A Complete Solution for Method Linearity in
HPLC and UHPLC
Frank Steiner, Fraser McLeod, Tobias Fehrenbach, and Andreas Brunner Dionex Corporation, Germering, Germany
Now sold under the
Thermo Scientific brand
IntroductIon
Validation for linearity requires the preparation and analysis of a set of several independently prepared solutions. As an example, according to ICH guidelines, HPLC method linearity is normally based on five concentration levels between 70% and 130% of the nominal concentration, each to be injected three times. The laboratory effort for this process can be substantial, especially if the method is required to quantify several substances simultaneously. To save time, it is common to prepare only one stock solution and subsequently dilute it to the different concentration levels. This procedure means that any error in the stock solution preparation is carried through all subsequent dilutions. In addition, it means that the preparation of the standards differs from the process used in day-to-day operation of the method, thus making comparison of the two processes difficult.
This study presents a fully automated gravimetric solution to prepare all solutions for a calibration or linearity test independently and without error-prone manual interference. All solutions prepared are immediately ready for injection onto the UHPLC analysis system. A comprehensive software package establishes the sequence for linearity validation automatically. When the HPLC results are available, linearity according to ICH guidelines is instantaneously calculated by the software, and all results are assessed against specification requirements. This complete solution accelerates the process significantly, excludes typical operator errors that occur during manual sample preparation and result calculation, and produces results of full statistical integrity.
ExpErImEntal
Instrumental Configuration
UltiMate® 3000 Basic Automated System supporting pressures up to 620 bar at flow rates up to 10 mL/min, consisting of the modules listed in Table 1.
Table 1. HPLC Module Configuration and
Connection Tubing
Pump LPG-3400SD with 200 µL SpinFlow mixer kit Sampler Column
Compartment ACC-3000
Detector DAD-3000 with semi-analytical flow cell (2.5 µL)
Capillaries Pump–Sampler 450 × 0.18 mm Viper™ Capillary Sampler–Column 550 × 0.18 mm Viper Capillary Column–Detector 450 × 0.13 mm Viper Capillary
2 Optimization of the Separation Power of 1-D Nano LC Analysis of Proteomics Samples A Complete Solution for Method Linearity in HPLC and UHPLC
UHPLC Method Parameters
The UHPLC method for analysis of soft drink ingredients used a core shell (fused core)-based 2.6 µm stationary phase material from Phenomenex. Detailed method parameters are listed in Table 2.
Table 2. UHPLC Column and Method Parameters
for Soft Drink Analysis
Column Phenomenex Kinetex 150 × 3.0 mm, 2.6 µm, C18 Precolumn Phenomenex KrudKatcher™, 0.5 µm
Injection Volume 20 µL Full-loop injection Temperature 45 °C Maximum Pres-sure 515 bar Data Collection Rate/Rise Time 50 Hz/0.1 s
Eluent A Water/0.075% formic acid (pH 3.65) Eluent B CH3CN/0.06% formic acid
Gradient 0.0 min 7% B 3.7 min 26% B 4.0 min 50% B 5.6 min 50% B 5.7 min 7% B 9.2 min 7% B
Instrument Control and Data Handling
The UltiMate 3000 system was controlled by Chromeleon® Chromatography Data System (CDS) software, version 6.80. The fully automated sequence generation, data processing, and validation report generation is based on the Chromeleon Extension Pack 2.
Sample Preparation
All standards for the linearity test of the soft drink analysis method were prepared using automation to the final concentration, without further dilution steps. Solid dosing was performed with the Quantos QB1-L system (Mettler-Toledo, Greifensee, Switzerland) using QH075-B dosing heads (Figure 1). The stacked dosing of all five compounds (acesulfame K, saccharin, caffeine, vanillin, and benzoate) was performed directly into a 100 mL flask. Dosing of the methanol/water 1:1 sample solvent was also performed gravimetrically with the Quantos QB liquid dosing option (Figure 1). After dosing, vials were closed with slotted septa. The samples were then dissolved and homogenized by manual shaking before the vials were placed into the autosampler trays.
Figure 1. Quantos QB1-L system for automated solid dosing and gravimetrically controlled liquid dosing.
rEsults and dIscussIon
Preparation of Samples for Linearity Validation
Statistically, it is recommended that each concentration level for a linearity test be individually prepared. This has the effect of randomizing potential sources of error (e.g., a possible incorrect weighing for one of the standards). This is a time consuming process, however, so many laboratories prepare a stock solution and then dilute it to the different concentration levels. The advantage of this approach is that it is a fast way of preparing standards, but the major disadvantage is that any error in the stock solution will be carried through to the diluted standards. A better approach is to automate preparation of the standards. This can be achieved using the new Quantos QB1-L system from Mettler-Toledo. The system can automatically dispense and weigh analytes into HPLC vials or volumetric flasks. It can also weigh in the appropriate amount of diluents in order to provide a gravimetric solution.This experiment tested the linearity of five analytes in a soft drink analysis. The Quantos system automatically weighed the correct amount of each analyte, then the correct amount of diluent to provide the concentrations shown in Table 3. The time required to prepare these standards was only 50 min. A manual preparation based on the stock solution approach would take approximately 2 h, or 4 h for the and manual individual preparation of all concentration levels, would require approximately 4 h.
