LC TROUBLESHOOTING
LLOQ: A case study
GC CONNECTIONS
Practical GC
COLUMN WATCH
The top 10 column myths
November 2013
Volume 16 Number 4
www.chromatographyonline.com
For analyzing biological, natural, and synthetic polymers
Field Flow
Fractionation
ES339292_LCA1113_CV1.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
THE FUTURE IS NOW SEPARATED FROM THE PAST.
Imagine a lab where all analytical scientists can acquire high-quality mass spectral data. On their own.
Within their existing workflows. Across every sample. Without training. A lab where uncertainty about compounds
is replaced by fast, efficient confirmation and the confidence that comes with crossing the LC/MS divide like never
before. Now imagine all this happening at the push of a button. This goes far beyond the power of mass detection.
This is the ACQUITY QDa™ Detector from Waters. SEPARATING BEYOND QUESTION.™ Visit waters.com/separate
Pharmaceutical & Life Sciences | Food | Environmental | Clinical | Chemical Materials
©2013 Waters Corporation. Waters and The Science of What’s Possible are registered trademarks of Waters Corporation.
ACQUITY QDa and Separating Beyond Question are trademarks of Waters Corporation.
Introducing the
ACQUITY QDa Detector
ES339517_LCA1113_CV2_FP.pgs 10.17.2013 21:34 ADV blackyellowmagentacyan
www.palsystem.com The PAL RTC is available through a global network of our Value Added Resellers (VAR). For further information please visit www.palsystem.com
PAL is a registered trademark of CTC Analytics AG | Switzerland
RTCNeed a more flexible Autosamplerfor your GC & GC/MS? Robotic Tool ChangeThe possibilities are endless - the limit is your imagination
Enhance your analytical capabilities
Robotic change between Liquid, Headspace and SPME methods Fully automated derivatisation prior to GC injection Use of multiple SPME fibers for method development Fits on any GC & GC/MS System
Ask for a PAL when you buy your next GC & GC/MS System!
ES339530_LCA1113_003_FP.pgs 10.17.2013 21:35 ADV blackyellowmagentacyan
LC•GC Asia Pacific November 2013
Editorial Policy:
All articles submitted to LC•GC Asia Pacific
are subject to a peer-review process in association
with the magazine’s Editorial Advisory Board.
Cover:
Original materials courtesy: Hong Li
Columns
17 LC TROUBLESHOOTING
What’s the Problem with the LLOQ? — A Case Study
John W. Dolan
Two different methods of calculating the LLOQ disagree. Which, if
either, is correct?
22 GC CONNECTIONS
Practical Gas Chromatography
John V. Hinshaw
Questions about how practical proposed gas chromatography (GC)
method changes are often come up during optimization for speed
and resolution, or while converting to a different carrier gas. Related
objective measurements such as the optimum practical carrier
gas velocity were defined more than 40 years ago. This instalment
reviews such metrics in the light of their relevance to today’s GC
challenges.
26 COLUMN WATCH
The Top 10 HPLC and UHPLC Column Myths
Ronald E. Majors
In any field, there are “misconceptions” or “myths” that arise and
are perpetuated and passed on to the next generation. These
myths are often driven by a lack of understanding of the real issues
by practitioners. In the first of a two-part feature from Ron Majors,
the top 10 high performance liquid chromatography (HPLC) column
myths are presented and attempts are made to demystify them by
offering some evidence that they are untrue. This part will feature
myths 10 to six.
Departments
33 Application Notes
COVER STORY
8 Field-Flow Fractionation for
Biological, Natural, and
Synthetic Polymers: Recent
Advances and Trends
Carmen Bria, Frédéric Violleau, and
S. Kim Ratanathanawongs Williams
A review of the latest trends in
field-flow fractionation (FFF) for
various types of polymer analysis.
November | 2013
Volume 16 Number 4
4
ES339284_LCA1113_004.pgs 10.17.2013 17:29 ADV blackyellowmagentacyan
Are Wyatt’s MALS instruments the product ofintelligent design or evolution? Yes.
© 2013 Wyatt Technology, Optilab, DAWN and the Wyatt Technology logo are registered trademarks of Wyatt Technology Corporation.
There’s no debate about it. We invented the first commercial Multi-Angle Light Scattering (MALS) detectors for GPC/SEC, then built on them to develop a complete family of related instruments, all of which provide unparalleled performance for our customers.
Our MALS products have multiplied and are the most widely-used on earth. Thousands of chemical, biotechnol-ogy, pharmaceutical, academic and government laboratories around the world rely on them to characterize proteins, polymers and macromolecules of all kinds. And many of our customers have published, resulting in nearly 10,000 peer-reviewed articles based on their work using Wyatt’s MALS instruments. These articles, application notes and other remarkable feedback in the scientific community have helped our 18+ PhD scientists and other innovators to refine our instruments more rapidly to meet our customers’ needs. Which oftentimes results in an unprecedented new product.
So, was it intelligent design or evolution that brought us Wyatt’s MALS detectors?Precisely.
DAWN® HELEOS. The most advanced multi-angle light scattering (MALS) detector for macromolecular charac-terization.
Optilab® T-rEX. The refrac-tometer with the greatest combination of sensitivityand range—and absolute refractive index, too.
ViscoStar®. The viscom-eter with unparalleled signal-to-noise ratios, stable baselines and a 21st-century interface.
Eclipse. The ultimate sys-tem for the separation of proteins and nanoparticles in solution.
DynaPro® Plate Reader. The only automated dynamic light scattering for 96 or 384 or 1536 well plates—now with an on-board camera!
ES339844_LCA1113_005_FP.pgs 10.18.2013 17:28 ADV blackyellowmagentacyan
6 LC•GC Asia Pacific November 2013
The Publishers of LC•GC Asia Pacific would like to thank the members of the Editorial Advisory Board
for their continuing support and expert advice. The high standards and editorial quality associated with
LC•GC Asia Pacific are maintained largely through the tireless efforts of these individuals.
LCGC Asia Pacific provides troubleshooting information and application solutions on all aspects
of separation science so that laboratory-based analytical chemists can enhance their practical
knowledge to gain competitive advantage. Our scientific quality and commercial objectivity provide
readers with the tools necessary to deal with real-world analysis issues, thereby increasing their
efficiency, productivity and value to their employer.
Editorial Advisory Board
Kevin AltriaGlaxoSmithKline, Harlow, Essex, UK
Daniel W. ArmstrongUniversity of Texas, Arlington, Texas, USA
Michael P. BaloghWaters Corp., Milford, Massachusetts, USA
Coral BarbasFaculty of Pharmacy, University of San
Pablo – CEU, Madrid, Spain
Brian A. BidlingmeyerAgilent Technologies, Wilmington,
Delaware, USA
Günther K. BonnInstitute of Analytical Chemistry and
Radiochemistry, University of Innsbruck,
Austria
Peter CarrDepartment of Chemistry, University
of Minnesota, Minneapolis, Minnesota, USA
Jean-Pierre ChervetAntec Leyden, Zoeterwoude, The
Netherlands
Jan H. ChristensenDepartment of Plant and Environmental
Sciences, University of Copenhagen,
Copenhagen, Denmark
Danilo CorradiniIstituto di Cromatografia del CNR, Rome,
Italy
Hernan J. CortesH.J. Cortes Consulting,
Midland, Michigan, USA
Gert DesmetTransport Modelling and Analytical
Separation Science, Vrije Universiteit,
Brussels, Belgium
John W. DolanLC Resources, Walnut Creek, California,
USA
Roy EksteenSigma-Aldrich/Supelco, Bellefonte,
Pennsylvania, USA
Anthony F. FellPharmaceutical Chemistry,
University of Bradford, Bradford, UK
Attila FelingerProfessor of Chemistry, Department of
Analytical and Environmental Chemistry,
University of Pécs, Pécs, Hungary
Francesco GasparriniDipartimento di Studi di Chimica e
Tecnologia delle Sostanze Biologica-
mente Attive, Università “La Sapienza”,
Rome, Italy
Joseph L. GlajchMomenta Pharmaceuticals, Cambridge,
Massachusetts, USA
Jun HaginakaSchool of Pharmacy and Pharmaceutical
Sciences, Mukogawa Women’s
University, Nishinomiya, Japan
Javier Hernández-BorgesDepartment of Analytical Chemistry,
Nutrition and Food Science University of
Laguna, Canary Islands, Spain
John V. HinshawServeron Corp., Hillsboro, Oregon, USA
Tuulia HyötyläinenVVT Technical Research of Finland,
Finland
Hans-Gerd JanssenVan’t Hoff Institute for the Molecular
Sciences, Amsterdam, The Netherlands
Kiyokatsu JinnoSchool of Materials Sciences, Toyohasi
University of Technology, Japan
Huba KalászSemmelweis University of Medicine,
Budapest, Hungary
Hian Kee LeeNational University of Singapore,
Singapore
Wolfgang LindnerInstitute of Analytical Chemistry,
University of Vienna, Austria
Henk LingemanFaculteit der Scheikunde, Free University,
Amsterdam, The Netherlands
Tom LynchBP Technology Centre, Pangbourne, UK
Ronald E. MajorsAgilent Technologies,
Wilmington, Delaware, USA
Phillip MarriotMonash University, School of Chemistry,
Victoria, Australia
David McCalleyDepartment of Applied Sciences,
University of West of England, Bristol, UK
Robert D. McDowallMcDowall Consulting, Bromley, Kent, UK
Mary Ellen McNallyDuPont Crop Protection,Newark,
Delaware, USA
Imre MolnárMolnar Research Institute, Berlin, Germany
Luigi MondelloDipartimento Farmaco-chimico, Facoltà
di Farmacia, Università di Messina,
Messina, Italy
Peter MyersDepartment of Chemistry,
University of Liverpool, Liverpool, UK
Janusz PawliszynDepartment of Chemistry, University of
Waterloo, Ontario, Canada
Colin PooleWayne State University, Detroit,
Michigan, USA
Fred E. RegnierDepartment of Biochemistry, Purdue
University, West Lafayette, Indiana, USA
Harald RitchieThermo Fisher Scientific, Cheshire, UK
Pat SandraResearch Institute for Chromatography,
Kortrijk, Belgium
Peter SchoenmakersDepartment of Chemical Engineering,
Universiteit van Amsterdam, Amsterdam,
The Netherlands
Robert ShellieAustralian Centre for Research on
Separation Science (ACROSS), University
of Tasmania, Hobart, Australia
Yvan Vander HeydenVrije Universiteit Brussel,
Brussels, Belgium
SUBSCRIPTIONS: LC•GC Asia Pacific is free to qualified readers in Asia Pacific. To apply for a free subscription, or to change your name or address, go to www.lcgceurope.com,
click on Subscribe, and follow the prompts.
To cancel your subscription or to order back issues, please email your request to
[email protected], putting LCGC Asia Pacific in the subject line.
Please quote your subscription number if you have it.
MANUSCRIPTS: For manuscript preparation guidelines, visit www.chromatographyonline.com or
call the Editor, +44 (0)1244 629 300. All submissions will be handled with reasonable care, but
the publisher assumes no responsibility for safety of artwork, photographs or manuscripts. Every
precaution is taken to ensure accuracy, but the publisher cannot accept responsibility for the
accuracy of information supplied herein or for any opinion expressed.
DIRECT MAIL LIST: Telephone: +44 (0)1244 629 300, Fax: +44 (0)1244 678 008.
Reprints: Reprints of all articles in this issue and past issues of this publication are available
(250 minimum). Contact Brian Kolb at Wright’s Media, 2407 Timberloch Place, The Woodlands, TX
77380. Telephone: 877-652-5295 ext. 121. Email: [email protected].
©2013 Advanstar Communications Inc. All rights reserved. No part of this publication may be re-
produced or transmitted in any form or by any means, electronic or mechanical including by photo-
copy, recording, or information storage and retrieval without permission in writing from the publisher.
Authorization to photocopy items for internal/educational or personal use, or the internal/educational
or personal use of specific clients is granted by Advanstar Communications Inc. for libraries and
other users registered with the Copyright Clearance Center, 222 Rosewood Dr. Danvers, MA 01923,
978-750-8400 fax 978-646-8700 or visit http://www.copyright.com online. For uses beyond those
listed above, please direct your written request to Permission Dept. fax 440-756-5255 or email:
[email protected]. Warning: The doing of an unauthorized act in relation to a copyright
work may result in both a civil claim for damages and criminal prosecution. LC•GC
Asia Pacific (ISSN 1471-6577) and the logo appearing on the cover of this magazine
are registered trademarks of Advanstar Communications Inc. LC•GC Asia Pacific
does not verify any claims or other information appearing in any of the advertisements
contained in the publication, and cannot take any responsibility for any losses or other
damages incurred by readers in reliance on such content.
Published by
10% Post
Consumer
Waste
Subscibe on-line at
www.chromatographyonline.com
Group Publisher
Mike Tessalone
Editorial Director
Laura Bush
Editor-in-Chief
Alasdair Matheson
Managing Editor
Kate Mosford
Assistant Editor
Bethany Degg
Sales Manager
Lindsay Jones
Sales Executive
Liz Mclean
Subscriber Customer Service
Visit (chromatographyonline.com)
to request or change a
subscription or call our customer
Service Department on
+001 218 740-6877
Bridgegate Pavilions, 4A
Chester Business Park,
Wrexham Road,
Chester, CH4 9QH
Tel. +44 (0)1244 629 300
Fax +44 (0)1244 678 008
Chief Executive Officer Joe Loggia
Chief Executive Officer Fashion Group, Executive Vice-PresidentTom Florio
Executive Vice-President, Chief Administrative Officer & Chief Financial OfficerTom Ehardt
Executive Vice-PresidentGeorgiann DeCenzo
Executive Vice-PresidentChris DeMoulin
Executive Vice-PresidentRon Wall
Executive Vice-President, Business SystemsRebecca Evangelou
Sr Vice-PresidentTracy Harris
Vice-President, Media OperationsFrancis Heid
Vice-President, LegalMichael Bernstein
Vice-President, Electronic Information TechnologyJ Vaughn
CORPORATE OFFICE641 Lexington Ave, 8th Fl. New York, New York 10022 USA
‘Like’ our page LCGC Join the LCGC LinkedIn groupFollow us @ LC_GC
ES339285_LCA1113_006.pgs 10.17.2013 17:29 ADV blackyellowmagentacyan
Sensational New Capabilities of impact HD from Single LC-Accurate Mass MS Analysis
High Defi nition 5 orders of magnitude dynamic range in one second from 50 GBit / sec sampling technology
Defi nitive trace analysis from complex, high-background matrices
Quantitation by Source Stability and HD dynamic range
Easy to use with intelligent self-optimizing MSMS routines for expert-caliber results fi rst time
Applications
Proteomics and Metabolomics Forensics and Drug screening
Food and Water Testing Synthetic Chemistry Support
impact HD™
For more information please visit www.bruker.comFor research use only. Not for use in diagnostic procedures.
LC-MS/MS
Innovation with Integrity
ES339515_LCA1113_007_FP.pgs 10.17.2013 21:34 ADV blackyellowmagentacyan
LC•GC Asia Pacià c November 20138
Macromolecules are ubiquitous in many areas of science and
technology. Depending on the macromolecule, it is important
to analyze properties like size, molar mass (MM), chemical
composition, degree of branching, and their respective
distributions to understand their behaviour. However, because
of the complex nature of polymers, current separation
techniques are not always capable of comprehensive analyses.
Size-exclusion chromatography (SEC) is widely regarded as the
workhorse for polymer characterization, but is limited by high
molar mass (HMM) macromolecules, weakly bound complexes
and aggregate species, and highly branched polymers.
Field-flow fractionation (FFF) is a versatile family of techniques
that complements SEC with additional separation capabilities
based on analyte size, mass, composition, or architecture
depending on the field used (Figure 1).
The open channel FFF design results in a soft separation
mechanism that is well suited for analysis of high and ultrahigh
MM polymers and samples containing microgel. Some key
advantages of FFF over SEC arises from its ability to separate
analytes over a broad size range (0.001 to 100 µm) using a
single channel, and the absence of column packing, which
greatly reduces shear degradation. SEC of protein aggregates
often requires the addition of cosolvents or preconditioning
of columns to reduce adsorption (1). However, addition of
cosolvents may induce aggregation, dissociate aggregates,
or cause sample specific adsorption (Figure 2) (2).
Preconditioning columns is often practised but not reported
in the literature, and even when preconditioning is used
poor recoveries and sample specific adsorption have been
observed (3). In FFF the ability to use formulation buffer allows
separations and measurements under solution conditions that
are more representative of actual use. For polymer analysis,
the shear degradation and co-elution of small and large
analytes observed in SEC for highly branched polymers
are attributed to effects caused by the column packing
material (4).
