82, 7150–7159 (2010).

(43) S. Bruns, J.P. Grinias, L.E. Blue, J.W. Jor-

genson, and U. Tallarek, Anal. Chem. 84,

4496–4503 (2012).

(44) P. Stevenson, F. Gritti, and G. Guiochon, J.

Chromatogr. A 1218, 8255–8263 (2011).

(45) F. Gritti and G. Guiochon, J. Chro-

matogr. A dx.doi.org/doi:10.1016/j.

chroma.2014.04.089 (2014).


1327, 49–56 (2014).

(47) F. Gritti and G. Guiochon, LCGC North Am.

30(7), 586–595 (2012).

(48) A. Daneyko, A. Höltzel, S. Khirevich, and U.

Tallarek, Anal. Chem. 83, 3903–3910 (2011).


1252, 31–44 (2012).


1252, 45–55 (2012).


1252, 56–66 (2012).

(52) D. Hlushkou, K. Hormann, A. Höltzel, S.

Khirevich, A. Seidel-Morgenstern, and U. Tal-

larek, J. Chromatogr. A 1303, 28–38 (2013).


1218, 5216–5227 (2011).

(54) K. Hormann and U. Tallarek, J. Chromatogr.

A 1312, 26–36 (2013).


1225, 79–90 (2012).


1169, 111–124 (2007).

(57) F. Gritti, A. dos Santos Pereira, P. Sandra,

and G. Guiochon, J. Chromatogr. A 1216,

8496–8504 (2009).

(58) S.M. Melnikov, A. Höltzel, A. Seidel-Mor-

genstern, and U. Tallarek, J. Phys. Chem.C

117, 6620–6631 (2013).

(59) S.M. Melnikov, A. Höltzel, A. Seidel-Mor-

genstern, and U. Tallarek, Anal. Chem. 83,

2569–2575 (2011).

(60) D. Hlushkou, F. Gritti, A. Daneyko, G.

Guiochon, and U. Tallarek, J. Phys. Chem C

117, 22974–22985 (2013).

(61) M. Camenzuli, H.J. Ritchie, J.R. Ladine,

and R.A. Shalliker, J. Chromatogr. A 1232,

47–51 (2012).

(62) R.A. Shalliker, M. Camenzuli, L. Pereira,

and H. J. Ritchie, J. Chromatogr. A 1262,

64–69 (2012).


1297, 64–76 (2013).

(64) F. Gritti, J. Pynt, A. Soliven, G.R. Dennis,

R.A. Shalliker, and G. Guiochon, J. Chro-

matogr. A 1333, 32–44 (2014).

(65) A. Daneyko, D. Hlushkou, S. Khirevich,

and U. Tallarek, J. Chromatogr. A 1257,

98–115 (2012).

(66) A. Daneyko, D. Hlushkou, S. Khirevich,

and U. Tallarek, J. Chromatogr. A 1257,

98–115 (2012).


3017–3035 (2013).


6488–6499 (2008).

(69) S. Fekete, I. Kohler, S. Rudaz, and D. Guil-

larme, J. Pharm. Biomed. Anal. 87, 105–119

(2014).

Fabrice Gritti is with the

Department of Chemistry at the

University of Tennessee in Knoxville,

Tennessee. Direct correspondence

to: [email protected] ◾

In Memoriam

This article was written in memory of Georges Andre Guiochon, Distin-guished Professor at the University of Tennessee Knoxville from 1987 to 2014. Prof. Guiochon was born on September 6, 1931, in Nantes, France, and passed away on Octo-ber 21, 2014, succumbing to an old enemy, neuromuscular failure due to post-polio syndrome. He was an exceptional and dedicated scientist in separation science for nearly six decades and received numerous in-ternational Awards for his immense contributions to the field. Beyond his professional achievements and lead-ership, his human qualities were no less remarkable. From the early ages of gas chromatography in the 1950s to the latest developments in super-critical f luid and liquid chromatog-raphy today, he was the mentor, the charming companion, and the loyal friend of several generations of scien-tists all over the world. In the names of all whose lives and professional ca-reers have been directly or indirectly influenced by Georges’ wisdom and advice, I would like to offer to his whole family our deepest sympathy. Georges will be missed and his mem-ory will live in us forever.

