May 2017 Volume 30 Number 5 www.chromatographyonline.com LC TROUBLESHOOTING Increasing resolution by changing selectivity Drug Target Discovery Combining HIC, SEC, and IEX with fluorescence polarization GC CONNECTIONS GC products review MULTIDIMENSIONAL MATTERS Miniaturized LC× LC and HRMS C E L E B R A T I N G O U R 3 0 t h Y E A R C E L E B R A T I N G O U R 3 0 t h Y E A R 1987-2017 1987-2017
52
Embed
Drug Target Discovery - files.alfresco.mjh.groupfiles.alfresco.mjh.group/alfresco_images/pharma/...Dec 19, 2018 · QUICK TUTORIAL Fundamentals of HPLC Video Training Course This
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
May 2017 Volume 30 Number 5
www.chromatographyonline.com
LC TROUBLESHOOTING
Increasing resolution by changing selectivity
Drug Target DiscoveryCombining HIC, SEC, and IEX with
fluorescence polarization
GC CONNECTIONS
GC products review
MULTIDIMENSIONAL
MATTERS
Miniaturized LC×LC and HRMS
CE
LEBRATING OUR 30th YEA
RCEL
EB
RATING OUR 30th YEAR
��
1987-20171987-2017
��
��
����
��
��
227www.chromatographyonline.com
‘Like’ our page LCGC Join the LCGC LinkedIn groupFollow us @ LC_GC
magnetic resonance (NMR) spectroscopy (9), microscale
thermophoresis (MST) (10), and fluorescence polarization
(FP) (11). Of these, FP is arguably the simplest, and is
quite inexpensive.
FP measurements are based on rotational differences
between an unbound and bound fluorescent
(FLU)-tagged molecule, and can be briefly explained
as follows: a drug rotates freely and rapidly in a
solution, but rotates slower when bound to a much
larger protein; the FP instrument can detect this
rotational difference and hence a binding. Parameters
that can affect the performance of FP measurements
include fluorescence intensity, sample viscosity, and
quenching of drug fluorescence. For more details on
the technique, see Figure 1 and references 12, 13,
and 14. Although FP is frequently used for measuring
binding strength and kinetics (15), it has not been used
in drug deconvolution with biosamples because it has
limitations regarding mixtures. FP should in principle
detect a drug-binding in a protein mixture, but cannot
tell which protein is involved. Therefore, we wanted to
test the approach of chromatographically resolving
proteins in a mixture before FP measurements. FP has
previously been described as a detection device with
capillary electrophoresis (CE) techniques (16–18), but
not with liquid chromatography (LC) (to our knowledge).
We examined a number of chromatographic principles
(size-exclusion chromatography [SEC], hydrophobic
interaction chromatography [HIC], and ion exchange
chromatography [IEX]) that are suited for separating
proteins without perturbing their biological activity and
could be combined with FP. We evaluated LC and FP with
two FLU-tagged drugs that antagonize the Wnt pathway
(“161-FLU” and “XAV-FLU”, see Figure 2[a and b]
and Figure 3), a signaling cascade system strongly
associated with, for example, colon cancer, and a current
focus in drug discovery (19,20). Emphasis was placed
Combining HIC, SEC, and IEX with Fluorescence Polarization for Drug Target DiscoveryTore Vehus1, Jo Waaler2, Stefan Krauss2, Elsa Lundanes1, and Steven Ray Wilson1, 1Department of Chemistry, University
of Oslo, Blindern, Oslo, Norway, 2Unit for Cell Signaling, Oslo University Hospital, Rikshospitalet, Oslo, Norway
Fluorescence polarization (FP) is a highly regarded technique for studying drug–protein interactions, but has limited value regarding protein mixtures. As a novel approach to drug target discovery, the possibility of combining FP with liquid chromatography (LC) was explored. Nondenaturing protein LC principles such as size-exclusion chromatography (SEC), hydrophobic interaction chromatography (HIC), and ion exchange chromatography (IEX) were found to be orthogonal and compatible with FP because the mobile phases used do not negatively affect detection. For simple protein mixtures, the SEC/HIC/IEX–FP approach was able to identify tankyrase as the target of a triazole-based inhibitor of the Wnt signaling pathway, which is heavily associated with colon cancer. However, the total peak capacity of the three LC dimensions was not sufficient to resolve at cell-proteome level, calling for higher resolution of intact proteins to enable stand-alone drug target discovery with LC and FP.
KEY POINTS• SEC, IEX, and HIC are orthogonal.
• Fluorescence polarization is compatible with SEC,
IEX, and HIC.
• SEC/IEX/HIC–FP enables protein–drug interaction
measurement in a mixture.
• The total peak capacity of SEC/IEX/HIC has to be
strengthened.
Ph
oto
Cre
dit: vs1
48
/Sh
utt
ers
toc
k.c
om
LC•GC Europe May 2017232
NovaFFF SoftwareNovaFFF Software
ICP-MSICP-MSRIRIUVUVDLSDLSMALSMALS
The FFF - MALS Platform Next Level Nano, Bio and Polymer Analysis
Contact us for more information: www.postnova.com
NEW
With SEC Option!
on evaluating the orthogonality of the three separation
principles (crucial for multidimensional separations prior
to FP), optimizing the peak capacity of the LC principles,
and ensuring that LC and FP are compatible.
ExperimentalChemicals: Type 1 water (resistivity [MΩ•cm @ 25 °C]
>18.0) was from a Milli-Q ultrapure water purification
system (Millipore). 10K cut-off 500 μL, 2 mL, or 5 mL
ultracentrifugation filters were also from Millipore.
the JW74-molecule (21), where an amine functionalized
polyethylene glycol was added. Labelling of 161 with
FLU was done according to the NHS-FLU manufacturer’s
protocol, and preparation of TNKS2a according to (22).
Cell Culturing: The SW480 wild type cell line was
cultured in Leibowitz’s L-15 medium containing 10%
(v/v) FBS and 1% (v/v) PS. To detach cells, T-EDTA
was used. Cells were harvested when they were ~90%
confluent. Prior to cell lysis, cells were counted using a
hemocytometer. Lysis of cells was performed by adding
400 μL cell lysis buffer per 107 cells with 50 mM PMSF.
Tubes were immersed in a sonication bath at 4 °C for
1 min × 5 with a 20 s delay between sonications. Cell
debris was removed with centrifugation at 13000 × g for
10 min and supernatant was transferred to eppendorf
tubes. Cell lysates were snap-frozen in liquid nitrogen,
and stored at -80 °C until use.
Liquid Chromatography: All protein LC experiments
were performed with a PerkinElmer Series 200 pump and
autosampler. UV absorption was measured at 280 nm
with a SPD-10AV UV detector (Shimadzu). Column
temperature was regulated with a Mistral-column oven
LC•GC Europe May 2017234
Vehus et al.
Polarized light
Drug
Depolarized light
Drug
Polarized light
Polarized light
Fast rotation
Slow rotation
Protein
(a)
(b)
Figure 1: Fluorescence polarization measurement on (a) a freely rotating fluorescent drug leading to depolarized light, and (b) a slowly rotating drug–protein complex maintaining a polarized light.
161-FLU/XAV-FLU/G007-LK
TNKS
Axin2 GSK3
CK1
Cell membrane
ß-catenin
ß-catenin
APC
ß-catenin destruction complex
Nucleus
Transcription factors
Prote
asom
e
Figure 3: Chemical structure of 161-FLU and XAV-FLU molecules.
(a) : 161-FLU
(b) : XAV-FLU
Figure 2: Wnt-pathway antagonism through inhibition of TNKS1/2 and stabilization of beta-catenin destruction complex leading to proteasomal degradation of beta-catenin and suppression of transcription factors.
DA – Unique Polyaromatic PhaseC18 –Traditional Reverse PhaseAqueous C18 – Traditional Reverse Phase with Enhanced SelectivityC8 – Less Retentive than Standard C18 ColumnPFPP – Operates in Reverse Phase, Normal Phase, or HILIC ModeEtG – Maximum retention for polar EtG/EtS alcohol metabolites
(Spark Holland) or a PerkinElmer Series 200 Peltier
column oven. Fraction collection was done with a Gilson
FC204 fraction collector.
IEX separations were performed on a 4.6 × 200 mm
PolyCATWAX mixed-bed ion-exchange column (PolyLC
Inc.) temperature regulated to 20 °C with an injection
volume of 100 μL. The column temperature was set to
4 °C. Mobile phase A (MP A) contained 20 mM MES
and mobile phase B (MP B) contained 20 mM MES and
0.8 M NaCl. For most separations the elution (at 1 mL/
min) started at 10% B for 10 min, then linear gradient
increased to 52.5% B in 20 min and held for 2 min at
100% B. Equilibration was done for 30 min at 10% B
between analyses.
HIC separations were performed on a 4.6 × 100 mm
ProPac HIC-10 column from Dionex (Thermo Fisher
Scientific). The column temperature was 20 °C. Injection
volume was 100 μL. MP A contained 2 M ammonium
sulfate, 0.05 M sodium phoshate pH regulated to 7.0, and
MP B contained 0.05 M sodium phoshate pH = 7.0. Most
gradients started at 0% B for 2 min with a linear gradient
elution to 75% B in 18 min and a hold for 3 min at 100% B.
Equilibration was done for 30 min at 0% B between analyses.
SEC separations were performed with a 4.6 × 35 mm
TSKgel-SuperSW guard column (Tosoh Corp.) coupled
in-line with a 4.6 × 300 mm TSKgel SuperSW3000
SEC column (Tosoh Corp.). Separations were done
isocratically with a flow rate of 0.35 mL/min and a mobile
phase containing 0.05 M sodium phosphate and 0.3 M
NaCl pH adjusted to 7.0 or 0.05 M sodium phosphate
and 0.3 M L-Arginine pH adjusted to 7.0.
The durations of fraction collections are shown in figure
legends and varied for each separation principle and
sample type. After fraction collection, the fractions were
(when needed) concentrated with cut-off filters of various
(Tecan Group Ltd.) using either 96- or 384-well plates
(Flat black polysterol, Greiner Bio One GmbH). For
fluorescein-labeled compounds, excitation was
performed at 485 ± 20 nm and emission measured at
535 ± 25 nm with cut-off filters. The number of flashes
was set to 25 and emitted light was integrated for 40 μs.
The settle time before measurements was 60 ms. Four
wells with water were used as instrument blanks. Aliquots
LC•GC Europe May 2017236
Vehus et al.
140
120
100
80
60
40
Tagged drug Tagged drug +binding protein
Tagged drug + non-binding protein
Flu
ore
scen
ce p
ola
riza
tio
n (
mP)
Figure 4: Fluorescence polarization drug binding identification of 200 μL 10 nM 161-FLU with 20 μL 0.2 mg/mL TNKS2a as binding protein and 20 μL 0.15 mg/mL ribonuclease B as non-binding protein added to a 96-well plate (n = 4).
240.0
200.0
160.0
120.0
80.0
40.0
Non-binding protein TNKS2a Cell lysate TNKS2a + Cell lysate
Flu
ore
scen
ce p
ola
riza
tio
n (
mP)
Figure 6: Fluorescence polarization between 161-FLU and non-binding protein (ribonuclease B), binding protein (TNKS2a), and SW480 wt cell lysate. 200 μL 10 nM 161-FLU and 10 μL protein samples in buffer were added. The TNKS2a and cell lysate concentration was 0.3 mg/mL and 21 mg/mL, respectively. In TNKS2a added cell lysate, the TNKS2a concentration was 0.3 mg/mL and the cell lysate concentration 21 mg/mL. Spread bars shown (n = 2).
Figure 7: Orthogonality plot for proteins separated in SEC, IEX, and HIC.
80
P < 0.018 pmole ≈ mLOD
P < 0.011 pmole > mLOD
8 pmole 1 pmole
70
60
50
Flu
ore
sce
nce
po
lari
za
tio
n (
mP
)
Non-binding protein (ribonuclease B)
Mole protein added to 2 pmole 161-FLU
Binding protein (TNKS2a)
Figure 5: Fluorescence polarization between 161-FLU and TNKS2a–ribonuclease B as a function of protein amount. The assay was performed in a 96-well plate with 200-μL 10 nM drug added to each well, and 20 μL protein dissolved in buffer so that the protein:drug ratio was 0.5 and 4. ** Significant difference at P < 0.01.
of 200 μL of the chromatographic fractions solution
were added to the 96-well plates, and 30 μL to the
384-well plates.
LC–MS/MS: LC–tandem mass spectrometry (LC–MS/MS)
experiments were performed as described in reference
23.
237www.chromatographyonline.com
Vehus et al.
(a) (b)SEC
IEX
(c) (d)
0
0
5 10
10 20 30 40
0 105 15 20 25 30
120
110
100
90
80
70
60
50
40
Fraction number
Flu
ore
scen
ce p
ola
rizati
on
(m
P)
180
160
140
120
100
80
90
80
163 17
70
60
50
40
60
40Flu
ore
scen
ce p
ola
rizati
on
(m
P)
Flu
ore
scen
ce p
ola
rizati
on
(m
P)
15 1 2 3 4 5 6 7 8
Fraction number
Fraction number
1 2 3 4 5 6 7
A (
280 n
m)
A (
280 n
m)
A (
214 n
m)
HIC
(e) (f)
Frac no #
tR (min)
tR (min)
tR (min)
Frac no #
Frac no #
Figure 8: (a) SEC–UV chromatogram of a 100 μL TP sample separated on a 4.6 mm × 300 mm size-exclusion column with flow rate of 0.35 mL/min. Mobile phase consisted of 0.05 M sodium phosphate + 0.3 M NaCl at pH 7. The column was temperature regulated at 20 °C. UV absorbance was measured at 280 nm. Eight fractions were collected from 4.5 min with 2 min in each. *TNKS2a containing fractions. (b) Fluorescence polarization binding study between XAV-FLU and TP fractions from SEC.
