An ionic liquid composed of purely functional sensing molecules: a colorimetrically calcium responsive ionic liquid

Yusuke Niwa , Tatsumi Mizuta , Kenji Sueyoshi , Tatsuro Endo and Hideaki Hisamoto *
Department of Applied Chemistry, Graduate School of Engineering, Osaka Prefecture University, 1-1 Gakuencho, Nakaku, Sakai, Osaka 599-8531, Japan. E-mail: hisamoto@chem.osakafu-u.ac.jp

Received 9th September 2019 , Accepted 17th October 2019

First published on 21st October 2019


A highly lipophilic ionic liquid (IL) consisting of stoichiometrically equal amounts of two purely functional chemical sensing molecules, an anionic calcium ionophore and a cationic dye, was synthesized for the first time, and used as a component for a poly(vinyl chloride)-based or neat liquid membrane optical sensor. This is the first report of an IL consisting of purely functional chemical sensing molecules, which is completely different from the previously reported ionic liquids typically consisting of imidazolium or other lipophilic cations. Since the present IL contained an extremely high concentration of dyes, the IL-based sensor showed dramatically enhanced sensitivity (13 times higher compared to that of a conventional sensor), and fully reversible and selective response to Ca2+. Preliminary investigation on the unique response characteristics of the present liquid membrane IL-based sensor was performed with the responses to Ca2+ and Na+.


Ionic liquids (ILs) including dye molecules have received much attention in physical chemistry fields for their unique optical characteristics such as photochromism,1 characteristic fluorescence properties,2 highly luminescence properties,3–6 thermochromism,7 and vapochromism.8,9 Based on these optical characteristics, ILs including dye molecules can be considered as an interesting material for use in optical chemical sensors. The application of various ILs in optical sensors has been recently reviewed by Muginova et al.10

On the other hand, a plasticized poly(vinyl chloride) (PVC) membrane is a versatile and practical sensing material used for ion-selective electrodes and optodes.11,12 Optodes are generally composed of PVC (30–33 wt%), a plasticizer (60–66 wt%), a lipophilic dye (∼5 wt%), and other components such as an ionophore (∼5 wt%). In this case, the sensitivity of an optode based on a plasticized PVC membrane is determined by the concentration of the dye and the membrane thickness. In order to obtain higher sensitivity for reliable naked eye-based detection, increasing the dye concentration or membrane thickness is necessary. However, the solubility of the dye in the plasticizer is usually limited to 5–50 mmol kg−1, and increasing the membrane thickness drastically increases the response time due to the slow diffusion of membrane components. Therefore, an alternative approach is needed to develop a highly sensitive optical sensor based on a plasticized PVC membrane.

Very recently, we applied an IL-based dye containing an unusually high concentration of a dye (296 mmol kg−1) to a rapid and highly sensitive anion optode (Fig. 1).13,14 However, ion-selective analysis was not achieved due to the lack of ion-recognition components.


image file: c9an01769k-f1.tif
Fig. 1 Concept of a colorimetrically Ca2+-responsive ionic liquid membrane for highly sensitive detection of Ca2+.

Here, we report a colorimetrically Ca2+-responsive IL as the first example of an IL composed of purely functional chemical sensing molecules for sensitive and selective analysis. The new IL comprises bis(4-n-octylphenyl) phosphate (OP2P), an anionic Ca2+ ionophore known as HDOPP,15,16 and a merocyanine dye (KD-M13), which forms a mono-cation by protonation (Fig. 2(a)).17 These original compounds are typically solids in the isolated form. However, surprisingly, we found that the ion pair formed by these compounds appeared as a liquid at room temperature (Fig. 2(b)), although the ion pair did not contain typical IL cations such as imidazolium or phosphonium salts. On the other hand, when a dye which has a bromo group instead of a tert-butyl group of KD-M13, KD-M11,17 was used for ion-pair formation with OP2P, the product appeared as a complete solid (Fig. S1). This interesting fact prompted us to further investigate the properties and functions of the [KD-M13][OP2P] liquid.


image file: c9an01769k-f2.tif
Fig. 2 (a) Structure of [KD-M13][OP2P]. (b) [KD-M13][OP2P] at 25 °C (scale bar represents 1 cm).

