ИСТИНА

Our aspiration was to design portable LC as a widely available ‘open platform’, moderate-pressure (over 100 bar), and importantly highly flexible modular system, based almost entirely on commercial off-the-shelf low-cost components. In this work, a modular portable LC has been designed, characterised and used, weighing under 2 kg (plus data acquisition). The backbone of the system was a breadboard assembled modular flexible microfluidic system (LabSmith), complemented with other off-the-shelf components, including an injection valve and on-capillary detectors, all operated through a PC. The system can accommodate six syringe pumps with 80, 20 and 5 µL syringe options, and maximum backpressures of up to ca. 150 bar, with two used for each A and B mobile phase in gradient operation and another pump for sampling and one in reserve for post-column derivatisation. The flow rate (typically 1 µL/min), which exhibited substantial pulsations due to the individual steps of the step motors driving the syringe pump, could be well dampened by using microbubbles of low-solubility inert gas (nitrogen, helium), automatically aspirated in each mobile phase draw cycle. Each of the two pairs of syringe pumps for mobile phases A and B were connected with microfluidic switching valves (two 4-port valves), and the A and B streams joined together with a Y-connector and a microfluidic pressure sensor, thus providing a system for low hold-up volume gradient formation. The mobile phase was led to a nano-LC sampling valve and to a capillary LC column. The detection was on-column (100 µm i.d. PTFE-coated fused silica) with an LED-based deep-UV-vis-NIR photometric detector, and end-column with an electrochemical detector (amperometric and/or potentiometric). The performance of the system was evaluated including the pumping accuracy, linearity of gradient formation, separation performance and sensitivity of detection. Typically analysis runs used 12-20 µL of solvent with overall reproducibility of 1 5 % and 3 6% RSD for retention times and peak area in gradient mode, respectively. Future avenues for miniaturisation of the data acquisition and system control are discussed.