Page 332 - Tandem Techniques

Page 316

They claimed that the use of this cell would reduce the detection limit (that is the minimum mass
required to produce a useful IR spectrum) to as little as 50 ng of material. A design of their flow cell is
depicted in Figure 8.21. The cell was fabricated from a crystalline block, of either calcium fluoride or
potassium bromide, 10 x 10 x 6 mm, with a 0.75 mm hole drilled through it. The cylindrical aperture in
the crystal carried the mobile phase from the column through the block and then to waste. The
collimated IR beam passes through the block, normal to the flow of the mobile phase, and is arranged to
transverse a section of the 'hole' conduit. Since the focal diameter of the FT/IR spectrometer was 3 mm,
and the hole in the cell only 0.75 mm, a beam condenser was used to reduce the focal diameter to that of
the hole. The authors observed that a maximum IR signal could be obtained when the peak maximum
was in the light path. Nevertheless, they also noted that the maximum signal-to-noise was obtained by
summing the spectra from scans taken across the peak, between ± 1.53 s of the Gaussian profile, as it
passed through the cell. They also reported a general point of interest when designing such cells. It was
found easier, and more effective, to modify optically the size of the IR beam to suit those of the flow
cell, than to try to fabricate the cell to have proportions that will accurately suit the dimensions of the IR
beam.

Sabo et al. [19] developed an LC/FTIR interface for both normal and reversed phase chromatography,
using an attenuated total reflectance cell. The flow-through cell is made from a cylindrical shaped zinc
selenide crystal, with cone-shaped ends, mounted in a stainless steel cell. The crystal, blazed at 45°,
gives 10 reflections of the IR beam down its length. The cell fits into an optical bench, which focuses
the incident beam onto the incident cone-face, and directs the radiation leaving the crystal onto the
detector.

The volume of the cell is rather large, vis. 24 µ1, which could adversely affect the resolution of a small-
bore column. On-the-fly spectra of the components from a 100 µl sample of a solution, containing 2%
of acetophenone and ethyl benzoate and 1% of nitrobenzene, gave clearly identifiable spectra.
Nevertheless, relative to other LC/FTIR systems, this was not a very sensitive device.