An Economical Multidimensional Chromatographic Approach to the Analysis of Switchgrass Extractives — ASN Events

An Economical Multidimensional Chromatographic Approach to the Analysis of Switchgrass Extractives (#91)

Paul G Stevenson 1 2 , Abhijit Tarafder 2 , Paul B Filson 2 , Nicole Labbe 2 , Georges Guiochon 2
  1. Deakin University, Geelong
  2. The University of Tennessee, Knoxville

While switchgrass (Panicum virgatum L.) is considered a front-running feedstock for bioenergy, the complexity of the cell wall and recalcitrance attributable to compounds such as lignin currently renders the utilisation of switchgrass for ethanol production at an economic disadvantage. Therefore, it is of utmost importance to develop separation methods to isolate as many chemical compounds for further analysis.

Like most natural products, the sample matrix of switchgrass extractives is too complex to be separated by conventional one-dimensional separation techniques. Multidimensional chromatography affords a much greater peak capacity than traditional forms of chromatography as, ideally, the peak capacity is the product of each separation dimension. However, the systems that will most likely provide the greatest differences in selectivity are often incompatible and are very difficult to couple, for instance the combination of reversed and normal phase high performance liquid chromatography (HPLC). This paper illustrates the progress over previous reports of analytical scale HPLC × supercritical fluid chromatography (SFC) and shows how the difficulties of direct coupling can turn into the advantage of great separation power.

In this paper we show how a multidimensional approach for the analysis of switchgrass extractives can achieve a peak capacity in the order of 12,500 maintaining a strong peak economy. This was obtained by coupling reversed phase HPLC with SFC in the off-line mode of multidimensional analysis. We were able to achieve a 97% use of the available separation space, measured via the geometrical approach to factor analysis, and resolve 13.0 peaks per minute. This greatly exceeded the separation power that was achieved by our one dimensional HPLC separation of 4.4 peaks per minute, and only slightly lower than what was achieved with SFC, 16.2 peaks per minute.