In practice, FFF offers users additional benefits. Prior to
SEC, filtering is often implemented as a sample preparation
step to remove large components and help prolong the life
of the column. Sample filtering has been shown to remove
soluble and insoluble microgels leading to erroneous MM
and polydispersity results (Figure 3) (5). Filtering is not
required in FFF and soluble polymers and microgels can
be simultaneously characterized. Many syntheses require
the addition of excess reagents, which may interfere with
subsequent product analyses. Such reagents or interfering low
MM sample components either elute in the void peak or can be
removed on-line through a semi-permeable membrane used in
some FFF techniques.
Separations in FFF are dependent on the strength of
an externally applied field which can be easily adjusted.
Therefore, resolution and separation speed are readily
controlled without the need to change channels. In addition,
the open channel design greatly reduces the chance of
contamination and inexpensive membranes can be replaced
when contaminated. Finally, FFF is easily coupled on-line with
detectors frequently used for SEC analysis, including multi-
angle light scattering (MALS), differential refractive index (dRI),
and mass spectrometry (MS) detectors. For those interested
in FFF, building a simple homemade system requires a FFF
channel and standard high performance liquid chromatography
(HPLC) components common to many laboratories. The recent
advances in FFF over the last three years are highlighted in this
review.
Field-Flow Fractionation for Biological, Natural, and Synthetic Polymers: Recent Advances and Trends
Carmen Bria1, Frédéric Violleau2, and S. Kim
Ratanathanawongs Williams1, 1Laboratory for Advanced
Separations Technologies, Department of Chemistry and
Geochemistry, Colorado School of Mines, Golden, Colorado,
USA, 2Université de Toulouse, INPT, Ecole d’Ingénieurs
de Purpan, Département de Sciences Agronomiques et
Agroalimentaires, Toulouse Cedex, France.
Field-fl ow fractionation (FFF) is a family of techniques that is increasingly used for separating and characterizing macromolecules. This review discusses recent advances in the characterization of biological, natural, and synthetic polymers. Applications of FFF are contrasted with size-exclusion chromatography to illustrate practical considerations when characterizing macromolecules. The use of different FFF à elds allows separations based on size, mass, composition, and architecture. The open channel design and subsequent low shear rate is well suited for analyzing weakly bound complexes, highly branched polymers, high molar mass analytes, and aggregates. Other beneà ts of FFF that are highlighted in this paper include simplià ed sample preparation, fl exibility in carrier fl uid choice, and on-line removal of low-molecular-weight contaminants.
ES339974_LCA1113_008.pgs 10.18.2013 20:05 ADV blackyellowmagentacyan
Contact us for more information
www.postnova.com
Groundbreaking new Materials meet pioneering Analytical Methods
Only the modular Postnova Field-Flow Fractionation Platform provides the perfect
matching FFF-instrument for many applications and can be extended in the future.
· More fi ndings: gain molar mass, confi rmation and branching by FFF-MALS
· High recovery: no unpredictable sample-matrix interactions on a column
· Signifi cantly less consumables: reduces operating expenses
ES339516_LCA1113_009_FP.pgs 10.17.2013 21:34 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201310
Williams et al.
Principles of FFFSeparation takes place in a thin, open, ribbon-like channel
where carrier fluid transports components down the separation
axis of the channel. Frictional drag at the channel walls creates
a parabolic flow profile across the channel thickness, w, with
the fastest flows in the middle of the channel and the slowest
flows near the walls (Figure 1). An external field (flow, thermal, or
sedimentation) is applied perpendicular to the separation axis
of the channel to drive components towards the accumulation
wall. This field-induced transport is counteracted by diffusion
of components away from the high concentration region near
the accumulation wall. Equilibrium is reached when the two
transport processes are balanced and there is no net flux of
sample in either direction. The equilibrium position is different for
each sample component depending on the magnitude of their
interaction with the applied field and their diffusion coefficient.
Components of smaller sizes diffuse further into the channel
than larger components based on the inverse relationship
between diffusion coefficient (D) and hydrodynamic diameter
(dh) given by the Stokes-Einstein equation for spherical analytes.
The smaller components experience the faster flows further
from the accumulation wall and therefore elute before larger
components in normal mode FFF. This normal mode elution
order is the reverse of that observed for SEC.
The main FFF techniques relevant to this review are
asymmetrical flow field-flow fractionation (AF4), hollow fibre flow
field-flow fractionation (HF5), and thermal field-flow fractionation
(ThFFF). AF4 utilizes a single permeable wall that allows a
crossflow to act as the perpendicular field (Figure 1[a]). The
permeable wall is composed of a porous frit covered with a
semipermeable membrane, the latter acting as the sample
accumulation wall. The retention time (tr) for AF4 is given in
equation 1:
w 2πηt 0v
c
tr=
2V0kTAF4d
h
.
[1]
where η is the carrier fluid viscosity, t0 is the void time, V.
c is the
crossflow rate, V0 is the void volume, k is Boltzmann’s constant,
and T is the temperature.
In HF5 a semipermeable hollow fibre membrane is used and
an outward radial flow acts as the perpendicular force (Figure
1[b]). The benefits of HF5 over AF4 are lower sample volumes
and a potentially disposable channel. Thermal FFF (ThFFF)
employs a hot and cold wall to create a temperature difference
(∆T) that subsequently induces thermal diffusion of components
towards the cold wall (in most cases) (Figure 1[c]). The retention
time is given in equation 2
DTΔπηt 0
tr= =
2kTThFFFd
h
DT Tt0
6D
Δ [2]
where DT is the thermal diffusion coefficient.
Biopolymers Biopolymers are a diverse class of macromolecules that
includes polypeptides, polynucleotides, and polysaccharides.
The versatility of AF4 separation and the characterization of
biopolymers is well established. Several review papers and a
book focusing on the analysis of biological polymers using FFF
have recently been published (6–8).
(a)
(b)
(c)
Separation axis
D
DT
D
W
Cross fow
Outward radial fow
Semi-permeablemembrane
Hot wall
Cold wall(Accumulation wall)
Porous frit
Semi-permeablemembrane
(Accumulation wall)
∆T
Figure 1: Types of FFF separation. (a) In AF4 a crossflow
passes through a semi-permeable membrane and porous frit. (b)
In HF5 a cylindrical semipermeable membrane is used and a
radial outward flow creates the perpendicular field. (c) In ThFFF
a temperature gradient (∆T) is formed between a hot wall and a
cold wall, and sample migrates towards the cold wall because of
thermal diffusion (DT).
(a) (b)
107 106
105
104
106
10 12 14 16 18 20
0.010
0.008
0.006
0.004
0.002
0.000
105
104
0 2 4 6 8 10 12 14 16 18 20 220.00
De
tect
or
vo
lta
ge
(V
)
De
tect
or
vo
lta
ge
(V
)
Mo
lecu
lar
we
igh
t (D
alt
on
s)
Mo
lecu
lar
we
igh
t (D
alt
on
s)
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18 0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
Elution time (min)
0 2 4 6 8 10 12 14 16 18 20 22
Elution time (min)
Figure 2: An IgG1 recombinant fully humanized monoclonal
antibody was analyzed by FFF in two different carrier fluids: (a)
0.1% acetic acid containing 50 mM magnesium chloride and (b)
10 mM phosphate buffer pH 7.1. High molar mass aggregates
(peak at ~18.5 min) present in (a) are absent in (b) as a result
of weak aggregate interactions stabilized by the magnesium
chloride. Adapted and reproduced with permission from B.
Demeule, M.J. Lawrence, A.F. Drake, R. Gurny, and T. Arvinte,
(2007), Biochim. Biophys. Acta-Proteins. Proteomics 1774,
146-153. © Elsevier.
ES339963_LCA1113_010.pgs 10.18.2013 20:04 ADV blackyellowmagentacyan
11www.chromatographyonline.com
Williams et al.
Characterizing protein—protein and
protein—macromolecule complexes
is important for understanding the
efficacy and functions of proteins.
AF4’s gentle separation mechanism is
well suited to analyze complexes with
weak interactions (Figure 4) (8). Current
techniques for characterizing protein
dissociation constants (Kd), such as
surface plasmon resonance (SPR),
analytical ultracentrifugation (AUC),
and SEC, are limited in their ability to
analyze more than two components
or to detect weak binding affinities
(Kd > µM). Protein—protein binding
between a neonatal Fc receptor (FcRn),
immunoglobulin (IgG), and human serum
albumin (HSA) was recently studied by
AF4 (9). FcRn is involved in removing
IgG proteins from lysosomal degradation
pathways and IgG transportation in the
body. AF4 separation of the IgG-FcRn
complex allowed for the determination
of a relatively low binding affinity (Kd
of 3.74 μM). In addition, FcRn, HSA,
IgG, and their associated complexes
were separated using AF4. By using an
internal standard curve, the formation
of multi-protein complexes were
determined, including a previously
unreported protein complex (HSA/FcRn/
IgG/FcRn, 303 kDa). The separation of
intact, weakly bound protein complexes
shows great promise for AF4 studies
of protein pharmacokinetics and
aggregation kinetics. Analysis of protein
aggregates, especially those in the
submicron size range, is of particular
interest in the development of therapeutic
proteins. Development of an AF4 method
for separating IgG monomer and
submicron IgG aggregates was recently
shown by Hawe et al. (3). Better size
resolution and recoveries of submicron
IgG aggregates were achieved by AF4
compared to SEC.
Lipoproteins are assemblies of
proteins and lipids that function as
carriers for lipids and cholesterols in
blood. AF4 has been used to analyze
low-density lipoproteins (LDL) and
high-density lipoproteins (HDL) (10).
LDLs have been associated with
an increased risk of coronary artery
disease (CAD). In addition to the
conventional AF4 channel, a hollow
fibre guard channel placed before
the AF4 channel was evaluated using
serum from healthy patients and
CAD patients. The guard channel
removed contaminants and improved
reproducibility in retention, and
fluorescence detection reduced
adsorption of serum proteins to the
membrane and reduced the amount
of serum required for each injection
(0.13 μL).
On-line coupling with a variety of
detection methods has expanded the
breadth of AF4’s characterization ability
in recent years. The use of MALS, dRI,
and quasi-elastic light scattering (QELS)
detectors has become more common
for characterizing macromolecules.
Characterization of dh, MM, radius of
gyration (rg), and chemical composition
was shown for a PEGylated protein
conjugate and its aggregates using
on-line AF4–UV-MALS–QELS–dRI,
SEC–UV-MALS–QELS–dRI, and
matrix-assisted laser-desorption/
ionization time-of-flight mass
spectrometry (MALDI–TOF -MS) as
complementary techniques (11).
PEGylated protein, unreacted protein
traces, and aggregated species were
detected by both AF4 and SEC, with
AF4 providing superior size resolution,
while MALDI–TOF-MS was unable to
detect aggregates. Detection by UV–
MALS-QELS–RI enabled chemical
composition characterization of
PEGylated proteins (1/1 PEG to protein
ratio) and allowed identification of
aggregates present using different
storage buffers.
Interest in characterizing protein
complexes by 2D off-line coupling of
AF4 with other separation techniques
and a variety of detection methods has
grown in recent years. Lectin-treated
N-linked glycopeptides in serum from
lung cancer and healthy patients were
separated by AF4 and subsequently
analyzed by nanoflow liquid
chromatography-electrospray ionization–
tandem mass spectrometry (nLC–
ESI-MS–MS) (12). Binding of various lectin
types to glycoproteins enabled a size-
based separation by AF4. Removal of
non-lectin bound glycopeptides and size
sorting of lectin-glycopeptide complexes
during AF4 allowed semi-quantitative
analysis and improved identification
of biomarkers by nLC–ESI–MS–MS.
Similarly, characterizing cholesterols and
triglycerides in lipoprotein complexes is
also important for understanding their
function in the body. Off-line coupling of
AF4 and gas chromatography (GC)–
Universal HPIC Analysis
less time, better results
Continuous high-pressure operation for capillary and analytical scale—true HPIC™ analy-
sis is here. The high-pressure capability of the new Thermo Scientif c™ Dionex™
ICS-5000+ HPIC system enables fast separations and high resolution with the new 4
µm particle IC columns. This Reagent-Free™ HPIC system simplifi es your analysis and
increases reproducibility by removing the inconsistencies of manually prepared eluents.
© 2
013 T
herm
o F
isher
Scie
ntifi c
Inc. A
ll rights
reserv
ed
. A
ll tr
ad
em
ark
s
are
the p
rop
ert
y of Therm
o F
isher
Scie
ntifi c
Inc. and
its s
ub
sid
iaries.
Forty inorganic ions and organic acids resolved in 40 minutes
WARRANT Y
IC
and Sample Prep
3 Year
• Learn about the new Dionex ICS-5000+ at thermoscientifi c.com/ics5000+
ES339973_LCA1113_011.pgs 10.18.2013 20:05 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201312
Williams et al.
improved ionization of the HDL and LDL lipids, and in a CAD
plasma sample 28 phospholipids, 18 triacylglycerides, and
six cholesteryl esters were identified.
Natural Polymers Starches are macromolecules essential to human beings and
are used in a variety of industrial and food applications. The
properties of starches (for example, digestion and thickening
abilities) are dependent on starch structure, which in turn is
dependent on the degree of branching. However, wide size
distributions and variation in branching makes characterizing
starches difficult (20). AF4 coupled with MALS and dRI
detectors has been successfully used to determine starch dh,
MM, and rg (21,22). An in-depth review of FFF characterizing
food macromolecules has been recently published (23).
Wahlund et al. demonstrated the power of AF4–MALS–RI to
rapidly separate amylose and amylopectin in maize, wheat, rice,
potato, and tapioca starches (24). Qualitative results for amylose
and amylopectin ratios demonstrate the feasibility for relatively
fast characterization of starches by AF4 and provide a starting
point for more extensive starch analyses. Studies by Juna et
al. have examined various starches (waxy maize, tapioca,
corn, sago) to better understand AF4 conditions and starch
processing parameters (25–30). Changes in size distributions
were observed with changes in AF4 conditions. For example,
at high cross flow rates, the dh, MM, and rg distributions of
tapioca, sago, and corn starch shifted to lower values because
of increased retention (or potential degradation of HMM
components). The effect of AF4 conditions is therefore important
and must be considered for accurate analyses of starches.
Coupling a separation technique with MALS and dRI
detectors can provide information on structural and branching
characteristics of starches. SEC is the most common separation
technique used to characterize starches, but low exclusion limits
and shear scission may bias results. AF4 has the potential to
reduce artifacts observed in SEC such as changes in MM and
size distributions as a result of shear degradation or aggregation
and large branched polymers that co-elute with smaller
components (4,20). A more in-depth comparison of AF4 and
SEC as separation techniques for starches is available (31). To
characterize size distributions and gain structural information
for a commercial starch and a waxy yam starch, Perez et al.
compared AF4–MALS–dRI and SEC–MALS–dRI (32). AF4 and
MS allowed cholesterols and triglycerides to be profiled from
human serum samples, and results showed agreement with
the current enzymatic determination methods (13).
The use of FFF has shown promise as a pre-MS separation
technique in proteomics analyses. To improve MS detection of
poorly soluble proteins, the effect of protein–SDS complexation
on protein solubility was examined by HF5 and nLC–MS (14).
SDS-denatured serum samples, unfractionated or fractionated,
showed improved solubility of the SDS–protein complexes,
and allowed for a greater number of proteins to be identified by
nLC–MS. Furthermore, the HF5 process was shown to remove
low MM (< 30 kDa) components, which subsequently led to
lowered background noise in the MS spectrum.
Protein phosphorylation is a post-translational modification
which plays an important role in protein regulation and can
be used as a biomarker for diseases like cancer. The 2D
on-line coupling of an isoelectric focusing (IEF) step and
AF4 step prior to nLC–ESI–MS–MS enabled separation of
phosphorylated proteins from a proteome sample based
on isoelectric point (pI) and dh (15). IEF-AF4 separation
was evaluated for unphosphorylated and phosphorylated
α-casein. Peptides with higher degrees of phosphorylation
eluted in the lower pH channels and at longer AF4 retention
times as expected. Relative abundances of phosphorylated
protein biomarkers were determined by IEF–AF4 and nLC–
ESI–MS-MS for a prostatic cancer line and a normal cell line.
In another study, improvements in direct on-line coupling of
AF4 with ESI–MS were shown in a small chip-type channel
for top-down proteomics that operates in the micro-flow rate
regime (Figure 5[a]) (16). The chip-type channel effectively
separated carbonic anhydrase (29 kDa) and transferrin (78
kDa) while using much lower, and more ESI–MS compatible,
channel flow rates (<12 µL/min) than previous on-line studies
(Figure 5[b]) (17,18). Resolution of monomer and aggregate
species as well as desalting during AF4 led to higher
signal-to-noise for ESI–MS detection (Figure 5[c] and 5[d]).
Lipodomic analysis of HDL and LDL from human serum
was also shown by chip-type AF4 (19). On-line desalting
0.8
Unfltered
Filtered0.6
0.4
0.2
0.0
0
Vl, mL
2
Vo
4 6 8 10
10
100
rrm
s, n
m
dR
I, V
1
Figure 3: ThFFF–MALS–dRI analysis of unfiltered (solid line,
black symbols) and 0.5-µm filtered (dashed line, grey symbols)
microgel-containing poly(vinyl acetate). The lines and symbols
represent the dRI fractograms and rg, respectively. Significant
polymer loss in the filtered sample is evident in the lower MM
distribution. Adapted and reproduced with permission from
D. Lee and S.K.R. Williams, (2010), J. Chromatogr. A. 1217,
1667-1673. © Elsevier.