*TNKS2a containing fractions. (c) IEX–UV chromatogram of 100 μL TP sample separated at 1 mL/min using gradient elution on a 4.6 mm × 200 mm mixed-bed ion-exchange column. The column was temperature regulated at 20 °C and UV absorbance was measured at 280 nm. MP A and MP B contained 20 mM MES and 20 mM MES + 0.8 M NaCl, both pH regulated to 6. Gradient start was at 10% B for 10 min, with a linear increase up to 52.5% B in 20 min, a wash-out step at 100% B for 2 min, and equilibration at 10% B for 30 min. Seven fractions were collected from 0 to 42 min, with 6 min in each. *TNKS2a containing fractions. (d) Fluorescence polarization binding study between XAV-FLU and TP fractions from IEX. *TNKS2a containing fractions. (e) HIC–UV chromatogram of 100 μL TP sample separated on a 4.6 × 100 mm Propac HIC-10 column. The column was temperature regulated at 20 °C and flow was set to 1 mL/min and the UV absorbance measured at 214 nm. The samples were eluted with 100% A for 2 min followed by a linear decrease to 25% A in 18 min, and a 3 min wash-out at 0% A. Equilibration for 30 min was carried out between analysis at 100% A. MP A contained 2 M ammonium sulfate + 0.05 M sodium phosphate, and MP B contained 0.05 M sodium phosphate, both pH adjusted to 7. A total of 22 fractions was collected in 1.5 min intervals from 0 to 33 min. *TNKS2a containing fractions. (f) Fluorescence polarization binding study between XAV-FLU and TP fractions from HIC. (Fractions with UV-signal from pure TNKS2a sample were collected, thus only three fractions were tested from the other samples to prove that TNKS2a retains its biological activity in a protein standard sample after HIC). *TNKS2a containing fractions.
Results and DiscussionFP as a Tool for Detecting Drug–Protein Interaction:
As a model for this initial investigation, we chose to use
FLU-tagged tankyrase inhibitors. Inhibition of tankyrase
allows axin proteins to engage in a so-called destruction
complex that degrades beta catenin for moderation of Wnt
signaling (Figure 3). TNKS2a (recombinant, binding part of
tankyrase) and ribonuclease B (non-binding) were added
to 161-FLU. For the TNKS2a–161-FLU solution, the FP
signal increased approximately 3.5 times, whereas with the
non-binding protein added, the signal did not significantly
increase (Figure 4). An additional step for verifying a
drug–protein interaction “FP event” is to add a competing,
non-fluorescent drug, which will reduce the FP signal as
a result of the unbinding of the FLU-tagged drug; this was
observed when adding G007-LK (22) (see Supplementary
Figure 1; supplementary information can be found at
polarization-drug-target-discovery), also performed
in subsequent experiments. Hence, FP was suited for
assessing binding between TNKS2a–161-FLU.
Sensitivity: An approach to detect drug–protein interactions
must be sensitive because biotargets may be present in
minute amounts. Varying amounts of TNKS2a were added
to 2 pmoles 161-FLU (Figure 5). It was found that the
addition of 4 times more moles TNKS2a compared to the
drug gave a significant change in FP value, thus indicating
that the mass limit of detection (mLOD) in this example was
about 8 pmoles; this sensitivity of the TNKS2a–161-FLU FP
experiment was considered satisfactory.
Binding in Complex Samples: To confirm that FP could
detect a binding protein in a complex sample such as a
cell lysate (but not tell which protein is binding), an FP
assay with 161-FLU in SW480 cell lysate (contains human
tankyrase-2) with and without TNKS2a was performed.
Indeed, the FP value increased when a whole cell lysate
was added to 161-FLU, implying that one or more proteins
present in the sample binds to 161-FLU (Figure 6). It is very
likely that this is at least partly a result of tankyrase binding,
because this drug variant is highly selective (21). When the
cell lysate was spiked with TNKS2a, the FP signal increased
further, implying that the drug was initially present at higher
mole ratios compared to the target protein. We concluded
that FP experiments can function for complex samples,
in that a binding can also be detected in the presence of
non-binding proteins.
FP Measurement Robustness in Common Solvents for
Protein LC: To pinpoint which protein in a complex mixture
is causing an FP-event with a drug, they must be separated
prior to measurement. To maintain the tertiary structure of
proteins (typically necessary for interaction with a drug),
nondenaturing chromatographic conditions must be used.
This is achievable with HIC, SEC, and IEX. Substantial
amounts of salts are present in the mobile phase used
for these principles, often varying in composition during
a solvent gradient. It was feared that FP signals could be
affected by varying salt amounts, because this could affect
viscoscity. However, FP-events were largely unaffected by
the mobile phases of HIC, SEC, and IEX (Supplementary
Figure 2). Hence, mobile phases of common nondenaturing
protein LC principles were quite compatible with FP.
Estimating the Orthogonality of HIC, IEX, and
SEC: Cell lysates constitute a typical matrix for drug
deconvolution. However, a single LC separation will not
be able to separate all proteins in a biological sample,
which contains thousands of different proteins. Hence, a
multidimensional approach is called for, which dramatically
increases the resolution if the dimensions have a degree
of orthogonality (24). The orthogonality of HIC, IEX, and
SEC for protein separation was therefore evaluated. Model
proteins chosen featured a range of isoelectric points (pI)
and molar masses (MM). An orthogonality plot (Figure 7)
shows an obvious deviation from linearity in normalized
retention times, indicating that the separation principles are
indeed highly orthogonal. Hence, HIC, IEX, and SEC were
LC•GC Europe May 2017238
Vehus et al.
(a)
(b)
(c)
5
1 2 3 4 5 6 7 8
10
12
100
90
80
70
60
50
10
8
6
20 30 40 50
Frac no #
A (
28
0 n
m)
Fraction number
Peptide ID 1
–COOH
H2N–
Peptide ID 2
Flu
ore
sce
nce
po
lari
za
tio
n (
mP
)
tR (min)
Figure 9: (a) SEC–UV chromatogram of 100 μL SW480 wt cell lysate separated on a 4.6 mm × 300 mm size-exclusion column with a flow rate of 0.35 mL/min. Mobile phase consisted of 0.05 M sodium phosphate + 0.3 M NaCl at pH 7. The column was temperature regulated at 20 °C and UV absorbance was measured at 280 nm. Eight fractions were collected from 6 min with 4.25 min in each. (b) Fluorescence polarization binding assay between 161-FLU and SEC fractions collected of SW480 cell lysate. FP assay performed in 384-well format, with 30 μL 1 μM 161-FLU and 10 μL of spin filtrated fractions added to each well (n = 1). (c) Partial amino acid sequence of human tankyrase 2 identified in SEC fraction 2 highlighting peptides (green) identified with LC–MS/MS.
suited for multidimensional LC separation of proteins for FP
measurements.
HIC/IEX/SEC–FP with a Simple Protein Mixture: A
protein mixture was chromatographed in parallel on HIC/
IEX/SEC, and fractions were mixed with 161-FLU and
FP-measured (Figure 8[a–f]). FP-event fractions were
digested with trypsin and a peptide separation and
identification was performed with LC–MS/MS, confirming
the presence of TNKS2a in all the FP-event fractions. None
of the other proteins were identified in all FP-event fractions
(Supplementary Figure 3), and this was in accordance with
the known retention times of the proteins in the various
dimensions. Hence, a simple proof of concept for identifying
a drug target in a mixture, using multidimensional LC and
FP, was demonstrated.
Peak Capacities: The previous experiment showed that
HIC/IEX/SEC–FP could be used to pinpoint drug-binding
proteins in simple mixtures. At this stage, the system can
be functional, for example, as a step following protein
pull-down with immobilized drug columns (25). We
wanted, however, to assess the potential for an LC and
FP-only approach to drug deconvolution of very complex
samples. For maximum resolution, peak capacity was
optimized. Regarding IEX, the peak capacity of a 180-min
long ammonium acetate gradient with the conditions
described was approximately 35 (Supplementary
Figure 4). In HIC, the peak capacity was about 20
(with a 20-min long gradient, and did not substantially
increase with longer gradients) for the 200-mm long
column when applying a decreasing ammonium sulfate
gradient (Supplementary Figure 5). The peak capacity
for SEC was estimated to be 5 with phosphate buffer and
sodium chloride in the MP (Supplementary Figure 6) (in
later experiments NaCl was exchanged with L-arginine
to increase recovery from SEC, but this did not affect the
peak capacity significantly [data not shown]). Combining
the maximum peak capacity from each column with a
generously estimated orthogonality of 0.9, the overall
peak capacity of the system was 20 × 35 × 0.9 × 5 ×
0.9 = 2835. This number is well below the number of
proteins in a biological samples (which can be on the
ten thousand-scale), and an additional separation and
isolation step seems to be necessary, or further peak
capacity enhancements. The approach was, however,
investigated on protein extract from cells, which provided
a strong FP-event signal in fraction number 2 for SEC–FP
(Figure 9).
ConclusionsAs a novel approach to drug deconvolution, FP has
been shown to be compatible with nondenaturing liquid
chromatography. HIC, SEC, and IEX provided orthogonal
separation prior to FP measurements, allowing for drug
targets to be pinpointed in a mixture. For the technique to
be a completely stand-alone technique for very complex
cell samples, the total peak capacity of protein
chromatography must be strengthened; the development
and application of higher resolving columns than
those used here are key. If this level is reached, the
technique can have potential for drug target discovery
for (fluorescent) compounds, even without the need for
immobilization-based approaches.
References(1) G.C. Terstappen, C. Schlupen, R. Raggiaschi, and G.
Gaviraghi, Nature Review Drug Discovery 6, 891–903
(2007).
(2) E. Tashiro and I. Imoto, Bioorganic & Medicinal Chemistry 20,
1910–1921 (2012).
(3) J.N.Y. Chan, C. Nislow, and A. Emili, Trends in
Pharmacological Sciences 31, 82–88 (2010).
(4) J.N.Y. Chan et al., Molecular & Cellular Proteomics 11(7),
M111.016642 (2012).
(5) G.M. West, L. Tang, and M.C. Fitzgerald, Analytical Chemistry
80, 4175–4185 (2008).
(6) J. Lee and M. Bogyo, Current Opinion in Chemical Biology 17,
118–126 (2013).
(7) R. Ghai, R.J. Falconer, and B.M. Collins, Journal of Molecular
Recognition 25, 32–52 (2012).
(8) J.A. Maynard et al., Biotechnology Journal 4, 1542–1558
(2009).
(9) A.L. Skinner and J.S. Laurence, Journal of Pharmaceutical
Sciences 97, 4670–4695 (2008).
(10) S. Patnaik et al., Journal of Medicinal Chemistry 55,
5734–5748 (2012).
(11) F. Perrin, Journal de Physique et le Radium 7, 390–401
(1926).
(12) A.M. Rossi and C.W. Taylor, Nature Protocols 6, 365–387
(2011).
(13) W.A. Lea and A. Simeonov, Expert opinion on drug discovery
6, 17–32 (2011).
(14) N.J. Moerke, Current Protocols in Chemical Biology 1(1), 1–15
(2009).
(15) T.G. Dewey, Ed., Biophysical and Biochemical Aspects of
Fluorescence Spectroscopy (Springer, 1991).
(16) Q.-H. Wan and X.C. Le, Analytical Chemistry 72, 5583–5589
(2000).
(17) E. Ban and E.J. Song, Journal of Chromatography B 929,
180–186 (2013).
(18) L. Ye, X.C. Le, J.Z. Xing, M. Ma, and R. Yatscoff, Journal of
Chromatography B: Biomedical Sciences and Applications 714,
59–67 (1998).
(19) J.N. Anastas and R.T. Moon, Nature Reviews Cancer 13, 11–26
(2013).
(20) B.T. MacDonald, K. Tamai, and X. He, Developmental cell 17,
9–26 (2009).
(21) J. Waaler et al., Cancer Research 71, 197–205 (2011).
(22) A. Voronkov et al., Journal of Medicinal Chemistry 56,
3012–3023 (2013).
(23) M. Rogeberg, S.R. Wilson, H. Malerod, E. Lundanes, N.
Tanaka, and T. Greibrokk, Journal of Chromatography A 1218,
7281–7288 (2011).
(24) M. Gilar, P. Olivova, A.E. Daly, and J.C. Gebler, Analytical
Chemistry 77, 6426–6434 (2005).
(25) S. Sato, A. Murata, T. Shirakawa, and M. Uesugi, Chem. Biol.
17, 616–623 (2010).
Tore Vehus is an Assistant Professor at Department
of Engineering Sciences, University of Agder, Norway.
He graduated with a M.Sc. in analytical chemistry
from University of Oslo (UiO) in 2012 and is currently
finishing a Ph.D. under Ass. Prof. Steven R. Wilson at
UiO. His work is currently focused on the development
of instrumentation and chromatographic columns for
fit-for-all analyses of complex biological samples.
Jo Waaler is a Researcher at Oslo University Hospital.
Stefan Krauss is a Professor at Oslo University
Hospital.
