An Enantioselective Potentiometric Sensor for 2-Amino-1-Butanol Based on Chiral Porous Organic Cage CC3-R

Porous organic cages (POCs) have attracted extensive attention due to their unique structures and tremendous application potential in numerous areas. In this study, an enantioselective potentiometric sensor composed of a polyvinyl chloride (PVC) membrane electrode modified with CC3-R POC material was used for the recognition of enantiomers of 2-amino-1-butanol. After optimisation, the developed sensor exhibited enantioselectivity toward S-2-amino-1-butanol (logKS,RPot = −0.98) with acceptable sensitivity, and a near-Nernstian response of 25.8 ± 0.3 mV/decade within a pH range of 6.0–9.0.


Introduction
Chirality is a general phenomenon and an important characteristic in naturally occurring molecules. For instance, most amino acids are levorotatory and sugars are dextrorotatory in biological systems. Consequently, chiral discrimination has attracted tremendous attention on account of its significance in pharmaceutical, biomedicine and chemical fields. Currently, chiral discrimination can be precisely achieved in many ways including gas chromatography (GC), high-performance liquid chromatography (HPLC) and high-performance capillary electrophoresis (HPCE). Although these methods have different advantages in terms of sensitivity or applicability, they suffer similar drawbacks including complicated operation and the need for expensive equipment. By contrast, ion-selective electrodes are simple, rapid and affordable, and have been widely applied to the enantioselective recognition and detection of chiral compounds in recent years [1][2][3][4][5][6][7][8][9].

Characterisation of the Synthesised CC3-R
The synthesised CC3-R crystals were characterised by Nuclear Magnetic Resonance (NMR), Powder X-ray diffraction (PXRD) and elemental analysis. As can be seen in Figure 3, the PXRD pattern of synthesised CC3-R crystals was consistent with the Singlecrystal simulation. Furthermore, CC3-R retained the same crystallinity and structure whether recrystallised from tetrahydrofuran or rinsed with water for 48 h, demonstrating excellent stability as chiral selector in the membrane electrode.
Elemental analysis was performed on CC3-R (C72H85N12); calculated = C 77. 31 [35][36][37][38]. In the present work, CC3-R was applied as a chiral selector in PVC membrane electrodes, resulting in impressive enantioselectivity for S-2-amino-1-butanol. Factors influencing the enantioselectivity of the CC3-R-based membrane electrode, such as the content of CC3-R, the category of plasticiser and the pH value of analyte solutions, were systematically investigated. 2-Amino-1-butanol ( Figure 2) is generally used as an intermediate in the synthesis of pharmaceuticals such as the bacteriostatic antituberculosis agent (S,S)-ethambutol [35][36][37][38]. In the present work, CC3-R was applied as a chiral selector in PVC membrane electrodes, resulting in impressive enantioselectivity for S-2-amino-1-butanol. Factors influencing the enantioselectivity of the CC3-R-based membrane electrode, such as the content of CC3-R, the category of plasticiser and the pH value of analyte solutions, were systematically investigated.

Characterisation of the Synthesised CC3-R
The synthesised CC3-R crystals were characterised by Nuclear Magnetic Resonance (NMR), Powder X-ray diffraction (PXRD) and elemental analysis. As can be seen in Figure 3, the PXRD pattern of synthesised CC3-R crystals was consistent with the Singlecrystal simulation. Furthermore, CC3-R retained the same crystallinity and structure whether recrystallised from tetrahydrofuran or rinsed with water for 48 h, demonstrating excellent stability as chiral selector in the membrane electrode.

Characterisation of the Synthesised CC3-R
The synthesised CC3-R crystals were characterised by Nuclear Magnetic Resonance (NMR), Powder X-ray diffraction (PXRD) and elemental analysis. As can be seen in Figure 3, the PXRD pattern of synthesised CC3-R crystals was consistent with the Singlecrystal simulation. Furthermore, CC3-R retained the same crystallinity and structure whether recrystallised from tetrahydrofuran or rinsed with water for 48 h, demonstrating excellent stability as chiral selector in the membrane electrode.
Elemental analysis was performed on CC3-R (C 72 H 85 N 12 ); calculated = C 77.  [35][36][37][38]. In the present work, CC3-R was applied as a chiral selector in PVC membrane electrodes, resulting in impressive enantioselectivity for S-2-amino-1-butanol. Factors influencing the enantioselectivity of the CC3-R-based membrane electrode, such as the content of CC3-R, the category of plasticiser and the pH value of analyte solutions, were systematically investigated.

