Infinity 2D-LC Solution Solution - United States Home | Agilent
Transcript of Infinity 2D-LC Solution Solution - United States Home | Agilent
Page 1
The Agilent 1290
Infinity 2D-LC Solution
Achieving highest separation power
with the Agilent 1290 Infinity 2D-LC
Solution
Page 2
2D-LC - What is it?
Peak capacities multiply for orthogonal separation mechanisms!
Two different modes:
Comprehensive 2D-LC (“LCxLC”)
Heart-cutting 2D-LC (“LC-LC”)
2DLC: Injecting the effluent or a part of the effluent of one column to a second
column, ideally with orthogonal separation behavior.
Purpose: increase total separation power.
Complex/unknown samples:
bio-pharma, food, polymers....
Known samples/improving
confidence: pharma, methdev...
Page 3
2D-LC - What is it?
Pump Autosampler Column Detector 1D-Chrom.
Standard 1-dimensional LC:
Page 4
2D-LC - What is it?
Pump Autosampler Column 1
Pump 2
Column 2
Injector
Detector
2-dimensional LC:
Page 5
2D-LC - comprehensive vs. heart-cutting 2D-LC
Comprehensive 2D-LC (LCxLC):
LC1
LC2
1st peak from
1st dimension
2nd peak from
1st dimension
3rd peak from
1st dimension
LC1
LC2
min
sec
Page 6
Only parts of the effluent of the first column will be
injected to the second column.
Heart-cutting 2D-LC (LC-LC):
2D-LC - comprehensive vs. heart-cutting 2D-LC
LC1
LC2
The gradients in the 2nd dimension can be much
longer as in comprehensive 2D-LC.
Loss of information.
But better quality data for peak of interest
Page 7
The 1290 Infinity 2D-LC Solution
• Based on the Agilent 1290 Infinity LC, thus highest performance for 2D-LC
– Excellent retention time precision
– High power range (up to 1200 bar, up to 5 ml/min)
– Highest sensitivity
• Highest flexibility on hardware set-up
– Different pumps, autosamplers and detectors supported
– Different detectors at different positions
– Innovative new 2D-LC valve head , different valve set-up possibilities supported
• Most easy-to-use 2D-LC acquisition SW
– Unique features like automatically shifted gradients or peak-triggered operation
– For comprehensive 2D-LC and heart-cutting 2D-LC
• High performance data analysis by partner-solution
Page 8
For heart-cutting 2D-LC by Agilent‘s OpenLAB CDS ChemStation ed.
For comprehensive 2D-LC Agilent recommends:
GC Image LCxLC Edition Software
Key features offered:
• Direct Agilent datafile import
• Peak identification, integration, annotation
• Comparative Analysis and Visualization
• Compound-libraries
• MS-data handling
Data-Analysis
Page 9
Hardware – Module-flexibility
1. Dimension
2. Dimension
1290 Infinity
Binary Pump
1290 Infinity
Binary Pump
1290 Infinity
Autosampler or 1260
HiP Autosampler
Optional
1260/1290
Infinity
Detector
One or two
1290 Infinity
TCC
Optional
1260/1290
Infinity
Detector
1260/1290
Infinity
detector
1260 Infinity Binary
or Quaternary Pump
1260 Infinity
Capillary Pump 1260 Infinity
Autosampler
For 1st
dimension
chromatogram
and peak-
triggering
To monitor
waste-line
Almost any Agilent pump or autosampler in the 1st dimension!
Almost any detectors are supported!
A 1290 Infinity Binary Pump for the 2nd dimension is required.
Page 10
Hardware – Valves, uniqueness and flexibility
1. Dimension
2. Dimension
waste
Loop1
Loop2
2pos/4port-duo
New Agilent-only special 2D-LC-QucikChange valve.
Single valve with fully symmetric flow-paths and symmetric fill/flush-out
behavior. Only valve that allows countercurrent flush-out of both loops.
Page 11
Advantages of the new Agilent 2D-LC QuickChange valve head
All flow paths are equal!
Symmetric contercurrent or co-current fill/analyze direction of loops possible
All in one valve
Loop1
Loop2
waste
2D-pump 2D-pump
1D-column
2D-column 2D-column
Fill direction Analyze direction
Valve
switching
1. 2.
Page 12
Hardware – Valve-flexibility
1. Dimension
2. Dimension
waste
Loop1
Loop2
2pos/
10port
valve
Most other valve configurations for comprehensive and heart-cutting
2D-LC are supported as well. Easy transfer of existing 2D-LC methods!