SECbindingaffnity
(Kd) 10-10M
lgG-antigen lgG-Fcγ RI7
lgG-FcRn-HSA
Proteasebinding
Lower affnityaggregates8,9
Higgher affnityaggregates9
lgG-Fcγ RII/lll7
lgG-FcRnFc-FcRn FcRn-HSA10-9M 10-8M 10-7M 10-6M 10-5M 10-4M 10-3M
FFFSPR
AUC10
Figure 4: A comparison of FFF to other currently used techniques
(analytical ultracentrifugation [AUC]; surface plasmon resonance
[SPR]) for protein–protein characterization. The open channel
FFF design and flow-based separation extends the current ability
to detect weak protein–protein interactions into the µM binding
affinity range. Adapted and reproduced with permission from
J. Pollastrini, T.M. Dillon, P. Bondarenko, and R.Y.T. Chou,
(2011), Anal. Biochem. 414, 88–98. © Elsevier.
ES339964_LCA1113_012.pgs 10.18.2013 20:04 ADV blackyellowmagentacyan
13www.chromatographyonline.com
Williams et al.
the physiological effects of soluble fibre. Several recent studies
have used AF4–MALS–dRI to analyze β-glucans (36–38). In
one of the studies, β-glucan aggregates under gastric digestion
conditions were disrupted while, after undergoing small
intestinal digestion, aggregates were reformed (Figure 6) (36).
The disruption and re-formation of aggregates is likely to impact
the behaviour and function of β-glucan. To demonstrate the
effect of processing and storage on aggregates, Ulmius et al.
subjected barley β-glucan samples to several conditions (such
as storage time, heating, freeze time, freeze-thaw, and change
in solution conditions) and performed AF4–MALS–dRI analysis
(37). Disruption, structural change, or elimination of β-glucan
aggregates was observed under most conditions. Properties of
individual and aggregated β-glucans from oat and barley were
also compared using AF4–MALS–dRI (38). Individual molecules
could be distinguished from supramolecular species based
on conformational differences across the size distribution. In
addition, dissolution of both β-glucans under harsh alkaline
conditions showed that barley β-glucan aggregates were not
dissolved as previously proposed.
AF4 has been applied to hyaluronan (HA) and sodium
hyaluronate (NaHA) polysaccharides, which have
important biological functions and industrial applications
(39). Characterization of HA MM and conformation by
AF4-MALS-dRI yielded results that were consistent with other
methods, including SEC–MALS–dRI. Both AF4 and SEC were
able to measure low MM (<1 × 106) samples. Molar mass
distributions, an important parameter for HA characterization,
were also similar between AF4 and SEC measurements.
NaHA is used commercially in pharmaceutical and cosmetic
products (40). Molar mass distributions and structural properties
of NaHA and commercially blended NaHA mixtures were
characterized and compared by frit inlet (FI) AF4–MALS–dRI.
Frit inlet is a particularly gentle FFF method without the initial
focusing step. Significant aggregation was not observed
SEC results both yielded smaller sizes for the commercial starch
than the waxy yam starch, while a more quantitative recovery
for AF4 (100%) was seen compared to SEC (62%). Structural
characterization of the starches was also accomplished by
SEC and AF4. Plotting the rg and MM of the same fraction, and
using the exponent, νg, from the equation rgi = KgMiνg where
Kg is a constant, the polymer shape can be described (νg of
0.3, 0.5–0.6, and 1 describe the polymer shape for a sphere,
a linear random coil, and a rod, respectively). Values for AF4
and SEC were all close to 0.4, which fell between a sphere and
a random coil. Rolland-Sabaté et al. examined the differences
between hydrodynamic chromatography (HDC)–SEC and
AF4 for characterizing starches (33). Better separation of
amylose and amylopectin was achieved with AF4 and allowed
determination of dh and MM distributions and better structural
characterization (especially for large amylopectin fractions). In
addition, the branching parameter distributions showed that
WTPS and WTRS amylopectins could be discerned by AF4, but
not HDC–SEC.
Characterizing aggregates is important for understanding the
solution behaviour and physical properties of polysaccharides.
Arabinoxylan and its aggregates were characterized by AF4–
MALS–dRI and SEC–MALS–dRI (34). Although aggregate
concentrations were low, co-elution of individual polymers and
aggregates in SEC led to larger molar masses and rg’s reported
compared to AF4. The MM, size, and conformation of dextrans
with varying amounts of α(1-3) glycosidic linkages has also
been investigated (35). Using νg values, dextrans containing
the most α(1-3) linkages were found to be the smallest and
densest while dextrans with the least amount displayed a
quasi-linear conformation. Understanding the physical and
structural properties of glucan allows for further development of
biomaterials.
β-glucan’s solution behaviour and ability to form aggregates
may be associated with beneficial health effects. For example,
understanding β-glucan digestion can aid in comprehending
1010 0.8
Mo
lar
mass
(g
/mo
l) 0.6
0.4
0.2
0.0
109
108
107
106
105
104
0 1 2 3Elution time (min)
Flu
ore
scen
ce-s
ign
al (V
)
4
Water
Gastric digestion
Small intestinal digestion
5 6to
Figure 6: AF4 fractograms of barley β-glucan (lines represent
fluorescence and symbols represent molar mass). Samples
dispersed in water (grey-dashed line, circles), after in vitro gastric
digestion (grey full line, squares), and undergoing additional
small intestinal digestion (black line, triangles) were analyzed
by AF4–MALS. Gastric digestion samples show a reduction in
aggregate species, while the re-formation of higher density is
shown after small intestinal digestion. Adapted and
reproduced with permission from M. Ulmius, S. Adapa, G.
Onning, and L. Nilsson,(2012), Food Chem. 130, 536–540.
Split
#1
#2
#3
transferrin
dimer
0 3 6 9Time (min)
15 1812
CA
ACN+acid
emitter
Rela
tive in
ten
sity
Rela
tive in
ten
sity
Rela
tive in
ten
sity
Crossfowout
MS
(a)
(c)
(b)
(d)
Pump
Ion count:
Ion count:1.4E6
1.4E6
+461696.6
+451734.4
a. AF4-ESI-MS of #2
Mr=78,008t
r=9.6~9.9 min
+451626.0
500 1000 1500 2000
b. Direct ESI-MS of transferrin
m/z2500 3000
500 1000 1500 2000m/z
2500 3000
+421857.8
+392001.4
+352229.8
Figure 5: (a) A schematic of the chip-type miniaturized AF4
channel interfaced with electrospray ionization mass spectrometry
(AF4–ESI–MS); (b) Base peak fractogram (BPF) of AF4–ESI–MS
for the separation of CA and transferrin (V. out/V
. c = 0.012/0.49 mL/
min; (c) Full scan ESI–MS for peak #2 (transferrin) after AF4 shown
in (b); (d) Full scan ESI–MS spectrum of transferrin (0.01 μg/
μL) without AF4. Considerably better S/N is observed in the
fractionated transferrin as a result of monomer/dimer resolution
and contaminant removal during AF4. Adapted and reproduced
with permission from K.H. Kim and M.H. Moon, (2011), Anal.
Chem. 83, 8652–8658. © American Chemical Society.
ES339971_LCA1113_013.pgs 10.18.2013 20:05 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201314
Williams et al.
degrees of branching and HMMs were analyzed. As a result
of the co-elution of large and small macromolecules in SEC, a
correct calculation of the MM distribution and the MM average
or branching ratio was not possible (Figure 7). In contrast, AF4
allows the precise determination of the MM distribution, the
MM averages, and the degree of branching because the MM
versus elution volume curve and the conformation plot were
not affected by the co-elution issues encountered in the SEC
analysis.
In addition, because of the absence of significant shear
degradation in the channel, characterization of linear and
branched HMM polyethylene by AF4 has been developed
under high temperature (HT) conditions (145 ºC) in organic
solvent (1,2,4-trichlorobenzene [TCB]) (46). Compared to
HT-SEC, HT–AF4 allows for a more complete separation of
highly branched polyethylene with limited co-elution of large
and small macromolecules. The HT–AF4 technique coupled
with MALS detection was used for quantification and size
determination of the co-eluting molecules. Furthermore, HT–AF4
induced lower shear and thermo-oxidative degradation of HMM
PE and PP than HT–SEC (47). As a consequence, the HMM
averages obtained from HT–AF4 are significantly higher than
those obtained from HT–SEC. It was shown that most of the
observed limitations of SEC could be overcome by using AF4.
AF4 has also been applied to dendritic polymer
characterization. Different poly(amidoamine) (PAMAM)
dendrimers have been characterized by AF4–MALS (48). The
separation between different generations (4 to 9) of PAMAM
particles has been shown under different pH conditions and
AF4 highlighted the presence of some impurities. Coupled with
other on-line characterization techniques (for example, MALS
or a differential viscometer), AF4 allows for a more detailed
physical characterization of each separated size fraction.
Aggregation and complexation of dendritic glycopolymers
used as drug delivery systems has been demonstrated using
AF4–MALS (49–51). In addition, removal of small sample
components through the ultrafiltration membrane during
AF4 can be used to quantitatively determine the amount of
complexed small guest dye molecules in core–shell polymers.
This feature of AF4 can potentially be used for the separation
and quantification of drugs encapsulated in polymers and
makes the AF4 technique very promising for the analysis of drug
delivery systems.
Other types of drug delivery systems such as micelles have
been characterized by AF4. Poly(ethyleneoxide-b-ε-caprolacto
ne) (PEO-b-PCL) self-assemblies in water were characterized by
AF4 with on-line MALS–dRI–UV–vis–QELS detection (52). This
study underlined the impact of the mass of the PEO and PCL
fragments on the micelle size. Hydrodynamic radii measured
by QELS were in good agreement with values calculated by
AF4 retention times. AF4 illustrated that in some instances the
number of self-assemblies present was very low compared
to the number of unassembled diblock copolymers. Finally,
quantification of photosensitizers used in photodynamic therapy
encapsulated by these micelles has been performed. This
approach was used to characterize several diblock copolymer
micelles (PEG-PVP, PEG-PLA, PEG-PLGA, and PEG-PCL)
and determine their in vitro half-lives in human serum (53). The
impact of human serum on the micelle size and stability was
shown by AF4. Indeed, micelle disassembly was observed for
PEG-PVP micelles, while PEG-PLA, PEG-PLGA, and PEG-PCL
micelles were far more stable.
while samples subjected to gamma ray sterilization showed a
significant breakdown of NaHA. Exudate gums are complex
polysaccharides with industrial applications. They are used
as emulsifiers and stabilizers and contain a small amount of
proteinaceous material. Molar mass, rg, dh, conformation,
apparent densities, and distribution of proteinaceous material
were determined for gum arabic (GA) and mesquite gum (MG)
by AF4–MALS–dRI (41). The separation of polysaccharide and
proteinaceous populations and the characterization of important
molecular data over the entire size range were demonstrated by
AF4. Using AF4, it was possible to conclude that GA-stabilized
emulsions were more stable against coalescence than
MG-stabilized emulsions.
The characterization of gelatine by AF4 has also been
demonstrated (42). In denatured native gelatine an increase
in MM during renaturation was attributed to α-, β-, and
γ-chain interactions. However, an increase in MM for thermally
pre-treated gelatine was not seen, indicating an inhibition of α-,
β-, and γ-chains in gelatine and therefore limiting renaturation.
The effect of available lysine (lysine with a hydrogen-bonding
amino group) on the formation of HMM compounds in gelatine
was also characterized by AF4 (43). A decrease in available
lysine with thermal treatment led to higher MMs.
Tannins play an important role in the colour, taste, and overall
quality of wine. Oxidized tannins formed macromolecules
and were characterized by AF4–MALS, showing soluble
and insoluble populations (44). Both AF4 and small-angle
X-ray scattering (SAXS) showed that the MM of insoluble
macromolecules was much higher than the soluble
macromolecules.
Synthetic Polymers In recent years, advances in the characterization of synthetic
polymers have included the introduction of an elevated
temperature AF4 instrument and new applications in AF4
and ThFFF. Low MM polyethylene samples and a number of
narrowly distributed polystyrene standards were analyzed
by AF4–MALS–dRI in organic solvent and compared to
SEC–MALS–dRI (45). At ambient temperature, low-density
polyethylene, polypropylene, and polybutadiene containing high
1011
1010
109
108
107
106
105
104
1011
1010
109
108
107
106
105
104
10 20 30
Retention time (min) Retention time (min)
MW
CSTR LDPE 1M
W CSTR LDPE 2
MW
CSTR LDPE 1
HT-AF4 HT-SEC(a) (b)
MW
CSTR LDPE 2
Mo
lar
ma
ss (
g/m
ol)
Mo
lar
ma
ss (
g/m
ol)
40 50 10 20 30 40 50
Figure 7: Separation of a low density polyethylene sample
(CSTR-LDPE 1) by (a) HT–AF4 and (b) HT–SEC, with MALS and
dRI detection. The abnormal curvature of the molar mass from
HT–SEC indicates co-elution as a result of the high branching
of the polymers. AF4 shows complete separation over the entire
size range into the ultrahigh molar mass range not detected by
SEC because of shear degradation. Adapted and reproducd
with permission from T. Otte, H. Pasch, T. Macko, R. Brull,
F.J. Stadler, J. Kaschta, F. Becker, and M. Buback, (2011), J.
Chromatogr. A. 1218, 4257–4267. © Elsevier.
ES339969_LCA1113_014.pgs 10.18.2013 20:05 ADV blackyellowmagentacyan
15www.chromatographyonline.com
Williams et al.
is proportional to DT/D. If D can be measured independently,
that is, by QELS, DT can be calculated. When on-line D
measurements are made, DT can be calculated as a function
of tr and subsequently correlated with polymer composition.
Using this premise, the DT was found to be independent of
MM for copolymers with similar compositions and dependent
on composition of copolymers with similar MM in a non-
selective solvent. The ThFFF–MALS–dRI–QELS combination
allowed rapid determination of copolymer MM and chemical
composition distributions. ThFFF has recently been coupled
to NMR off-line (61) and on-line (62) in the analysis of triblock
copolymers and PS, poly(methyl methacrylate) (PMMA),
polyisoprene (PI), and PS-b-PMMA block copolymers,
respectively. NMR provided an independent measurement
of copolymer composition and confirmed compositional
separation by ThFFF.
To date, ThFFF method development has been predominantly
through trial-and-error based on other published work. A
recent paper demonstrated that a theoretical approach
based on temperature-dependent osmotic pressure gradient
and polymer–solvent interaction parameters can be used
to successfully estimate DT and retention times for different
polymer–solvent pairs (57). Experiments confirmed the
calculation of poly(n-butyl acrylate) (PBA), poly(methyl acrylate)
(PMA), and PS retention times in different solvents. This provides
a potential route to predicting good solvents for polymer
retention.
Thermal diffusion is an intriguing phenomenon with hidden
potential for other important analyses. A recent development
has shown that the correlation between theoretical and
experimental DT values can provide information about the
number of chain ends for branched polymers (63). The
uniqueness of this study lies in the fact that the chain ends can
be determined without the need for a linear polymer analogue.
The ThFFF–MALS–dRI–QELS combination allows simultaneous
determination of MM, composition, and number of chain ends.
ConclusionsFFF is a versatile family of techniques for characterizing
biological, natural, and synthetic macromolecules. As
a complementary technique to SEC, more detailed
macromolecule characterizations are possible using both
FFF and SEC. The open channel design and soft separation
mechanism of FFF make it a powerful technique for analyzing
weak macromolecule interactions, polymer aggregates, and
HMM and highly branched polymers. The benefits to users are
also evident in simplified sample preparations, ultrafiltration of
contaminants during separation, and flexibility in carrier fluid
choice among others. A conference dedicated to this subject
— The 16th International Symposium on Field- and Flow-Based
Separations (FFF2013) — was held in Pau, France in July, and
FFF2014 will be held in Salt Lake City, Utah, USA in October
next year. Further interesting developments are anticipated,
along with a flurry of associated publications.
AcknowledgementsCB and SKRW thank the National Science Foundation
CHE-1013029 for financial support.
References1. T. Arakawa, D. Ejima, T.S. Li, and J.S. Phil, J. Pharm. Sci. 99(4),
1674–1692 (2010).
ThFFF has been mainly used to fractionate and characterize
lipophilic polymers in organic solvents. The applied force is a
temperature gradient that causes thermal diffusion of analytes.
The magnitude of the thermal diffusion coefficient DT has been
empirically observed to depend on the polymer—solvent
interface and other factors (54–56). Thermal diffusion in liquids
is a complex phenomenon that is not yet fully understood (57–
60). However, its usefulness in ThFFF polymer separations has
been demonstrated and new interesting capabilities are being
developed. For example, the observation that different polymer
chemistries in the same solvent or the same polymer chemistry
in different solvents can have different DT and hence tr (see
Equation 2) allows for chemical composition (in addition to size)
analyses of polymers.
