Using the principles discovered during the CLC laboratory experiments, the Engineering Prototype was designed and built to perform CLC. It was designed using Solid Works 3D engineering design software (Dassault Systemes SolidWorks Corp.) and specification drawings were generated for each component. The parts were fabricated by various machine shops and the instrument was assembled at TerraSep. The instrument was made to fit onto the spindle of a commercially available super speed floor model centrifuge (maximum 21,000 rpm).
The Engineering Prototype is shown in the schematic diagram. The platter was designed to hold 24 radially arranged columns (5 mm ID X 125 mm length), but only 12 were finally used in the finished prototype. The yellow arrow shows where the TSD is attached to the centrifuge and the direction of rotation. Solvents are introduced via pumps in the core of the device and enter the columns driven by centrifugal force. After passing through the column the fluid flows back to the core of the device and is deposited in a solvent recovery tank. The proprietary configuration of the instrument allows the solvent flow to be regulated independently of the centrifugal force generated by the rotating instrument.
The next figure shows a cut away view of the platter that holds the columns through an individual column. Two solvents are pumped individually into the core solvent chamber by peristaltic pumps that are controlled by a software program that was created at TerraSep. The solvent flows are independently controlled to obtain gradients of different concentrations of the two solvents. The configuration of the mixing chamber ensures excellent mixing of the two solvents before entry into the columns.
The solvent then flows through the column and flow cell located at the end of the column. The solvent then returns to the core of the instrument where it flows into the solvent recovery tank. The configuration allows flow rate to be controlled by solvent input rate.
At the end of the column the solvent flow enters a flow cell where compounds that have been separated are detected. As shown on the diagram to the left, optical glass bounds the flow cell on the top and bottom allowing light to pass from the source below the platter through the solvent flow to a light detector located above the flow cell. The Engineering Prototype has a high output wavelength ultraviolet and visible range light source. The photomultiplier detects changes in the transmitted light. The digitized information from the detector is collected, analyzed and displayed in a standard chromatographic format.
The figure to the left shows the positioning of the components of the detection module. The light source is underneath the platter and is not shown in this diagram. The light from the light source is focused onto one end of a fiber optic cable. The other end of the fiber optic cable is positioned above the platter in close approximation to the top surface of the platter and at the edge of the platter directly above the flow cells. The second fiber optic cable collects the light that has traversed the flow cell and transmits the light to the photomultiplier detector. The single fixed detector collects information from all flow cells as the platter rotates and each passes under the fiber optic cable. Proprietary computer software was written to process the data from the detector and chromatograms constructed using a commercially available program. This design will allow multiple detectors for different types of light sources and wavelengths to be arrayed at positions around the circumference of the platter. In this way, multiple different modes of detection can be used to measure compounds in the solvent flow effectively simultaneously.
Using standard laboratory equipment, experiments were performed at TerraSep to examine the separation efficiency of CLC. Reverse phase column chromatography was performed using normal and reversed phase packing material of various particle diameter, column diameter, and column lengths. A proprietary CLC procedure was devised to separate common dye compounds (visible to eye) using various solvent mixtures. HPLC was performed on the same dyes using the identical column packing material and the results directly compared to CLC.
The results were startling.
HPLC using 3-micron particles achieved a separation efficiency of 1.6 X 105 plate number per meter. In contrast, CLC using
the same packing material achieved a separation efficiency of 1.1 X 109 plate number per meter. Therefore, the separation
efficiency of CLC was nearly 7,000 times more efficient than HPLC using a crude experimental procedure for CLC. The ability
to use particles as small as 0.05 microns, a controlled sample introduction method, and to control flow rates over the entire
process will allow this to be increased at least one hundred fold. This increase in separation efficiency is unprecedented
in separation sciences.
We directly compared Centrifugal Column Liquid Chromatography (CLC) and High Pressure Liquid Chromatography (HPLC) under isocratic conditions using the same columns packed with the same material using the same mobile phase at the same mobile phase linear velocity using the same sample (analyte) with the same injection volume (or nearly).
Under the conditions of analysis, TerraSep's prototype CLC is 43 times better than a modern commercial HPLC.