3
Table 3. Concentrations (mg/100 g) of Soft Drink Analytes Prepared by Quantos QB1-L
Analyte Standard 1 Standard 2 Standard 3 Standard 4 Standard 5
Acesulfame K 7.390 9.730 9.930 11.450 13.650
Saccharin 2.695 3.480 3.890 4.305 5.085
Caffeine 2.800 3.700 3.995 4.355 5.220
Vanillin 6.285 8.110 8.995 9.860 11.655
Benzoate 14.050 18.100 19.665 21.910 25.855
Speeding the Analysis Method
Using classic HPLC, a single soft drink analysis may take approximately 30 min. This means that the total run time for the linearity experiment will be 7.5 h (5 calibration levels × 3 injections per level × 30 min). With UHPLC, however, it is possible to signficantly decrease run time. Figure 2 shows the analysis of all five analytes, with all peaks separated in less than 4 min. This means that all injections can be performed in just 1.25 h. The sequence generation with three injections at five different concentration levels is performed automatically by the chromatography data system, thus reducing labor time.
Figure 2. UHPLC analysis of five soft drink ingredients in less than 5 min.
Data Analysis and Reporting of the Method Linearity
Validation
Once all data has been acquired, it is necessary to calculate the results. In keeping with ICH guidelines, the following values must be reported: correlation coefficient, y-intercept, slope of the regression line, and residual sum of squares. Laboratories must also confirm that the correlation coefficient is within the limits expected of the method (typically ≥0.999).
Performing these calculations can be a time consuming task. Some laboratories use Excel spreadsheets in an effort to speed the process, but even this can be inefficient because users typically manually transcribe values into the spreadsheet, then have another person review the transcription. For this particular analysis, the estimated time for the use of spreadsheets and the associated review step is 2 h.
28277 Minutes 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Acesulfame K Saccharin Caffeine Vanillin Benzoate
4 Optimization of the Separation Power of 1-D Nano LC Analysis of Proteomics Samples A Complete Solution for Method Linearity in HPLC and UHPLC
The Chromeleon CDS software immediately generates results for all analytes and creates the complete report automatically. All that is required by the operator is to name the peaks and enter their concentrations—a process that takes just 5 min. Figure 3 shows the automatically generated report for the analyte saccharin.
The combination of the Quantos QB1-L and the Dionex UltiMate 3000 system produced outstanding results for the linearity experiment. Table 4 shows the correlation coefficient for all analytes, each of which has an R2 value >0.999.
Table 4. Correlation Coefficients of All Analytes
Analyte R2 Acesulfame K 0.99973 Saccharin 0.99989 Caffeine 0.99939 Vanillin 0.99971 Benzoate 0.99964
Save Time with Transfer to New Process
As Table 5 shows, the time to perform the experiment using the traditional process is 13.5 h. This process involves manual sample preparation with individual weighing, HPLC analysis at conventional speed, manual data processing, and report creation. In contrast, the fully automated new UHPLC-based process can complete the required steps in just over 2 h, which corresponds to a more than 6-fold overall productivity increase.
Table 5. Time to Perform Linearity Experiment:
Traditional vs New Automated Process
Step Time Taken
(Traditional Process) (New Process)Time Taken
Sample Prep 240 min 50 min
Analysis 450 min 75 min
Results 120 min 5 min
Total 810 min 130 min
Figure 3A
5
Speed
•Simplicity
•Solutions
Chromeleon and UltiMate are registered trademarks, and Viper is a trademark of Dionex a Thermo Fisher Scientific Corporation. KrudKatcher is a trademark of Phenomenex, Inc.
north america U.S. / Canada (847) 295-7500 south america Brazil (55) 11 3731 5140 Europe Austria (43) 1 616 51 25 Benelux (31) 20 683 9768; (32) 3 353 4294 Denmark (45) 36 36 90 90 France (33) 1 39 30 01 10 Germany (49) 6126 991 0 Ireland (353) 1 644 0064 Italy (39) 02 51 62 1267 Sweden (46) 8 473 3380 Switzerland (41) 62 205 9966 United Kingdom (44) 1276 691722
asia pacific
Australia (61) 2 9420 5233 China (852) 2428 3282 India (91) 22 2764 2735 Japan (81) 6 6885 1213 Korea (82) 2 2653 2580 Singapore (65) 6289 1190 Taiwan (886) 2 8751 6655 dionex products 1228 Titan Way P.O. Box 3603 Sunnyvale, CA 94088-3603 (408) 737-0700 www.thermoscientific.com/dionex LPN 2867-01 3/11 ©2011 Thermo Fisher Corporation, Inc.
summary
• The automated Quantos system enables a 4-fold reduction in sample preparation time, relative to manual weighing and liquid dosing. This approach encourages individual weighing for all concentration levels (rather than diluting from a stock solution) yet avoids human error and provides full electronic documentation.
• Transfer of the analysis to the new UHPLC process can accelerate separation time by factor of 6, and yield chromatograms with excellent peak resolution for the highest accuracy of analytical results. • The greatest potential for time saving is in the fully automated data
processing and reporting step. Using Chromeleon Extension Pack 2 for ICH-conforming method validation, this task can be performed up to 24× faster than the manual approach. In addition, the new process excludes possible errors associated with transfer of data from the chromatography software to the data evaluation and reporting software.
