15
strength gradient profile is shown in Table 1. Peak tracking was accomplished with PeakMatch®
, DAD spectroscopy and
comparison of UV spectral matches with a pre- constructed spectral library from individual analyte injections. Integrated data, which included retention time, peak area and peak width at half-height, were exported into Microsoft Excel and arranged in a table, in which one peak was located with all its data in one single horizontal line, the table was then copied and pasted into the simulation software. Alternatively, Analytical Instrument Association file extensions (AIA files) were exported from the ChemStation data capture programme into Peak Match®
for peak
tracking purposes. Time (min)
%B
05 30 35 36 51
100 100 5 5
Table 1. Example of a typical gradient input conditions employed (30 minute gradient shown).
2.4 Software 2.4.1 Chromatography modelling and prediction software
Peaks were identified and aligned based on peak areas using the PeakMatch®
software
(v. 3.6.3, Molnár-Institute Berlin, Germany), which became part of DryLab®
2010, having
user friendly tools, such as peak turnover and peak splitting functions, which greatly reduce the ubiquitous problem of peak misassignment. Virtual experimentation with HPLC runs (“modelling”) was performed in DryLab®
a recently developed 3-D-device for Design Space visualisation. Predictions were compared with the original experiments to control the validity of the modelling process. Generation of 3-D resolution models was carried out with a new proprietary algorithm.
2.5.1 Experiments for modelling Four initial input data runs were acquired under the following conditions: Gradient times (tG
) of
15 and 45 min, temperatures of 40°C and 60°C, eluent A and B as described in section 2.3. The organic modifier in eluent B consisted of either MeOH, AN or mixtures of both: (MeOH:AN) (1:1)(V/V). 12 binary and ternary mobile phase conditions were chromatographed as shown in Fig. 7. Input data, 3 x 4 experimental runs were performed overnight. After the chromatograms were collected, they were
Figure 1. Screenshot of the experimental chromatograms and tabulated peak assignments in PeakMatch®
: Retention times and
peak areas were obtained using the following chromatographic conditions: Gradient times, temperature, column, gradient range, flow rate and eluent A as stated in the experimental section. Eluent B was B1 as described in section 2.3. Binary gradients for the MeOH-tG
–T plane are as follows: Chromatograms correspond on the left and right sides to tG min respectively, whereas the top and bottom ones correspond to T: 60° and 40°C respectively. exported into Peak Match® for peak tracking.
The data were then finally transferred by one mouse click to DryLab®
2010 (Fig. 2).
3. Results and discussion The approach of two dimensional modelling of gradient time and temperature is possibly the most widespread type of modelling and optimisation protocol performed within the pharmaceutical industry [30]
. Acetonitrile is 2010 v. 3.9, (Molnár-Institute) including
commonly used as the organic component of the mobile phase due to its low viscosity and UV cut off – however, recent shortages of AN and its associated high costs, have forced chromatographers to re-assess their need for AN. The use of selectivity differences of “ternary eluent systems”, and hence varying chromatographic resolution and selectivity between the organic modifiers acetonitrile and methanol in RP binary gradient chromatography is well documented [3,24,29]
.
Several of the commercially available software packages can model the retention of analytes using gradient chromatography as a function of ternary mobile phase composition [29]
. In a recent publication, Molnár et al[28] have
shown that this approach can be extended to 3-dimensional modelling of tG
, temperature
(T) and ternary composition or pH. Only twelve experimental input runs - two differing gradient times, two differing temperatures and three differing ternary eluent compositions have been shown to be necessary in order to reliably create 3- dimensional retention models of these factors within DryLab®
2010.
Fig. 1 highlights the usefulness of PeakMatch® in graphically displaying the chromatograms with strongly different selectivity’s and their associated data (i.e. peak assignments, tR and peak areas), in an orderly manner. Each analyte should be tabulated in a single horizontal row to aid peak assignment. The numbers above the peaks correspond to peak areas, which are obtained directly from the integration software and are then reduced by a factor of 105
to
This paper will highlight the use of this approach in the method development of the separation of a range of twenty pharmaceutically relevant bases and two neutral components of widely differing chemical/physical properties in a rapid and semi-automated fashion. The work flow involved the design of the experiments, automated collection of the input data (12 experimental runs – i.e. 3 x tG
– T models),
semi-automated peak tracking, automated creation of the 3-dimensional resolution space (i.e. retention models), validation of the model using three extra experimental runs, exploration and evaluation of the Design Space for optimum, robust regions of the model to work in and confirmation of the desired modelled chromatogram within the Design Space to that obtained experimentally.
3.1 Peak tracking using PeakMatch® The chromatograms are imported directly into the PeakMatch®
software from the
ChemStation data files as compatible AIA/AnDI files (*.cdf).
:15 min and 45
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