Elsa Lundanes is a Professor at the Department of
Chemistry at the University of Oslo.
Steven Ray Wilson is an Associate Professor at
the Department of Chemistry, University of Oslo.
Research interests have included miniaturization,
multidimensional separations, and on-line
hyphenatations with MS and NMR. Application fields
are in cancer research (diagnostics and drug discovery)
and neuroscience.
239www.chromatographyonline.com
Vehus et al.
LC•GC Europe May 2017240
Ph
oto
Cre
dit: B
an
an
a R
ep
ub
lic Im
ag
es/S
hu
tte
rsto
ck.c
om
(U)HPLC: The Shape of ThingsTo Come
Gert Desmet, Department of Chemical
Engineering, Free University of
Brussels, Belgium
A recent argument was raised in the scientific press that
in pursuit of greater speed and separation resolution,
ultrahigh performance liquid chromatography (UHPLC) is
faced with practical limitations and will struggle with its
own version of Moore’s law (1).
This empirical law was first proposed to describe
the long-term progress made in the micro-electronics
industry. Moore’s law states that speed and memory
storage capacity are roughly doubling every two years.
Progress is occurring by shrinking the distance between
the transistors on the chips to cram even more of them on
the same surface. However, the current spacing between
the transistors is already down to a dazzling small 22 nm,
and most theoretical models predict that the fundamental
laws of physics will prevent the distance being reduced
below 10–7 nm. It is clear that Moore’s law will one day
run into a hard stop and bring a halt to the advances in
speed and data storage if the electronics industry does
not find a new paradigm to store and manipulate data.
A gloomy parallel was drawn with (U)HPLC to
emphasize that this field has been witnessing a Moore’s
law-type of progress in speed and resolution over the
past decade. This progress was essentially realized
by making increasingly smaller particles, and it was
suggested that (U)HPLC is also facing the end of
practical progress with its own version of Moore’s law.
Most specialists agree that with pressure limits entering
a range where the compressibility of the liquid makes
it harder to precisely control the flow rate and where
viscous heating threatens to become unacceptably high,
we have now reached the stage of what can practically
be achieved by particle size reduction.
Slip flow technology has been suggested as a possible
way out of this, but its promises still need to be achieved
in practice (2). Sub-micron particles may also be able
to realize the ultra-rapid separations (in the order of a
few seconds) needed in the final dimension of the best
possible three-dimensional LC (3D LC separations, but
this is likely to remain a very niche application for a
long time.
However, the limits of Moore’s law in (U)HPLC only
relate to packed beds of spherical particles. We should
Hot Topics in Separation ScienceA series of short articles exploring current trends in separation science that will be addressed at the HPLC 2017 conference in Prague, Czech Republic, from 18–22 June 2017.
not forget the sphere is only one of the many shapes
that are possible. Just think of monoliths, perfusion
particles, and pillar arrays. Measured by Golay’s and
Knox’s separation impedance number, these are far
better shapes than the packed bed of spheres and hold
the promise of a 10-fold increase in efficiency (for the
same time) and even a 100-fold reduction of the analysis
time (for the same efficiency). These approaches have
not delivered their promise yet, some because of the
lack of order and some because the size of the individual
elements is still too large to reach their performance limit
in a range of practical times or efficiencies—and some
still suffer from both problems.
However, with new materials engineering possibilities,
such as silicon micromachining and 3D printing, rapidly
gaining widespread availability, it is highly possible
we will one day see a commercially viable production
technology that will be able to produce the perfect
chromatographic column, breaking away from Moore’s
law by trading our spherical particles for supports with a
much more advantageous shape as measured by Golay’s
and Knox’s separation impedance.
Let us not forget how this field recently defeated
Moore’s law already, with the (re-)introduction of
core–shell particles (representing a fundamental
change of the particle design) leading to a large gain
in speed and resolution. So, let us be optimistic and
consider that maybe the next 50 years will be the era
of support shape, rather than of support size. And with
exciting contributions on the possibilities of silicon
micromachining and 3D printing on the programme,
the HPLC 2017 conference could be the start of this
new era.
References(1) M.S. Reisch, C&E News 94(24), 35–36 (2016).
(2) B.A. Rogers, Z. Wu, B. Wei, X. Zhang, X. Cao, O. Alabi, and
M.J. Wirth, Anal. Chem. 87, 2520−2526 (2015).
The Role of LC–MS in Lipidomics
Gerhard Liebisch, Institute of Clinical
Chemistry and Laboratory Medicine,
University of Regensburg, Germany
Lipidomics, the analysis of lipids by mass spectrometric
methods, revolutionized lipid science (1). It provides
LC–MS provides separation of lipids and reduces the
complexity of the matrix. It typically provides a higher
sensitivity than shotgun and offers retention time as an
additional parameter to identify lipid species. Therefore,
low abundant lipid mediator species are typically
analyzed by targeted LC–tandem mass spectrometry
(MS/MS) (5). Lipidomic methods apply both reversed
phase with nonpolar as well as normal phase and
hydrophilic interaction chromatography (HILIC) with polar
selectivity.
Reversed phase separates lipids species based on
their hydrophobic moiety, that is, the hydrocarbon chain
for most lipid classes. This allows the separation of
hundreds of lipid species based on their chain lengths
and double bond number (6,7). The retention behaviour
follows certain rules and increases the confidence of
lipid species identification. This permits the separation
of isomeric lipid species with different acyl chains like
PC 18:1_18:1 (phosphatidylcholine with two acyl chains
containing 18 carbon atoms and 1 double bond) and
PC 18:0_18:2 (18:0 and 18:2 acyl chain). However,
quantification in lipidomics typically relies on lipid
species not present in the samples. In reversed-phase
chromatography most of the lipid species and internal
standards elute at different times, thus experience
different matrix effects and different solvent composition,
which influences their ionization and may result in
inaccurate quantification (8).
The polar selectivity of normal phase and HILIC
provides lipid class-specific separation. This has great
advantages in terms of quantification because analytes
and internal standards show similar retention times.
Moreover, identification of lipid species comprising a lipid
class is straightforward. In contrast to reversed-phase
methods, separation of acyl chain isomers is not usually
possible by normal phase and HILIC methods, but
lipid class-specific separation may resolve isomers like
bis(monoacylglycero)phosphate and phosphatidylglycerol
(9). A promising approach is the application of
polar stationary phases with ultrahigh-performance
supercritical fluid chromatography (UHPSFC), which
offers ultrafast separation for quantitative analysis of
multiple lipid classes (10).
Today, an increasing number of studies are reporting
poor quality lipidomics data with misidentification
and inaccurate or inappropriate quantification of lipid
molecules. These studies primarily use untargeted
metabolomics approaches (11) and the reasons for the
poor data quality include analytical, bioinformatics,
lipid species and opens new possibilities to gain an
insight into lipid biology. This helps not only to explain
the vital role of lipid species as membrane building
blocks, but also to unravel their bioactive functions.
Thus, lipid species can act as signaling molecules and
modulate membrane properties, which are essential for
organelle and membrane protein function. Moreover,
the first examples demonstrated their potential as novel
biomarkers to monitor human health (2).
Lipidomics research is based on two main approaches:
direct infusion mass spectrometry (DIMS) analysis
(shotgun lipidomics) and liquid chromatography (LC)–MS
analysis. In direct infusion analysis a crude lipid extract
is infused into the mass spectrometer and lipid species
identification relies on specific precursor ion, neutral
loss scans (3). The main advantage of infusion-based
analysis is its simplicity and the straightforward way
of quantification. Analytes and internal standards are
present in the same sample matrix and thus experience
the same ion suppression and matrix effects. Shotgun
analysis is therefore able to provide comprehensive,
quantitative lipidomes, as for example demonstrated for
yeast (4). The application of high mass resolution, MSn,
and derivatization–gas phase reactions can provide
detailed lipid structures. However, the application
of shotgun approaches is limited in sensitivity and
separation of isomeric lipid species. Moreover, in-depth
characterization of lipid species in crude lipid extracts
may be complicated by co-isolation of precursor ions.
Lipidomics revolutionized lipid science. It provides detailed quantitative information on hundreds of lipid species and opens new possibilities to gain an insight into lipid biology. This helps not only to explain the vital role of lipid species as membrane building blocks, but also to unravel their bioactive functions.
243www.chromatographyonline.com
Hot Topics in Separation Science
UHPLC Coupled with Accurate Mass and High Resolution Mass Spectrometry for Complex Environmental Analyses
E. Michael Thurman and
Imma Ferrer, Laboratory
of Environmental Mass
Spectrometry, University of
Colorado, Boulder, Colorado,
USA
Environmental analyses of food, soil, and water have
changed dramatically over the last decade. Topics such
as pesticides, food additives, and natural products
have become important as food products are globally
grown and distributed (1). Monitoring their quality is
critical to international business. Pharmaceuticals,
fluorinated surfactants, and endocrine disruptors in
water are major new topics, where not only parent
compounds are unknown but also their metabolites
and degradation products are often more important
or more abundant than the parent compound (2). New
environmental issues, such as hydraulic fracturing
and its wastewater, have captured our attention as the
production of oil and gas has increased exponentially
in the past decade (3). With this technology comes
the problem of wastewater disposal and groundwater
contamination. These environmental issues have greatly
benefited from the combination of ultrahigh-performance
liquid chromatography (UHPLC) mated to high resolution
mass spectrometry (HRMS). Because suppression by
matrices creates challenges in environmental analysis,
both sample preparation, such as solid-phase extraction
(SPE), and UHPLC make important contributions to
eliminating or reducing suppression.
The analysis of environmental samples has challenged
our ability to separate the thousands of compounds
that are present in a food or water extract. Furthermore,
the salts and metal ions associated with these extracts
further complicate the analytical challenges. HRMS,
such as time-of-flight MS and Orbitrap MS, has been
adopted by many laboratories to address these pressing
environmental issues. To gain the most from HRMS,
UHPLC has been rapidly accepted as a separation
method. In particular, the use of sub-2-μm particles in
a variety of packing materials has enabled the mass
spectroscopist to fully appreciate the power of HRMS
and accurate mass by separating compounds of isobaric
mass, as well as isotopes of various compounds that
have the same identical mass. Even the highest resolving
power in mass spectrometry will not separate two isomers
that have the same formula; thus, UHPLC plays a critical
role in separation and identification of environmental
targets. A separation by UHPLC then allows the use of
MS/MS followed by accurate mass for identification.
For example, pharmaceuticals in water may have not
only identical formulas (isomers) but may also have
nearly identical MS/MS spectra. The analysis problem
of tramadol and desvenlafaxine are just that problem
(4). The use of UHPLC using a C-8 column was easily
able to separate these two isomers, such that they
could be identified correctly (4). The importance of
these pharmaceuticals is that they may contribute to
the formation of dimethylnitrosoamine (NDMA), which is
an important new chlorination product created in water
treatment (5).
A valuable mass spectrometry technique, auto
MS/MS, is available from many vendors of high-resolution
mass spectrometers. The peak capacity of the analytical
column used in UHPLC is valuable for this type of
analysis because it gives the auto MS/MS spectra a more
easily interpreted accounting of the unknowns present in
a food or water extract. The MS/MS spectra may then be
correlated to various libraries currently available without
the interference of matrix materials.
Another reason to increase the resolution in
chromatography before mass spectrometry is the
current availability of complex databases. With accurate
mass analysis, it is possible to create a database of
accurate masses for any compounds that one would
like to investigate in an environmental sample, that is,
of course, if that compound will ionize in either positive
or negative ion electrospray. However, the limitation of
HRMS does not stand alone in the conundrum of environmental analysis. The power of separation methods brings us ever closer to fully characterizing the environmental pollutants in food, soil, and water.
and educational aspects. Therefore, it is necessary
to implement reporting standards for lipidomics data
to share with the scientific community (12). These
standards need to cover both shotgun and LC–MS
approaches. Only the application of both approaches
in a complementary and confirmatory way permits a
comprehensive and accurate coverage of the lipidome.
References(1) K. Yang and X. Han, Trends Biochem. Sci. 41(11), 954–969
(2016).
(2) S. Sales et al., Sci. Rep. 6, 27710 (2016).
(3) X. Han, K. Yang, and R.W. Gross, Mass Spectrom. Rev. 31(1),
134–78 (2012).
(4) C.S. Ejsing et al., Proc. Natl. Acad. Sci. U.S.A 106(7),
2136–2141 (2009).
(5) G. Astarita et al., Biochima. et Biophys. Acta 1851(4), 456–68
(2015).
(6) M. Ovcacikova et al., J. Chromatogr. A 1450, 76–85 (2016).
(7) K. Sandra and P. Sandra, Curr. Opin. Chem. Biol. 17(5), 847–53
(2013).
(8) S. Krautbauer, C. Buechler, and G. Liebisch, Analytical
Chemistry 88(22), 10957–10961 (2016).
(9) M. Scherer, G. Schmitz, and G. Liebisch, Analytical Chemistry
82(21), 8794–8799 (2010).
(10) M. Lisa and M. Holcapek, Analytical Chemistry 87(14), 7187–95
(2015).
(11) G. Liebisch, C.S. Ejsing, and K. Ekroos, Clinical Chemistry
61(12), 1542–1544 (2015).
(12) G. Liebisch et al., Biochimica. et Biophysica. Acta (2017).