For more information on this topic, please visit our homepage at:



THE APPLICATION NOTEBOOK

Call for Application Notes

LCGC is planning to publish the next issue of T e

Application Notebook special supplement in June.

T e publication will include vendor application

notes that describe techniques and applications of

all forms of chromatography and capillary elec-

trophoresis that are of immediate interest to users

in industry, academia, and government. If your

company is interested in participating in these

special supplements, contact:

Michael J. Tessalone, Group Publisher,

(732) 346-3016

Edward Fantuzzi, Associate Publisher,

(732) 346-3015

Stephanie Shaf er, East Coast Sales Manager,

(774) 249-1890

Lizzy T omas, Account Executive,

(574) 276-2941

Application Note Preparation

It is important that each company’s mate-

rial f t within the allotted space. T e editors

cannot be responsible for substantial editing

or handling of application notes that deviate

from the following guidelines:

Each application note page should be no more

than 500 words in length and should follow

the following format.

Format

• Title: short, specif c, and clear

• Abstract: brief, one- or two-

sentence abstract

• Introduction

• Experimental Conditions

• Results

• Conclusions

• References

• Two graphic elements: one is the company

logo; the other may be a sample chromato-

gram, f gure, or table

• T e company’s full mailing address,

telephone number, fax number,

and Internet address

All text will be published in accordance with

LCGC ’s style to maintain uniformity through-

out the issue. It also will be checked for gram-

matical accuracy, although the content will not

be edited. Text should be sent in electronic for-

mat, preferably using Microsoft Word.

Figures

Refer to photographs, line drawings, and

graphs in the text using arabic numerals in

consecutive order (Figure 1, etc.). Company

logos, line drawings, graphs, and charts must

be professionally rendered and submitted as

.TIF or .EPS f les with a minimum resolution

of 300 dpi. Lines of chromatograms must be

heavy enough to remain legible after reduc-

tion. Provide peak labels and identif cation.

Provide f gure captions as part of the text,

each identif ed by its proper number and title.

If you wish to submit a f gure or chromato-

gram, please follow the format of the sample

provided below.

Tables

Each table should be typed as part of the main

text document. Refer to tables in the text by

Roman numerals in consecutive order (Table I,

etc.). Every table and each column within the

table must have an appropriate heading. Table

number and title must be placed in a continu-

ous heading above the data presented. If you

wish to submit a table, please follow the format

of the sample provided below.

References

Literature citations must be indicated by arabic

numerals in parentheses. List cited references

at the end in the order of their appearance. Use

the following format for references:

(1) T.L. Einmann and C. Champaign, Science

387, 922–930 (1981).

T e deadline for submitting application notes for the June issue of T e Application Notebook is:

April 16, 2015

T is opportunity is limited to advertisers in LCGC North America. For more information, contact:

Mike Tessalone at (732) 346-3016, Ed Fantuzzi at (732) 346-3015, Stephanie Shaf er at (774) 249-1890, or Lizzy T omas at (574) 276-2941.

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-pression problem was originally discov-ered. The ion suppression trace is shown on the bottom. Column: 75 mm × 4.6 mm ODS-3; mobile-phase A: 0.05% heptaf uo-robutyric acid in water; mobile-phase B: 0.05% heptaf uorobutyric acid in aceto-nitrile; gradient: 5–30% B in 4 min. Peaks: 1 = metabolite, 2 = internal standard, 3 = parent drug.



ES561703_LCGCAN0215_CV4_FP.pgs 01.29.2015 21:22 ADV blackyellowmagentacyan

the application - informa...

Documents