Page 13
Hardware – Valve-flexibility
1. Dimension
2. Dimension
waste
Loop1 Loop2
2x
2pos/6port
valves
Also dual valve-head configurations supported with automatic
synchronisation of valve-drives!
Page 14
Dashboard: All modules in one dashboard, all system
parameters immediately visible
2D-LC Acquisition Software
Page 15
2D-LC Acquisition Software - supported 2D-LC modes
comprehensive LCxLC
Heart-cutting LC-LC
Standard-repeating
Time or peak-triggered
Time-triggered
Peak-triggered
Page 16
Most easiest 2D-LC System Configuration - “One screen for the entire system”
Page 17
Most easiest set-up of complex 2D-LC methods - 2D-LC specific parameters of the 2D-pump
Page 18
Most easiest set-up of complex 2D-LC methods - Time-segmenting for the 2nd dimension
Page 19
Time-segmenting in comprehensive 2D-LC - Reducing cost!
1D Gradient/
2D Gradients
Idle flow rate
2D Pump flow
rate
2D Pump flow rate
Solvent saving!!!
2D Time-segments
set in software idle
idle
idle
Time-triggered
Peak-triggered
Valve toggling
Increase life time!
%B (2D)
%B (1D)
F (2D)
1D Chromatogram
Page 20
Time-segmenting in comprehensive 2D-LC - Reducing cost!
1D Gradient/
2D Gradients
Idle flow rate
2D Pump flow
rate
2D Pump flow rate
Solvent saving!!!
2D Time-segments
set in software idle
idle
idle
Time-triggered
Peak-triggered
Valve toggling
Increase life time!
%B (2D)
%B (1D)
F (2D)
1D Chromatogram
Page 21
Time-segmenting in comprehensive 2D-LC - Reducing cost!
1D Gradient/
2D Gradients
Idle flow rate
2D Pump flow
rate
2D Pump flow rate
Solvent saving!!!
2D Time-segments
set in software idle
idle
idle
Time-triggered
Peak-triggered
Valve toggling
Increase life time!
%B (2D)
%B (1D)
F (2D)
1D Chromatogram
Page 22
Time-segmenting in comprehensive 2D-LC - Reducing cost!
1D Gradient/
2D Gradients
Idle flow rate
2D Pump flow
rate
2D Pump flow rate
Solvent saving!!!
2D Time-segments
set in software idle
idle
idle
Time-triggered
Peak-triggered
Valve toggling
Increase life time!
%B (2D)
%B (1D)
F (2D)
1D Chromatogram
Page 23
Example: graphical editing of a gradient shift - replace editing of large timetables by a few mouse operations -
1 Use context menu to enter the editing mode
3 Draw a straight line by dragging the mouse to a new %B value at a
specified runtime of the 1st dimension
2 Timetable entries are marked with circles
4 When releasing the mouse, a new TT entry is made and the gradient
rollout is automatically updated
6 Insert / Delete shift points (mouse cursor and context menu changes
near to a shift line)
5 Repeat step 3 + 4 with other TT entries
Page 24
2D-LC Acquisition Sofware - supported 2D-gradients modes
Isocratic Advancing Isocratic
Std. Gradient Shifted Gradient
…and most combinations!
Page 25
2D-LC Acquisition Sofware - supported 2D-gradients modes
Isocratic Advancing Isocratic
Std. Gradient Shifted Gradient
• Start and end time
• Peak-triggered
• Start and end time
• Peak-triggered
• Stepwise or gradual
change of isocratic cond.
• Start and end time
• Peak-triggered
• Start and end time
• Peak-triggered
• constantly shifted %B2D
• constantly shifted %B2D and
shifted Δ%B2D
Page 26
Agilent 1290 Infinity 2D-LC Solution
Performance and Applications
Page 27
Performance
Optimization of a 2D RP-RP LC separation
red line: 1st dimension gradient
blue line: repeating 2nd dimension gradient
Current state-of-the-art 2D-LC – narrow spread of peaks
2nd dimension gradient repeats between 5% and 95% organic solvent
Page 28
Performance
Optimization of a 2D RP-RP LC separation
First adjustment of the second dimension gradient to improve
separation by lower organic in the second dimension for the
compounds of higher polarity.
Page 29
Performance
Optimization of a 2D RP-RP LC separation
Second adjustment of the second dimension to the same maximum
organic content as given from the first dimension.
Page 30
Increase in organic composition during the runtime to improve separation
and focus the compounds by better usage of separation time in the 2nd
dimension
Performance
Optimization of a 2D RP-RP LC separation
Page 31
Performance
Optimization of a 2D RP-RP LC separation
1st Dimension (red line): Gradient: 0 min 5% B – 30 min
95% B, 40min stop-time.
