Saturday, December 18, 2010
Method Development in UHPLC
Method Development in UHPLC
The rules for developing a new method in UHPLC are slightly different from those of conventional HPLC because it is necessary to account for the backpressure constraint generated by the use of small particles.
3.1. Choice of column dimensions
Depending on the supplier, it is possible to find some columns dedicated to UHPLC with internal diameters of 1, 2.1 and 4.6 mm. As discussed previously, the 4.6 mm I.D. column is not of great interest because of the significant frictional heating generated by high mobile phase flow rates, generating a lack of repeatability of retention times and some potential difficulties in transferring methods from conventional HPLC. In addition, the consumption of organic solvent is critical as the flow rate should be in the range 3-5 mL/min. Concerning the 1mm I.D column, the effect of frictional heating should be neglected even at 1000 bar but the compatibility between this column geometry and any UHPLC instrument is extremely critical (particularly for tubing geometry). Due to these statements, the 2.1 mm I.D. column should be considered as optimal for UHPLC operation.
Regarding the column length, it should be selected according to the required efficiency (in isocratic mode) or peak capacity (in gradient mode). It has to be noted that there is no real limitation in UHPLC column length. Numerous suppliers propose some 150mm columns which can be coupled in series using low volume tubing if an experiment has to be performed with very long columns.
In the isocratic mode, it is well known from the basic equations of chromatography that efficiency is directly proportional to the column length. Therefore, a 50x2.1 mm, 1.9 µm column should generate around 10,000 plates, while the efficiency is increased to 20,000 and 30,000 plates with a 100 and 150x2.1 mm, 1.9 µm column respectively. Therefore, the column length should be chosen according to the requirement of the separation and the longest column always provides the highest efficiency. However, with 150 mm or longer columns, the mobile phase flow rate should be adapted to avoid over-pressurizing the analytical system.
In the gradient mode, the relationship between column length and chromatographic performance (expressed as peak capacity in gradient mode) is not trivial. In fact, the peak capacity depends both on efficiency and column dead time, but each to a different extent. Therefore, the column length should be adapted according to the gradient time and the longest column doesn’t necessarily provide the highest peak capacity. It can be demonstrated that a 50x2.1 mm, 1.9 µm column has to be selected for gradient times lower than 5 minutes. The 100x2.1 mm, 1.9 µm column gives optimal performance for gradient times between 5 and 20 minutes and finally, the 150x2.1 mm, 1.9 µm column should only be used with gradients longer than 20 minutes.
3.2. Choice of mobile phase conditions
In UHPLC, the mobile phase flow rate has to be selected according to the Van Deemter curve (similarly to conventional HPLC) but also, according to the backpressure generated. For compounds with molecular weights in the range 100-400 g.mol-1 , the optimal flow rate in isocratic mode for a 2.1 mm I.D. column packed with 1.9 µm particles is around 400-600 µL/min. Due to the low mass transfer resistance generated by small particles (because of the reduced diffusion path), it is even possible to work up to 1000 µL/min, with a limited loss in efficiency (around 20%). When dealing with larger molecules, the mobile phase flow rate should be reduced to 200-400 µL/min due to a reduction of diffusion coefficients.
The rules in gradient mode are different, where the highest flow rate always provides the best peak capacity because it is dependent on the column dead time and, to a lesser extent, on efficiency. Therefore, the flow rate for gradient UHPLC experiments should be elevated but at the maximum equal to 80-90% of the maximum pressure withstood by the instrument. This solution is recommended to attain a sufficient level of robustness and to handle unexpected changes in column backpressure after hundreds of injections.
Regarding the mobile phase temperature, it could be valuable to work in UHPLC at a mobile phase temperature of 40-50°C instead of room temperature. With this strategy, the mobile phase viscosity is reduced and the backpressure diminishes by about 30% (at 50°C for an acetonitrile-water mobile phase) without affecting chromatographic performance.
Finally, it is well known that the viscosity of acetonitrile-water hydro-organic mixtures is on average 1.5 to 2-fold lower than that of methanol-water. Therefore, in the case of method development, the initial choice for mobile phase is acetonitrile as it generates significantly lower backpressure and more possibilities in UHPLC
compared to methanol (particularly for the choice of column length).
3.3. Decision tree for method development
On the basis of the above discussion, it is possible to establish some generic conditions for the UHPLC method development: the column should be initially a C18 with geometry of 50x2.1 mm, 1.9 µm operating at a temperature of 40-50°C. The mobile phase should c onsist in a mixture of acetonitrile and buffer. It is generally better to begin the experiments in the gradient mode at a mobile phase flow rate close to the maximal pressure accepted by the UHPLC instrument. Regarding the choice of gradient time, it is extremely different in UHPLC compared to regular HPLC. Therefore, table 1 summarizes the optimal gradient time for various sets of UHPLC conditions.
If the selectivity with the generic conditions previously described is not acceptable, it is possible to adapt various parameters (e.g. mobile phase pH, nature of the stationary phase and organic modifier). In UHPLC, the first parameter to change is the mobile phase pH, then the column chemistry and finally the organic modifier nature (because of the constraint in pressure with UHPLC).