ThFFF coupled with MALS–dRI–QELS was used
to simultaneously determine the MM and composition
of polystyrene–poly(n-butyl acrylate) (PS-PBA) and
polystyrene-poly(methyl acrylate) (PS-PMA) copolymers (Figure
8[a] and 8[b]) (56). Equation 2 shows that the retention time
100(a)
(b)
Th
FFF P
S w
eig
ht
perc
en
tTh
FFF P
S w
eig
ht
perc
en
t
100
80
80
y=.99x +.17
R2=.98
y=1.0x +1.21
R2=.98
60
60
Nominal PS weight percent in PS-PBA
40
40
20
100
80
60
40
20
20
1008060
Nominal PS weight percent in PS-PMA
4020
Figure 8: Weight percent composition of (a) PS-PBA and (b)
PS-PMA copolymers were determined through averaged on-line
DT measurements. ThFFF weight percent values are consistent
with the nominal weight percent values. Adapted and reprinted
with permission from J.R. Runyon and S.K.R. Williams, (2011), J.
Chromatogr. A. 1218, 6774 –6779. © Elsevier.
ES339970_LCA1113_015.pgs 10.18.2013 20:05 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201316
Williams et al.
2. B. Demeule, M.J. Lawrence, A.F. Drake, R. Gurny, and T. Arvinte,
Biochim. Biophys. Acta-Proteins. Proteomics 1774(1), 146–153
(2007).
3. A. Hawe, S. Romeijn, V. Filipe, and W. Jiskoot, J. Pharm. Sci.
101(11), 4129–4139 (2012).
4. S. Podzimek, P. Lebedat, and C. Johann, LCGC North America
27(1), 62–69 (2009).
5. D. Lee and S.K.R. Williams, J. Chromatogr. A. 1217(10), 1667–1673
(2010).
6. R.N. Qureshi and W.T. Kok, Anal. Bioanal. Chem. 399(4), 1401–1411
(2011).
7. G. Yohannes, M. Jussila, K. Hartonen, and M.L. Riekkola, J.
Chromatogr. A. 1218(27), 4104–4116 (2011).
8. S.K.R. Williams and K.D. Caldwell, Eds., Field-Flow Fractionation in
Biopolymer Analysis (Springer Verlag, Wien, 2012).
9. J. Pollastrini, T.M. Dillon, P. Bondarenko, and R.Y.T. Chou, Anal.
Biochem. 414(1), 88–98 (2011).
10. J.Y. Lee, D. Choi, C. Johan, and M.H. Moon, J. Chromatogr. A.
1218(27), 4144–4148 (2011).
11. K. Rebolj, D. Pahovnik, and E. Zagar, Anal. Chem. 84(17), 7374–83
(2012).
12. J.Y. Kim, S.-K. Kim, D. Kang, and M.H. Moon, Anal. Chem. 84(12),
5343–5350 (2012).
13. R.N. Qureshi, E. Kaal, H.G. Janssen, P.J. Schoenmakers, and W.T.
Kok, Anal. Chim. Acta 706(2), 361–366 (2011).
14. K.H. Kim, J.Y. Lee, and M.H. Moon, Analyst 136(2), 388–392 (2011).
15. K.H. Kim, J.Y. Kim, M.O. Kim, and M.H. Moon, J. Proteomics. 75(8),
2297–2305 (2012).
16. K.H. Kim and M.H. Moon, Anal. Chem. 83(22), 8652–8658 (2011).
17. M. Hassellov, G. Hulthe, B. Lyven, and G. Stenhagen, J. Liq.
Chromatogr. Relat. Technol. 20(16–17), 2843–2856 (1997).
18. P. Reschiglian, A. Zattoni, B. Roda, L. Cinque, D. Parisi, A. Roda,
F. Dal Piaz, M.H. Moon, and B.R. Min, Anal. Chem. 77(1), 47–56
(2005).
19. K.H. Kim, J.Y. Lee, S. Lim, and M.H. Moon, J. Chromatogr. A.
1280(0), 92–97 (2013).
20. M.J. Gidley, I. Hanashiro, N.M. Hani, S.E. Hill, A. Huber, J.-L. Jane,
Q. Liu, G.A. Morris, A. Rolland-Sabaté, A.M. Striegel, and R.G.
Gilbert, Carbohydr. Polym. 79(2), 255–261 (2010).
21. S. You, S.G. Stevenson, M.S. Izydorczyk, and K.R. Preston, Cereal.
Chem. 79(5), 624–630 (2002).
22. M. Van Bruijnsvoort, K.G. Wahlund, G. Nilsson, and W.T. Kok, J.
Chromatogr. A. 925(1), 171–182 (2001).
23. L. Nilsson, Food Hydrocolloids 30(1), 1–11 (2013).
24. K.G. Wahlund, M. Leeman, and S. Santacruz, Anal. Bioanal. Chem.
399(4), 1455–1465 (2011).
25. S. Juna, P.A. Williams, and S. Davies, Carbohydr. Polym. 83(3),
1384–1396 (2011).
26. S. Juna and A. Huber, Starch-Starke 64(1), 18–26 (2012).
27. S. Juna and A. Huber, Starch-Starke 64(2), 87–96 (2012).
28. S. Juna and A. Huber, J. Chromatogr. A. 1219, 161–172 (2012).
29. S. Juna and A. Huber, Starch-Starke 64(3), 171 (2012).
30. S. Juna, M. Damm, C.O. Kappe, and A. Huber, Starch-Starke 64(8),
(2012).
31. H. Ruebsam, M. Krottenthaler, M. Gastl, and T. Becker, Starch-
Starke 64(9), 683–695 (2012).
32. E. Perez, O. Gibert, A. Rolland-Sabate, Y. Jimenez, T. Sanchez,
A. Giraldo, B. Pontoire, S. Guilois, M.C. Lahon, M. Reynes, and D.
Dufour, J. Agric. Food Chem. 59(1), 263–273 (2011).
33. A. Rolland-Sabate, S. Guilois, B. Jaillais, and P. Colonna, Anal.
Bioanal. Chem. 399(4), 1493–1505 (2011).
34. L. Pitkanen, M. Tenkanen, and P. Tuomainen, Anal. Bioanal. Chem.
399(4), 1467–1472 (2011).
35. R. Irague, A. Rolland-Sabate, L. Tarquis, J.L. Doublier, C. Moulis, P.
Monsan, M. Remaud-Simeon, G. Potocki-Veronese, and A. Buleon,
Biomacromolecules 13(1), 187–195 (2012).
36. M. Ulmius, S. Adapa, G. Onning, and L. Nilsson, Food Chem.
130(3), 536–540 (2012).
37. M. Ulmius, G. Onning, and L. Nilsson, Food Hydrocolloids 26(1),
175–180 (2012).
38. A. Hakansson, M. Ulmius, and L. Nilsson, Carbohydr. Polym. 87(1),
518–523 (2012).
39. T. Luan, Y.P. Fang, S. Al-Assaf, G.O. Phillips, and H.B. Zhang,
Polymer. (Guildf). 52(24), 5648–5658 (2011).
40. M. Ali, E. Hwang, I.H. Cho, and M.H. Moon, Anal. Bioanal. Chem.
402(3), 1269–1276 (2012).
41. J. Alftren, J.M. Penarrieta, B. Bergenstahl, and L. Nilsson, Food
Hydrocolloids 26(1), 54–62 (2012).
42. K. Rbii, O. Surel, N. Brambati, A.M. Buchert, and F. Violleau, Food
Hydrocolloids 25(3), 511–514 (2011).
43. K. Rbii, F. Violleau, N. Brambati, A.M. Buchert, and O. Surel, Food
Hydrocolloids 25(6), 1409–1412 (2011).
44. A. Vernhet, S. Dubascoux, B. Cabane, H. Fulcrand, E. Dubreucq,
and C. Poncet-Legrand, Anal. Bioanal. Chem. 401(5), 1559–1569
(2011).
45. T. Otte, H. Pasch, T. Macko, R. Brull, F.J. Stadler, J. Kaschta, F.
Becker, and M. Buback, J. Chromatogr. A. 1218(27), 4257–4267
(2011).
46. T. Otte, T. Klein, R. Brull, T. Macko, and H. Pasch, J. Chromatogr. A.
1218(27), 4240–4248 (2011).
47. T. Otte, H. Pasch, R. Brull, and T. Macko, Macromol. Chem. Phys.
212(4), 401–410 (2011).
48. I.J. Suarez, R. Rosal, A. Rodriguez, A. Ucles, A.R. Fernandez-Alba,
M.D. Hernando, and E. Garcia-Calvo, TrAC Trends. Anal. Chem.
30(3), 492–506 (2011).
49. A. Lederer and S. Boye, LCGC Europe 24(12), 620–628 (2011).
50. S. Boye, N. Polikarpov, D. Appelhans, and A. Lederer, J.
Chromatogr. A. 1217(29), 4841–4849 (2010).
51. S. Boye, D. Appelhans, V. Boyko, S. Zschoche, H. Komber,
P. Friedel, P. Formanek, A. Janke, B.I. Voit, and A. Lederer,
Biomacromolecules 13(12), 4222–4235 (2012).
52. J. Ehrhart, A.F. Mingotaud, and F. Violleau, J. Chromatogr. A.
1218(27), 4249–4256 (2011).
53. T. Miller, R. Rachel, A. Besheer, S. Uezguen, M. Weigandt, and A.
Goepferich, Pharm. Res. 29(2), 448–459 (2012).
54. R.M. Sisson and J.C. Giddings, Anal. Chem. 66(22), 4043–4053
(1994).
55. M.E. Schimpf and J.C. Giddings, J. Polym. Sci Part. B: Polym. Phys.
27(6), 1317–1332 (1989).
56. J.R. Runyon and S.K.R. Williams, J. Chromatogr. A. 1218(38),
6774–6779 (2011).
57. J.R. Runyon and S.K.R. Williams, J. Chromatogr. A. 1218(39),
7016–7022 (2011).
58. M.E. Schimpf and S.N. Semenov, J. Phys. Chem. B. 104(42),
9935–9942 (2000).
59. E.P.C. Mes, W.T. Kok, and R. Tijssen, Int. J. Polym. Anal. Charact.
8(2), 133–153 (2003).
60. A.C. Wurger, Phys. Rev. Lett. 102(7), (2009).
61. C.A. Ponyik, D.T. Wu, and S.K.R. Williams, Anal. Bioanal. Chem. In
Press (2013).
62. W. Hiller, W. van Aswegen, M. Hehn, and H. Pasch,
Macromolecules. 46(7), 2544–2552 (2013).
63. J.R. Runyon, D.T. Wu, and S.K.R. Williams, Characterizing branched
Polymers Using Thermal Field-Flow Fractionation — In Preparation.
Carmen Bria is a PhD student in the Chemistry Department at the
Colorado School of Mines (Colorado, USA). His research focuses
on the use of FFF and light scattering to characterize proteins
and probe protein aggregation processes. He is also working on
developing improved membrane surfaces for flow FFF.
Frédéric Violleau graduated from ENSCT (INPT – University
of Toulouse, France) with a “Diplôme d’Ingénieur en chimie”
(equivalent to a MSc in chemistry) and from the National
Polytechnic Institute of Toulouse (University of Toulouse) with
a PhD in organic chemistry. He joined Ecole d’Ingénieurs
de PURPAN (EI Purpan – INPT – University of Toulouse) in
2003 and he is currently vice head of the Agricultural and
Food Sciences Department. He has experience in using
AsFlFFF technology for various applications involving proteins,
polysaccharides, polymers, and particles.
S. Kim R. Williams is a professor of chemistry and the Director
of the Laboratory for Advanced Separations Technologies at
the Colorado School of Mines. She began her journey with FFF
as a postdoctoral fellow with the late J. Calvin Giddings at the
University of Utah (Utah, USA) and has acquired more than
25 years of experience in this field. Research in the Williams
group focuses on developing new capabilities for nanoparticle
and macromolecular analyses using FFF and related methods.
Dissemination of these new technologies are done through
collaborations with scientists at universities, companies, and
national laboratories. She recently edited a book entitled Field-
Flow Fractionation in Biopolymer Analysis.8
ES339972_LCA1113_016.pgs 10.18.2013 20:05 ADV blackmagentacyan
17www.chromatographyonline.com
LC TROUBLESHOOTING
Recently, a reader emailed me with a
problem he was having determining the
lower limit of quantification (LLOQ) for his
method, which had a target LLOQ of 0.01
µg/mL for his analyte. He compared the
LLOQ calculated using the International
Committee on Harmonization guidelines
(ICH) (1) with replicate injections of a
reference standard and found that the
two differed by more than an order of
magnitude. He came to me to help him
figure out what was wrong. The method
was proprietary, and the reader needed
to stay anonymous, so I’ve disguised
things a bit, but this case study helps us
to better understand how to evaluate a
calibration curve.
The ICH (1) presents a formula to
calculate what they call the quantitation
limit (QL), but what most users call the
limit of quantification (LOQ) or LLOQ:
QL = 10σ/S [1]
where σ is the standard deviation of the
response (the standard error [SE]) and S
is the slope of the calibration curve. This
is calculated easily from the regression
statistics generated in Microsoft Excel or
your data system software. Let’s see how
this works.
Table 1 includes the initial data
from the calibration curve. The user
injected eight concentrations of his
analyte, ranging from 0.01 to 1.0 µg/
mL, generating the peak areas shown
in the “Response” column of Table 1. I
used Excel’s regression tool to generate
the regression statistics, part of which
I’ve included in Table 2. These include
the coefficient of variation (r2), the
standard error of the curve (SE‑curve),
the y‑intercept (intercept‑coefficient),
the standard error of the y‑values
(intercept‑SE‑y), and the slope of the
curve (X variable). Calculated values for
these variables are shown in the second
two columns of Table 2, headed “With
1.0 µg/mL”.
The user used equation 1 with the
standard error of the curve (SE‑curve)
and slope, and found that the LLOQ was
predicted to be ~0.15 µg/mL (summarized
as the first entry of Table 3). (Here I’ll
pause to remind you that I’ve rounded
and truncated numbers in the tables
for ease of viewing; if you try to repeat
my calculations, your results may differ
slightly.) Yet, when he injected n = 10
replicates of a 0.01 µg/mL solution, he
found the percent relative standard
deviation (%RSD) was 1.1% (last entry,
Table 3), which he felt indicated the LLOQ
was considerably lower than the 0.15‑µg/
mL prediction using the ICH technique. At
this point he contacted me.
Examine the Calibration CurveThe calibration curve shown in
Figure 1(a) was supplied to me with the
data set. You can see that the value
of r2 = 0.9986 is excellent. The linear
regression line is shown in blue; at first
glance, this looks good too. However, a
closer examination of the regression line
shows that it is above the data points
at low concentrations and below the
data points at the high concentrations,
passing through the data points at middle
concentrations. This kind of behaviour
tends to send up a caution flag for me
because the higher concentrations tend
to dominate the calculation. It is time to
examine the data a little more carefully.
Although it is part of the reporting
requirements for most methods, we
should be a little careful about putting
too much confidence in values of r2.
The reason for this is that the coefficient
of variation is meant to be used with
homoscedasic data; that is, data in which
the standard deviation is approximately
the same throughout the data range.
Chromatographic data, however, are not
homoscedastic, but heteroscedastic. The
relative standard deviation (%RSD) tends
to be constant throughout the range. In
plain English, chromatographic data don’t
have, for example, standard deviations of
±1 ng/mL throughout the concentration
range, but they might instead have
±0.1% RSD throughout the range.
The coefficient of variation, r2, doesn’t
describe heteroscedastic data very well,
so if we use r2 as our sole determinant
of the goodness of a calibration curve,
we may be misled. This all means that
r2 = 0.9986 for these data does not
guarantee that all is well.
Back to Table 1. I’ve used the
regression equation to calculate the
expected response at each concentration
and compared this to the actual response
to determine the percent error. These
values are listed in the third column of
Table 1 (%‑error; with 1.0 µg/mL). You can
see that the deviations from the expected
values increase at lower concentrations,
as expected, but they are also larger at
high concentrations than in the middle
of the curve. One technique to find out
if there is a problem with the highest
concentration is to drop it from the data
set and repeat the calculations. I did
this by dropping the 1.0‑µg/mL point;
the data are shown in column four of
Table 1 (%‑error, without 1.0 µg/mL).
Notice how this reduces the deviations
from the expected values. Also, the error
increases at the lower concentrations,
as expected, but is very small at higher
concentrations (with the exception of 1.0
µg/mL). The regression results for the data
without the 1.0‑µg/mL point are shown in
the last two columns of Table 2 (headed
“without 1.0 µg/mL”). You can see that
the SE‑curve, SE‑y, and y‑intercept are
all reduced by approximately an order
of magnitude, yet r2 changes very little
(0.9986 versus 1.000). In Figure 1(b),
I’ve plotted the revised regression line,
What’s the Problem with the LLOQ? — A Case StudyJohn W. Dolan, LC Resources, Walnut Creek, California, USA.