E. Michael Thurman
Imma Ferrer
LC•GC Europe May 2017244
Hot Topics in Separation Science
the database lies in the fact that sometimes as many as
a thousand isomers may exist for a formula, such as a
simple fungicide of the elements, C, H, N, and O. How
does one tackle this problem? One powerful technique
is slow chromatography with high peak capacity and
reproducibility of retention time. This allows one to use
retention time in the mass spectrometry database to
accurately pull out targeted compounds and radically
decrease the false positives caused by isomeric
compounds.
Another important advance in mass spectrometry is
the use of ion mobility for the separation of surfactants
associated with wastewater from hydraulic fracturing.
The complexity of surfactants adds to the hundreds of
ions associated with a single group of compounds. These
surfactants are used as clay stabilizers and emulsifiers to
move oil and gas from deep underground to the surface.
The combination of a heatmap generated by UHPLC
versus IM drift time is a powerful visual tool to see and
identify new groups of compounds present in wastewater
samples. This is especially important in that these
surfactants may contribute to earthquake occurrence
when these wastewaters are disposed of by deep well
injection, a common technique in the United States (6).
Thus, HRMS does not stand alone in the conundrum
of environmental analysis. The power of separation
methods, including UHPLC and advances in sample
preparation (solid-phase extraction and other sample
preparation tools), brings us ever closer to fully
characterizing the environmental pollutants in food, soil,
and water. HPLC 2017 will highlight and encourage us
in the field of environmental analysis to continue this
interesting journey in high resolution chromatography.
References(1) E.M. Thurman et al., Anal. Chem. 78, 6703–6708 (2006).
(2) M. Strynar et al., Environ. Sci. Technol. 49, 11622–11630 (2015).
(3) Y. Lester et al., STOTEN 512, 637–644 (2015).
(4) I. Ferrer and E.M. Thurman, J. Chromatogr. A 1259, 158–166
(2012).
(5) D. Hanigan et al., Environ. Sci. Technol. Lettrs. 2, 151–157
(2015).
(6) W.L. Ellsworth, Science 341, 142 (2013).
Advances in Glycomics in Biology and Medicine
Milos V. Novotny, Department of Chemistry,
Indiana University, Bloomington, Indiana, USA
and Regional Centre for Applied Molecular
Oncology Masaryk Memorial Oncological
Institute, Brno, Czech Republic
The importance of glycosylated structures in modern
biology and medicine has been beyond dispute for
many years, but there are still gaps in biochemical
understanding. The current realization that virtually
all major human diseases have been associated with
glycosylation changes demands in-depth structural
studies of these highly complex glycobiomolecules.
Glycoscience with its many directions and a broad
scope in both prokaryotic and eukaryotic systems is
currently securing its place at the centre stage of modern
biological research (1).
However, the enormous complexity of glycoconjugate
molecules and the abundance of glycosylated proteins
in biological fluids and tissues present significant
challenges to modern analytical methods and
measurement technologies. As underscored in the
2012 report of the National Research Council to the
U.S. National Academies (2), developing new tools for
glycoscience is one of the highest priorities of the general
scientific inquiry.
The rapidly growing fields of glycomics and
glycoproteomics reflect these on-going efforts.
Biomolecular mass spectrometry (MS) is today the central
identification and measurement technique for glycans
and glycopeptides. The contemporary MS features
state-of-the-art instrumentation in terms of ionization
and fragmentation techniques, highly resolving mass
analyzers, and very sensitive detection. However, the
field of analytical glycobiology also urgently needs
high-performance separation methods to assist MS
measurements because of (a) the sheer complexity
of the mixtures generated during various cleavages
of glycoprotein molecules; and (b) glycan isomerism,
which is both biologically relevant and methodologically
difficult.
The role of separation scientists to develop better
analytical methods and reliable protocols to study
differences in glycosylation (for example, “normal” versus
“aberrant” glycosylation levels) seems secure for a long time
to come. Starting with sample preparation, fractionation, and
preconcentration, and ending with the discrete resolution
of glycoconjugates before MS measurements, many of
these tasks are accomplished through chromatographic
principles. Rapid advances are increasingly seen in
many of these vitally important tasks of glycomics and
glycoproteomics. Carbohydrate derivatization at microscale,
such as permethylation or fluorescent labeling, are often
desirable to enhance identification and measurements.
Recent advanced methods in analytical glycoscience have
been reviewed (3,4).
Glycomic profiling measurements have been
particularly important in a search for disease biomarker
candidates. Additionally, glycosylation analysis of
therapeutic glycoproteins is becoming increasingly
essential to evaluate their bioactivity, safety, immune
response, and solubility. While deglycosylation protocols
for N -linked glycans have been reliably developed
during recent years, the same still cannot be said about
quantitative analysis of O -linked oligosaccharides,
although some progress has recently been made. The
general profiling procedures may involve matrix-assisted
For investigation of biomolecular interactions, the
following ACE methods are available: The first mode is
nonequilibrium ACE of equilibrated mixtures of analyte and
ligand at different analyte–ligand ratios in the background
electrolyte (BGE) free of ligand and analyte can be used
for strong or slowly dissociating complexes. From the peak
areas of the analyte, ligand, and analyte–ligand complex,
their equilibrium concentrations and the stability constant of
the complex can be determined.
The second mode is the dynamic equilibrium ACE of
an analyte in the BGE containing a free ligand at several
distinct concentrations. Electrophoretic migration of the
analyte is retarded as a result of the formation of the
analyte–ligand complex. The advantage of this mode, the
“so-called” mobility shift assay, is that the analyte need not
be perfectly pure (the admixtures can be separated during
the ACE experiment) and concentration of the analyte
need not be exactly known since estimation of the stability
constant is based on measurement of analyte effective
mobility.
Partial filling ACE (PF-ACE) is a special ACE mode in
which only a part of the capillary is filled with ligand solution
in the BGE. This technique has several advantages over
classical ACE. Instead of adding the ligand at several
concentrations to the BGE in the whole capillary and in one
or both electrode vessels, the use of a short ligand zone
means that the consumption of the valuable ligand is very
low. The binding constants can be calculated from a slope
of linear dependence of analyte migration time changes on
the substance amount of ligand in the ligand zone in the
capillary.
In frontal analysis ACE (FA-ACE), a long zone of
equilibrated mixture of analyte–ligand complex at different
analyte–ligand ratios is introduced into the capillary, and
in the applied electric field the complex dissociates and
the zones of free analyte, analyte–ligand complex, and
free ligand are formed. From the heights of the analyte or
ligand zones on the electropherograms, their equilibrium
concentrations, binding constants, and stoichiometry of the
complexes can be determined. These parameters can also
be determined by a special mode of FA-ACE, continuous
FA-ACE, in which the analyte–ligand complex at different
analyte–ligand ratios is electrokinetically introduced in the
capillary.
In ACE with immobilized ligand, the ligand is covalently
or by physical sorption attached to the inner capillary wall
and the analyte is electrophoretically or electroosmotically
transported through this affinity open-tubular column. The
strength of the analyte–ligand interaction is evaluated from
the reduced electrophoretic mobility of the analyte as a
result of its interaction with the ligand.
References(1) R.K. Harstad et al., Anal. Chem. 88, 299–319 (2016).
(2) H.M. Albishri et al., Bioanalysis 6, 3369–3392 (2014).
(3) P. Dubský, M. Dvorák, and M. Ansorge, Anal. Bioanal. Chem.
408, 8623–8641 (2016).
(4) S. Štepánová and V. Kašicka, J. Sep. Sci. 38, 2708–2721
(2015).
The 45th International
Symposium on High Performance
Liquid Phase Separations and Related
Techniques (HPLC 2017) will be
held in Prague, Czech Republic,
from 18–22 June 2017.
Website: www.hplc2017-prague.org
Václav Kašicka, Institute of Organic
Chemistry and Biochemistry of the Czech
Academy of Sciences, Prague, Czech
Republic
Affinity Capillary Electrophoresis—A Powerful Tool to Investigate Biomolecular Interactions
FOR MORE INFORMATION ABOUT OUR GPC INSTRUMENTS, COLUMN TECHNOLOGY AND APPLICATION
SOLUTIONS FOR SEC/GPC ANALYSIS CONTACT US AT +49 6155 7043700 OR VISIT WWW.ECOSEC.EU
OVER 40 YEARS OF EXPERIENCE IN GPC & SEC
COLUMNS FOR AQUEOUS & ORGANIC SOLVENTS
EcoSEC - 7th GENERATION COMPACT GPC INSTRUMENT
EcoSEC HT HIGH TEMPERATURE SYSTEM UP TO 220 °C
YOUR SPECIALIST IN SEC/GPC ANALYSIS
LC•GC Europe May 2017250
LC TROUBLESHOOTING
This is the third in a series of “LC
Troubleshooting” instalments
that consider how we can use
the fundamental properties of
chromatographic separations to
estimate the impact of different
variables on liquid chromatography
(LC) separations. In the first
instalment (1) we looked at the
column plate number, N, and saw
that it was not as useful a tool as
we might guess, but concluded
that a column with a plate number
of N ≈ 10,000 is a good place to
start. Last month (2), we considered
retention, expressed as the retention
factor, k. It is best to adjust retention
for 2 ≤ k ≤ 10. If the sample won’t fit
in 1 ≤ k ≤ 20, a gradient method is
likely a better choice.
As in the previous two instalments,
we’ll use the fundamental resolution
equation as a guide:
Rs = ¼N0.5 (α-1) [k/(1+k)] [1]
(i ) (ii ) (iii )
This month we’ll focus on
selectivity, α, (or peak spacing) as
expressed in term ii of equation
1, where α is the selectivity factor
between two peaks with k-values of
k1 and k2:
α = k2/k1 [2]
Several variables can be adjusted
to change α. It is not surprising that
some of these variables are more
effective and some are easier to
change than others. By looking at
the effects of different variables
on the separation, we can “count
the cost” of different choices and
choose a balance of the various
costs (effectiveness, time invested,
expense, and so forth) that fit our
requirements.
Orthogonal SeparationsWhen we focus on selectivity during
LC method development, we are
looking for ways to move peaks
relative to each other. This means
changing values of α (equation 2) by
changing the k-value of one or both
peaks under consideration. When
more than two peaks are present, it
may be necessary to move several
peaks relative to each other so that
satisfactory separation is obtained.
As part of the process of improving
selectivity, we’d like to choose a
variable (for example, solvent type
or pH) that has a high probability
of making the desired change.
One way to compare changes in
selectivity is to plot the retention of
the various sample components,
expressed as log k, for two different
variables, as in Figure 1.
In Figure 1(a), the retention
of the sample components
(blue diamonds) falls close to a
straight line, with a coefficient
of determination, r2, of 0.98. If
the slope of the line were 1.0, all
components would have the same
retention with either variable 1
or variable 2. If r2 is close to 1,
but the slope is not, the relative
retention would be approximately
the same, but little or no change
in α would be observed. When
the coefficient of determination is
close to one, we can refer to the
separation conditions as equivalent.
We might see this if the retention
on two different C18 columns were
compared. This approach would be
desirable if we want two columns
that can be used interchangeably,
with one of them designated as
a backup column for a method.
However, this is not a desirable
situation if we want to choose
a variable that will improve the
separation of a hard-to-separate
peak pair.
Figure 1(b) shows more scatter
for the points around the trend
line than Figure 1. This might
occur if we changed the organic
solvent (B-solvent) type in a
reversed-phase method, where
variable 1 is acetonitrile and variable
2 is methanol. Note that when two
adjacent points lie on opposite sides
of the trend line, a retention reversal
has occurred for those two peaks.
For example, the point circled in
red is eluted before the one circled
in green with variable 1, whereas it
comes out second with variable 2.
As the scatter about the trend line
increases r2 drops, the likelihood
of significant changes in relative
retention (selectivity) increases,
also. When the separations under
two conditions are quite different,
we refer to the separation as
orthogonal. (Yes, technically to
be orthogonal, the coefficient of
Count the Cost, Part 3: Increasing Resolution by Changing SelectivityJohn W. Dolan, LC Troubleshooting Editor
Several variables can be used to change selectivity in a liquid chromatography (LC) separation. Here we compare the variables in an effort to prioritize which experiments will be most effective.
HPLC 2017 in Prague
Visit us at booth #13
Kromasil columns are available with
small particles to achieve the most
selectivity and resolution possible when
running on an HPLC system. With sizes
down to 1.8 μm, you can also enable the
full column efficiency using UHPLC
systems.
Small particles are available in many of
the Kromasil product lines:
• Classic premium silica phases for RP
chromatography and ClassicShell
columns with solid-core silica particles.
• EternityXT long-life columns for harsh
conditions and their EternityShell solid-
core particles equivalent.
• SFC columns for green technologies.
Visit the Kromasil website or ask your
nearest distributor for the full details on
our small-particles columns.
www.kromasil.com/small-particles
Beyond expectations
Small particles
for HPLC and
UHPLC
LC•GC Europe May 2017252
LC TROUBLESHOOTING
determination should be zero, but
this is extremely unlikely with two
chromatographic conditions.)
Orthogonal LeverageAlthough we could use r2, we get
a better measure of the scatter of
the points, as in Figure 1(b), as a
function of the standard error of
the curve, which for the present
discussion I’ll call “orthogonal
leverage”. Orthogonal leverage is
the power of a variable to make
significant selectivity changes for
reversed-phase separations.