Flow rate: 0.1 mL/min.
2nd Dimension (blue line): Initial Gradient: 0 min - 5% B, 0.5
min - 15% B, 0.51 min - 5% B,
0.65 min - 5% B.
Modulation-time: 0.65 min
Gradient development:
5% B at 0 min to 50% B at 30 min
15% B at 0.5 min to 95% B at 30 min
5% B at 0.51 min to 50% B at 30 min
5% B at 0.65 min to 50% B at 30 min
Flow rate: 3 mL/min.
Page 32
Performance
Test mix of 20 compounds – 2D-Precision
0.00
0.50
1.00
1.50
2.00 Retention Time Precision RSD (%)
0.00
1.00
2.00
3.00
Peak Volume Precision RSD (%)
15 compounds <1% RSD
all compounds <2% RSD
8 compounds <1% RSD
16 compounds <2% RSD
Page 33
Applications
Separation of polyphenoles in juice and wine
by 2D RP-RP LC Compound Compound Name Peak I (min) Peak II (sec)
1 Gallic acid 7.15 7.56
2 Esculin 9.75 18.63
3 3,4 HO benzoic acid 9.75 9.74
4 HO Phenacetic acid 11.70 13.55
5 6,7 HO coumarin 12.35 19.66
6 HO Benzoic acid 12.35 16.51
7 Syringic acid 13.00 25.78
8 Rutin 13.65 33.63
9 Naringin 14.95 32.43
10 Coumaric acid 14.95 25.44
11 Hesperidin 15.60 34.89
12 Ferulic acid 15.60 27.63
13 Myricetin 17.55 30.20
14 Morin 18.85 30.20
15 Resveratrol 18.85 26.65
16 Salicylic acid 19.50 18.55
17 Luteolin 19.50 32.79
18 Quercetin 20.15 30.20
19 Kaempferol 21.45 30.96
20 Apigenin 22.10 25.84
21 Naringenin 22.10 27.65
22 Hesperetin 22.75 28.74
23 7 HO Flavone 22.75 34.29
24 Pinosylvin 24.70 18.81
25 Chrysin 27.30 27.49
26 Flavone 28.60 26.33
•3-Dimensional display of the separation
of the 26 compound standard mixture.
•40 minutes 1st dimension separation
•39 seconds for each 2nd dimension
separation
Page 34
1
2
3
4
5
6
7
8 9
10
11
12
13 14
15
16
17
18 19
20
21 22
23
24
25 26
2D-LC plot of the optimized separation of
26 polyphenolic compounds standard
mixture
Applications
Separation of polyphenoles in juice and wine
by 2D RP-RP LC
2D-LC plot with software detected peak
annotation
Page 35
Applications
Performance data of selected compounds of
the used standard mixture.
Retention time RSD [%]
typically better than 0.6
Peak volume RSD [%]
typically better than 3%.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
RS
D [
%]
Retention time precision
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
RS
D [
%]
Peak volume precision
Page 36
Gallic acid
Coumaric
acid
Syringic acid
Gallic acid
O
OH
OH
OH
OH
O
OH
OH
O
O
CH3
CH3
O O
OH
The main components are typically hydroxyllic benzoic
acid compounds like Gallic acid
(insert: UV spectrum of gallic acid).
Kivilompolo M., Oburka V.,
Hyötyläninen T., Comprehensive
two-dimensional liquid
chromatography in the analysis of
antioxidant compounds in wines
and juices. Anal. Bioanal. Chem.
2008, 391, 373-380.
Applications
Red Grape juice separated by 2D-LC.
Page 37
•Besides the simple hydroxy benzoic acid derivatives more
complex glycosidic compounds like Rutin and Esculin were
identified.
•Mixed antioxidant juice containing: Red and green grape,
apple, black current, cherry, cranberry, pomegranate, bilberry
Rutin
O
O
OOH
OHOH
OH
O
OH
OH
OH
OOCH3
OH
OH
OH
Coumaric acid
Ferulic acid
Gallic acid
3,4 HO Benzoic acid
Esculin
Rutin
6,7 HO Coumarin
Syringic acid
Applications
2D-LC separation of a mixed antioxidant juice
Page 38
•The 2D plot shows a pattern which is typically for the early
eluting polar hydroxy benzoic acid derivatives.
•But there is also one very well known antioxidant compound
typically inherent in red wine: Resveratrol.
OH
OH
OH
Resveratrol
Applications
2D-LC separation of Merlot red wine.