Two methods of calculating the lower limit of quantifi cation (LLOQ) disagree. Which, if either, is correct?
ES339297_LCA1113_017.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201318
LC TROUBLESHOOTING
which visibly fits the data better than the
original if the 1.0‑µg/mL point is ignored.
Another way to evaluate these differences
is to compare the absolute values of the
%‑error, as shown in the last two columns
of Table 1. The sum of these absolute
values is shown at the bottom. Notice that
eliminating the 1.0‑µg/mL point from the
regression calculation reduced the total by
more than 2.5‑fold from 54% to 20%. This
is definitely a better fit of the data.
An additional way to visualize the data
is shown in Figure 2, where I’ve taken just
the lower (Figure 2[a]) and higher (Figure
2[b]) portions of the concentration curve
and expanded the scale. Now the original
regression (blue line) is obviously an
inferior fit to the revised one (red line) at
both ends of the scale.
At this point, it might be interesting to
determine what the problem is with the
1.0‑µg/mL point, but I don’t have any
additional information to help me with this
task. It would be nice to make several
replicate injections to be sure the 1.0‑µg/
mL data point isn’t an outlier. If the problem
persists over replicate injections, a new
preparation of the standard should be
checked to eliminate the possibility of
formulation errors. Another possibility is
that the peak is large enough to cause a
slightly nonlinear behaviour of the detector,
which often happens as the detector signal
nears its upper limit. In any event, I think
it is prudent to exclude this point from the
regression without further indications that it
should be included as a valid point.
Another question that often comes up
is whether the calibration curve should be
forced through x = 0, y = 0 or not. This
is a simple test that was discussed in an
earlier “LC Troubleshooting” column (2).
If the value of the y‑intercept calculated
from the regression process is less than
the standard error of the y‑intercept, it
means that the y‑value is within 1 standard
deviation (SD) of the 0,0 point. Most
statistical tests will tell you that there is no
statistical difference between a point <1
SD from the mean and the mean, so the
curve can be forced through zero. How do
you check this? The data are in the Excel
regression summarized in Table 2 on the
line labelled “intercept.” The “coefficient”
column lists the calculated value of the
y‑intercept, so if this is less than the
standard error (SE‑y), you can force the
curve through zero. You can see that in
both cases (with or without 1.0 µg/mL
included), the y‑intercept is greater than
the standard error, so the curve should not
be forced through zero.
Double-Check the CalculationsNow that we’ve decided to exclude 1.0 µg/
mL from the regression calculations, let’s
see why the ICH method predicted such
a large LLOQ. When I tried to reproduce
the user’s results, I found the problem. He
was using the standard error of the curve
(SE‑curve, line 2 of Table 2) instead of
the standard error of the y‑intercept. The
SE‑curve value represents the variability
around the regression curve throughout
the whole range of the curve. But for
determination of the LLOQ, we want to use
the standard error in that region instead,
so SE‑y is more appropriate. Otherwise we
often find that the variability of the larger
concentrations overpowers the variability of
the lower ones and gives an unrealistically
high value of the LLOQ. When I used the
SE‑y value with equation 1, the LLOQ was
reduced by approximately two‑fold with
Figure 1: Plot of data of Table 1 with overlay of regression lines. Regression (a)
including and (b) excluding the 1.0‑µg/mL point.
0 0.2 0.4 0.6 0.8 1.0
Concentration (μg/mL)
0
4
8
12
16
(a)
Resp
on
se (
x10
-6)
r2 = 0.9986
y = 14.9E6x+133832
0
4
8
12
16
0 0.2 0.4 0.6 0.8 1.0
Concentration (μg/mL)
r2 = 1.0000
y = 15.4E6x+39211
Resp
on
se (
x10-6
)
(b)
Table 1: Input data and error calculations.
Concentration (µg/mL)
Response%-Error Absolute %-Error
With 1.0 µg/mL
Without 1.0 µg/mL
With 1.0 µg/mL
Without 1.0 µg/mL
0.01 207,028 37% ‑7% 37% 7%
0.05 853,543 3% ‑5% 3% 5%
0.10 1,548,352 5% 2% 5% 2%
0.20 3,096,704 1% 1% 1% 1%
0.40 6,193,568 ‑1% 0% 1% 0%
0.60 9,290,112 ‑2% 0% 2% 0%
0.80 12,386,816 ‑2% 0% 2% 0%
1.00 14,686,085 3% 5% 3% 5%
Sum 54% 20%
ES339300_LCA1113_018.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
The inventors of the Valco gas-tight valve for chromatography
45 years of experience in valves for chromatography
s�&OR�INJECTION��STREAM�SELECTION��TRAPPING�
and column switching
s�"ORES�FROM������MM��������TO���MM�������
s�:ERO�DEAD�VOLUME�lTTINGS�FOR���������������
� ������OR������TUBING�
s�!LLOYS�AND�POLYMER�COMPOSITES�TO�MEET
virtually any system requirement
s�-ANUAL��PNEUMATIC��OR�ELECTRICALLY ACTUATED
Valco Instruments Co. Inc.tel: 800 367-8424
fax: 713 688-8106
North America, South America, and Australia/Oceania contact:
�
VICI AG Internationaltel: Int + 41 41 925-6200
fax: Int + 41 41 925-6201
Europe, Asia, and Africa contact:
Request a
catalog
�
®
Multiposition stream selectors
Diaphragm valves
Sampling and switching valves
ES339843_LCA1113_019_FP.pgs 10.18.2013 17:28 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201320
LC TROUBLESHOOTING
be near or prepared at the quantitation
limit.” In the last line of Table 3, you can
see that the n = 10 replicate injections
at 0.01 µg/mL gave imprecision of 1.1%
RSD, an excellent value at the LLOQ for
most methods. This strongly suggests
that the method will perform adequately
at the desired LLOQ of 0.01 µg/mL of the
target analyte.
SummaryThis data set has served as a good
example of how easy it is to misinterpret
the results of a calibration curve. We saw
that the value of r2 can be misleading
about how good the calibration curve is.
the original dataset (0.15 versus 0.08 µg/
mL), as shown in the first two lines of Table
3. When the SE‑y of the revised calibration
curve (without 1.0 µg/mL) is used, the
predicted LLOQ drops to 0.011 µg/mL. As
mentioned above, the revised calibration
curve generates values of SE‑curve and
SE‑y that are approximately an order of
magnitude smaller than the original data
set (Table 2).
The predicted LLOQ that we just
calculated using the ICH method is not
sufficient, however. The ICH document
(1) clearly states, “the limit should be
subsequently validated by the analysis of
a suitable number of samples known to
It was shown that it is useful to examine
both a visual and tabular expression of
the data. The original plot (Figure 1[a])
suggested that the highest concentration
might be biasing the regression, and
when this point was eliminated, the new
trend line (Figure 1[b]) fits all the other
points better. Expanding the scale on
the plots (Figure 2) also helped to get
a better picture of what is happening.
Comparing the sum of absolute values
of the deviations of experimental data
points from those calculated from the
regression is a simple way to see if a new
data treatment reduces the overall error.
In the present case, error was reduced by
more than 2.5‑fold simply by dropping the
highest concentration point (Table 1).
When using estimating techniques,
such as the ICH method used here, it is
imperative to use the correct coefficients
or the wrong conclusions may be
drawn. Fortunately, the user noticed that
something was wrong and searched for
further help. If, instead, he believed the
calculations, he might have discarded a
good method or spent unnecessary time
trying to improve an already acceptable
method. Finally, regression curves,
percent‑error tables, and data plotting
techniques are merely tools to help us
understand the data better. When it
comes to determining the LLOQ, there
is nothing that can compare with the
measured performance from multiple
injections at the target LLOQ.
References(1) Validation of Analytical Procedures: Text and
Methodology Q2(R1), International Conference
on Harmonization, Nov. 2005, http://www.ich.
org/LOB/media/MEDIA417.pdf.
(2) J.W. Dolan, LCGC North Am. 27(3), 224–230
(2009).
John W. Dolan is the vice president of
LC Resources, Walnut Creek, California,
USA. He is also a member of the LC•GC
Asia Pacific editorial advisory board. Direct
correspondence about this column should
go to “LC Troubleshooting”, LC•GC Asia
Pacific, 4A Bridgegate Pavilion, Chester
Business Park, Wrexham Road, Chester,
CH4 9QH, UK, or e‑mail the editor‑in‑chief,
Alasdair Matheson, at amatheson@
advanstar.com
Table 2: Summary of Excel regression statistics.
Concentration
(µg/mL)With 1.0 µg/mL Without 1.0 µg/mL
r2 0.9986 1.0000
SE‑curve 222,989 28,842
Coefficient SE‑y Coefficient SE‑y
Intercept 133,832 119,416 39,211 16,245
X variable 14,934,035 15,417,430
Figure 2: Expanded sections of Figure 1(a) (blue) and 1(b) (red): (a) 0.01–0.2 µg/
mL region, (b) 0.6–1.0 µg/mL region.
0 0.05 0.10 0.15 0.20 0
10
20
30
Concentration (μg/mL)
Resp
on
se (
x10
-5)
(a)
0.6 0.8 1.0 8
10
12
14
16
Concentration (μg/mL)
Resp
on
se (
x10
-6)
(b)
Table 3: Summary of LLOQ calculations.
Technique LLOQ (10σ/S’)
SE‑curve 0.149 µg/mL
SE‑y with 1.0 µg/m 0.080 µg/mL
SE‑y without 1.0 µg/mL 0.011 µg/mL
0.01 µg/mL, n = 10 1.1% RSD
ES339295_LCA1113_020.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
waters.com
©2013 Waters Corporation. Waters, ACQUITY, Advanced Polymer Chromatography and The Science of What’s Possible are registered trademarks of Waters Corporation.
APC is a trademark of Waters Corporation.
APC vs. GPC with narrow standard polystyrene Mp = 510
Minutes
1.60
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
1.80 2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00
μRIU
22.0
20.0
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
Minutes
4.204.00 4.40 4.60 4.80 5.00 5.20 5.40 5.60 5.80 6.00 6.20 6.40
μRIU
THE PATH TO INNOVATION.WITH ADVANCED POLYMER CHROMATOGRAPHY, IT’S VERY CLEAR.
More information about your polymers in less time. Every time. That’s what it takes to bring
innovation into advanced Chemical Materials markets. And that’s exactly what you can expect
from the truly unique instrumentation and column chemistries of the new Waters® ACQUITY
Advanced Polymer Chromatography®
(APC ™) System. To learn more about why polymer
chromatography will never be the same again, visit waters.com/APC
ES339512_LCA1113_021_FP.pgs 10.17.2013 21:35 ADV blackyellowmagentacyan
LC•GC Asia Pacià c November 201322
GC CONNECTIONS
One of the classical trade‑offs in gas
chromatography (GC) separations
lies between speed of analysis and
peak resolution. Chromatographers
can increase the speed of analysis
in a number of ways, including the
use of shorter and narrower columns,
higher temperatures and temperature
programme ramp rates, and faster
flow rates, but higher speeds do
not guarantee equal or better peak
resolution. The relationships between
flow or velocity and resolution have
recently received attention in the
context of a drive towards faster
separations, and the on‑going
substitution of hydrogen carrier gas for
helium in many laboratories also fuels
the discussion. This instalment of “GC
Connections” discusses the effects of
increased carrier‑gas flow or velocity
in an example separation that includes
two pairs of solutes.
Optimum Practical Velocity One of the more neglected separation
metrics is the optimum practical
carrier‑gas velocity (OPGV). This
idea is not new: The pioneers of
gas chromatography formulated the
OPGV as one way to measure the
trade‑offs between speed of analysis
and resolution. As the carrier‑gas flow
increases above an optimum value,
peaks become broader and their
resolution starts to decline but they
are eluted sooner in proportion to
the higher flow. Scott and Hazeldean
(1) proposed that an optimum
compromise between the two could
be found by increasing the flow until
the corresponding increase in a plot of
plate height versus average carrier‑gas
velocity becomes essentially linear. An
optimum velocity would be reached
at the point where additional losses of
resolution because of further increases
in velocity could not be compensated
for by a corresponding increase in
column length.
Without experimental measurements
from multiple columns, the OPGV has
been considered as the velocity at
which a tangent line from the origin
meets a plot of measured values of the
plate height, Hmeas, versus the average
linear carrier‑gas velocity, u–. This is the
velocity at which the quantity H/u– hits
a minimum (2). A plot of experimental
Hmeas versus u– departs from the
linear at higher velocities because of
secondary gas‑compression effects
at higher pressures and extracolumn
broadening from the detector if the
peaks become narrow enough. The
basic Golay equation, however,
neglects such effects. A plot of the
theoretical plate height, Htheor, versus u–
will never become completely linear:
H = (B/u–) + Cu– [1]
where H is the height of one
theoretical plate, u– is the average
carrier‑gas linear velocity, B describes
the broadening of a peak because
of gas diffusion along the direction
of carrier‑gas flow, and C describes
broadening because of the effects
of solute molecules entering and
leaving the stationary phase. As the
linear velocity (flow) increases, a
decreasingly small fraction of the total
theoretical plate height is a result of
the B term, and the C term dominates.
These effects are shown in Figure 1
for the basic Golay equation using a
25 m × 0.53 mm column. In this, and
the subsequent column treatments,
the influence of the stationary phase
on solute broadening is minimal; the
stationary‑phase film thickness used
was 0.4 µm, which influenced this plot
by less than 2%. Plot (a) is the total
theoretical plate height; plot (b) is the
B term contribution; and plot (c) is
the C term contribution. Plot (c) also
represents a tangent that meets the
total plate height (a). That this junction
occurs at infinite linear velocity shows
the fundamental difficulty with using
the basic Golay equation this way for
OPGV calculations.
The basic Golay equation yields an
infinite linear velocity if a tangent‑line
construction is used to find the OPGV
because the theoretical relationship
neglects the effects on plate height
of operating at higher inlet pressures
and of producing potentially very
narrow peaks. The theoretical plot
does not curve away from a linear
relationship at elevated velocities,
but experimental data do. Although
it is a convenient way to explain
idealized column band‑broadening
behaviour without making arduous
measurements of Hmeas versus u– data
for multiple columns, widespread
use of the basic Golay equation has
resulted in the neglect of OPGV as a
means of expressing a practical upper
limit for average linear velocity in
specific separations.
A simple approach to determining
a finite value for OPGV from the
basic Golay equation is to choose an
arbitrary point that sets the OPGV at
the velocity where gas–gas diffusion
contributes a fixed percentage of
the overall band broadening. Figure
1 illustrates an example at the
point labelled (d), where gas–gas
diffusion contributes 10% of the total
band‑broadening and u– = 117 cm/s.
The optimum velocity, u–opt — the point
at which H is at a minimum — is shown
as well at point (e), where u– = 39.2
Practical Gas ChromatographyJohn V. Hinshaw, BPL Global Ltd, Hillsboro, Oregon, USA.
Questions about how practical proposed gas chromatography (GC) method changes are often come up during optimization for speed and resolution, or while converting to a different carrier gas. Related objective measurements such as the optimum practical carrier gas velocity were deà ned more than 40 years ago. This instalment reviews such metrics in the light of their relevance to today’s GC challenges.
ES339301_LCA1113_022.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
23www.chromatographyonline.com
GC CONNECTIONS
cm/s. But this idealized and arbitrary
OPGV point is not connected to
physical properties that accommodate
the effects of an increasing inlet
pressure gradient on peak broadening;
the velocity of 117 cm/s seems too
high for a reasonable upper limit.
Extended band‑broadening theories
that do include such effects can
produce a better theoretical picture of
the effects of increasing the velocity.
Practical TheoryA number of authors have proposed
more‑complete theories, including
Golay himself, although his extended
equations address porous‑layer
open‑tubular (PLOT) and support‑
coated open‑tubular (SCOT) columns
and do not consider how gas–gas
diffusion is affected by carrier‑gas
compression inside the column. A
relationship proposed by Giddings
(3) works well as a theoretical model
for determining OPGV values that are
closer to experimental data. Such
calculations are only as good as the
model and the accuracy of the applied
physical parameters, of course, but
they can provide useful insight for the
selection of practical operating gas
velocities or flows.
The B and C terms of the Golay
equation (equation 1) are proportional
to the rate at which solutes diffuse
through the carrier gas. Thus,
band‑broadening increases as the
gas–gas diffusion rates increase.
The Golay equation considers these
diffusion rates at the column outlet
pressure (atmospheric pressure) alone
1.0
0.8
(e)
(d)
(a)
(c)
(b)
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120 140 160 180
H (
mm
)
u (cm/s)
Figure 1: Plot of the basic Golay equation for n‑hexane: (a) total plate height, (b) B
term contribution to the plate height, (c) C term contribution to the plate height, (d)
OPGV where the B term contribution accounts for 10% of the total plate height, and
(e) optimum carrier‑gas velocity. Theoretical column parameters: 25 m × 0.53 mm,
0.4‑µm nonpolar stationary‑phase film thickness, 130 °C, helium carrier gas.
