AbstractThis Application Note demonstrates the use of Agilent 1260 Infi nity Analytical Supercritical Fluid Chromatography coupled to an Agilent 6130B Single Quadrupole mass spectrometer (SFC/MS) to study chiral and structurally related isomers with high analysis speed and excellent separation effi ciency. The quantitative performance and analytical reproducibility were evaluated and proved to be highly reliable. This study also illustrates that modifi er composition has a great effect on both chromatographic separation and MS signal response in ESI positive ion mode, thus optimization of modifi er is crucial for successful SFC/MS analysis.
Fast Analysis of Chiral and Structurally Related Isomers Using Supercritical Fluid Chromatography Mass Spectrometry
Application Note
AuthorsYuqin Dai and Lisa ZangAgilent Technologies, Inc.Santa Clara, CA, USA
2
IntroductionLC/MS is used extensively in drug discovery applications. However, some limitations including separation effi ciency and speed of analysis for chiral and structural related isomers have been noted. SFC/MS offers an excellent alternative or orthogonal separation to LC/MS. Advantages of SFC include very low system back pressure with liquid CO2 and organic modifi ers, which allow high throughput analysis with high fl ow rate and unique selectivity of SFC for some compound types. Additionally, SFC has proven to be cost effective and is considered a green technique by reducing organic solvent consumption.
This Application Note describes the utility of SFC/MS for fast analysis of a number of chiral and structurally related isomers. The performance of an Agilent 1260 Infi nity SFC coupled to an Agilent 6130B Single Quadrupole MS, in terms of the separation effi ciency, speed of analysis, reproducibility, and quantitative aspect is demonstrated.
ExperimentalChemicals and methodSeveral chiral and structurally related isomers were selected for the evaluation. The structures and molecular information of these compounds are shown in Figure 1. Samples of individual test compounds were prepared in methanol. Chromatographic separation was optimized using the chiral Lux Cellulose-1 column and an Agilent ZORBAX Rx-SIL column respectively. Research grade CO2 gas (from Airgas) was used as the supply for mobile phase A and methanol with 20 mM ammonium formate (mobile phase B) was used as the modifi er of supercritical liquid CO2.
Full MS or SIM
scans in positive ion mode were used for qualitative and quantitative analyses of the analytes of interest.
Figure 1. Structures of test compounds.
WarfarinC19H16O4 (MW = 308.3) C15H25NO3 (MW = 267.4)
C25H35N3O6 (MW = 473.5)
C21H28O5 (MW = 360.4)
C10H15NO (MW = 165.1)
Metoprolol
Atenolol impurity F
Ephedrine hydrochloride (1S, 2R)
Prednisolone and Cortisone
CH3
CH3
OH
•HClHN
(1R, 2S)
CH3
NHCH3
OH
•HCl
O
OOH
O
HOH
H3CO
CH3
CH3
O N
OHHO
O O
H2N NH2
CH3H3C
O ON
HO
H
H H
HO OH
O
OOH
OHH
H HO
OO
3
InstrumentationAnalyses were performed using the Agilent 1260 Infi nity Analytical SFC system coupled to the Agilent 6130 Single Quadrupole mass spectrometer. The SFC/MS system consists of the following modules:
• Agilent 1260 Infi nity SFC control model (G4301A)
• Agilent 1260 Infi nity SFC binary pump (G3402A)
• Agilent 1260 Infi nity High Performance Degasser (G4225A)
• Agilent 1260 Infi nity SFC Autosampler (G4303A)
• Agilent 1290 Infi nity TCC (G1316C) with a 2-column selection valve
• Agilent 1260 Infi nity isocratic pump,
• Agilent 1260 Infi nity DAD (G1315C) with high pressure SFC fl ow cell
• Agilent 6130B Single Quadrupole Mass Spectrometer with ESI source.
The instrument confi guration is illustrated in Figure 2. Agilent OpenLab ChemStation Edition software version C.01.05 was used to control the SFC/MS instrument. In this study, the UV detector was bypassed. The pre-MS heating was achieved by fl ow through an additional TCC’s 6-µL preheating block with temperature set at 60 °C. As previously described1, the heating device prevents freezing of the lines caused by the expansion of CO2 upon exiting backpressure regulator (BPR). Additionally, a make-up fl ow was introduced prior to the BPR through an Agilent zero dead volume T-connector. This confi guration provides the best retention time and peak area reproducibility, thus it is used for qualitative and quantitative analysis.
The SFC/MS experimental conditions are summarized in Table 1.
Figure 2. Agilent 1260 Infi nity Analytical SFC/6130B Single Quadrupole mass spectrometer confi guration.
Agilent 1260 Isocratic Pump
Pre-MS heating
Agilent 1260 Infinity Analytical SFC System
Aurora A5
Agilent 6130B Single Quadrupole LC/MS
BPR
UV detector
SS capillary tubing 0.12 × 105 mm
60 °C
SS capillary tubing 0.12 × 400 mm
Table 1. SFC/MS conditions.