The detailed synthetic procedure is described in the ESI. In brief, the ionophore and dye were dissolved in CH2Cl2, and ion-exchange experiments were performed to remove inorganic ions. After evaporation of the solvent and complete drying under vacuum, the ionic liquid purely composed of the ionophore and dye was obtained. The 1H NMR spectrum presented in Fig. S2 clearly shows the stoichiometric formation of OP2P and KD-M13 in a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 ratio. Fig. S3 shows the results of the differential scanning calorimetry (DSC) measurements. Since a weak, but reproducible endothermic peak that is supposed to indicate the melting temperature was found at 22 °C, we could adjudge this new material to be an IL. Firstly, we used this material as a plasticized PVC membrane to investigate its ion-sensing function.

The plasticized PVC membrane consisting of PVC (10 wt%) and [KD-M13][OP2P] (90 wt%) was prepared by the spin coating method. This membrane contained very high concentrations of the dye and ionophore (872 mmol kg−1), 17–170 times higher than that in a conventional PVC membrane (5–50 mmol kg−1 of dye and ionophore). In the absorption spectra, the absorbance of the maximum absorption wavelength (630 nm) increased with an increase in the Ca2+ concentration (pH 7.0, [Ca2+]: 10−6 to 10−1 M, see Fig. 3), and these changes were fully reversible (Fig. S4). The response mechanism can be considered as the ion-exchange equilibrium of protons and cations at the interface of the PVC membrane and an aqueous solution. This mechanism is similar to that in a conventional ion-selective optode membrane since the response curve shift characteristics upon changing the pH of the solution were almost the same as those of the conventional one (see Fig. S5). Thus we applied the conventional optode response theory for evaluating the results. However, since the membrane components are fully functional molecules except PVC, the molecular crowding conditions might affect various sensing characteristics. Here, ion selectivity, response time, and sensitivity were quantitatively evaluated.


image file: c9an01769k-f3.tif
Fig. 3 Absorption spectra of the [KD-M13][OP2P] PVC membrane (thickness: 240 nm, sample solutions: 50 mM Tris-HCl buffer (pH 7.0) containing 10−6–10−1 M CaCl2).

Ion selectivity was investigated by using four typical cations (Na+, K+, Mg2+, and Ca2+). This membrane exhibited selectivity in the order Ca2+ > Mg2+ ≫ K+ ≈ Na+, indicating that the Ca2+-selective response was successfully achieved (Fig. 4). For control experiments, the control compound where the OP2P anion of [KD-M13][OP2P] was replaced with a common ionic liquid anion, bis(1,1,2,2,3,3,4,4,4-nonafluoro-1-butanesulfonyl)imide, was independently prepared in the same manner and used for the same experiments, and it was found that no ion selectivity was observed under the same experimental conditions (Fig. S6 and S7). This result suggested the importance of the ionophore anion in [KD-M13][OP2P] to obtain calcium ion selectivity. Even though 1 mM Mg2+ coexisted in the sample solution, more than millimolar levels of Ca2+ can be quantified (Fig. S8(iii)). Ion-exchange constants were obtained by fitting the theoretical response curves calculated using the theoretical equations for ion-exchange equilibrium, according to a previous report,14 and selectivity constants, KoptCa, interfering ion, were obtained as KoptCa,Mg = 1.0 × 10−1 and KoptCa,K and KoptCa,Na ≦ 1.0 × 10−5.


image file: c9an01769k-f4.tif
Fig. 4 Response curves for different cations. (The vertical axis, α, represents the deprotonation ratio of the dye defined by the minimum and maximum absorbances obtained by 1 M H3PO4 and 1 M NaOH solutions. Sample solutions: 50 mM Tris-HCl buffer (pH 7.0) containing 10−6–10−1 M chloride salts of each cation.)

