🔬 condensed matter

Transitions between liquid crystalline phases investigated by dielectric and infra-red spectroscopies

This study investigates the phase transitions of the liquid crystalline 11OS5 compound from isotropic liquid to crystal using broadband dielectric and infra-red spectroscopies, supported by density-functional theory calculations, to identify molecular interactions and detect significant dynamic changes such as the slowing of flip-flop relaxation at the smectic C to hexagonal smectic X transition.

Original authors: Aleksandra Deptuch, Natalia Osiecka-Drewniak, Anna Paliga, Natalia Górska, Anna Drzewicz, Katarzyna Chat, Mirosława D. Ossowska-Chruściel, Janusz Chruściel

Published 2026-02-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Aleksandra Deptuch, Natalia Osiecka-Drewniak, Anna Paliga, Natalia Górska, Anna Drzewicz, Katarzyna Chat, Mirosława D. Ossowska-Chruściel, Janusz Chruściel

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a crowd of people at a party. Sometimes they are milling about randomly (a liquid), sometimes they are standing in neat rows facing the same direction (a crystal), and sometimes they are in a weird middle state where they are lined up but can still slide past each other (a liquid crystal).

This paper is about studying a specific chemical molecule called 11OS5 as it goes through these different "mood swings" or phases. The scientists used two main tools to watch the party: Dielectric Spectroscopy (listening to how the molecules react to an electric "shout") and Infrared Spectroscopy (taking a "fingerprint" of how the molecules vibrate).

Here is the story of what they found, explained simply:

1. The Two Tools: Listening and Fingerprinting

The Electric Shout (Dielectric Spectroscopy): Imagine shouting at the crowd. If the people are free to move, they turn their heads quickly. If they are stuck in a rigid formation, they can't move at all. The scientists measured how fast the molecules could "flip" or turn around when an electric field was applied.
- What they found: In the warmer, looser phases, the molecules were flipping around like happy dancers. But as the temperature dropped and the molecules got stuck in a specific, ordered pattern (the Smectic X phase), they suddenly stopped flipping. It was like the music stopped, and everyone froze in place. The "flip" slowed down so much it became almost invisible.
The Fingerprint (Infrared Spectroscopy): Every molecule vibrates like a guitar string when hit by light. Different vibrations create different "notes." By listening to these notes, the scientists could tell if the molecules were standing alone or hugging their neighbors.

2. The "Crystal" Mystery: How do they sit?

When the molecules finally turned into a solid crystal, the scientists wanted to know: How are they sitting next to each other?

The Theory: They used a supercomputer to simulate two ways the molecules could sit:
1. Head-to-Head: Like two people shaking hands face-to-face.
2. Head-to-Tail: Like two people standing in a line, one behind the other.
The Detective Work: They compared the computer's "predicted notes" with the "real notes" from the experiment.
The Verdict: The real notes matched the Head-to-Tail scenario. It turns out the molecules prefer to stand in a line, with the "head" of one molecule interacting with the "tail" of the next. The "Head-to-Head" idea didn't fit the data.

3. The "Blurry" Transition

One of the trickiest parts of the study was a transition between two very similar liquid crystal phases (Smectic A and Smectic C).

The Problem: These two phases are like twins. In one, the molecules stand straight up; in the other, they tilt just a tiny bit (only 11 degrees). It's so subtle that standard tools couldn't tell them apart.
The Solution: The scientists tried using a fancy math trick called K-Means Clustering (basically, a computer sorting algorithm that groups similar things together).
The Result: The computer was good at spotting the big changes (like when the party turned into a solid block), but it got confused by the tiny tilt. It couldn't clearly separate the two "twin" phases. This taught the researchers that while math is powerful, sometimes you need to know the "story" of the phases beforehand to interpret the data correctly.

4. The "Stretching" Band

The scientists focused on one specific "note" in the fingerprint: the C=O stretch (a carbon-oxygen bond vibrating).

