Jun
Yoshida
*a,
Shuhei
Tamura
a,
Hidetaka
Yuge
a and
Go
Watanabe
*b
aKitasato University, Department of Chemistry, School of Science, Kitasato 1-15-1, Minami-ku, Sagamihara, Kanagawa 252-0373, Japan. E-mail: yoshidaj@kitasato-u.ac.jp
bKitasato University, Department of Physics, School of Science, Kitasato 1-15-1, Minami-ku, Sagamihara, Kanagawa 252-0373, Japan. E-mail: g00325@kitasato-u.ac.jp
First published on 29th November 2017
A liquid crystal host–guest system composed of achiral organic molecules (host) and colored chiral metal complexes (guest) was fabricated to sense both right- and left-handed circularly polarized light (r- and l-CPL), depending on the guest (dopant) concentration. The CPL-sensing can be reversibly turned on and off upon mechanical stress and heating.
A chiral nematic liquid crystal (N*-LC) with a helicoidal structure (Fig. 1) is another useful material for producing and sensing CPL. N*-LCs with a right-handed helical structure (P helix) selectively reflect r-CPL upon the incidence of linearly polarized light (PL, Fig. 1), and vice versa for left-handed N*-LCs (M helix). The easy fabrication of N*-LCs and their thin films is promising for application in optical devices.10 Additionally, the helical pitch of N*-LCs can be tuned by applying external stimuli such as photo-irradiation, heating, and voltage.11 Thus, the wavelength of selective reflection (SR), that is the wavelength of sensing CPL, can be finely tuned. However, the sensing of both r- and l-CPLs based on N*-LCs also requires the preparation of both P and M helical structures. Although temperature-induced and photo-induced helical switching of N*-LCs has been reported,12,13 the switching is limited to special cases.
Herein, we propose the sensing of both r- and l-CPL based on a colored liquid crystal host–guest system. While N*-LCs are usually prepared by doping colorless chiral dopants (guests) with a colorless nematic liquid crystal (host) based on the requirements for optical devices, we prepared colored N*-LCs by doping colored octahedral metal complex dopants with commercially available colorless nematics. The selective reflection (SR) of CPL in N*-LCs can be monitored as a square pattern by electronic circular dichroism (CD) spectroscopy that detects the different absorptions of l-CPL and r-CPL (Fig. 1). Here, we have also focused on the induced circular dichroism (ICD) phenomenon. ICD is a signal based on the extrinsic absorption of helically arranged molecules.14–20 In our case, ICD signals based on host nematics and guest dopants can be observed in the ultraviolet and visible light regions, respectively. ICD and SR yield CD signals with opposite signs (Fig. 1) when λ0 (the reflective wavelength of the helical pitch band) is longer than λab (the wavelength of the absorption band).16 Hence, CD patterns are potentially “programmable” from the combination of ICD and SR, as schematically shown in Fig. 1(b). Saeva and co-workers reported the co-existence of SR (UV region) and ICD (visible region) bands in their pioneering work on the CD spectrum of an N*-LC sample.16 However, SR and ICD phenomena have been treated independently. One reason is that the doping of N*-LCs with dyes (usually achiral) disturbs the formation of the helicoidal structure and leads to the extension of the helical pitches.21
A red-colored ruthenium complex with Δ,Λ chirality (Δ,Λ-Rutrop-1, Fig. 1) was used in this study as a chiral dopant in combination with two nematic (N) liquid crystals, 4-cyano-4′-pentylbiphenyl (5CB) and 4-pentylphenyl-4-methoxybenzoate (PPMB). Δ,Λ–Rutrop-1 has been recently developed by us as a chiral dopant with strong absorption in the visible region (ε = ∼20000 at 530 nm in chloroform), high helical twisting powers (HTPs), and acceptable miscibility with nematics, in order to investigate dopant–host interactions by Raman spectroscopy.22 Such properties in Δ,Λ-Rutrop-1 can also be used for the induction of ICD in the visible light region.
Fig. 2(a) shows the CD spectra of a binary mixture of 5CB and ∼0.04 mol% enantiomeric Rutrop-1, injected in a homogeneous cell with 25 μm gap (Grandjean state). The mixture exhibits a strong signal below 350 nm, attributed to ICD based on the helical arrangements of the 5CB host molecules, while it also exhibits broad positive signals ranging from 350 to 550 nm. Considering that Δ-Rutrop-1 in chloroform solution exhibits both positive and negative CD signals in the range of 350–550 nm (Fig. 2(b)), the only positive signal in the region for a Δ-Rutrop-1/5CB mixture is attributed to ICD based on the guest Δ-Rutrop-1. A Δ-Rutrop-1/PPMB sample (∼0.04 mol%) also shows positive ICD in the region of 350–550 nm, and vice versa, for a Λ-isomer sample (Fig. S1, ESI†). The effect of linear dichroism was found to be negligible in judging the sign of ICD from the measurements with the samples rotated 90 degrees (Fig. S1 and S2, ESI†). The pitch length of Δ-Rutrop-1/5CB (0.04 mol%) was determined to be ∼25 μm based on the Cano method.11,23 Hence, SR cannot be observed in the visible region at this dopant concentration. The wavelength of selective reflection (λsr) is represented by λsr = npcosα, where n, p, and α represent the average refractive index (∼1.5 for 5CB at RT), the helical pitch, and the angle at which the helicoidal structures are viewed, respectively.