SFCColumns 1) Agilent ZORBAX Rx-SIL, 4.6 × 100 mm, 1.8 µm (p/n 828975-901)
2) Phenomenex Lux Cellulose-1, 4.6 × 150 mm, 3 µmColumn temperature 40 °CInjection volume (full loop) 5 µLBPR pressure 140 barBPR temperature 60 °CPre-MSD heating 60 °CMake-up fl ow 0.15 mL/min at 100 % BSupercritical fl uid (A) CO2
Modifi er (B) B1) Methanol with 20 mM ammonium formateB2) Methanol with 0.1 % formic acid
SFC fl ow and isocratic conditions
1) 4 mL/min at 20 % B for warfarin2) 2 mL/min at 35 % B for metoprolol3) 4.5 mL/min at 40 % B for impurity F (Cellulose-1 column)4) 4.5 mL/min at 40 % B for impurity F (Qual., Rx-SIL column)5) 4.5 mL/min at 30 % B for impurity F (Quant., Rx-SIL column)6) 3 mL/min at 20 % B for prednisolone and cortisone
SFC fl ow and gradient (for ephedrine mixture)
5 % B (initial), 5–10 % B (0-7 minutes), 10–40 % B (7–7.5 minutes), hold 40 % B for 0.5 minutes, 40–5 % B (8–8.5 minutes)
MSIon mode positiveCapillary voltage 3,000 VDrying gas 12 L/min at 350 °CNebulizer 50 psi
MS scan (200–450) with 20 % cycle time for SIM (warfarin, metoprolol, and the mixture of prednisolone and cortisone)MS scan (100–250) with 20 % cycle time for SIM (Ephedrine)SIM only (ion 474.3) (quantitative analysis for impurity F )
Peak width 0.03 minutesFragmental voltage 120
4
Results and DiscussionSeparation effi ciencySFC/MS provides several advantages over LC/MS, including high separation effi ciency, shorter analysis time, and reduced organic solvent consumption. Figure 3 shows the isocratic chromatographic separation of three representative chiral and structurally related isomers using either Lux Cellulose-1 or ZORBAX Rx-SIL columns by SFC/MS. All three pairs of isomers were well separated in less than 3.5 minutes.
Warfarin
Warfarin
1.010
20
40
60
80
100
120
140
160×103
1.5 2.0
1.751
2.856
2.5 3.0 3.5 min
A
Atenolol impurity F
Atenolol impurity F
0.60
10
20
30
40
50
60
70
×103
0.8 1.0
1.196
1.211
1.2 1.6 1.8 min
B
1.4
Cortisone
Prednisolone
0.20
10
30405060
8090
×103
0.4 0.6
0.597
0.775
0.8 1.2 1.4 min
C
70
20
1.0
Figure 3. SIM chromatograms of (A) 500 ng/mL of warfarin using the Lux Cellulose-1 column at 20 % B1, (B) 200 ng/mL of atenolol impurity F using the Agilent ZORBAX Rx-SIL column at 40 % B1, and (C) 200 ng/mL of cortisone and prednisolone mixture using the Agilent ZORBAX Rx-SIL column at 20 % B1.
5
modifi er B2 compared with modifi er B1 (Figure 4C). This modifi er effect was also observed for atenolol impurity F using ZORBAX Rx-SIL column (data not shown). It has been reported that the SFC elution strength of MeOH can be adjusted by acid-salt additives and their concentrations in separation of acidic enantiomer compounds2. Our fi nding showed a similar trend, in that the acidic additive (formic acid) did not help in chiral separation of the compounds studied. We found ammonium formate additive provided better separations.
containing 0.1 % formic acid). It is clear that a modifi er had a signifi cant effect on both chromatographic separation and MS signal response. Overall, modifi er B1 provided higher MS signals than modifi er B2 for all the isomers studied. However, the effect on the chromatographic separation was compound dependent. For example, modifi er B2 signifi cantly reduced separation effi ciency of metoprolol (Figure 4A), atenolol impurity F (Figure 4B), and a mixture of (1R, 2S) and (1S, 2R)-ephedrine isomers (data not shown), while similar chromatographic separation of warfarin was obtained using
Effect of Additive in MethanolChiral discrimination is a very complex phenomenon. It is almost impossible to predict which chiral stationary phase and modifi er combination will provide the best separation. Optimal conditions can vary greatly and are compound specifi c. In this study, two types of modifi ers (B1 and B2) were selected for the evaluation. Figure 4 shows separation profi les of three representative chiral molecules using Lux Cellulose-1 column and mobile phase CO2 in combination with either modifi er B1 (methanol containing 20 mM ammonium formate) or B2 (methanol
Figure 4. SIM chromatographic separation profi les of (A) metoprolol, (B) atenolol impurity F, and (C) warfarin using the Lux Cellulose-1, modifi er B1 (top panel) and B2 (bottom panel)
35 % B1
Metoprolol
00
50
150200250300
0.5 1.0
1.221
1.706
1.5 2.0 2.5 min
A
350400450
100
050
150200250300350400450
100
×103
35 % B2
0 0.5 1.0 1.5 2.0 2.5 min
×103
20 % B1
Warfarin
020
6080
100
1.25 1.75
1.752
2.860
1.711 2.896
2.0 2.5 3.0 min
C
120140160
40
050
150200250300350400450
100
1.5 2.25 2.75 3.25
1.0 1.25 1.75 2.0 2.5 3.0 min1.5 2.25 2.75 3.25
×103
20 % B2
×103
40 % B1
Atenolol impurity F
00
50
150200250300
1.0 1.4
1.406 1.602
1.6 2.0 2.2 min
B
350400450
100
050
150200250300350400450
100
1.2 1.8 2.4
0 1.0 1.4 1.6 2.0 2.2 min1.2 1.8 2.4
×103
40 % B2
×103
6
ReproducibilityReproducibility is critical for quantitation. Figure 5 shows the overlaid SIM chromatograms of six consecutive injections of warfarin (5A), metoprolol (5B), atenolol impurity F (5C), and a mixture of (1R, 2S) and (1S, 2R) ephedrine isomers (5D). Excellent reproducibility was obtained with RSDs of less than 0.1 % for retention time and less than 5 % for peak area.