11708
©2013 Sigma-Aldrich Co. LLC. All rights reserved. SIGMA-ALDRICH and SUPELCO are trademarks of Sigma-Aldrich
Co. LLC, registered in the US and other countries. Ecoporous and Titan are trademarks of Sigma-Aldrich Co. LLC.
N 1.9 μm UHPLC ColumnsS l ® d T ™ C UHPLC l b dSSupelco® introduces Titan™ C18 UHPLC columns, basedS
on 1.9 μm totally porous, monodisperse silica particles. o
These particles are the result of a newly developed, patent-T
pending, Ecoporous™ silica manufacturing process.p
For new product information, ordering
and real time availability, visit
sigma-aldrich.com/titan
ES339302_LCA1113_023.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201324
GC CONNECTIONS
accommodates the net effect
of increasing diffusion rates on
band‑broadening as solutes progress
along the column. The correction is
larger with increasing inlet pressure
and, thus, follows the influence of
higher carrier‑gas velocities. (The
details of this correction factor and the
Giddings equation are available for
interested readers as supplementary
material on‑line at http://wiki.hrgc.com.)
In addition to the influence of the
carrier‑gas pressure, diffusion rates
are different for different solutes. They
grow smaller with increasing molecular
weight. The diffusion rate of n‑hexane,
for example, in helium (or in any other
and does not consider the effect of
the higher pressures and gradient
inside the column itself. Physical
measurements as well as theoretical
treatments of gas diffusion show that
diffusion is inversely proportional to
pressure: Diffusion slows down as
pressure increases. The effect of
intracolumn carrier‑gas pressure on
diffusion rates is not very large at low
inlet pressures, such as in a 0.53‑mm
i.d. column. As inlet pressures rise the
effect becomes more significant — for
example, with a 0.25‑mm i.d. column
— especially at higher linear velocities.
Giddings’ equation applies a
pressure correction factor that
carrier gas) is about 40% faster than
that of n‑dodecane. Temperature plays
a role too; gases diffuse more rapidly
at higher temperatures. The current
discussion is limited to isothermal
conditions so variable temperature
isn’t a concern, but its influence is
significant when temperatures or
programming rates are changed as
part of optimization.
Figure 2 shows a series of Giddings
theoretical plate height versus u–
curves for n‑C6, n‑C8, n‑C10, and
n‑C12 on a 25 m × 0.25 mm column
at 130 °C. Theoretical diffusion
coefficients for each solute were used
as listed by Ettre (4). The column
pressure drops were calculated
from theory as well, using the same
relationships found in instrumental
electronic pneumatics.
The influence of the individual solute
diffusion rates on plate height is quite
clear. Each solute takes on its own
optimum average linear velocity, as
marked on each plot in Figure 2. That
different solutes have different u–opt
values is not new information, although
the span of the optima in this case
— from 34 cm/s for n‑C12 to 44 cm/s
for n‑C6 — appears wider than might
be expected. Without considering
the OPGV for the moment, the optima
plainly show that biasing the carrier‑
gas velocity on the high side appears
to be a good idea: Small losses in
efficiency are taken while achieving a
faster separation. But what about the
high end of linear velocity for speed
optimization purposes?
With the Giddings equation a tangent
line from the origin intersects each
curve at a well‑defined point, and
these points correspond to the original
definition of the OPGV. The tangent
line for n‑C6 is drawn in Figure 2; it
intersects the plot at u– = 68 cm/s.
Figure 3 illustrates another way of
expressing the OPGV by plotting the
number of plates generated per second
as a function of the average carrier‑gas
velocity. The maxima of these plots
correspond to each solute’s OPGV.
Similar to the span of u–opt, the OPGV
values range from 54 cm/s for n‑C12 up
to 68 cm/s for n‑C6.
Taken together, Figures 2 and 3
would define a range for minimum and
maximum optimized average velocities
across the scope of the idealized
normal hydrocarbons that were
employed, from the highest optimum
0 10 20 30 40 50 60 70 80 90 100
400
300
200
100
0
(d)
(c)
(b)
(a)
N/s
u (cm/s)
Figure 3: Plots of theoretical plates per second from the Giddings equation:
(a) n‑hexane, (b) n‑octane, (c) n‑decane, and (d) n‑dodecane. Vertical tick marks
on each plot show the maximum, at the OPGV. Theoretical column parameters,
same as in Figure 2.
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60
OPGV
70 80 90 100
H (
mm
)
(d)
(c)
(b)
(a)
u (cm/s)
Figure 2: Plots of the Giddings equation for a series of hydrocarbons: (a) n‑hexane,
(b) n‑octane, (c) n‑decane, and (d) n‑dodecane. Dashed line: tangent that intersects
plot (a) at its OPGV. Theoretical column parameters, same as Figure 1 except for
column internal diameter of 0.25 mm.
ES339298_LCA1113_024.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
25www.chromatographyonline.com
GC CONNECTIONS
velocity of 44 cm/s to the lowest OPGV
of 54 cm/s. Compared to operating
at the lowest optimum of 34 cm/s, a
velocity of 54 cm/s would decrease
the analysis time by roughly 38% while
sacrificing about 20% of the theoretical
plates generated for the later‑eluted
n‑C12. Running the separation at 44
cm/s would restore much of the plate
number while sacrificing some of the
gain in analysis time.
These trade‑offs in column
efficiency for speed are also not
new information, but here the OPGV
values provide a more meaningful
upper end for a range of velocities
than do arbitrary velocities based on
simple percentages or multipliers,
while preserving most of the column’s
separating power. And as always,
the best procedure is to determine
actual experimental performance for
the analytes of interest. Theoretical
predilections provide useful guidelines
and place boundaries on the range of
practical conditions, but they are no
substitute for real data.
Practical Resolution Chromatography is all about peak
resolution, not just the efficiency of
individual solutes. The discussion so
far has considered solutes standing
alone, ones that are well separated
under almost any conditions at that.
Applying the Giddings equation to
pairs of closely eluted solutes provides
some more information and guidance
for selecting optimized velocities by
putting the modelled separation into
a context of resolution. Two pairs
of peaks were chosen so that the
resolution, Rs, of each pair will range
around 2.0. This exceeds the minimum
“baseline” resolution of 1.5 that is
often considered good enough, but a
resolution of 2.0 does provide some
working room for eventual performance
losses because of column degradation
and also makes for a more robust
method. Many separations do exhibit
more than one critical pair of peaks,
those that are resolved at close to the
minimum, so it is useful to consider
what happens to solute pairs at the
beginning and end of a separation as
the velocity is optimized.
Figure 4 shows the resolution that
would be obtained in theory between
two pairs of solutes, n‑hexane with
a closely following hypothetical
analogue, and n‑dodecane with
another closely eluted analogue.
Each solute pair is separated from its
neighbour with a separation factor, α,
of 1.03. Each pair of solutes shows
an optimum resolution at close to
the optimum linear velocity, which is
expected because they both share
nearly identical properties. The later‑
eluted pair (Figure 4[b]) does gain
some resolution at optimum over the
earlier pair (Figure 4[a]), which is
also expected because the later pair
simply has more time in the column for
resolution to develop. A bias towards
higher velocities up to the OPGV is
again clear, even more so than is
apparent for the individual solutes’
efficiencies as shown in Figure 2. It is
also interesting to see that both pairs’
resolution declines linearly with nearly
the same slope when the carrier‑gas
velocities are pushed higher above
their OPGV levels. If a minimum
resolution of 1.5 were the goal for
these peak pairs, then a velocity of
around 82 cm/s would be acceptable
and would realize a gain in speed of
analysis of approximately 2.5‑fold.
ConclusionBoth efficiency and resolution are
critical for obtaining acceptable
separations. With extra resolution
available, a column can be pushed to
higher linear velocities or flows while
still obtaining a minimum performance
level. Short of experimental data,
theoretical treatment of a separation
can yield useful information about
how performance would be affected
by increasing the speed of analysis,
but only to the degree that the
theory reflects the chromatographic
process. Application of an extended
theory such as the Giddings equation
provides additional insight beyond the
basic Golay equation, and this helps to
frame practical limits on optimization
for speed of analysis. But in cases
where peak resolution is minimal,
there is no substitute for careful
experimental evaluation of a faster
separation scheme.
References(1) R.P.W. Scott and G.S.F. Hazeldean, in
Gas Chromatography 1960, R.P.W. Scott,
Ed. (Butterworths, London, UK, 1960), pp.
144–161.
(2) W. Jennings, Analytical Gas
Chromatography (Academic Press,
Orlando, Florida, USA, 1987), pp. 77–79.
(3) J.C. Giddings, S.L. Seager, L.R. Stucki,
and G.H. Stewart, Anal. Chem. 32,
867–870 (1960).
(4) L.S. Ettre and J.V. Hinshaw, Basic
Relationships of Gas Chromatography
(Advanstar, Cleveland, Ohio, USA, 1993),
p. 47.
John V. Hinshaw is a senior scientist
at BPL Global Ltd., Oregon, USA,
and is a member of the LC•GC Asia
Pacific editorial advisory board.
Direct correspondence about this
column should be addressed to “GC
Connections”, LC•GC Asia Pacific, 4A
Bridgegate Pavillion, Chester Business
Park, Chester, CH4 9QH, UK, or email
the editor‑in‑chief, Alasdair Matheson,
0 10 20 30 40 50 60 70 80 90 100
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Rs (a)
(b)
u (cm/s)
Figure 4: Plots of resolution for two pairs of solutes, using the Giddings equation:
(a) n‑hexane and a closely eluted analogue and (b) n‑dodecane and a closely
eluted analogue. Separation factor for both pairs α = 1.03. Theoretical column
parameters, same as in Figure 2.
ES339299_LCA1113_025.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
LC•GC Asia Pacià c November 201326
COLUMN WATCH
In any field there are often
“misconceptions” or “myths” that are
perpetuated and passed on to the next
generation. These myths are often driven
by a lack of understanding by practitioners
of the real issues, and can change as time
moves on. Originally, seven years ago, in
a “Column Watch” instalment (1), the 10
most popular myths of the time around
high performance liquid chromatography
(HPLC) column technology were
demystified by discussing the issues at
hand. Because HPLC is approaching
its 50th year, many column myths have
already been passed down to two
generations of liquid chromatographers.
Recently, ultrahigh-pressure liquid
chromatography (UHPLC) has come into
its own and a new set of myths are arising.
The purpose of this instalment of “Column
Watch” is to revisit and update readers on
the most popular column myths of today
and try to dispel some of these myths
before they get perpetuated. This column
is an adaptation of an oral presentation at
the HPLC2013 conference in Amsterdam,
the Netherlands (2). In keeping with the
“countdown theme,” I will start with number
10 and work my way up to the top myth.
Myth 10: Air Will Kill an HPLC ColumnFalse: HPLC and UHPLC columns are
shipped with plugs of either stainless steel
or polymeric construction installed at both
end. Users are told that a column should
always be capped tightly after the column
is disconnected from the instrument. The
thought is that large amounts of air can
get inside the column, perhaps damaging
the packing material, causing bubbles in
the detector flow cell when installed into
the HPLC system in the future, and maybe
disrupting the packed-bed morphology.
One should first realize that the tiny hole
in the endfitting is less than 0.02 in. in
diameter and therefore has an extremely
small cross-sectional area. If left open, the
small amount of air that diffuses into the
column could hardly cause irreparable
damage. Depending on the volatility of
the solvent used to store the column,
there could be some evaporation near the
end of the column. But large quantities
of air would have a hard time diffusing
through the microparticles in the packed
bed seeing that we need thousands of
pounds per square inch of pressure to
push liquid mobile phases through these
micrometre-sized particles. The small
amount of air that could conceivably enter
into the ends of the column would be
immediately dissolved once the system
was pressurized or, at least be flushed out
in the initial pressurization in a short time
and should not cause any problems with
the chromatography later on. However,
if you feel more secure by capping the
endfittings, by all means do so.
Myth 9: All C18 (L1) Columns Are the SameFalse: All of our HPLC column surveys
have shown that C18 is, by far, the most
popular bonded phase in existence (3).
Because pharmaceutical manufacturers
were the earliest adopters of HPLC, the
United States Pharmacopeial Convention
(USP), not wanting to favour any particular
manufacturer of HPLC columns,
developed a classification system that
gave a generic description for each
type of bonded phase column that was
submitted under a new drug application.
For HPLC columns, an “L” designation
was given, and because C18 is used for a
majority of submittals, its designation was
“L1.” As additional phases came along,
they were given their own “L” number such
as C8 (L7), CN (L10), phenyl (L11), and so
on. The implication with this system was
that each C18 column that was submitted
also designated as L1, was the same as
the last L1 column. Unfortunately, this
system proved to be unreliable because
columns from different manufacturers,
produced from different base silicas and
bonded with different silane reagents
using different synthetic routes, were not
chromatographically the same and one
could therefore not be substituted for
another. With more than 800 different L1
columns introduced into the marketplace,
it has proven to be a confusing system.
Several approaches, including the use of
the hydrophobic subtraction model (4,5)
that gives a more detailed classification
of reversed-phase columns, have
been proposed but to this day the “L”
classification is still in widespread use.
Thus, some chromatographers who
do not really understand the issues still
believe that “all C18 columns are the
same.” Simple examples that this is not
the case are shown in Figures 1 and
2. In Figure 1, four different C18 silica
bonded phase columns are shown for
the same separation under the same
operating conditions; each phase
provides a different chromatogram.
The Top 10 HPLC and UHPLC Column Myths: Part 1Ronald E. Majors, Agilent Technologies, Wilmington, Delaware, USA.
Webster’s New Collegiate Dictionary deà nes a myth as “an ill-founded belief held uncritically, especially by an interested group.” Could that group be misinformed chromatographers? In the à rst of a two-part feature from Ron Majors, the top 10 high performance liquid chromatography (HPLC) column myths are presented and attempts are made to demystify them by offering some evidence that they are untrue. This part will feature myths 10 to six. Since ultrahigh-pressure liquid chromatography (UHPLC) has come about, new myths are popping up and these shall also be dealt with here.
ES339304_LCA1113_026.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
27www.chromatographyonline.com
COLUMN WATCH
To demonstrate that the L system also
doesn’t hold for other bonded phases
as well, Figure 2 provides an example
of three different C8 (L7) columns, one
of which (Figure 2[b]) was very similar
to the original chromatogram and could
probably be substituted in an HPLC
method while the third column (Figure
2[c]) is quite different and might even
be considered as orthogonal to the first
two columns. The Fs designation shown
alongside each chromatogram is a
numerical classification of how “close”
of a fit columns are to one another
(4,5). Close Fs numbers are potentially
replacement columns while large values
of Fs imply that the column would not be
a “drop-in” replacement in a particular
HPLC method and, in fact, might be
a useful column when first performing
method development because it offers a
different selectivity to the other columns.
So, the bottom line is: All C18 (L1) and
other reversed phase columns are not the
same.
Myth 8: Never Use 100% Water with a Reversed-Phase LC ColumnFalse: This myth was brought about by
users who experienced a phenomenon
popularly known as “phase collapse”
when using reversed-phase columns with
a low percentage of organic solvent or
100% water as a mobile phase. Phase
collapse really is a misnomer as the
phenomenon was better explained as
phase dewetting. Phase dewetting is
highly undesirable since retention times
decrease and are not reproducible,
peaks may become distorted and
reequilibration times may be quite long.
Earlier, we published two detailed papers
on this subject (6,7). The phase dewetting
conditions most often occur when users
are trying to increase the retention of very
polar compounds in reversed-phase LC
by decreasing the percentage of organic
solvent in the mobile phase to low values
to increase the retention of these polar
compounds, which have a tendency to
be eluted very early in the chromatogram.
Nowadays, this problem is frequently
addressed by using hydrophilic interaction
liquid chromatography (HILIC).
With the help of Figure 3, I will try to
explain phase dewetting. Figure 3 shows
two situations: Situation A, where the
aqueous mobile phase has a significant
portion of water-soluble organic solvent,
such as methanol or acetonitrile — a
densely chemically bonded C18 (or other
hydrophobic bonded phase) prefers to be
solvated with organic solvent (for example,
like-like relationship); and situation B,
when the mobile phase has a very low
percentage of organic content (<10%)
or even 100% water. A very simplistic
visualization of phase collapse can be
observed in the upper portion of Figure 3.
Situation A shows the C18 bonded group
being solvated with methanol and in this
state the hydrophobic moieties are able
to interact with the hydrophobic portions
of solute molecules and provide retention.