The conventional PVC membrane with the same dye and ionophore (PVC: 32 wt%, DOPP: 63 wt%, KD-M13 and HDOPP: 2.5 wt% (51 mmol kg−1), thickness: 2.0 μm; for more details, see the ESI) gave better selectivity constants (KoptCa,Mg = 4.0 × 10−3, KoptCa,K and KoptCa,Na ≤ 1.0 × 10−6). This might be attributed to the use of DOPP, which works as a carrier solvent for calcium chelates of alkyl phosphate.18 The response time of the [KD-M13][OP2P]-based PVC membrane was longer than that of the conventional PVC membrane (Table S1). This was caused by the increase in the viscosity of [KD-M13][OP2P] upon extraction of the calcium ions, since the response time gradually increased with the increase in Ca2+ concentration (see Table S1). An increase in Ca2+ concentration causes the membrane components to revert to the simple mixture of the original solid materials, namely, [OP2P]2Ca and deprotonated KD-M13. It should be noted that visible precipitation was not observed in this process, and background absorbance was also not increased. Thus, we speculate that the present IL exists as a highly viscous liquid at a high Ca2+ concentration in the aqueous phase. Deviation of the highest concentration plots in Fig. 4 (10−1 M Ca2+ and Mg2+) from the theoretical response curves might support this hypothesis. Interestingly, treatment of the membrane with 1 M NaOH solution resulted in complete deprotonation without increasing the response time (see Fig. 3 and Fig. S4), but for a very high concentration of alkaline Ca2+ solution (pH: 10, [Ca2+]: 1 M), the deprotonation rate (α) was almost saturated at 0.8 and the response was much slower compared to that observed with NaOH. Hence, the viscosity of the present PVC membrane might be affected by the interactions between the extracted ion and the ionophore. This fact suggested that the membrane viscosities both ‘before’ and ‘after’ the response to ions should be taken into account for designing ion sensors under molecular crowding conditions.

Sensitivity, evaluated as the absorbance for a 100 nm thick membrane, was significantly enhanced 13 times (ΔA = 0.26) as compared to that for a conventional PVC membrane (ΔA = 0.020). This high sensitivity is promising for applications in naked eye-based detection. Recently, it was reported that physicochemical properties such as viscosity can be controlled by the alkyl chain position and molecular design motif,19 implying that further improvement of the response time can be achieved by molecular design of the ion pair.

[KD-M13][OP2P] could also be used as a neat membrane due to its high viscosity, and in this case, the concentrations of the dyes and ionophores reached 990 mmol kg−1 each. This neat membrane reversibly responded to 1 M H3PO4 and 1 M NaOH solutions similar to the case of the [KD-M13][OP2P]-PVC membrane (Fig. S9), with slightly enhanced sensitivity (ΔA = 0.27). After soaking the neat membrane in calcium sample solutions (pH 7.0, [Ca2+]: 10−6 to 10−1 M), a systematic response was observed. Image analysis was conducted using a flat-bed scanner. Plotting the colour difference (ΔE) of CIE1976 based on the scanned image (1264 pixels) allowed us to obtain a good calibration line (R2 = 0.989) in the log [Ca2+] range of −5 to −1 (Fig. 5).


image file: c9an01769k-f5.tif
Fig. 5 Response plots vs. Ca2+ concentration (in the log scale) and images of the neat membrane after soaking in Ca2+ solutions (thickness: 200 nm, sample solutions: 50 mM Tris-HCl buffer (pH 7.0) containing 10−6–10−1 M CaCl2).

In this work, we successfully synthesized an IL composed of purely functional chemical sensing molecules, a lipophilic ionophore and a lipophilic pH indicator dye, for the first time. This material has sufficient fluidity for the reversible diffusion of ions, leading to fully reversible and selective colour change in response to Ca2+. Although some improvements in the response time are still needed, the preliminary results presented in this paper are promising to open a new direction in the development of highly sensitive optical sensors using functional dyes.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors thank Professor Akikazu Matsumoto and Assistant Professor Yoshihito Suzuki of Osaka Prefecture University for the DSC measurements and fruitful discussions. This work was partially supported by the Takahashi Industrial and Economic Research Foundation, the Nakatani Foundation, the Izumi Science and Technology Foundation, and the Masuya Memorial Basic Research Foundation.