Fig. 2 CD spectra of (a) a binary mixture of Rutrop-1 (∼0.04 mol%) and 5CB, and (b) Rutrop-1 in chloroform. In (a), the samples were injected into a homogeneous alignment cell of 25 μm thickness. |
To achieve the co-existence of SR and ICD in the visible light region, binary mixtures with higher dopant concentrations (1.0, 1.4, 2.4, and 3.0 mol% for Δ-Rutrop-1/5CB; and 0.6, 1.2. 1.8, and 2.4 mol% for Δ-Rutrop-1/PPMB) were prepared. The samples were sandwiched between a glass slide and a cover glass with no surface treatment and their CD spectra were measured (Fig. 3 and Fig. S3–S8, ESI†).‡ The samples were set to be in a Grandjean state by mechanical stress as described in the following section. At 1.0 mol%, SR is still not observed in the visible light wavelength, while the SR band is observed at ∼1200 nm by transmission spectroscopy (Fig. S9, ESI†). In contrast, a huge positive band is observed in the range of 400–550 nm (red line in Fig. 3), which is attributed to ICD signals based on ruthenium complex dopants. In the 1.4 mol% sample, negative saturated CD signals attributed to SR appear from 550 to 750 nm.§ The SR of the right-handed helix corresponds to pseudo-absorption of right-handed CPL, affording negative CD signals. The SR phenomenon was also confirmed by transmission and reflection spectroscopy (Fig. S9–S11, ESI†). While the CD signals saturated at 1.4 mol% with our equipment, they were found to be significant in the wavelength region longer than 340 nm by monitoring the voltage applied to the photomultiplier tube (Fig. S4, ESI†). When the dopant concentration increases to 2.4 mol%, strong negative and small positive CD signals are observed in the region of 400–550 nm and in the region below 400 nm, respectively. The SR band is proposed to shift to a shorter wavelength because of the shortened pitch length upon the increase of dopant concentration. The pseudo-absorption of r-CPL (selective reflection) is larger than the actual absorption of l-CPL, affording a negative CD signal. In the 3.0 mol% sample,¶ only the negative band is observed at a further blue-shifted region (Fig. 3), reflecting further shortening of the pitch length. Blue phases are observed at higher dopant concentrations (3.5 and 4.0 mol%).22 Comparison of the 1.0, 1.4, 2.4, and 3.0 mol% samples clearly indicates that the inversion of CD signals occurs just by changing the dopant concentration. Similar CD signal changes are also observed in the Δ-Rutrop-1/PPMB system, upon the increase of dopant concentration (Fig. S7, ESI†). In addition, the use of the Λ isomer of Rutrop-1 certainly affords inverted ICD and SR signals (Fig. S8, ESI†). Although concentration-dependent circularly polarized luminescence has been recently reported by Ito and co-workers,24 concentration-dependent sensing of opposite-handed CPL has not yet been established. Our host–guest methodology of combining ICD and SR can also avoid the time-consuming and laborious preparation of both enantiomers.
During the course of this study, we found that Δ-Rutrop-1/5CB mixtures are mechano-responsive at least for the samples containing more than 1.0 mol% dopants. The Δ-Rutrop-1/5CB (1.4 mol%) mixture sandwiched by a glass slide and a cover glass exhibited a focal conic texture with randomly distributed helicoidal domains (Fig. 3(b), 4(a) and Fig. S7, ESI†), when the sample was cooled from the isotropic phase. The focal-conic state was stable under ambient conditions, while it easily turned into a Grandjean texture with a planar alignment upon application of mechanical stress (just by pushing with a spatula, Fig. 3(b) and 4). The Grandjean texture changed back to the focal-conic texture again upon heating (Fig. S12 and a movie file in the ESI†). Responding to the reversible focal-conic/Grandjean transformation, the Rutrop-1/5CB system exhibits enhanced and diminished CD signals in the visible light region. In the focal conic state of Δ-Rutrop-1/5CB (1.4 mol%), the SR and ICD signals are not clearly observed, or are weak (red lines in Fig. 4(b)). In contrast, enhanced CD signals appear upon the application of mechanical stress (blue lines in Fig. 4(b)). The reversibility was confirmed at least five times (Fig. 4(c)). The appearance/disappearance of the ICD and SR signals could be explained by forward light scattering in the focal-conic state and Bragg reflection in the Grandjean state (Fig. 4(a)).
The response to mechanical and shear stresses is widely known in N*-LCs.25–31 However, the composition and dopant concentrations of N*-LCs that exhibit mechano-responses are known only empirically. Thus, mechano-response has not been applied to control and switching of CD signals in N*-LCs. The modulation of ICD signals by changing the sample states (focal conic and Grandjean states) is limited to the report by Saeva and co-workers,16 to the best of our knowledge. Additionally, the mechano-responsive properties were observed in the sample just sandwiched by two glasses. There is no necessity for surface alignment in the system and it is hence an accessible method for switching CD signals.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7sm01975k |
‡ The samples were also injected into a homogeneous 5 μm cell (Fig. S6, ESI†), while more saturated signals were obtained compared to the samples just sandwiched by glass slides. |
§ The maximum scale measurable in our CD equipment (JASCO J-720) is ±2000 mdeg. |
¶ The color of the 3 mol% cell looks purple due to the reflection of light around 400 nm and the absorption of red-colored ruthenium dopants. |
This journal is © The Royal Society of Chemistry 2018 |