On the other hand, for the right hand side
of the upper portion of Figure 3, situation
B shows a C18 phase in a 100% water
mobile phase. The C18 functionality
prefers to be in a self-associated state
(like prefers like) and folds upon itself in
a collapsed state. The bottom portion of
Figure 3 shows the situation as it actually
happens. Most of the interactions with
an LC stationary phase occur inside the
pores (rather than on the outer surface),
so when an organic solvent is present at
higher concentrations (greater than 10%),
the pores are filled with the water–organic
mixture that allows the C18 bonded
groups to be solvated, and everything
behaves normally. However, when the
solvent within a pore becomes unfriendly
(for example, very low %B or 100% A),
there is a tendency to force the water out
of the pore, which results in a dewetting
phenomenon. This dewetting doesn’t
occur instantaneously, but can happen
over a number of column volumes as
the organic solvent is leached out of
the solvated bonded phase. As this
is happening, the retention of organic
solutes may decrease with time and
retention times also decrease accordingly.
Selectivity can change during this time
and peak shapes can become distorted.
Phase dewetting most often occurs with
very hydrophobic, very dense chemically
bonded phases. Phases that are highly
1 2
3
4
1 2 3 4 5
1 2
3
4
1 2 3 4 5
1 2,3
4
1 2 3 4 5
1 2 3
4
1 2 3 4 5
(d)
(c)
(b)
(a)
Time (min)
Figure 1: Chromatograms obtained using C18 bonded phases with the same base
material but different chemistries: (a) Zorbax Eclipse Plus C18 (different surface
treatment for same base silica, double endcapping, same bonding chemistry as
Eclipse XDB-C18); (b) Zorbax StableBond SB-C18 (same base silica, sterically
protected C18 phase, no endcapping); (c) Zorbax Eclipse XDB-C18 (same base
silica, monomeric bonding chemistry, double endcapping); (d) Zorbax Extend-C18
(same base silica, bidentate bonding chemistry, double endcapping). Column
dimensions: 50 mm × 4.6 mm, 1.8-µm dp; mobile phase: 69:31 acetonitrile-water;
flow rate: 1.5 mL/min; temperature: 30 °C; detection: single-quadrupole
electrospray ionization MS, positive mode scan. Peaks: 1 = anandamide,
2 = palmitoylethanolamide, 3 = 2-arachinoylglycerol, 4 = oleoylethanolamide.
ES339303_LCA1113_027.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201328
COLUMN WATCH
after the stationary phase solvated with an acetonitrile-buffered water mixture. Then, the mobile phase was 90% ammonium dihydrogen phosphate buffer and 10% acetonitrile, which allowed solvation of the hydrophobic stationary phase. Next, the column was rinsed for a period of time with an aqueous buffer (no organic). The mobile phase was returned to the original conditions and the sample was reinjected. Note the greatly reduced elution times and the change in selectivity that occurred with the sample chromatogram (Figure 4[b]). Clearly, this separation is different from what would be expected. The water rinse caused a change in the retention characteristics of the stationary phase most likely by a phase dewetting phenomenon. Next, the column was treated with a 50:50 mixture of aqueous buffer and acetonitrile, followed by the mobile phase. Indeed, the chromatogram returned to the original one shown in Figure 4(a).
Table 1 provides a list of phases that do not show the phase dewetting phenomenon. These phases all have a polar functional group of some kind close to the surface, near or on the chemically bonded phase. Polar embedded phases are among the most popular of these special phases. In this case, a polar functional group is located on the alkyl phase itself usually only a few carbon atoms removed from the silica surface. Different commercial phases utilize different embedded functional groups, the most popular being amide, urea, and carbamate. With these polar groups, the water in the mobile phase can interact and solvate the phase so that collapse or dewetting doesn’t occur. Some columns have incorporated polar functional groups in other ways such as endcapping with a polar functionality (for example, diol). These phases are usually given an AQ designation. Very short chain phases (such as C2) do not show phase dewetting because they may allow residual silanol groups to hydrogen bond with the aqueous component of the mobile phase. Surprisingly, very long alkyl chain phases do not show phase dewetting, most likely because the steric requirement allow surface silanols to remain on the bonded silica; hence, water can interact with these silanols and allow surface solvation. Phases with wide pore diameters (for example, 300 Å) do not show phase dewetting because the pores are wide enough not to force water out of them, although I am not aware of
endcapped with non-polar silane reagents may encourage the situation. The % organic in which dewetting may occur varies with a number of parameters including type of bonded phase, bonded phase coverage (density), pore size, and the presence and availability of residual surface silanol groups among others. Phase dewetting does not permanently damage the column and it can be recovered as described below. However, over the years many chromatographers have been totally baffled by the presence of phase dewetting and much time has been lost trying to solve the problem of shifting retention times. Hence, they believe that one shouldn’t run reversed-phase columns in highly aqueous media.
There are two approaches to overcome the phase dewetting phenomenon: Subject the column to a high back pressure according to the Laplace-Young equation (see reference
6 for an example of this approach); or resolvate the stationary phase with a higher % organic in an organic–water mixture or mobile phase.
The first approach is inconvenient and requires a lot of experiments to get the right back pressure. The second approach is the easiest to perform because the column is already installed in the instrument and the experimental conditions can be adjusted to ensure that sufficient organic solvent is present to resolvate the phase. Of course, a third approach is to use a phase that is solvated under all mobile phase conditions (see below).
To illustrate what can happen in a phase dewetting situation, Figure 4 provides an example of the separation of procainamides on a very hydrophobic-C8 phase. The sequence of the experiments is outlined in the figure caption and will not be repeated here. Figure 4(a) depicts the normal isocratic separation that occurs
1
1
(a)
(b)
(c)
0 2
2
2
3
3
3
4
4
4
5
5
5
Time (min)
6
6
6
0 2
2
+
+
1
4
4
6
0 2 4 6
6
7
7
7
8
8
8
9
9
9
Fs= 0.0
Fs= 3.1
Fs= 37
Figure 2: Separation of the same mixture on three reversed-phase columns under the same conditions: (a) Ace C8 (Advanced Chemical Technologies); (b) Precision C8 (Mac-Mod); (c) Inertsil C8 (GL Sciences). Column dimensions: 15 cm × 4.6 mm; flow rate: 2.0 mL/min; temperature: 35 °C; mobile phase: 50:50 30 mM potassium phosphate buffer (pH 2.8)–acetonitrile. Peaks: 1 = N,N-diethylacetamide, 2 = nortriptyline, 3 = 5,5-diphenylhydantoin, 4 = benzonitrile, 5 = anisole, 6 = toluene, 7 = cis-chalcone, 8 = trans-chalcone, 9 = mefenamic acid. (Courtesy of Lloyd Snyder and John Dolan, LC Resources).
ES339306_LCA1113_028.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
29www.chromatographyonline.com
COLUMN WATCH
any studies on the wettability of these
wide-pore phases.
Figure 5 shows the use of an AQ-type
phase in a situation with a low percentage
of organic mobile-phase content. The
sequence of experiments with this
column was very similar to that shown
in Figure 4. Because this phase was
developed to work in mobile phases with
a low percentage of organic content and
100% water, it did not undergo phase
dewetting when subjected to the same
conditions of the highly hydrophobic
phase of Figure 4. Such phases are to be
recommended for the separation of small
polar compounds that are lowly retained
on many reversed-phase chromatography
columns.
Myth 7: It Takes a Minimum of 10 Column Volumes to Reequilibrate an LC ColumnFalse: Equilibration time is very important
in gradient chromatography because
it is a limiting factor in the throughput
of the technique. At the conclusion of
gradient, the column must be returned to
its original state before another injection
can be made. The longer it takes for
this reinstatement to occur, the longer
the overall gradient run. In addition, if
one takes longer to reequilibrate the
column than is actually required, solvent
is wasted. In modern two-dimensional
liquid chromatography (2D LC×LC), the
throughput is dictated by the speed of the
secondary column because the flow on
the primary column is not stopped during
the second chromatographic step. If 10
column volumes are required instead
of just a few, then the comprehensive
chromatography, already a fairly slow
process, is made even slower. Finally, if
reequilibration time is too short and the
column has not been stabilized, then
repeatability of retention time, important
when this parameter is used to help
identify components, may be limited.
There have been a number of studies
of the reequilibration times, but for the
purposes of brevity, I would like to cite
two of the more comprehensive ones
(8,9). Schellinger, Stoll, and Carr (8)
studied high-speed gradient elution in the
reversed-phase LC of neutral compounds
and bases in buffered eluents with regard
to retention repeatability and column
reequilibration and conducted a follow-up
study of full equilibrium conditions (9).
There are many variables affecting the
reequilibration in reversed-phase LC. In
isocratic LC, there is no equilibration time
at the conclusion of a chromatographic
run, but there may be a considerable
waiting time when changing solvent
composition. In gradient LC, the eluent
composition, the bonding density of
the reversed-phase LC packing, the
instrumental design (particularly, the
gradient delay volume also known as the
dwell volume), the flow rate, the use or
lack of use of bonded phase additives
to wet the bonded phase, and the types
of solutes (ionic, ionizable, neutral) and
flushing times all play a part.
I would like to summarize the major
outcomes of these reequilibration
studies. First, the instrument gradient
delay volume, although important in
the real world, must be subtracted from
the total volume of the eluent passed
through the column because the delay
volume itself has little to do with the
reequilibration time that actually occurs
within the chromatography column.
Earlier liquid chromatographs often had
several millilitres of delay volume. The
delay volume is the total volume from
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
CH3OH
CH3OH
CH3OH
CH3OH
CH3OH
CH3OH
CH3OH
CH3OH
CH3OH
SiO2
SiO2
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
Situation A Situation B
Water forced out of pore Pores are wetted with methanol
In A, analytes are properly retained
In B, analytes partially retained or unretained
Time (min) 0 5 10
1
2
3
Time (min) 0 5 10
1
2
4
5 4
5
3
(a) (b)
Figure 3: Phase collapse (or more correctly, phase dewetting).
Figure 4: Inconsistent retention in a highly aqueous mobile phase as demonstrated
by the separation of procainamides on a hydrophobic C18 column. Sequence
of events: Condition with a 50:50 mixture of phosphate buffer and acetonitrile for
15 min; run mobile phase for 5 min; inject the sample and obtain the chromatogram
shown in (a); switch to 100% aqueous for 30 min; switch back to mobile phase for
5 min; inject the sample and obtain the chromatogram shown in (b); repeat the
first three steps; chromatogram returns to (a). Column: 150 mm × 4.6 mm Eclipse
XDB-C8; mobile phase: 90% 50 mM KH2PO4 (pH 3.5), 10% acetonitrile; flow rate:
1.0 mL/min; temperature: room temperature. Peaks: 1 = uracil, 2 = procainamide,
3 = N-acetylprocainamide, 4 = N-propionylprocainamide, 5 = caffeine.
ES339296_LCA1113_029.pgs 10.17.2013 17:38 ADV blackyellowmagentacyan
LC•GC Asia Paciàc November 201330
COLUMN WATCH
the point of solvent mixing to the head
of the column. There are many volumes
to flush out before the actual gradient
reaches the column. For high-pressure
mixing, the volume of the mixing tee,
mixer (if present), pulse damper, pressure
transducer, all the connecting tubings,
injector including the loop or by-pass
volume, and guard column can be
substantial. For low-pressure gradient
systems, the proportioning valve, inlet
and outlet check valves, the pump piston
chamber, and various tubing adds even
more to the gradient delay volume. More
recently, newer UHPLC instruments have
addressed these problems by greatly
decreasing the instrumental contributions
to gradient delay volume as well as
extracolumn volumes.
The reequilibration study came up with
several conclusions. There are two types
of equilibrium: Repeatable equilibrium
and full equilibrium. Repeatable
equilibrium means that full equilibrium
may not have been achieved, but on
a practical basis, if the retention time
repeatability on subsequent runs is less
than 0.002 min then for non-ionizable
solutes in unbuffered eluents and for
basic compounds using the popular
trifluoroactic acid and formic acid
additives, repeatable equilibrium can be
achieved within two column volumes.
For a non-endcapped phase, 1% (v/v)
n-butanol added to the mobile phase was
required to achieve rapid full equilibrium in
two column volumes. For an endcapped
phase, the n-butanol was not required.
Myth 6: Superàcially Porous (Solid-Core) Particles Have a Signiàcantly Lower Sample Loading Capacity Compared to Totally Porous ParticlesFalse: The sample capacity of an HPLC
packing material is proportional to the
available surface area, which, of course,
is related to the amount of bonded phase
chemically attached to the available
silanols through monomeric bonding.
If one goes through the mathematical
calculations of the volume of a 2.7-µm
spherical totally porous particle (TPP)
and compares the volume of a 2.7-µm
superficially porous packing (SPP) with
a 0.5-µm porous shell, the total volume
available on the SPP particle is about 25%
less than the TPP. This assumes that the
porous portion of both particles has the
same characteristics, which may not be
the case. Nevertheless, the loss in surface
volume of the SPP is nowhere near that
of the pellicular packings of yesteryear
in which the shell thickness was 1–2 µm
and the particle size was 45–50 µm. If the
surface area of the SPP is actually larger
than that of the TPP, then the difference in
sample capacity between the two could
be less (this may be the case).
Rather than relying on mathematical
estimates, experiments were actually
performed on comparing the TPP
and SPP with similar chromatographic
conditions for basic compounds. For the
basic compound dextromethorphan,
successively larger concentrations were
injected onto the four columns. Three of
the columns were SPP columns Poroshell
120 EC-C18 (100 mm × 3.0 mm, 2.7-µm
dp, Agilent Technologies); Ascentis
Express C18 (100 mm × 3.0 mm, 2.7-µm
dp, Sigma Aldrich/Supelco); and Kinetex
C18 (100 mm × 4.6 mm, 2.6-µm dp,
Phenomenex). The TPP column was the
Zorbax Eclipse Plus C18 (100 × 3.0 mm,
1.8-µm dp, Agilent Technologies). At the
time, only a 100 mm × 4.6 mm Kinetex
column was available, so adjustments
were made in the experimental
conditions to accommodate the different
column dimensions.
One definition of overload is when
the sample size injected causes a 10%
dropoff in efficiency (or alternatively a 10%
increase in peak width). One can view
the results of the experiment in Figure 6,
which shows a plot of the peak width
versus sample concentration injected
at a constant volume. A 10% increase
in peak width for the SPP columns
occurred at roughly the same sample
loading as for the totally porous column
at a concentration value of 0.05 mg/
mL indicating a comparable sample
capacity for both types of columns.
Interestingly, a recent similar study
by Fallas and colleagues (10) came
to the same conclusion. Table 2 is an
abbreviated extract of data from their
work, which shows that, for both basic
and acidic compounds on the Poroshell
120 EC-C18 and the totally porous Zorbax
Eclipse Plus C18 columns of the same
dimensions and same conditions used,
has nearly the same sample capacity.
Interestingly, other SPP columns and
porous particle columns were evaluated
in the same study. All of them weren’t very
Table 1: Phases to address the dewetting problem in reversed-phase chromatography.
Polar-embedded alkyl phases (such as amide, urea, carbamate, ether, other polar
function)
Hydrophilic, polar-endcapped, and polar-enhanced stationary phases (such as AQ,
hydroxyl, amide)
Non-endcapped, short chain alkyl phases
Long chain alkyl phases (such as C30)
Wide-pore diameter phases
0 10 15 5 0 10 15 5
(a) (b)
Time (min) Time (min)
Figure 5: Consistent retention of procainamides. Sequence of events: condition with
a 50:50 mixture of phosphate buffer and acetonitrile for 15 min; run mobile phase for
5 min; inject the sample and obtain the chromatogram shown in (a); switch to 100%
aqueous for 30 min; switch back to mobile phase for 2 min; inject the sample and
obtain the chromatogram shown in (b). Column: 150 mm × 4.6 mm, 5-µm dp Zorbax
SB-Aq; mobile phase: 90% 50 mM KH2PO4 (pH 3.5), 10% acetonitrile; flow rate:
1.0 mL/min; temperature: room temperature. Peaks: 1 = uracil, 2 = procainamide,
3 = N-acetylprocainamide, 4 = N-propionylprocainamide, 5 = caffeine.