In Table 1, I’ve summarized
orthogonal leverage data for the
common variables we use to control
reversed-phase separations. This
table is based on a large database
of chemically diverse compounds
under different chromatographic
conditions (see reference 3 for
more discussion). When the
orthogonal leverage for a specified
change in a variable is ≥1, it is a
good choice to try to change α for
hard-to-separate peaks. Yes, this is
an average number that may or may
not apply to a specific sample, but
in our “count the cost” series, we’re
seeking knowledge that will allow us
to choose conditions that increase
the probability of success as well
as learn which conditions are likely
to be a waste of time. In each case,
data are shown for an arbitrary, but
experimentally reasonable, change
in the variable (3). Let me next
briefly interpret the data of Table 1.
%B and tG: %B refers to the
percentage of organic solvent in a
reversed-phase mobile phase, most
commonly methanol or acetonitrile.
In the discussion of k (2), the Rule of
Three was mentioned: We can
expect approximately a threefold
change in k for a 10% change in
the B-solvent. This percentage is
a reasonable amount of change
without being excessive. We can
see that this gives an orthogonal
leverage value of 0.8, which is less
than our target value of ≥1. However,
80% of the way to our goal isn’t bad,
and as we know from experience
and the discussion of k last month
(2), many times a change in %B is
sufficient to pull two peaks apart.
Also, this change is very easy, so
even though it is not as effective as
some other variables, it usually is
worth pursuing. A threefold change
in the gradient time, tG, will have a
similar effect on a gradient method
as a change of 10%B does in an
isocratic one (4). The two changes
also have approximately the same
effect on orthogonal leverage (0.7
versus 0.8). A change in gradient
time is easy and reasonably
powerful, so again we choose easy
over powerful and often implement a
change in gradient time early in the
method development process.
Column Temperature: A change in
the column temperature can change
the selectivity of a separation (5)
because of the affect of temperature
on k and mobile-phase pH. A
change in temperature of 20 °C
gives us an orthogonal leverage of
0.7. Once again, this value is less
than the desired ≥1, but because it
is easy, similar to a change in %B or
tG, it is often a variable we try early
in the method development process.
Another reason that we often
choose to investigate %B or tG
and °C early in our investigations
is that they are easily modelled
based on two experimental values,
much like we saw for k in (2). Thus
Table 1: Comparing “orthogonal leverage” for reversed-phase selectivity
%B: percent organic solvent in the mobile phase; tG: gradient time; Temperature: column temperature; Methanol (acetonitrile): changing from methanol to acetonitrile as the B-solvent; column: a change in column chemistry; Fs: quantitative comparison of column selectivity described in (7); pH: the pH of the aqueous component of the mobile phase; [buffer]: concentration of the buffer used to control mobile-phase pH.
*Desired to be ≥1.0; here “orthogonal leverage” is defined as 10 × |δlog α|avg or 14 × standard error (SE) as defined in (3)
†Only for ionics
Based on data of reference 3.
(a)
(b)
0.04
0.04
-0.04
-0.04
-0.4
0.04
0.40
0
0.08
0.08
-0.04
log
k,
va
ria
ble
1lo
g k
, va
ria
ble
1
log k, variable 2
log k, variable 2
r2 = 0.98
r2 = 0.25
0
0
Figure 1: Comparing retention between
two conditions. (a) Two variables with
similar (“equivalent”) retention; (b) two
variables with quite different (“orthogonal”)
retention. See text for details.
By looking at the effects
of different variables
on the separation, we
can “count the cost” of
different choices and
choose a balance of the
various costs that fit our
requirements.
253www.chromatographyonline.com
LC TROUBLESHOOTING
versa) as the mobile-phase organic
component usually has a significant
change in the peak spacing of
the chromatogram. This effect is
reflected in the value for orthogonal
leverage in Table 1 of 2.0, twice
the minimum desired target of ≥1.
However, although the solvent
swap is powerful, it can create
such large selectivity changes that
we may have trouble figuring out
which peaks correspond between
experimental runs at two values
of %B, tG, or °C allow prediction
of k and α for other values of that
variable. Counting the cost of the
experiments, these are very high
value, low cost choices.
Methanol–Acetonitrile: We all
know from experience, as well
as other “LC Troubleshooting”
instalments (for example, reference
6) that a change from using
methanol to acetonitrile (or vice
the methanol and acetonitrile
chromatograms. We have to balance
this challenge against the power
of moving the peaks. As a result,
we often delay investigation of the
effect of solvent type changes until
we have checked out the easier, but
less-effective variables noted above.
Column: A change from one column
to another can have a wide range
of results, varying from little or no
change in the separation to large
changes in α. The label on the
column does not necessarily give
us enough information to make
the decision. However, there are
now on-line databases, such as
the one discussed in reference 7,
Table 2: Ranking the variables
VariableChange
in αUniversal Convenient
Low-UV/LC–MS
Robustness Equilibration
%B 0 + + + + +
Temperature 0 (+) + + + + +
Solvent type + + + 0 + 0
Column type 0 (+) + 0 + + +
pH ++ - 0 0 - +
+: effective variable, positive characteristic; 0: less-effective variable, less desirable, still
CHROMacademy’s video training coursesHPLC Method Development - Nov
GC Method Development - Jan
Fundamentals of HPLC - March
Fundamentals of GCMS - May
Fundamentals of GC - July
Fundamentals of LCMS - Sept
All video courses run in a yearly cycle and are now included
in the price of a single Premier Membership, all for a fee of $399 per year.
LC•GC Europe May 2017256
GC CONNECTIONS
For its 69th session 5–9 March 2017,
the Pittsburgh Conference on Analytical
Chemistry and Applied Spectroscopy
(Pittcon) headed north to McCormick
Place in Chicago, Illinois, USA. In sharp
contrast to the many previous subzero
sessions held in Chicago, the weather
was unusually mild, so much so that I
was able to enjoy a Sunday morning run
along the Lakefront Trail. I passed by
the 17th Annual Chicago Polar Plunge
where hundreds of bewildered swimmers
encountered water colder than the air.
Attendance was up slightly compared to
last year with more than 14,000 registered
participants, 22% of whom came from 89
countries outside the United States. The
exposition hosted 78 exhibitors from 37
countries, 141 of which displayed their
products for the first time.
The technical programme continued
to be a strong part of the conference:
more than 2000 sessions were presented
in 72 symposia, 89 oral sessions, 19
contributed sessions, six workshops, 56
poster sessions, and 15 awards. Among
these were the 2017 LCGC Lifetime
Achievement in Chromatography Award
presented to Professor Pat Sandra
(Research Institute for Chromatography),
and the LCGC Emerging Leader in
Chromatography Award presented to Dr.
Deirdre Cabooter (University of Leuven).
Pittcon 2018 will meet in Orlando,
Florida, 26 February to 1 March at the
Orange County Convention Center, where
participants likely, but not necessarily, will
enjoy even balmier weather. The 2019
session is scheduled for 17–21 March in
Philadelphia, Pennsylvania.
This annual “GC Connections”
instalment reviews gas chromatography
(GC) instrumentation, columns, and
accessories shown at this year’s Pittcon
or introduced during the previous year.
For a review of new products in other
areas of chromatography, columns,
and related accessories, please see
the “Column Watch” and “Perspectives
in Modern HPLC” column instalments
from the April 2017 issue of LCGC
Europe (1,2) as well as the “Sample Prep
Perspectives” column in the May issue of
LCGC North America (3).
The information presented here is
based on manufacturers’ replies to
questionnaires, as well as on additional
information from manufacturers’ press
releases, websites, and product literature
about the past year’s products, and not
upon actual use or experience of the
author. Every effort has been made to
collect accurate information, but because
of the preliminary nature of some of
the material LCGC Europe cannot be
responsible for errors or omissions. This
column instalment cannot be considered
to be a complete record of all new GC
products introduced this year at Pittcon or
elsewhere because not all manufacturers
chose to respond to the questionnaire,
nor is all of the submitted information
necessarily included here because of the
limited available space and the editors’
judgment as to its suitability.
GC: 2016–2017 As the gas–solid variety of GC attains its
70th anniversary in 2017, this past year
had a number of notable advances in
GC technology that clearly demonstrate
its ongoing viability. Far from those
early university experiments, some of
the newest developments help remove
a number of significant obstacles in
routine GC while other advances deliver
even higher performance to the gas
chromatography–mass spectrometry
(GC–MS) realm. The most significant
development in GC instrumentation this
year has to be Agilent’s new Intuvo 9000
GC system, designed for reduced time
spent on routine tasks with minimized
maintenance risks. The core Intuvo
developments are outlined in the following
tables. Laboratories that adopt the system
will deploy columns in instrument-specific
directly heated modules, but this should
prove not to be a significant barrier to the
multiyear cycle of instrument upgrades
and replacements. Ellutia’s 500-Series
GC systems implemented a hybrid
air-oven column heating system, which
supports resistively heated as well as
conventionally heated columns. Qmicro
displayed application-specific versions of
their micro-GC platform, which is based
on microelectromechanical systems
New Gas Chromatography Products for 2016–2017John V. Hinshaw, GC Connections Editor
In this instalment, John Hinshaw presents an annual review of new developments in the field of gas chromatography (GC) seen at Pittcon 2017 and other venues in the past year.
Table 1: Companies introducing new
products in 2016–2017
Company Name
Agilent Technologies
CDS Analytical
Dani Instruments
Ellutia
LECO Corporation
Markes International
Peak Scientific
Phenomenex
Proton OnSite
Qmicro B.V.
Restek Corporation
Scion Instruments
Shimadzu Corporation
The 4S Company
Thermo Fisher Scientific
VICI DBS Ltd.
VUV Analytics
Valves, fittings, detectors, and more for chromatography and liquid handling
Markes International introduced its new xr series of thermal desorption (TD) instruments: The
TD100-xr 100-tube automated thermal desorber; the UNITY-xr single-tube thermal desorber;
the ULTRA-xr 100-tube autosampler for UNITY-xr; and the Air Server-xr on-line volatile organic
chemical (VOC) monitoring system. The samplers are designed for use with GC–MS analyzers.
A key new feature is an extended capability for automated sample splitting and re-collection
that allows valuable samples to be reanalyzed for method development and compliance with
standard methods. The samplers can recover compounds from C2 to C44, and they incorporate
a new water-management module for on-line monitoring of humid air streams. Markes also has
redesigned the instrument control software.
Qmicro B.V.
Explosion
proof micro GC
analyzer
This explosion proof (IECEx/ATEX certified) gas analyzer from Qmicro is based on the company’s
micro GC platform. Several custom, specifically configured applications are enabled by the
four-channel GC cartridge. The transmitter performs on-line gas analysis and heating value
computations and has external dimensions of 29 × 26 × 12 cm3. The replaceable GC cartridge
encompasses the injector and sample loop, MEMS micro thermal conductivity detectors with a
500 ppb detection limit, GC column, and zone heating up to 180 °C at a maximum ramp rate
of 5 °C/s. The micro GC platform is designed for fast reliable gas mixture analysis in on-line
monitoring. It allows fully autonomous gas analysis and can run unattended preconfigured
analysis methods, including peak identification, integration, and result communication via industry
standards. Other features include automated calibration, low consumption of consumables, and
less maintenance. The micro GC platform can be configured for a variety of applications, such as
variable natural gas or biogas applications, as well as others.
259www.chromatographyonline.com
GC CONNECTIONS
the past year. Continuing their
advancements in vacuum ultraviolet
spectroscopic detection for GC,
VUV Analytics expanded the
wavelength range of its VGA-101
detector as well as enabling series
coupling with some conventional GC
detectors. From Dani Instruments,
the DiscovIR–GC Solid Phase GC–
FTIR detector produces solid-phase
transmission infrared (IR) spectra of
eluted GC components.
A number of sampling accessories
round out the newest instrumental
offerings, including the XR Series of
thermal desorption instruments and
HiSorb extraction probes from Markes,
the 6000 Series Pyroprobe pyrolyzers
from CDS Analytical, and the Master
MTAS robotic autosampler platform
from Dani, all of which advance their
respective sampling tasks with new or
improved capabilities.
(MEMS) for the inlets and detectors plus
direct column heating. All three of these
systems make use of customized column
formats.
For GC–MS, the Exactive GC Orbitrap
GC-MS system from Thermo-Fisher
Scientific, the GCMS-QP2020
GC–MS system and GCMS-TQ8050
triple-quadrupole GC–MS system
from Shimadzu, and the Pegasus BT
GC–time-of-flight MS system from
LECO represent the latest advances
in GC–MS technologies. Two new
application-specific offerings from Agilent
in this area, the SureTarget GC/MS Water
Pollutants Screener and the GC/MS Arsine
Phosphine Analyzer, plus the EPA 8270D
analyzer kit from Thermo Fisher Scientific,
all emphasize the ongoing march towards
routine application of high-performance
GC–MS for standardized methods.
Two optical spectroscopy GC
detectors were announced during
The zone of general GC accessories
was not left out this year, either. The
ADM flowmeter from Agilent has
some new calibration capabilities that
minimize downtime. New Thermolite
septa and Topaz inlet liners from Restek
are designed to extend trace-level
analysis capabilities. Restek also now
offers an electron ionization (EI) filament
replacement for a number of Agilent MS
detectors. There was a large crop of new
gas generators, too: VICI DBS entered
this segment with several offerings for GC
systems, Proton OnSite showed a new
G-Series family of hydrogen generators,
and Peak Scientific has a new nitrogen
generator.
In GC columns, Agilent and
Phenomenex introduced some new polar
columns that offer increased stability
and upper temperature limits as well as
application-specific selectivity tailoring.
Agilent’s new Intuvo System modular
Table 2: New GC instrument systems Contd....