Page 39
R2=0.9985
Calibration of 2D-LC quantification
for resveratrol , 1 – 50 µg/mL
Applications
Quantification of resveratrol in Merlot red wine
by 2D-LC separation
OH
OH
OH
Found in Merlot red wine:
4.5 µg/mL, 4.5 mg/L
Page 40
min 19.5 20 20.5 21 21.5 22 22.5 23 23.5
Main Compound – Pure?
Main Compound
1st dimension separation
2nd dimension separation
Additional impurities!
Impurity
Impurity
Impurity
Applications – heart-cutting 2D-LC - impurity profiling of an active pharmaceutical ingredient
Page 41
Summary
The Agilent 1290 Infinity 2D-LC solution brings the power of two-dimensional LC-
separation to you at a never before experienced ease-of-use!
Highest separation power by 2D-LC combined with outstanding performance of the
1290 Infinity LC – the ideal tool for complex samples form biological origin, polymers,
food-extracs, and many more.
Innovative new hardware and software features for ease-of-use, reduced operation
costs and highest performance.
Upgradability or re-use of existing Agilent LC equipment.
Page 42
Achieving highest separation power with
the Agilent 1290 Infinity 2D LC Solution and GC Image® LCxLC Informatics
Stephen E. Reichenbach1,2
1GC Image, LLC, USA 2University of Nebraska – Lincoln, USA
Page 43
LCxLC Informatics
Non-targeted, multi-sample analyses require comparing every
constituent of every sample and so are the most challenging.
• Effective data preprocessing (baseline correction)
• Accurate peak detection
• Correct compound identification
• Robust cross-sample feature matching & analysis
• Automation
Adapt 10+ years of GCxGC informatics R&D.
Same informatics are applicable to targeted & single sample analyses.
Page 44
Data Preprocessing
Baseline correction is required to accurately detect and quantify peaks.
• Estimate baseline (possibly multichannel)
• Remove baseline from data (by subtraction)
LCxLC data exhibits two-dimensional baseline.
GC Image LCxLC Informatics performs baseline correction by two-
dimensional statistical modeling [1,2].
Page 45
Baseline Correction
Example of LCxLC baseline correction [3]:
• Before baseline correction: μ=3.55, σ=0.81
• After baseline correction: μ=-0.01, σ=0.04
Before (top) and after (bottom) baseline correction.
Page 46
Peak Detection
Peaks are two-dimensional, with (hopefully) several D2 chromatograms
across each peak.
GC Image LCxLC Informatics performs true two-dimensional peak
detection by Drain algorithm [4] (based on Watershed [5]).
Coeluting peaks may require unmixing, e.g., GC Image LCxLC
Informatics provides deconvolution by Parallel Factor Analysis
(PARAFAC) [6].
Page 47
Peak Detection & Identification
x Peaks in standards mix. Identifications based
on retention times/indices and/or UV/MS
spectra.
# Compound # Compound
1 Gallic acid 14 Morin
2 Esculin 15 Resveratrol
3 3,4 HO Benzoic acid 16 Salicylic acid
4 HO Phenacetic acid 17 Luteolin
5 6,7 HO Coumarin 18 Quercetin
6 HO Benzoic acid 19 Kaempferol
7 Syringic acid 20 Apigenin
8 Rutin 21 Naringenin
9 Naringin 22 Hesperetin
10 Coumaric acid 23 7 HO Flavone
11 Hesperidin 24 Pinosylvin
12 Ferulic acid 25 Chrysin
13 Myricetin 26 Flavone
Page 48
Compound Identification (HR-MS)
―Centroid
―Profile
―Isotopic Ratio
(A) Chromatographic image view; (B) Resveratrol peak apex spectrum; (C)
Formula calculator uses accurate mass. [7]
Page 49
Template Matching
Template matching is a powerful tool for automated identification [8].
Templates record a priori patterns (including retention times and spectra)
and rules with chemical logic for peak matching (CLIC™ [9]).
Template matching uses advanced pattern recognition to identify the
template pattern in new chromatograms.
Template metadata (including identities) are applied to the new
chromatogram.
Page 50
Template Matching
Open circles
are template
peaks.
Filled circles
are matched
target peaks.
Most peaks are matched, but some peaks may be undetected (trace or
co-eluted peaks), unmatched (retention-times distance or spectral
mismatch), or mismatched (nearer peak).
Page 51
Automation
Automated non-targeted multi-sample analysis:
1. Preprocessing & peak detection.
2. Pairwise template matching to find reliable peaks
(consistent across samples).
3. Align chromatograms & create composite
chromatogram (all peaks, all samples).
4. Define peak-region features, one per peak in
composite chromatogram.
Page 52
Automation
5. Create feature template with reliable peaks and
composite peak-regions.
