Temperature Effects on the Optical Path Length of Infrared Liquid Transmission Cells

Abstract

Infrared (IR) spectroscopy is widely used for studies of temperature-dependent properties of liquids and solutions, such as thermal denaturation of proteins and other molecules of biological interest. The variation of the spectroscopic signals with temperature can be affected by the changes in the optical path length due to the thermal expansion of the components of the sample cell. In this report we investigate the temperature dependence of the optical path length for a liquid IR sample cell of a design typical for aqueous solution experiments. The path lengths were measured from the interference fringes, both in dry cells and with cells partially filled with water. We found that the optical path length variations are significant, on the order of several percent within the temperature range used (0-87 °C). Several commercially available spacers (Teflon, mylar, and lead) and gaskets (Teflon, lead, silicone rubber, Viton, and neoprene) were tested to find materials with either the smallest or most reproducible effect. Teflon, due to its phase transition (known as the "knee point") near room temperature, leads to abrupt changes in path length when used as either spacer or gasket component. On the other hand, Teflon is preferred for its inertness, while several of the other tested materials, most notably lead, are not practically usable due to adhesion to the cell windows upon heating and contact with the aqueous sample. The combination that yielded the most reproducible results, with minimal complications due to adhesion, was Teflon spacer with neoprene gaskets. The implications of the optical path length changes for the temperature-dependent IR experiments and their possible corrections are discussed.

Temperature dependence of amino acid side chain IR absorptions in the amide I' region

“…25 The amino acid concentrations were 16.7 6 0.2 mg mL 21 , determined by weight, except for acidic amino acids (Asp, Glu), for which 8.25 6 0.1 mg mL 21 was used. 21 Temperature was controlled automatically by an external water bath interfaced with the FTIR control software (Opus, Bruker Optics). All IR spectra were collected on a Bruker Tensor 27 FTIR spectrometer equipped with an RT-DLaTGS detector.…”

“…Three to five samples of each model compound were prepared to ensure reproducibility and for subsequent statistical analysis. 21 The spectra were collected as averages of 256 scans at a resolution of 4 cm -1 nominally from 0 to 87 C in 3 C increments (calibrated temperatures from 1 to 83 C) with a 3-min equilibration time preceding each measurement, which was experimentally determined to be sufficient for the sample to reach thermal equilibration. Samples were placed in a custom made IR liquid cell with CaF 2 windows separated by a 50 mm (nominally) Teflon spacer.…”

“…Samples were placed in a custom made IR liquid cell with CaF 2 windows separated by a 50 mm (nominally) Teflon spacer. 21 For determining absolute absorbances, the exact optical pathlength was calibrated as a function of temperature before and after the measurement of each sample using an empty cell, as detailed in previous work. 21 Temperature was controlled automatically by an external water bath interfaced with the FTIR control software (Opus, Bruker Optics).…”

“…Second, due to the interference from the most commonly used denaturants, urea and guanidinium, 14 the means for inducing peptide or protein unfolding are essentially limited to temperature. 21 Another complication for the analysis of amide I' IR spectra arises from the interference of amino acid side chain absorption bands. As we [15][16][17] and others [18][19][20] have shown, the amide I' of peptides in solution is strongly temperature dependent, mainly due to the interaction of the amide oscillator with the solvent.…”

“…For accurate analyses of thermally induced structural changes in peptides and proteins, the amide I' must be corrected for this intrinsic temperature dependence, along with additional experimental factors, such as the variations in the spectral pathlengths due to the thermal expansion of the sample cell components. 21 Another complication for the analysis of amide I' IR spectra arises from the interference of amino acid side chain absorption bands. Six amino acids: aspartate (Asp), glutamate (Glu), asparagine (Asn), glutamine (Gln), arginine (Arg), and tyrosine (Tyr) have non-negligible side chain IR absorptions that overlap the amide I' region.…”