Company Product Description
Scion Instruments
Scion Analytical GC Analyzer Solutions
Scion Instruments announced GC analyzers that tailor the capabilities of GC systems to meet specific analytical requirements. Based on the company’s 436 and 456 gas chromatography platforms, these systems can be configured with multiple columns, switching valves, and temperature-controlled ovens. The Scion SPT (sample preconcentration trap) helps chromatographers perform low level determinations from environmental to gas purity analysis. Available configurations include Simulated Distillation (SIMDIST), Detailed Hydrocarbon Analysis (DHA), Refinery Gases (RGA), Oxygenates, Natural Gas Analyzers (NGA), and Transformer Oil Gas Analysis (TOGA).
Shimadzu
Corporation
GCMS-
QP2020
GC–MS
system
The Shimadzu GCMS-QP2020 GC–MS system offers Advanced Scanning Speed
Protocol, which allows scans up to 20,000 μ/s. The system features a new large-capacity
turbomolecular pump with heightened exhaust efficiency for all carrier gases, including
nitrogen, and the system enables simultaneous scan and single-ion monitoring (SIM) for
qualitative and quantitative data in a single run. The new Smart SIM creation function
automatically creates a program that enables a staggered SIM of multiple components,
resulting in higher SIM sensitivity, while the Quick-CI function allows users to introduce
reagent gas while using the EI source to look for the molecular ion. The ion source is
accessible from the front of the instrument.
GCMS-
TQ8050
triple-
quadrupole
GC–MS
system
The Shimadzu GCMS-TQ8050 MS detector enables detection of femtogram-level concentrations.
It utilizes a new turbomolecular pump that is designed to achieve a higher vacuum and yield
higher sensitivity, accuracy, and stability. Shimadzu’s UFsweeper technology helps to conduct
multiple reaction monitoring (MRM) analysis speeds up to 800 transitions/s, while the company’s
Smart MRM technology helps accurately create methods for ultra-trace analysis and ensures
high sensitivity for MRM measurements. A high-efficiency ion source generates and transmits
ions directly to the detector for higher sensitivity and improved repeatability. Alongside the mass
spectrometer, Shimadzu’s LabSolutions Insight software provides analysts with multianalyte data
review, colour-coded quantitative flags, and a status review function. The system has a stated
instrument detection limit (IDL) of 0.36 fg OFN.
Thermo
Fisher
Scientific
Exactive GC
Orbitrap
GC–MS
system
The Thermo Scientific Exactive GC Orbitrap GC–MS system is designed to provide sensitive,
routine-grade performance for both targeted and nontargeted analyses. The system reportedly
offers the quantitative power of a GC triple-quadrupole mass spectrometer combined with the
advantages of Orbitrap high-resolution accurate mass technology. The new system is designed
for scientists working in routine environments who are looking to increase their reach beyond
targeted quantitation in analysis. The GC–MS system has a resolving power of up to 50,000
(FWHM) at m/z 272, routine sub-part-per-million mass accuracy, and an instrument detection
limit of less than 6 fg OFN. The electron ionization/chemical ionization (EI/CI) ExtractaBrite
ion source is removable under vacuum through a vacuum interlock. The system is capable of
vent-free column exchange with a source plug.
LC•GC Europe May 2017260
GC CONNECTIONS
Table 3: New GC accessories
Company Product Description
Agilent Technologies
ADM flowmeter The Agilent ADM flowmeter provides an external reference for verifying flows and is intended for use when troubleshooting detectors or other GC problems. The flowmeter measures flow volumetrically, which eliminates the need to select a gas type and allows for composite gas streams. The flowmeter incorporates a removable calibrated cartridge. Instead of returning the meter to a third party for recalibration, the cartridge can be replaced regularly, once a year, to keep the meter compliant. Range: 0.5–750 mL/min; accuracy: ±2% of reading or 0.2 mL/min—whichever is greater. The meter has a USB port and can record up to four flows on screen.
Intuvo Flow Chips
Agilent’s Intuvo Flow Chips are modular components that enable flexible configuration of the Intuvo 9000 GC system flow path. These application-specific chips provide simplified connections between the inlet and column (via a guard chip), and column to detector without the need for ferrules. The chips are fitted with smart keys that plug directly into the instrument and automatically configure it for backflush, flow splitting, or MS detection.
Intuvo gasket
Agilent’s Intuvo 9000 GC system uses ferrule-free face seals called gaskets for all fittings within the sample gas flow path. These gaskets take the place of ferrules throughout the GC system, providing a face seal between components of the flow path. These connections are reportedly easily replaced, provide leak-free connections, and enable click-and-run column changes. They are available in three types: polyimide, nickel, and as a plug. The polyimide gasket is designed for standard use up to 350 °C, and the nickel gasket provides a solution for applications at temperatures as high as 450 °C. The plug gasket can be used to check for leaks and for troubleshooting the flow path.
Intuvo Guard Chip
The Intuvo Guard Chip is a simple disposable chip that contains flow channels that connect the inlet of the Intuvo 9000 GC system to the Intuvo column via an inlet flow chip. The guard chip acts as a guard column within a single, disposable chip to prevent unwanted material from depositing on and damaging the head of the column and is designed to be easily installed and replaced. It provides almost 1 m of sample flow path just before the Intuvo GC column. This protection eliminates the need for retention time adjustment and the need to trim a column.
CDS Analytical
6000 Series Pyroprobe pyrolyzers
CDS Analytical introduced its 6000-series Pyroprobe pyrolyzers in two models: the 6150 base model and the 6200 with analytical trap, reactant gas, and sorbent tube capability. Both pyrolyzers can heat to 1400 °C at up to 20 °C/ms using up to 10 stored temperature profiles. The interface temperature is settable as well, up to 400 °C, and is programmable up to 100 °C/min. A heated sample line and valve oven maintain temperatures over the entire gas flow path. The 6200 trap can be heated at up to 1000 °C/min up to 400 °C. An available dynamic headspace option for the 6200 Pyroprobe can sample from 25-mL test tubes or from an 800 mL vessel, at up to 300 °C. A liquid nitrogen cryotrap is also available. The 6000-Series instruments use a LCD touchscreen and a Windows-based software package.
Dani Instruments
DiscovIR–GC Solid Phase GC–FTIR detector
Dani Instruments’ DiscovIR–GC Solid Phase GC–FTIR detector couples a gas chromatograph to Fourier-transform infrared (FT-IR) spectroscopic detection for the identification of GC eluants by depositing them in a spiral track onto a -40 °C cryogenically cooled rotating sample collection disk. Trapped components are then spectrally scanned by a FT-IR interferometer to acquire searchable solid-phase transmission spectra for identification. The spectrometer operates at 4 cm-1 resolution. The FT-IR detector is compatible with the company’s Master GC gas chromatograph and Master AS autosampler.
Master MTAS autosampler
Dani Instruments’ Master MTAS robotic autosampler platform can be configured for conventional autosampling, dual injection, or automated solid-phase microextraction (SPME). The Master Dual-AS autosampler performs simultaneous injection into two GC inlets, allowing analyses on two columns or detectors at the same time. Beyond decreasing overall analysis times for otherwise serially performed dual-column GC, the dual injection mode can provide increased selectivity and sensitivity through the use of two different columns or detectors. The autosampler can hold up to two trays of 80 vials apiece and has a liquid injection volume range of 0.1–500 μL.
Master SHS Robotic autosampler with standard addition
Dani Instruments’ Master SHS autosampler for static headspace can now be used for standard, surrogate, and reagent addition. In one example, in which the autosampler was configured with the company’s Master GC and Master TOF-MS Plus, detection limits of 1 μg/L of formaldehyde in cosmetic products via standard addition and derivatization with PFBHA were reported, with a linear dynamic range to 100 μg/L.
Markes International
HiSorb extraction probes
Markes’ HiSorb extraction probes are a sampling system for the analysis of volatile and semivolatile organic compounds (VOCs and SVOCs) that can be used for immersive or headspace sampling of liquids and solid samples. They are compatible with thermal desorption (TD)GC–MS analysis using industry-standard tubes on all leading commercial systems. The probes feature detection limits lower than for SPME because of the larger capacity of their polydimethylsiloxane (PDMS) sorbent. Cryogen-free preconcentration by TD before automated GC–MS analysis reportedly improves sensitivity. Markes’ HiSorb Agitator heats and agitates HiSorb probes in standard 10- or 20-mL sample vials. The probes are then washed, dried, and inserted into a conventional thermal desorption tube for automated TD-GC–MS analysis.
Peak ScientificSolaris nitrogen generator
Peak Scientific's Solaris nitrogen generator has been engineered and designed as a gas delivery solution that can reduce downtime and increase workflow efficiencies for compact mass spectrometer instruments or for evaporative light scattering detection (ELSD). Built in the company’s ISO-9001 manufacturing facility, the Solaris nitrogen generator can provide up to 10 L/min of high purity nitrogen (up to 99.5%). Developed with a space-saving design, Solaris can be placed on a benchtop and paired with an additional air compressor unit to provide air supply for laboratories without an in-house air supply or who wish to contain their gas supply in a single system.
Proton OnSite’s G Series benchtop hydrogen generators utilize proton-exchange membrane (PEM) technology to produce ultra-high purity hydrogen on-site. The generators sense demand and adjust gas production accordingly. The G-series family is available with flow rates of 200, 400, and 600 mL/min and produce 99.9995% pure hydrogen at output pressures from 43–119 psig (3–8 barg). The G600-HP model scrubs the hydrogen to a 99.99999% purity level. The G4800 model provides 99.9999+% purity hydrogen at up to 4.8 L/min at pressures up to 200 psig (13.8 barg).
Restek Corporation
EI filament replacement part for Agilent MS detector
Restek’s EI filament replacement part is designed for Agilent 5972, 5973, 5975, and 5977 GC–MS systems. The filaments meet or exceed original manufacturer’s performance and are subjected to quality control (QC) tests including heat, electrical current, and resistance. In addition, samples from each filament manufacturing lot are installed in a mass-selective detector for in situ testing.
Restek Methanizer
Restek’s Methanizer is an aftermarket add-on for Agilent 5890, 6890, and 7890 GC FID systems. A methanizer allows parts-per-billion-level determination of CO and CO2 by converting them to methane upstream of an FID system. The system incorporates temperature control to ensure complete conversion of CO and CO2 to CH4. A separate installation kit includes all parts needed for installation into any Agilent GC system.
Thermolite Plus Septa for GC inlets
Restek’s Thermolite Plus septa are usable with inlet temperatures as high as 350 °C, and reportedly have ultra-low bleed levels. The septum incorporates a new plasma coating that eliminates sticking in the injection port. Some of the septa have a CenterGuide design to minimize coring, and the 5-mm septa are partially predrilled for improved puncturability. The septa come preconditioned and ready to use, packaged in ultraclean blister packs. Each batch is reportedly GC–FID tested.
Topaz inlet liners
Restek’s Topaz inlet liners feature an improved deactivation designed to help push detection limits downward for reactive compounds, which also yields better reproducibility and enables longer liner lifetime. Topaz liners are available in clean blister packs for most laboratory GC inlet systems.
The 4S Company
GC-SOS gas chromatography simulation and optimization software
The 4S Company’s GC-SOS simulation and optimization software is an effective tool for developing highly efficient GC methods that reportedly can reduce development time from hours to minutes and can produce more efficient methods. This new version features flexible input with one to three training runs and up to five temperature segments. In many cases an existing method can be used as a training run. The software uses an auto-optimization proprietary numerical algorithm to provide a highly optimized method in seconds, and it has an animation viewer that provides visualization of separations and can be used for teaching as well.
Thermo Fisher Scientific
EPA 8270D analyzer kit
Thermo Scientific’s EPA 8270D kit is designed for use with Thermo Scientific ISQ single-quadrupole GC–MS systems coupled with the Thermo Scientific TRACE 1300 Series gas chromatographs. The kit allows laboratories updating their current GC–MS system to take advantage of key features, including a single-column method and modular GC injectors and detectors, as well as a removable ion source under vacuum. The EPA 8270D analyzer kit includes column, liners, septa, and ferrules specifically designed for EPA semivolatile analysis and a CD with specific instrument and data processing methods, e-workflow, compound retention time database, environmental method specific reports, and an instructional user guide. A video tutorial for method setup ensures that the instrument is up and running with EPA Method 8270D immediately following system installation. Both the Thermo Scientific Dionex Chromeleon chromatography data system (version 7.2 SR4 MUB or newer) and the Thermo Scientific TraceFinder software (EFS version 4.1 or newer) are compatible. The kit also features a dynamic range of 0.2–200 ppm with a single column and liner, plus reduced helium usage with Thermo Scientific’s Instant Connect Helium Saver module.
VICI DBS Ltd.
FID Plus Hydrogen Gas and Zero Air Generator
The VICI DBS range of FID gas generators combines hydrogen and zero air generators into one system. Available in high and ultrahigh purity for all GC detector and carrier gas applications, the generator has software control via USB and alarm capability. It is available in two styles: the FID Station Plus is flat for placement under a GC system, and the FID Tower Plus is a tower configuration for benchtop placement next to instruments. The FID gas generators are available with H2 flow ranges up to 1 L/min and 150 psig (10.5 barg) and air output flow at a maximum of 1.5 L/min.
N2 TOWER Plus nitrogen gas generator
The VICI DBS N2 Tower Plus is a high-purity nitrogen gas generator that produces up to 99.999% pure N2 from a range of models at flow rates up to 4 L/min. An optional catalytic furnace reportedly reduces total hydrocarbon levels to below 0.1 ppm. An external air compressor is required.