ES339384_LCA1113_030.pgs 10.17.2013 18:18 ADV blackyellowmagentacyan
Please fill in ovals as shown: ●1. I am employed by: (fill in ONE only)0001 ◯ Private Industry0002 ◯ University/College0003 ◯ Government0004 ◯ Hospital0005 ◯ Medical Center0006 ◯ Research Lab/Institute/Foundation0007 ◯ Independent Analytical Lab0008 ◯ Utility Company0220 ◯ Other (please specify) ___________________________
2. My function is: (fill in ONE only)0010 ◯ Research & Development0020 ◯ Quality Control/Assurance, Validation0030 ◯ Purchasing0040 ◯ Manufacturing/Processing0050 ◯ Teaching0060 ◯ Marketing/Sales0070 ◯ Corporate Management0080 ◯ Regulatory0090 ◯ Analysis0100 ◯ Technical Services0120 ◯ Lab Management0150 ◯ Other (please specify) ___________________________
3. My primary field of work is: (fill in ONE only)L ◯ PharmaceuticalF ◯ EnvironmentalE ◯ Energy/Petroleum/FuelsJ ◯ Medical/Biological/ClinicalG ◯ Forensics/NarcoticsM ◯ Plastics/Polymers/RubberA ◯ Agricultural/Food/Beverage3 ◯ Inorganic Chemicals4 ◯ Organic chemicalsH ◯ Instrumentation Design/
DevelopmentV ◯ Other (please specify)
_____________________________
4. The chromatographic techniques I use are: (fill in ALL that apply)A ◯ HPLCW ◯ UHPLCB ◯ Gas ChromatographyC ◯ Thin Layer ChromatographyD ◯ Gel ElectrophoresisE ◯ Ion ChromatographyF ◯ Size-Exclusion ChromatographyG ◯ Supercritical Fluid ChromatographyH ◯ Preparative/Process hromatographyI ◯ Low-Pressure ChromatographyJ ◯ Chiral ChromatographyK ◯ Affinity ChromatographyL ◯ Capillary ElectrophoresisM ◯ Solid Phase ExtractionN ◯ Supercritical Fluid ExtractionR ◯ LC-MSS ◯ GC-MSX ◯ Flash ChromatographyQ ◯ Other (please specify)
_____________________________
Signature Date
Mr/Mrs/Ms First name Family name
Job Title Company
Address
Postcode City Country
E-Mail.
Tel.
Fax
Sign here if you wish to receive (continue to receive) LC•GC Asia Pacific FREE
Print your name and address
Advanstar Communications provides certain customer contact data (such as customers’ names, addresses, phone numbers and e-mail addresses) to third parties who wish to promote relevant products, services and other opportunities which may be of interest to you. If you do not want Advanstar Communications to make your contact information available to third parties for marketing purposes, please contact us by emailing [email protected] or by mail at Advanstar Communications, 13/F, 88 Hing Fat Street, Causeway Bay, Hong Kong.
LC TROUBLESHOOTING
Problem solving
GC–MS
Quantitative analysis of catalyst
poisoners
THE ESSENTIALS
Multidimensional GC
September 2012
Volume 15 Number 3
www.chromatographyonline.com
Detecting herbicides in tap water
Biomolecule AnalysisUsing reversed-phase liquid
LC TROUBLESHOOTING
The role of the injection solvent
GC CONNECTIONS
Developments in SPME
COLUMN WATCH
Achiral stationary phases
November 2012
Volume 15 Number 4
www.chromatographyonline.com
FREE*
Please let us have your email address – so we can keep our renewal costs down
*Free to qualifying readers in Asia Pacific.
Answer all questions by filling in the circles
1113HA
Visit our new e-learning site dedicated to the analytical chemistry
community:www.CHROMacademy.com
Please quote your referencenumber if you have it:
Apply today foryour subscriptionSubscribe FREE by completing and mailing this form today! Mail completed page to LCGC Asia Pacific,131 W. 1st Street, 5th Floor,
Duluth, MN 55802, USA –OR– Fax completed page to 1-218-740-6417 — or — 1-218-740-6437
◯ YES! I wish to receive (continue receiving) LC/GC Asia. ◯ No◯ YES! I would also like to receive LCGC’s E-newsletter - LCGC Asia Pacific Update E-newsletter
ES339513_LCA1113_031_FP.pgs 10.17.2013 21:34 ADV blackyellowmagentacyan
LC•GC Asia Pacià c November 201332
COLUMN WATCH
Phyllis Brown: A
Story of Serendipity,
Perseverance, Vision,
and Chromatography
Karyn M. Usher
interviewed Phyllis
Brown, a pioneer at the forefront of high
performance liquid chromatography in
the 1960s. Article and podcast available
on-line. bit.ly/PhyllisBrown
Security Vetting:
Detecting Medicine
Residues in Meat
The Column spoke
to John Points from
LGC about analysing
veterinary residues in the food chain,
the role of the National Reference
Laboratory, and recent food scares in
the media. bit.ly/13CeGW8
Analysis of the State of the
Art: Liquid Chromatography
Instrumentation
A panel of experts were invited to assess
the current state of the art of liquid
chromatography
instrumentation and
to try and predict
future developments.
bit.ly/16HTrsJ
The LCGC Blog:
An Excel Tutorial
for Modeling
Chromatographic
Separations
In this LCGC Blog post,
Kevin Schug describes a chromatography
tutorial paper written by Sandy Dasgupta
and Akinde Kadjo in which an Excel
macro (developed by Kadjo) that is
capable of visualizing the development of
chromatographic separations is explained.
bit.ly/17wUn0A
www.chromatographyonline.com
‘Like’ our page LCGC
Stay Connected With LCGC
Follow us @ LC_GC
Join the LCGC LinkedIn group
The highlights of UHPLC/HPLC editorial
content from chromatographyonline.com
LCGC ONLINE
different, indicating the sample capacity
of a number of porous and SPPs when
tested under the same conditions were
essentially the same. However, with some
of the newer SPPs with a thinner shell
thickness, there may be different sample
capacities than TPPs of the same size.
References(1) R.E. Majors, LCGC North Am. 24(11),
1172–1182 (2006).
(2) R.E. Majors, “Top Ten LC Column Myths,
Lecture PL2” presented at HPLC 2013,
Amsterdam, Amsterdam, The Netherlands,
2013.
(3) R.E. Majors, LCGC North Am. 25(1), 31–39
(2012).
(4) L.R. Snyder and J.W. Dolan, LCGC North Am.
22(12), 1146–1152 (2004).
(5) L.R. Snyder and J.W. Dolan, LCGC North Am.
23(2), 118–127 (2005).
(6) M. Przybyciel and R.E. Majors, LCGC North
Am. 20(6) 516–523 (2002).
(7) R.E. Majors and M. Przybyciel, LCGC North
Am. 20(7), 584–593 (2002).
(8) A.P. Schellinger, D.R. Stoll, and P.W. Carr, J.
Chromatogr. A 1192(1), 41–53 (2008).
(9) A.P. Schellinger, D.R. Stoll, and P.W. Carr, J.
Chromatogr. A 1192(1), 54–61 (2008).
(10) M.M. Fallas, S.M.C. Buckenmaier, and D.V.
McCalley, J. Chromatogr. A 1235, 49–59
(2012).
“Column Watch” Editor Ronald E. Majors
is a senior scientist at the Columns and
Supplies Division, Agilent Technologies,
Wilmington, Delaware, USA, and is a
member of the LC•GC Asia Pacific editorial
advisory board. Direct correspondence
about this column should be addressed
to “Column Watch”, LC•GC Asia Pacific,
4A Bridgegate Pavilion, Chester Business
Park, Wrexham Road, Chester, CH4 9QH,
UK, or e-mail the editor-in-chief, Alasdair
Matheson, at [email protected]
Table 2: Sample capacity for two C18 columns.
20 mM ammonium formate (pH 3)
Column C0.5 (mg/L) Nortriptyline C0.5 (mg/L) 2-NSA
Zorbax Eclipse Plus C18 (TPP, 1.8 µm) 97 163
Poroshell 120 EC-C18 (SPP, 2.7 µm) 100 128
100 mM ammonium formate (pH 3)
Column C0.5 (mg/L) Nortriptyline C0.5 (mg/L) 2-NSA
Zorbax Eclipse Plus C18 (TPP, 1.8
µm)300 472
Poroshell 120 EC-C18 (SPP, 2.7 µm) 360 451
Extracted and adapted from reference 10
0.8
0.7
0.6
0.5
0.4
Peak w
idth
(s)
0.3
0.2
0.1
0.001 0.01
HBr OCH3
CH3
N
pKa = 8.3
Concentration of dextromethorphan (mg/mL)
0.1 1
0
Figure 6: Sample loading of a basic compound (dextromethorphan) onto superficially
porous and sub-2-µm totally porous columns. A 10% increase in peak width for
the superficially porous particle columns occurs roughly at the same loading as
the 1.8-µm totally porous particle column. Mobile phase: 80% 25 mM Na2HPO4
buffer (pH 3.0), 20% acetonitrile; detection: UV absorbance at 205 nm; temperature:
30 °C. Columns: blue diamonds: 100 mm × 3.0 mm, 2.7-µm dp Agilent Poroshell 120
EC-C18; orange squares: 100 mm × 3.0 mm, 2.7-µm dp Supelco Ascentis Express
C18; green triangles: 100 mm × 4.6 mm, 2.6-µm dp Phenomenex Kinetex C18; yellow
X’s: 100 mm × 3.0 mm, 1.8-µm dp Agilent Zorbax Eclipse Plus C18.
ES339305_LCA1113_032.pgs 10.17.2013 17:39 ADV blackyellowmagentacyan
LC•GC Asia Pacifi c November 2013 33
ADVERTISEMENT FEATURE
Gamma-Hydroxybutyrate (GHB) is a naturally occurring
substance found in the human central nervous system. It is also
easily synthesized. GHB has been used medically as a general
anesthetic, as well as to treat conditions such as insomnia, clinical
depression, narcolepsy, and alcoholism. Illegally, it has been used
as an intoxicant or as an agent in Drug Facilitated Sexual Assaults
(DFSA) or to improve athletic performance. GHB has a very short
window of detection in urine and blood making it diff cult to detect
in support of date rape cases. Using hair samples has been shown
to be a viable alternative for detecting the presence of GHB above
endogenous levels (1). This application note describes the extraction
and subsequent analysis of GHB from decontaminated hair samples.
After incubation in methanol, the extract is dried and then dissolved in
deionized water (D.I.) prior to sample clean-up using anion exchange
SPE (CUQAX156). The recoveries are greater than 90%, and matrix
effects are less than 5%.
Sample Preparation
To a clean glass tube, add 100 mg of decontaminated hair
sample.
Add 1 mL of CH3OH and internal standard, vortex mix.
Incubate the sample at 40 °C for approximately 12 h.
Centrifuge sample at 3000 rpm for 10 min.
Transfer organic phase to a clean glass tube.
Evaporate to dryness <40 °C.
Dissolve residue in 3 mL of D.I. H2O (pH 7).
Vortex mix.
Sample Extraction
Condition the CLEAN-UP extraction column with 3 mL of CH3OH
followed by 3 mL D.I. H2O. Aspirate at <3 in. Hg to prevent the
sorbent bed from drying.
Load the sample onto the column at 1–2 mL/min.
Wash the column with 3 mL D.I. H2O followed by 3 mL of CH
3OH.
Dry the column for 10 min at >10 in. of Hg.
Analyte Elution
Elute the GHB with 2 mL × 3 mL aliquots of CH3OH w/ 6% acetic
acid. The eluate collection rate should be 1–2 mL/min.
Dry the Eluate
Evaporate the extract to dryness under nitrogen <40 °C. Then
reconstitute in 100 μL of mobile phase.
Determination of Gamma-Hydroxybutyrate (GHB) in Hair Samples Using Solid-Phase Extraction and LC–MS–MSJeffery Hackett, UCT
UCT, LLC 2731 Bartram Road, Bristol, Pennsylvania19007, USA
Tel: (800) 385 3153
E-mail: [email protected]
Website: www.unitedchem.com
Sample Preparation Products
CUQAX156CLEAN-UP® SPE Column - Quaternary Amine w/
Chloride Counter Ion, 500 mg/6 mL
LC–MS–MS Method
System: Agilent 1200 LC
Injection: 10 μL
LC column: Thermo Fisher Gold C18, 50 mm × 2.0 mm, 1.9 μm
Column temp: 40 ºC
Mobile phase: Acetonitrile w/ 0.1% formic acid: D.I. H20 w/ 0.1% formic
acid; (50:50)
Flow rate: 0.2 mL/min
LC–MS–MS Conditions
Detector: API 4000 MS/MS
Conclusion
This method offers analysts working in the area of DFSA a viable,
eff cient method for determining the presence of GHB in hair
samples. The isolation and quantif cation of this drug (performed
by SPE and LC–MS–MS) is a robust alternative to GC–MS where
chemical derivatization is required for the analysis of this compound.
Reference
(1) P. Kintz, V. Cirimele, C. Jamey, and B. Ludes, J. Forensic
Science 48(1), 195–200 (2003).
Ion Source ESI
Ion Mode Negative
Ion Spray Voltage - 4500V
Curtain Gas 10
Gas 1 40
Gas 2 40
CAD Gas Medium
Source Temp 650 °C
Mode Positive
XIC of -MRM(4 pairs): 103.020/57.000 Da ID: gnb from Sample 7 (H3) of GHB-SSQAXhair1123010.wiff (Turbo Spray)
XIC of -MRM(4 pairs): 109.130/90.000 Da ID: gnb-d6 from Sample 7 (H3) of GHB-SSQAXhair1123010.wiff (Turbo Spray)
Max. 1.7e4 cps
Max. 7490 ps
1.6e4
Inte
nsi
ty (
cps)
Inte
nsi
ty (
cps)
1.4e4
1.2e4
1.0e4
8000.0
6000.0
74907000
6000
5000
4000
3000
2000
1000
0
2000.0
1.18
1.23
4000.0
0.00.5 1.0 1.5 2.0 2.5
Time (min)3.0 3.5 4.0 4.5
0.5 1.0 1.5 2.0 2.5Time (min)
3.0 3.5 4.0 4.5
Figure 1: GHB chromatogram.
Table 1: Mass spec table.
Compound RT (min) Precursor Product 1 Product 2
GHB 1.23 103.0 57.0 84.0
GHB-D6 1.18 109.1 90.0 60.9
ES339341_LCA1113_033.pgs 10.17.2013 17:40 ADV blackyellowmagentacyan
34 LC•GC Asia Pacifi c November 2013
ADVERTISEMENT FEATURE
Polydimethylsiloxane (PDMS) is the world’s most common silicone.
Its applications range from contact lenses and medical devices
to elastomers, caulking, lubricating oils, and heat resistant tiles.
For all of its applications, the weight-average molar mass (and
its distribution) is directly associated with the performance of the
product. A DAWN multi-angle light scattering (MALS) detector
coupled with a size-exclusion chromatograph (SEC) provides the
perfect tool for making molecular weight determinations without
reference to standards or column calibration.
For this application note, a polydimethylsiloxane sample was
analysed by SEC in toluene, using Wyatt Technology’s DAWN
and an Optilab refractometer as the respective MALS and
concentration detectors.
Figure 1 shows the chromatograms of polydimethylsiloxane with
signals from the light scattering at 90° (top) and the RI (bottom)
detectors. The RI signal is negative because the refractive index
increment (dn/dc) of polydimethylsiloxane in toluene is negative.
A positive signal can be obtained if the polarity of the signal output
is reversed. Because the light scattering signal is proportional to
dn/dc squared, its signal is positive.
By combining the DAWN and Optilab data, the absolute
molar masses of this siloxane were calculated without making
any assumptions about the polymer’s conformation or elution
time.
A polystyrene standard with a molar mass of 200 kD was
analysed under the same conditions, as it is frequently used to
calibrate columns for conventional chromatography. Both results
are plotted in Figure 2. Even though polydimethylsiloxane is a linear
polymer, just as this polystyrene standard is, the molar masses at
the same elution time are not identical for the two polymers.
If polystyrenes had been used as calibration standards, the molar
mass for polydimethylsiloxane would have been erroneous. The
results once again demonstrate the power of MALS in determining
absolute molar masses of polymers without any reference to
calibration routines or polymer standards — even when those
polymers appear to share the same conformation as the standards.
SEC-MALS of Silicones Wyatt Technology Corporation
Wyatt Technology Corporation6300 Hollister A venue, Santa Barbara, California 93117, USA
Tel: +1 (805) 681 9009 fax: +1 (805) 681 0123
Website: www.wyatt.com
90˚ light scattering signal
Optilab RI signal(negative dn/dc)
Figure 1: Chromatograms obtained by SEC of a PDMS sample with signals from the DAWN (top, red) and the Optilab RI (bottom, blue).
90˚ LS for
PDMS
260K
PDMS
200K
PS
PSSTD
Figure 2: Plots of the molar mass versus elution time superimposed over the signals from the DAWN, for the PDMS sample and the polystyrene “standard”, showing the large errors associated with conventional column calibration.
ES339340_LCA1113_034.pgs 10.17.2013 17:40 ADV blackyellowmagentacyan
ES339540_LCA1113_CV3_FP.pgs 10.17.2013 21:37 ADV blackyellowmagentacyan
www.gerstel.com
Perfect Flavor and Perfectly Safe?
Dynamic Headspace(DHS), Headspace
Extraction, SPE,addition of standards
SPME and Headspace
Thermal Desorptionand Twister™ (SBSE)
Olfactory DetectionPort (ODP)
MAESTRO PrepAheadproductivitiy
Effi ciently automated solutions
for GC/MS and LC/MS analysis:
Flavor and Fragrance
Off-fl avors
Pesticide Residues (QuEChERS)
Glyphosate / AMPA
Contaminants, Mycotoxins
Migration from packaging
For the highest product quality, you can rely on
GERSTEL solutions for GC/MS and LC/MS
ES339514_LCA1113_CV4_FP.pgs 10.17.2013 21:34 ADV blackyellowmagentacyan