Soo Ryeon Ryu

Two-dimensional correlation analysis and waterfall plots for detecting positional fluctuations of spectral changes.

In this study, we demonstrate the potentials and pitfalls of using various waterfall plots, such as conventional waterfall plots, two-dimensional (2D) gradient maps, moving window two-dimensional analysis (MW2D), perturbation-correlation moving window two-dimensional analysis (PCMW2D), and moving window principal component analysis two-dimensional correlation analysis (MWPCA2D), in the detection of the existence of band position shifts. Waterfall plots of the simulated spectral datasets are compared with conventional 2D correlation spectra. Different waterfall plots give different features in differentiating the behaviors of frequency shift versus two overlapped bands. Two-dimensional correlation spectra clearly show the very characteristic cluster pattern for both band position shifts and two overlapped bands. The vivid pattern differences are readily detectable in various waterfalls plots. Various types of waterfall plots of temperature-dependent infrared (IR) spectra of ethylene glycol, which does not have the actual band shift but only two overlapped bands, and of Fourier transform infrared (FT-IR) spectra of 2 wt% acetone in a mixed solvent of CHCl3/CCl4 demonstrate that waterfall plots are not able to unambiguously detect the difference between real band shift and two overlapped bands. Thus, the presence or lack of the asynchronous 2D butterfly pattern seems like the most effective diagnostic tool for band shift detection.

Two-dimensional heterospectral correlation analysis of X-ray photoelectron spectra and infrared spectra for spin-coated films of biodegradable poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) copolymers.

We investigated the thermal behavior of spin-coated films of biodegradable poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (P(HB-co-HHx)) copolymers at the molecular level. To better understand details of thermal behavior of spin-coated films of P(HB-co-HHx) copolymers, we applied two-dimensional (2D) correlation analysis to the spectra of P(HB-co-HHx) (HHx = 12.0, 10.0, and 3.8 mol %) copolymers during the heating process from 30 to 150 degrees C as obtained by temperature-dependent infrared-reflection absorption spectroscopy and X-ray photoelectron spectroscopy (XPS). 2D IR and 2D XPS correlation spectra of spin-coated films of P(HB-co-HHx) copolymers clearly revealed the sequence of intensity changes with increasing temperature: an amorphous band increases first and then a band for less ordered secondary crystals decreases before a band for well-ordered primary crystals. Furthermore, the synchronous 2D heterospectral XPS/IR correlation spectrum elucidated the correlation between the IR and XPS bands, confirming their band assignments. The asynchronous 2D heterospectral correlation spectrum revealed the probe-dependent asynchronicity between XPS and IR signals arising from the same species even under identical perturbation conditions because of the different scales of molecular changes probed. It clearly provides a complete interpretation of the phase transition phenomenon of P(HB-co-HHx) copolymers, which could not have been obtained through XPS or IR study alone, and also, the results obtained thereof offer a new insight into the molecular interactions as well observed by two different probes.

What is the origin of positional fluctuation of spectral features: true frequency shift or relative intensity changes of two overlapped bands?

We investigated what is really meant by so-called positional or frequency fluctuation of spectral features. To show the difference between the true frequency shift of a single band and apparent peak maximum shift caused by relative intensity changes of overlapped adjacent bands, we analyzed infrared (IR) spectra of the OH stretching band of ethylene glycol during the heating process and the C=O stretching band of acetone in a mixed solvent CHCl3/CCl4 with varying solvent compositions. These spectra are well-known examples of so-called “band shift” phenomena often interpreted as the manifestation of gradual changes in the IR frequency associated with a specific chemical bond under the influence of molecular interactions. Analyses of IR spectra showed that the apparent positional shifts of peak maxima in these systems are actually due to relative contribution changes of two overlapped bands, instead of the gradual frequency shift of a single band induced by the change in the strength of molecular interactions. To further clarify our interpretation of “peak maximum shifts”, we also analyzed simulated spectral datasets, comparing the true band frequency shift and change in the relative contributions of overlapped bands. It was found that principal component analysis (PCA) is a surprisingly sensitive tool to distinguish the two possible mechanisms of peak maximum shift. The new insight revealed by this study should help the interpretation of molecular interactions probed by vibrational spectroscopy.