The Observation: As the molecules got colder and packed tighter, this note got lower in pitch (a "redshift").
The Analogy: Imagine a rubber band. If you pull it tight, it vibrates differently than when it's loose. The change in pitch told the scientists that as the molecules got colder, they started forming weak "handshakes" (hydrogen bonds) with their neighbors, tightening the whole structure.

The Big Takeaway

This paper is a detective story about how molecules behave when they get cold.

They freeze: As they cool down, they stop spinning and get locked into a rigid crystal.
They line up: In the crystal, they stand in a "head-to-tail" line, not face-to-face.
They are tricky: Some changes are so subtle that even smart computers need a human's help to spot them.

By combining the "electric listening" and the "vibrational fingerprinting," the scientists built a complete picture of how this molecule transforms from a flowing liquid to a structured solid.

1. Problem Statement

The study focuses on the compound 11OS5 (4-pentylphenyl-4'-undecyloxythiobenzoate), a liquid crystal that exhibits a complex phase sequence including isotropic (Iso), nematic (N), smectic A (SmA), smectic C (SmC), a metastable hexagonal smectic (SmX), and crystal (Cr) phases.

The Challenge: While macroscopic phase transitions are often detectable via calorimetry or optical microscopy, subtle structural changes between specific mesophases (particularly the SmA $\to$ SmC transition and the SmC $\to$ SmX transition) are difficult to characterize.
Specific Gaps: The SmA $\to$ SmC transition in 11OS5 involves a very small tilt angle change ( $\approx 11^\circ$ ) and is barely visible in differential scanning calorimetry (DSC). Furthermore, the molecular arrangement and intermolecular interactions within the crystal phase (specifically whether molecules adopt a head-to-head or head-to-tail configuration) require detailed spectroscopic and theoretical validation.

2. Methodology

The authors employed a multi-modal approach combining experimental spectroscopy with computational modeling:

Broadband Dielectric Spectroscopy (BDS):
- Measurements were taken from 373 K to 273 K (cooling) in the frequency range of 0.1– $10^7$ Hz.
- Data analysis involved fitting complex permittivity spectra using the Cole-Cole model to extract relaxation times ( $\tau$ ), dielectric strengths ( $\Delta\varepsilon$ ), and ionic conductivity parameters.
Infra-Red (IR) Spectroscopy:
- Experimental: FT-IR spectra were collected in transmission mode (KBr pellets) during cooling and in reflection mode at room temperature.
- Data Analysis: Two advanced statistical methods were applied to detect subtle spectral changes:
  1. Correlation Coefficient Matrix: To quantify similarity between spectra at different temperatures.
  2. K-means Cluster Analysis: To group spectra into distinct phases based on Euclidean distance, testing various spectral sub-ranges.
Density Functional Theory (DFT) Calculations:
- Performed using Gaussian 16 to interpret experimental IR bands.
- Levels of Theory: Various basis sets (def2SVP, def2SVPP, def2TZVP, def2TZVPP, 6-31+G(d), 6-311+G(d,p)) and functionals (B3LYP-D3(BJ), BLYP-D3(BJ)) were tested.
- Scaling: Calculated frequencies were scaled against experimental data to determine Root Mean Square Error (RMSE) and computational efficiency.
- Dimer Modeling: To determine crystal packing, IR spectra were calculated for head-to-head and head-to-tail dimers to simulate intermolecular interactions.

3. Key Contributions

Comprehensive Phase Characterization: Provided a complete dielectric and IR profile of 11OS5 across all its phases, including the metastable SmX phase.
Methodological Benchmarking: Systematically compared the performance (accuracy vs. computational cost) of various DFT basis sets and functionals for predicting IR spectra of liquid crystalline thiobenzoates.
Advanced Data Mining: Demonstrated the utility of correlation matrices and k-means clustering in distinguishing liquid crystal phases where traditional peak-shifting analysis fails (e.g., SmA vs. SmC).
Structural Insight: Resolved the molecular configuration in the crystal phase by correlating experimental band shifts with DFT dimer calculations.