Figure 5. Overlay of the SIM chromatograms from six replicate injections of (A) 500 ng/mL of warfarin, (B) 100 ng/mL of metoprolol, (C) 200 ng/mL of atenolol impurity F, and (D) 200 ng/mL of (1R, 2S) and (1S, 2R) ephedrine mixture using the SFC/MS. The separation was performed using the Lux-Cellulose-1 column and the mobile phase CO2 in combination with the modifi er B1.
Peak 1
Peak 2
N = 6RSD (RT)RSD (peak area)
Peak 10.051.2
Peak 20.030.8
020
6080
100
1.0 2.0 2.5 3.0 min
A
120140
40
1.5 3.5
×103 Peak 1
Peak 2
N = 6RSD (RT)RSD (peak area)
Peak 10.041.1
Peak 20.031.8
010
304050
1.0 1.0 1.5 2.0 min
B
6070
20
0.5 2.5
×103
Peak 1
Peak 2
N = 6RSD (RT)RSD (peak area)
Peak 10.081.0
Peak 20.000.7
05
152025
0 1.0 1.5 2.0 min
C
3035
10
0.5 2.5
×103
(1R, 2S)
(1S, 2R)
N = 6RSD (RT)RSD (peak area)
(1R, 2S)0.084.9
(1S, 2R)0.073.7
40
80
100
120
4.0 5.0 5.5 6.0 min
D140
60
4.5 6.5
×103
7
Quantitative performanceThe ZORBAX Rx-SIL column provided similar separation effi ciency for atenolol impurity F as Lux Cellulose-1 column. Since the silica column offers a cost benefi t compared to the chiral column, quantitative performance was further evaluated for atenolol impurity F using the ZORBAX Rx-SIL column and the SFC/MS in SIM mode only. As shown in Figure 6, the lower limit of quantitation (LLOQ) was 1.0 ng/mL with a linear dynamic range of 1.0–250 ng/mL.
ConclusionsThe Agilent 1260 Infi nity Analytical SFC System coupled to the Agilent 6130B Single Quadrupole System provided excellent analytical performance for analyzing chiral and structurally related isomers. The resulting data showed good sensitivity, high separation effi ciency, excellent retention time and peak area reproducibility, and reliable quantitation. Optimization of modifi er composition was important and methanol containing 20 mM ammonium formate proved to enable the successful SFC/MS analyses of all isomers in this study. Furthermore, the Agilent SFC/MS system uses standard grade CO2 gas instead of liquid SFC grade C CO2 gas, which results in 10–15x lower operating costs.
Figure 6. Quantitative analysis of atenolol impurity F using the Agilent ZORBAX Rx-SIL column and 30 % B1 at 4.5 mL/min. (A) calibration curve of Peak 1, (B) calibration curve of Peak 2, (C) SIM chromatogram of impurity F at 1.0 ng/mL (LLOQ).
A
160
×103 ×103
140120100806040200
0
1234
5
6
R2 = 0.99912y = 714.12x – 1,635.2
7
100Amount (ng/mL)
200Ar
ea
C
900
950
850
800
750
700
650
600
550
5001.0
Peak 2Peak 1
1.51.25Time (min)
2.01.75 2.25 2.5 2.75 3.0 3.25
Area
B2001751501251007550250
0
1234
5
6
R2 = 0.99909y = 815.95x – 1,941.6
7
100Amount (ng/mL)
200
www.agilent.com/chem
This information is subject to change without notice.
© Agilent Technologies, Inc., 2014Published in the USA, February 3, 20145991-3859EN
References1. M. Dunkel, G. Vanhoenacker, F. David,
P. Sandra, M. Vollmer “The Agilent 1260 Infi nity SFC/MS Solution” Agilent Technologies Publication Number 5990-7972EN, 2011.
2. T. Zhang, D. Nguyen, P. Franco (Chiral Technologies, Europe IIIkirch, France), M. Vollmer (Agilent Technologies, Waldbronn, Germany) “Enantiomer Separation of Acidic Compounds” Agilent Technology Publication Number 5990-9594EN, 2012.