NM-H2 Plus Hydrogen Gas Generator
The VICI DBS NM-H2 Plus is a high-purity hydrogen generator that produces 99.999999% pure hydrogen gas for GC carrier gas and flame ionization detectors. The device uses a proton-exchange membrane (PEM) purifier that does not employ palladium membranes. A cascading configuration allows multiple generators to be connected for scalable laboratory expansion. The generators are available in various models with output flow rates ranging from 100 to 1000 mL/min, and the outlet pressure is adjustable from 1 to 160 psig (0.1–11 barg).
VUV AnalyticsVGA-101 gas chromatography detector
The VGA-101 vacuum ultraviolet (VUV) benchtop spectrometer from VUV Analytics is designed to meet the needs of scientists with advanced GC applications. The VGA-101 features a wavelength spectrum of ~120–430 nm, which provides selectivity for complex structures such as polyaromatic hydrocarbons (PAHs). The operating temperature of up to 450 °C allows the deconvolution and analysis of high boiling-point compounds. The detector can be placed in series with MS and other GC detectors. The detector reportedly provides sampling rates up to 100 Hz. Applications include the analysis of high-boiling-point fuel samples containing complex hydrocarbon mixtures, and characterization of isomeric compounds with extensive branching or ring structure that are difficult to distinguish with alternative methodologies.
LC•GC Europe May 2017262
GC CONNECTIONS
columns are being made available in
a variety of common dimensions and
stationary phases; the modules feature a
quick-connect planar cage that does not
use conventional ferrules.
2016–2017 was another very active
year in GC that again emphasized the
pivotal role that small-molecule and
volatile component analyses fulfil in
the fields of separation science. As we
spin around to Pittcon 2018, I expect to
be pleasantly surprised by more new
developments and innovations.
Acknowledgements I would like to thank the manufacturers
and distributors that kindly furnished
the requested information, which
allowed a timely report on new product
introductions over the past year. For
those manufacturers who did not
receive a “New Products” questionnaire
this year and would like to receive one
and be considered for early inclusion
into the 2018 new GC and related
product introductions review, as well as
the other related review articles to be
published in LCGC, please send the
name of the primary company contact
plus the mailing and e-mail addresses
to Laura Bush, Editorial Director, LCGC
and Spectroscopy, UBM Americas,
485 Rte. 1 South, Bldg. F, Suite 210,
Iselin, NJ 08830, USA, Attn: 2018
New Chromatography Products. The
questionnaire will be sent out later in
2017.
References(1) D.S. Bell, LCGC Europe 30(4), 196–207
Norman J. Dovichi, University of Notre Dame, Indiana (USA)
Capillary zone electrophoresis as a tool for eukarotic
proteomics
Pat Sandra, Research Institute for Chromatography, Kortrijk (Belgium)
The LC toolbox for biopharmaceutical characterization
Zoltán Takáts, Imperial College, London (UK)
Ambient and LC-MS lipidomic profi ling of clinical
samples – new era in cancer diagnostics
Peter A. Willis, NASA, Pasadena (USA)
Development of liquid phase separation systems
for spacefl ight missions of exploration
March 6, 2017 Abstract deadline for Best Poster Award
March 20, 2017 Deadline for early registration payment
April 17, 2017 Abstract deadline for poster presentations
Key Dates
Early Regular
Academic 650 € 790 €
Industry 900 € 1 080 €
Student 240 € 340 €
Student + 2 Short Courses 360 € 460 €
Registration Fees
SPECIAL SUPPORT
for students and young
scientists
LC•GC Europe May 2017264
MULTIDIMENSIONAL MATTERS
The development of separation
systems on the basis of on-line,
comprehensive two-dimensional
liquid chromatography (LC×LC)
is a highly complex task, not only
because of the high number of
variable operation parameters, but
also because of the high demands
on the instruments. Recently, the
commercial availability of new high
performance liquid chromatography
(HPLC) systems specifically
designed for LC×LC operation
has attracted much interest in
the academic and industrial
community. The latest innovations
in multidimensional separations
were collected in a special issue
of a journal dedicated to this topic
(1). Although LC×LC appears to
have matured, there are some
specific problems still present that
hamper the widespread use of this
technology. One key aspect is the
coupling of an on-line LC×LC system
to a mass spectrometer. Generally,
on-line LC×LC is based on a very
fast second dimension separation
to achieve low cycle times (2). This
often results in flow rates that are far
above the optimum for electrospray
ionization mass spectrometry
(ESI-MS). In order to circumvent
the necessity for flow-splitting, a
miniaturized LC×LC system with
nano-LC in the first dimension and
micro-LC in the second dimension
was described previously (3). This
month’s “Multidimensional Matters”
explores the benefits of coupling
miniaturized comprehensive 2D LC
to a hybrid high-resolution mass
spectrometer (HRMS) with a focus
on its application in environmental
(water) analysis.
The Selection of Suitable
Stationary Phases—The First
Dimension: The selection of a
suitable stationary phase in on-line
LC×LC not only encompasses
the need for a high orthogonality,
but also the appropriate column
dimensions. Although there are
numerous combinations of different
column chemistries to enhance
orthogonality (4), it has become
common practice for the selection
of two reversed-phase stationary
phases in on-line LC×LC to use
a less retentive column in the
first dimension (1D) to avoid the
necessity of a very strong eluent
in the first dimension that would
be a strong eluent in the second
dimension (2D) as well (2). However,
this practice is inconsistent with the
need for good sensitivity because
on-column focusing increases with
the retentivity of the stationary phase
material (5). This is especially true
for a nano-LC column where the
sample volume has to be adapted
in conjunction with the internal
diameter. Moreover, polar analytes
that experience no retention on
a classical silica-based reversed
phase stationary phase cannot
be trapped. It is these analytes,
however, that play a pivotal role in
environmental analysis (6). Porous
graphitic carbon (PGC) is ideally
suited to trap very polar compounds
that would elute at the void volume
on a silica-based reversed stationary
phase. Leonhardt et al. recently
demonstrated that 5-fluorouracil,
which has almost no retention
on a “classical” reversed-phase
stationary phase, could be eluted
with a retention factor of 146 on
a PGC stationary phase with an
internal diameter of 75 μm (5). The
authors noted that small fluctuations
The Benefits of Coupling Miniaturized Comprehensive 2D LC with Hybrid High-Resolution Mass SpectrometryJuri Leonhardt1, Jakob Haun1, Torsten C. Schmidt2, and Thorsten Teutenberg1, 1Institut für Energie- und Umwelttechnik
e. V., Duisburg, Germany, 2University Duisburg-Essen and Centre for Water and Environmental Research (ZWU), Essen, Germany
Comprehensive two-dimensional liquid chromatography (LC×LC) is evolving and becoming more commonly used in practice, but there are some specific problems still present that hamper the widespread use of this technology. One key aspect is the coupling of an on-line LC×LC system to a mass spectrometer. Generally, on-line LC×LC is based on a very fast second dimension separation to achieve low cycle times. This often results in flow rates that are far above the optimum for electrospray ionization mass spectrometry (ESI-MS). This month’s “Multidimensional Matters” looks at the benefits of miniaturization in the first and second dimension for coupling with a high-resolution mass spectrometer (HRMS) and describes an environmental analysis application.
265www.chromatographyonline.com
MULTIDIMENSIONAL MATTERS
in the composition of the injection
solvent could lead to fronting
effects. This means that the method
is most suited for a large-volume
direct injection of an aqueous
sample without the need for further
preconcentration. This strategy is
currently applied in most laboratories
dealing with water analysis, because
the availability of very sensitive mass
spectrometers allows for a direct
injection of large sample volumes
(for example, 1000 μL injected onto
a 4.6 mm i.d. column [7]). Leonhardt
et al. successfully increased the
absolute injection volume to 5 μL
on a 12 mm × 0.075 mm PGC
nano-LC column coupled to a
50 mm × 0.1 mm reversed-phase
C18 core–shell stationary phase.
Moreover, West et al. described
essential differences in retention
behaviour of compounds separated
on PGC phases compared to
common reversed-phase C18
materials (8). At that time this
contradicted the popular opinion
that a PGC phase is simply a
hydrophobic phase comparable
to C18 phases. From this, it
was decided to use PGC as the
stationary phase with an internal
diameter of 100 μm for the first
dimension that would include a
large volume injection to counteract
analyte dilution—a major problem in
on-line LC×LC approaches (9).
The Selection of Suitable
Stationary Phases—The Second
Dimension: With regard to the
differences in retention behaviour to
PGC phases, a reversed-phase C18
Small 2dc
Large 2d
c
Reduction ofextra-column delayand band broadeningin 2D(due to higher 2Dflow rates)
Figure 2: Dependence of gradient delay time on the flow rate for different delay volumes (V_GD).
World leaders in thermal desorption
for environmental air monitoring
Tailor your TD system to your requirements – canisters, on-line and sorbent tubes
ɵ Maximise efficiency by running VVOC to SVOC on one platform
ɵ Sample splitting and re-collection for easier validation and method compliance
ɵ Reduce your costs – avoiding use of liquid cryogen
Find out more http://chem.markes.com/XR
LC•GC Europe May 2017266
MULTIDIMENSIONAL MATTERS
material was chosen for the second
dimension separation. In order to
obtain a fast second dimension
separation, the use of elevated
temperatures is recommended to
operate a column at high efficiency
even if the flow rate is far above the
van Deemter minimum (10). In terms
of ultimate temperature stability,
bridged ethyl hybrid particles have
proven to be extremely temperature
resistant, even if the temperature is
increased to 150 °C (11). However,
it is an essential requirement that
the column hardware in which the
stationary phase is packed is also
stable at the applied temperatures.
Unfortunately, this point turned
out to be the Achilles heel for the
application of very high-temperature
LC in the 2D. The main problem was
that the available capillary column
hardware was either based on
packed fused-silica capillaries that
needed plastic parts for the fittings
or even based on packed PEEK
capillaries. Often, steel sheathings
covered the packed capillaries to
mimic the outer appearance of a
standard HPLC column. In both
cases, high temperatures cause
deformation and carry a high
potential of hardware failure. For this
reason, the temperature was set to
60 °C. In order to further increase
the separation speed in the 2D, a
core–shell stationary phase with
a particle diameter of 2.6 μm was
chosen instead of a fully porous
sub-2-μm particle packed column.
Since the 2D column has to fulfill
significantly more requirements
compared to the 1D column, the
main arguments for choosing either
a small or a large 2D column internal
diameter are visualized in Figure 1.
As can be seen, the preselection
of the 2D column internal diameter
is dominated by questions of speed
and dilution effects. A large diameter
increases analyte dilution simply by
additional dispersion that occurs
within the large column volume. On
for increased flow rates. It can be
concluded that a smaller column can
enhance overall speed by increasing
the speed of the separation itself,
whereas the use of higher flow rates
in combination with a larger column
internal diameter decreases the
contribution of extra-column volumes
to the analysis time. This means that
for an optimum speed, the stationary
phase in combination with the
column internal diameter should be
chosen so that the optimum flow rate
is high enough to keep the influence
of the extra-column volumes low,
but low enough to allow the use of
a small internal diameter column to
optimize towards linear velocity. An
essential requirement for a linear
velocity optimization is that the
latter does not result in significant
efficiency losses, which depend
on the stationary phase and the
the other hand, the transferred 1D
solvent will be far better diluted as
well, so the analyte bands potentially
can be better refocused on the 2D
stationary phase. The latter process
counteracts analyte dilution (12).
The question of speed, however,
is of greater importance when fast
LC is applied. As demonstrated in
the theory, chromatographic speed
is proportional to the average
linear mobile phase velocity (u).
This means linear velocity can
be increased by decreasing the
column diameter at a constant flow
rate. An increase of the column
diameter implies that significantly
higher flow rates are necessary
to keep u constant, so solvent
consumption increases as well. On
the other hand, extra-column delay
times, such as for example the
gradient delay time, are reduced
(a) (b)
Figure 3: The use of PEEK tubing sleeves when mounting a 50-μm i.d. 360-μm o.d. fused-silica capillary to a column end fitting for 1/16” capillaries. (a) Ideal arrangement before tightening the nut. (b) Void volume formation: dislocated inner sleeve and fused-silica capillary after tightening the nut. Grey: connection bore of the column end fitting. Brown: fused-silica capillary. Orange: PEEK sleeve 1/32” A 360 μm o.d. Blue: PEEK sleeve 1/16” A 1/32” o.d. Arrows: approximate grip of the ferrule.
0.6 P 7
P 6
P 55
4
2
3
1
6
P 4
P 3
P 2
P 8
0.5
0.4
0.3
0.2
0.1
2D
Rete
nti
on
Tim
e (
min
)
1D Fraction Number
0 10 20 30 40 50 60 70 80
Figure 4: The areas used for the calculation of the surface coverage. Adapted with permission from reference 3.
Although LC×LC appears to have matured, there are some specific problems still present that hamper the widespread use of this technology.
267www.chromatographyonline.com
MULTIDIMENSIONAL MATTERS
experimental conditions. However,
in the case of a sub-3-μm core–
shell stationary phase at elevated
temperature, this requirement is
fulfilled (13).
The influence of the extra-column
volume on analysis time will be
discussed by using the example of
the gradient delay volume (Vdwell).