4. Key Results

A. Dielectric Spectroscopy Findings

Relaxation Processes: Two processes were identified in Iso, N, SmA, and SmC phases:
1. Maxwell-Wagner-Sillars (MWS): Low-frequency process related to interfacial polarization.
2. Molecular s-process (Flip-flop): High-frequency rotation of molecules around their short axes.
Phase Transitions:
- SmA $\to$ SmC: Detected only as a very small step in dielectric dispersion at 1 kHz; the relaxation parameters remained largely unchanged.
- SmC $\to$ SmX: A dramatic change occurred. The s-process relaxation time slowed down by a factor of 46 (shifting to lower frequencies), and the activation energy increased from 94 kJ/mol (SmC) to 144 kJ/mol (SmX). This indicates the onset of long-range positional order within layers.
- Crystallization: No relaxation processes were detected in the crystal phase, indicating a complete loss of orientational and conformational disorder.

B. IR Spectroscopy and DFT Analysis

DFT Performance:
- The def2TZVPP/B3LYP-D3(BJ) level yielded the lowest RMSE (11.6 cm $^{-1}$ ) but required the longest computation time (156.9 h).
- The def2TZVP/B3LYP-D3(BJ) level offered a near-identical RMSE (11.9 cm $^{-1}$ ) in half the time (85.8 h).
- The 631+Gd/B3LYP-D3(BJ) level provided a good balance with an RMSE of 12.5 cm $^{-1}$ and a significantly shorter time (17.6 h).
Spectral Analysis of Transitions:
- SmA $\to$ SmC: The correlation matrix and k-means clustering failed to distinguish these phases, confirming that the structural difference (tilt angle) is too subtle to alter the intra-molecular vibrations significantly.
- SmC $\to$ SmX: Detected via specific bands (860–930 cm $^{-1}$ , 1065–1290 cm $^{-1}$ , 1482–1625 cm $^{-1}$ ). The C=O stretching band showed a negligible redshift ( $<1$ cm $^{-1}$ ), suggesting that while positional order increased, orientational disorder (flip-flop motion) persisted.
- Crystallization: Marked by a significant redshift of the C=O band ( $\approx 8$ cm $^{-1}$ ) and a blueshift of the C-S/Ph-alkoxy band ( $\approx 4$ cm $^{-1}$ ).

C. Molecular Configuration in Crystal Phase

DFT calculations on dimers revealed that the head-to-tail configuration (where the C=O group interacts with the alkoxy chain of a neighbor) causes a redshift in the C=O stretching frequency.
The head-to-head configuration (C=O near S atom) causes a blueshift.
Conclusion: Since the experimental C=O band redshifts upon crystallization, the molecules in the crystal phase adopt a head-to-tail (or intercalated) configuration, not head-to-head.

5. Significance

Understanding Mesophase Dynamics: The study clarifies that the transition from SmC to SmX involves a massive increase in rotational energy barriers (activation energy) due to the locking of molecular positions, even though the tilt angle change is small.
Methodological Validation: It establishes that while k-means clustering is powerful for detecting major phase changes (Iso $\to$ N, SmX $\to$ Cr), it requires prior knowledge of phase sequences and careful selection of spectral windows to detect subtle transitions like SmA $\to$ SmC.
Computational Efficiency: The paper provides a practical guide for researchers, showing that the 631+Gd/B3LYP-D3(BJ) level is sufficient for accurate IR assignment in these systems, saving significant computational resources compared to triple-zeta basis sets.
Structural Determination: The ability to deduce the specific packing geometry (head-to-tail) of liquid crystals in the solid state purely from IR spectral shifts and DFT modeling offers a non-invasive alternative to X-ray diffraction for systems where crystallization is rapid or difficult to analyze.