Notes and references

  1. L. C. Branco and F. Pina, Chem. Commun., 2009, 6204–6206 RSC.
  2. C. A. B. Rodrigues, C. Graça, E. Maçôas, A. Fedorov, C. A. M. Afonso and J. M. G. Martinho, J. Phys. Chem. B, 2013, 117, 14108–14114 CrossRef CAS PubMed.
  3. Y. Yoshida and G. Saito, Phys. Chem. Chem. Phys., 2010, 12, 1675–1684 RSC.
  4. P. S. Campbell, M. Yang, D. Pitz, J. Cybinska and A.-V. Mudring, Chem. – Eur. J., 2014, 20, 4704–4712 CrossRef CAS PubMed.
  5. J. M. Delgado, A. Raymundo, M. Vilarigues, L. C. Branco and C. A. T. Laia, Chem. – Eur. J., 2015, 21, 726–732 CrossRef CAS.
  6. H. Zhu, G. Zhang, M. Chen, S. Zhou, G. Li, X. Wang, Q. Zhu, H. Li and J. Hao, Chem. – Eur. J., 2016, 22, 6286–6293 CrossRef CAS.
  7. B. Monteiro, M. Outis, H. Cruz, J. P. Leal, C. A. T. Laia and C. C. L. Pereira, Chem. Commun., 2017, 53, 850–853 RSC.
  8. K. Y. Yung, A. J. Schadock-Hewitt, N. P. Hunter, F. V. Bright and G. A. Baker, Chem. Commun., 2011, 47, 4775–4777 RSC.
  9. Y. Funasako, T. Mochida, K. Takahashi, T. Sakurai and H. Ohta, Chem. – Eur. J., 2012, 18, 11929–11936 CrossRef CAS PubMed.
  10. S. V. Muginova, D. A. Myasnikova, S. G. Kazarian and T. N. Shekhovtsova, Anal. Sci., 2017, 33, 261–274 CrossRef CAS PubMed.
  11. E. Bakker, P. Bühlmann and E. Pretsch, Chem. Rev., 1997, 97, 3083–3132 CrossRef CAS.
  12. P. Bühlmann, E. Pretsch and E. Bakker, Chem. Rev., 1998, 98, 1593–1688 CrossRef PubMed.
  13. T. Mizuta, K. Sueyoshi, T. Endo and H. Hisamoto, Sens. Actuators, B, 2018, 258, 1125–1130 CrossRef CAS.
  14. T. Mizuta, K. Maeno, K. Sueyoshi, T. Endo and H. Hisamoto, Anal. Sci., 2018, 34, 517–519 CrossRef CAS PubMed.
  15. H. M. Brown, J. P. Pemberton and J. D. Owen, Anal. Chim. Acta, 1976, 85, 261–276 CrossRef CAS.
  16. Y. Ermolenko, T. Yoshinobu, Y. Mourzina, K. Furuichi, S. Levichev, M. J. Schöning, Y. Vlasov and H. Iwasaki, Talanta, 2003, 59, 785–795 CrossRef CAS.
  17. H. Hisamoto, H. Tohma, T. Yamada, K. Yamauchi, D. Siswanta, N. Yoshioka and K. Suzuki, Anal. Chim. Acta, 1998, 373, 271–289 CrossRef CAS.
  18. J. Ruzicka, E. H. Hansen and J. C. Tjell, Anal. Chim. Acta, 1973, 67, 155–178 CrossRef CAS.
  19. F. Lu, T. Takaya, K. Iwata, I. Kawamura, A. Saeki, M. Ishii, K. Nagura and T. Nakanishi, Sci. Rep., 2017, 7, 1–12 CrossRef.

Footnote

Electronic supplementary information (ESI) available: Experimental section, synthesis of the present material, and results of DSC, response time, and reversibility. See DOI: 10.1039/c9an01769k

This journal is © The Royal Society of Chemistry 2019