This is defined as the volume from
the point of gradient mixing to the
column head. The gradient delay
time (tdwell) needed to flush this
volume can be calculated by using
the fundamental equation that
defines the flow rate as volume per
time, thus:
[1]tdwell =
Vdwell
F
where F is the volumetric flow
rate. If Vdwell is held constant as
in Figure 2, functions of the type
f(x) = b/x are obtained on the basis
of equation 1.
The maximum value of the time
scale in Figure 2 has been chosen
as 60 s. This is the intended time
for a complete 2D cycle in this
example. As can be seen from
Figure 2, the gradient delay time
drastically increases at low flow
rates. A delay time of 2 s (3.3%)
was set as a significance level. A
delay below this tolerable value
guarantees that the gradient delay
time is not a significant part of
the cycle time, keeping in mind
that the extra-column volume from
the column end to the detector
will also be added to the analysis
time. As shown in Figure 2, this
significance level can only be
reached by gradient delay volumes
much less than 10 μL within the
given flow rate range. Moreover,
it can be seen that the gradient
delay volumes have to be reduced
together with the flow rate at
an equal rate to keep the delay
constant. A gradient delay volume of
50 μL is a typical value for modern
ultrahigh-pressure LC (UHPLC)
instruments with small standard
mixers (usually ~ 35–45 μL).
Miniaturized mixers (1–25 μL) are
available for capillary-UHPLC
systems and columns. Accordingly,
smaller gradient delay volumes
below 10 μL can be obtained if the
tubing dimensions are selected
appropriately. The lowest delay
volumes of around 1 μL can be
obtained if no mixer is used.
However, this potentially results in
a baseline ripple for conventional
piston-based pumps that affects
the sensitivity of the detector. The
pump systems used for this study
are pneumatic pumps that do not
need a separate mixer post to the
tee-connector that unifies and mixes
the flow of both gradient channels.
Thus, gradient delay volume is easily
reduced to 1 μL for micro-LC. From
Figure 2 it can be deduced that the 2D column internal diameter should
allow a flow rate of at least 30 to
40 μL/min to avoid a too strong
influence of the gradient delay in a
60 s cycle time.
It was therefore decided to
use a stationary phase for the
second dimension with an internal
diameter of 300 μm and a length
of 50 mm. The application of even
smaller internal diameters was not
considered with respect to the
already high pressure drop and the
need for high loading capabilities as
a result of the transfer volume from
the 1D. It can therefore be concluded
that the final diameters were 0.1 mm
and 0.3 mm for the first and second
dimension, respectively. In order
to compare these values to that of
typical conventional on-line LC×LC
setups, which usually use 1.0 mm or
2.1 mm in the 1D and 4.6 mm in the 2D, the relative difference between
the column internal diameters of the
two coupled LC dimensions (Δ i.d.)
was calculated using equation 2:
[2]i.d. =
—2dc
2dc
1dc𝚫
where 2dc is the internal diameter
of the second dimension, and 1dc
is that of the first dimension. The
results are listed in Table 1.
As can be seen from Table 1, the Δ
i.d. used for the miniaturized on-line
LC×LC system lies in between the
Δ i.d. of typical conventional on-line
LC×LC systems. This indicates that
the contribution to analyte dilution
caused by the difference in internal
diameter is not expected to be
significantly higher or lower than in
conventional LC×LC systems.
Note on Column Fittings: Usually,
there is no discussion on column
fittings as the capillary outer
diameter (o.d.) is standardized to
1/16” in conventional HPLC. Several
column manufacturers, however,
still pack micro-LC columns with
the corresponding fittings for
standard capillaries, which are very
large in comparison to the internal
Table 1: Comparison of the Δ i.d. between non-miniaturized on-line LC×LC and the miniaturized approach described in this
article. Conventional setups 1 and 2 show typical inner diameters of non-miniaturized on-line LC×LC systems that do not use a
flow split between the dimensions.
1D column i.d. 2D column i.d. ∆ i.d.
Conventional setup 1 1.0 4.6 0.78
Conventional setup 2 2.1 4.6 0.54
Miniaturized LC×LC system 0.1 0.3 0.66
The selection of a suitable stationary phase in on-line LC×LC not only encompasses the need for a high orthogonality, but also the appropriate column dimensions.
Chromatographic speed is proportional to the average linear mobile phase velocity (u). This means linear velocity can be increased by decreasing the column diameter at a constant flow rate.
LC•GC Europe May 2017268
MULTIDIMENSIONAL MATTERS
diameter of the column and the
capillaries that are usually used in
micro-LC (o.d.: 1/32” or 360 μm).
Consequently, PEEK sleeves—for
360-μm o.d. capillaries there are
often two—are frequently used to
bridge the gap between the outer
diameter of the capillary and the
connection bore size of the column
end fitting (see Figure 3).
Significant extra-column band
broadening can result from unwanted
void volumes at the column end
fittings. The void volumes can be
formed as shown in Figure 3, or
between the stacked sleeves or the
sleeve and the capillary up to the
grip point. Accordingly, columns with
end fittings for 1/32” o.d. capillaries
should be used to avoid peak
broadening. Additionally, the use of
zero dead-volume fitting assemblies
is recommended where possible.
The Simplified Heating
Concept: To keep the setup of the
multidimensional system as simple
as possible, it was decided to heat
both LC dimensions equally by using
the air-bath oven of the column
compartment. A detrimental point,
however, is the stability of the valves
used for the modulation, which
were also included in the heating
compartment. Since both extended
pressure and temperature might
lead to a continuous abrasion of the
rotor, 60 °C was chosen as the oven
temperature for isothermal heating of
both LC dimensions to increase the
lifetime of the valves as well as the
stationary phases.
Selection of Mobile Phases: For
a further optimization of selectivity,
different mobile phases should
be used in the first and second
dimension. Therefore, methanol was
selected as protic solvent in the 1D
and acetonitrile was selected as
aprotic solvent in the 2D. The reason
for the chosen order is the different
viscosity maxima of binary solvent
systems consisting of
water–acetonitrile and water–
methanol. A mixture of water–
methanol exhibits a much higher
pressure maximum when a solvent
gradient is applied than a mixture
of water–acetonitrile (14) and is
therefore more suitable to be used
in the 1D. In the 2D, a very high
linear velocity has to be achieved
to reduce the cycle time. Hence,
a mixture of water–acetonitrile has
been used in the 2D.
Gradient Programming: If an
unknown sample has to be analyzed
or the information about the sample
is limited, a generic gradient should
be used. This means that for both
dimensions, a linear gradient
that covers the full range from,
for example, 5% to 95% B should
be applied. Many studies have
described more advanced gradient
programming for LC×LC methods
including so-called shift-gradients
(15). A shift-gradient usually refers to
a change of the starting conditions
of the second dimension separation
during the linear gradient of the first
dimension separation. The potential
benefit of shift-gradients is that the
gradient window for consecutive 2D runs can be adapted so that the
resolution of compounds eluting
during these gradients is higher
when compared to full-gradients.
Unfortunately, this concept does
not consider that the approach
is no longer generic. Many users
would prefer easy-to-use generic
methods without changing the
gradient parameters. The technical
feasibility to apply shift-gradients
is an advantage if fine-tuning of
a separation needs to be done
to optimize a two-dimensional
separation. For screening methods,
the application of full-gradients
seems to be a better way because it
is difficult or impossible to anticipate
all theoretical combinations of a
complex sample. The gradient
programming has therefore been
kept as simple as possible for
a generic approach. For the 1D
separation, a linear gradient was
programmed with an isocratic
plateau at the end of the gradient.
The overall analysis time for each
injection is about 110 min. In the 2D
separation, the cycle time was 1 min.
Hyphenation to Mass
Spectrometry: The hyphenation of
a miniaturized LC×LC system to a
mass spectrometer is very critical
in terms of the extra-column dead
volume. This refers to the transfer
line connecting the column outlet
with the ion source of the mass
spectrometer as well as the emitter
tip of the ion source itself. Whereas
a short connection between the LC
99
4844
16
99
65 64
31
0A B DC
50
100
Nu
mb
er
of
dete
cted
targ
ets
1D LC2D LC
Figure 5: Overview of the identified analytes by 1D HPLC–MS and 2D nLC×μLC–MS. Detailed list of detected targets is given in reference 18. A: Detected targets in reference standard by < 5 ppm; B: detected targets in wastewater sample by < 5 ppm; C: detected targets in wastewater sample by < 5 ppm and retention time < 2.5%; D: detected targets in wastewater sample by < 5 ppm, retention time < 2.5% and MS/MS hit. Adapted with permission from reference 18.
A gradient delay volume of 50 μL is a typical value for modern ultrahigh-pressure LC (UHPLC) instruments with small standard mixers (usually ~ 35–45 μL).
269www.chromatographyonline.com
MULTIDIMENSIONAL MATTERS
system and the mass spectrometer is
not the main obstacle, an unsuitable
internal diameter of the emitter tip
can be devastating in terms of the
observed separation efficiency.
To reduce the band broadening
behind the column, an emitter tip
with an internal diameter of 50 μm
was installed instead of the classical
100-μm i.d. emitter tip. While the
classical tip is made of stainless
steel, the modified miniaturized
emitter tip is based on a PEEKSil
capillary. To ensure ionization, at
the top of the PEEKSil capillary a
stainless steel tip with the respective
internal diameter was installed. With
regards to the connection technique
it should be noted that the PEEKSil
emitter is designed for 1/32” fittings.
High pressure-resistant fittings are
screwed to a 1/32” union. This union
offers the advantage of being able
to be used as the grounding point.
The change of the emitter tip can be
easily accomplished in a few minutes.
In the next section, we describe
the application of a miniaturized,
on-line dual-gradient LC×LC
system coupled to hybrid HRMS
detection. A 99-component standard
mixture and a complex wastewater
sample were used to demonstrate
the performance of this approach.
Moreover, a comparison was made
between the miniaturized 2D LC
approach and a conventional 1D LC
approach that is usually used for
suspected target screening of
environmental samples.
Results and DiscussionCalculation of the Surface
Coverage as a Measure of
Orthogonality: First of all, the
surface coverage for the LC×LC
separation of the 99-component
standard mixture was calculated
using the convex hull that includes
all analyte spots (see Figure 4).
In this case, the convex hull is
an eight-sided irregular polygon
described by the points P1 to P8.
The area of this convex hull was
calculated by a vector method that
Dück et al. used in their work (16).
Accordingly, the area of the convex
hull can be divided in six scalene
triangles that are numbered in
Figure 4. Each of these triangles
can be described by two vectors
that have the same corner point as
origin and the other corner points as
heads. The vectors that were used to
determine the six areas are marked
red in Figure 4. P1 was chosen
as the origin of all vectors. The
complete calculation can be found
in reference 3. The surface coverage
can now be approximated by the
area ratio of the convex hull and the
rectangular area, which is ~0.61 (or
61%). This result clearly underlines
that both dimensions are only weakly
correlated. According to Gilar et al.,
a coverage of 60% of the available
separation space can be considered
very high (4). The authors even state
that for most practical applications
a surface coverage higher than 63%
cannot be achieved.
Comparison of 2D LC with 1D
LC: A higher peak capacity is
usually obtained in two-dimensional
liquid chromatography compared
with one-dimensional separations
resulting in a higher number
of compounds that can be
chromatographically resolved.
The user, however, is not primarily
interested in whether it is
possible to obtain a higher peak
capacity. For practical purposes
it is more important whether a
two-dimensional system will also
lead to a significantly higher
number of detected or identified
compounds when compared to
a one-dimensional separation.
Therefore, the sample itself and
not the theoretical peak capacity
should be the basis for this
evaluation. A direct comparison
between a one-dimensional and a
two-dimensional separation would be
useful for an objective judgement of
the performance of two-dimensional
separation approaches. However,
there is only a very limited number
of dedicated studies dealing
with this issue (17). Leonhardt et
al. recently made a comparison
between a one-dimensional and a
two-dimensional chromatographic
approach coupled to hybrid
high-resolution mass spectrometry
(18). The comparison was based
on a screening approach in
environmental analysis, where three
criteria for compound identification
were employed. First, the accurate
mass with a deviation of less than
5 ppm was used for identification.
If the first criterion was fulfilled, the
retention time should not deviate
more than 2.5% from the retention
time measured in the reference
standard. If this criterion was also
fulfilled, MS/MS spectra of the
reference standard were compared
with that of the sample. Figure 5
summarizes the results of this
comparison.
All chosen compounds could
be detected with both approaches
on the basis of the accurate mass
in the reference standard. When
a real sample was analyzed,
48 compounds were found with
the 1D LC approach, while 65
compounds could be detected
with the miniaturized 2D LC
approach. Using the retention
time as an additional filter for the
elimination of false positive hits,
four compounds had to be removed
in the 1D LC approach, while only
one compound had to be eliminated
for the 2D LC approach. It should
be emphasized that retention time
is a powerful parameter to ensure
the identification of a compound.
Of course, a further criterion for an
unambiguous identification should
be considered. MS/MS information
can help to distinguish even isobaric
compounds that have the same
mass-to-charge ratio, m/z. The
possibility of acquiring additional
MS/MS information also depends
on the peak width. As can be seen
from Figure 5, MS/MS information is
only obtained for a small number of
analytes that have been identified
on the basis of the first and second
criterion. The reason is that the
For a further optimization of selectivity, different mobile phases should be used in the first and second dimension.
A direct comparison between a one-dimensional and a two-dimensional separation would be useful for an objective judgement of the performance of two-dimensional separation approaches.