6. Apply feature template to characterize each
chromatogram.
7. Use features for cross-sample analysis:
– Sample classification
– Chemical fingerprinting
– Monitoring
– Sample clustering
– Chemical marker discovery
Page 53
Example Data
LCxLC chromatograms for samples of three drinks.
Objective: Compare chemical compositions of three drinks:
red grape, antioxidant, merlot
– Four concentrations: 1, 2, 5, 10 μL
– Two replicate injections, 24 chromatograms
Red grape Antioxidant Merlot
Page 54
Example Results
Blue circles indicate drink-specific average is greater than overall average;
White circles indicate drink-specific average is less than overall average;
Areas indicate magnitudes.
• Differences for peak-region features between drink-specific
averages & overall averages.
• Analysis reveals differing compositions.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Red grape
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Merlot
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Antioxidant
Page 55
Example Results
Blue circles indicate first average is greater than second average; White
circles indicate second average is greater than first average; Areas indicate
differences.
• Pairwise differences for peak-region features between
drink-specific averages.
• Analysis reveals compositional differences.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Merlot – Antioxidant
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Antioxidant – Red grape
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Merlot – Red grape
Page 56
Example Results
Blue circles indicate first average is greater than second average; White
circles indicate second average is greater than first average; Areas relative
to sqrt of summed variances.
• Pairwise relative differences for peak-region features between drink-
specific averages.
• Analysis reveals varying relative compositions.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Antioxidant – Red grape
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Merlot – Red grape
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00 5.00 10.00 15.00 20.00
Merlot – Antioxidant
Page 57
Conclusions
• Automated workflow for LCxLC requires many operations.
• GC Image LCxLC informatics adapted from 10+ years of GCxGC
informatics R&D.
• Automated data processing allows fast, robust non-targeted, multi-
sample analyses.
• Example analysis shows capabilities for classification, fingerprinting,
etc.
Page 59
References 1. S. Reichenbach, M. Ni, D. Zhang, E. Ledford. "Image Background Removal in Comprehensive Two-Dimensional Gas
Chromatography." Journal of Chromatography A, 985(1-2):47–56, 2003.
2. S. Reichenbach, P. Carr, D. Stoll, Q. Tao. "Smart Templates for Peak Pattern Matching with Comprehensive Two-Dimensional Liquid
Chromatography." Journal of Chromatography A, 1216(16):3458–3466, 2009.
3. Data courtesy of P. Carr et al., University of Minnesota, 2011.
4. S. Reichenbach, M. Ni, V. Kottapalli, A. Visvanathan. "Information Technologies for Comprehensive Two-Dimensional Gas
Chromatography." Chemometrics and Intelligent Laboratory Systems, 71(2):107–120, 2004.
5. S. Beucher, C. Lantuejoul. "Use of watersheds in contour detection.", Int’l Workshop on Image Processing, Real-Time Edge and
Motion Detection/Estimation, pp. 17–21, 1979.
6. V. van Mispelaar, A. Tas, A. Smilde, P. Schoenmakers, A. van Asten. "Quantitative analysis of target components by comprehensive
two-dimensional gas chromatography." Journal of Chromatography A, 1019:15–29, 2003.
7. Data courtesy of E. Naegele et al., Agilent Technologies, 2012.
8. S. Reichenbach, P. Carr, D. Stoll, Q. Tao. "Smart Templates for Peak Pattern Matching with Comprehensive Two-Dimensional Liquid
Chromatography."Journal of Chromatography A, 1216(16):3458-–3466, 2009.
9. S. Reichenbach, V. Kottapalli, M. Ni, A. Visvanathan. "Computer Language for Identifying Chemicals with Comprehensive Two-
Dimensional Gas Chromatography and Mass Spectrometry (GCxGC-MS)." Journal of Chromatography A, 1071(1-2):263–269, 2005.
10. S. Reichenbach, X. Tian, Q. Tao, E. Ledford, Z. Wu, O. Fiehn. "Informatics for Cross-Sample Analysis with Comprehensive Two-
Dimensional Gas Chromatography and High-Resolution Mass Spectrometry (GCxGC-HRMS)." Talanta, 83(4):1279-1288, 2011.
11. S. Reichenbach, X. Tian, C. Cordero, Q. Tao. "Features for non-targeted cross-sample analysis with comprehensive two-dimensional
chromatography."Journal of Chromatography A, 1226:140-148, 2012.
Page 60
Learn more
For more information about the Agilent 1290 Infinity 2D-LC
Solution, visit www.agilent.com/chem/infinity-2D-LC to watch
video and get Technical Overviews.
60