Characterisation of the Synthesised CC3-R
The synthesised CC3-R crystals were characterised by Nuclear Magnetic Resonance (NMR), Powder X-ray diffraction (PXRD) and elemental analysis. As can be seen in Figure 3, the PXRD pattern of synthesised CC3-R crystals was consistent with the Singlecrystal simulation. Furthermore, CC3-R retained the same crystallinity and structure whether recrystallised from tetrahydrofuran or rinsed with water for 48 h, demonstrating excellent stability as chiral selector in the membrane electrode.

Optimisation of Membrane Components
The nature and amount of chiral selector and plasticiser contained in the membrane can strongly influence the selectivity and sensitivity of the membrane electrode. Consequently, the potential response characteristics of multiple electrodes with different quantities of CC3-R and three types of plasticiser (o-NPOE, DOS and DBP) were evaluated. Figure 4 shows the potential response characteristics of membrane electrodes with different CC3-R mass percentages. The performance of the membrane electrode improved with increasing CC3-R content, and the best enantioselectivity toward S-2-amino-1-butanol was achieved with 3% CC3-R (by weight). However, the enantioselectivity and sensitivity decreased slightly when the amount of CC3-R reached 4%. It is possible that the PVC membrane becomes saturated, hence the number of recognition sites does not increase proportionately with the chiral selector. Moreover, excess CC3-R could affect the ion-exchange capacity of the membrane electrode.

Optimisation of Membrane Components
The nature and amount of chiral selector and plasticiser contained in the membrane can strongly influence the selectivity and sensitivity of the membrane electrode. Consequently, the potential response characteristics of multiple electrodes with different quantities of CC3-R and three types of plasticiser (o-NPOE, DOS and DBP) were evaluated. Figure 4 shows the potential response characteristics of membrane electrodes with different CC3-R mass percentages. The performance of the membrane electrode improved with increasing CC3-R content, and the best enantioselectivity toward S-2-amino-1-butanol was achieved with 3% CC3-R (by weight). However, the enantioselectivity and sensitivity decreased slightly when the amount of CC3-R reached 4%. It is possible that the PVC membrane becomes saturated, hence the number of recognition sites does not increase proportionately with the chiral selector. Moreover, excess CC3-R could affect the ion-exchange capacity of the membrane electrode. The influence of the type of plasticiser is shown in Figure 5. DOS and DBP were clearly inferior to o-NPOE in terms of detection limit and enantioselectivity coefficient for S-2-amino-1-butanol.

Effect of pH on the Electrode
In order to investigate the effect of pH on the response performance of the optimised membrane electrode, the potential response value of the 2-amino-1-butanol solution (1.0 × 10 −3 mol·L −1 ) was measured at different pH values (pH 2.0-12.0). As shown in Figure 6, the potential response value was stable within a pH range of 5.0-9.0. Furthermore, a large difference between the two enantiomers The influence of the type of plasticiser is shown in Figure 5. DOS and DBP were clearly inferior to o-NPOE in terms of detection limit and enantioselectivity coefficient for S-2-amino-1-butanol.

Optimisation of Membrane Components
The nature and amount of chiral selector and plasticiser contained in the membrane can strongly influence the selectivity and sensitivity of the membrane electrode. Consequently, the potential response characteristics of multiple electrodes with different quantities of CC3-R and three types of plasticiser (o-NPOE, DOS and DBP) were evaluated. Figure 4 shows the potential response characteristics of membrane electrodes with different CC3-R mass percentages. The performance of the membrane electrode improved with increasing CC3-R content, and the best enantioselectivity toward S-2-amino-1-butanol was achieved with 3% CC3-R (by weight). However, the enantioselectivity and sensitivity decreased slightly when the amount of CC3-R reached 4%. It is possible that the PVC membrane becomes saturated, hence the number of recognition sites does not increase proportionately with the chiral selector. Moreover, excess CC3-R could affect the ion-exchange capacity of the membrane electrode. The influence of the type of plasticiser is shown in Figure 5. DOS and DBP were clearly inferior to o-NPOE in terms of detection limit and enantioselectivity coefficient for S-2-amino-1-butanol.

Effect of pH on the Electrode
In order to investigate the effect of pH on the response performance of the optimised membrane electrode, the potential response value of the 2-amino-1-butanol solution (1.0 × 10 −3 mol·L −1 ) was measured at different pH values (pH 2.0-12.0). As shown in Figure 6, the potential response value was stable within a pH range of 5.0-9.0. Furthermore, a large difference between the two enantiomers

Effect of pH on the Electrode
In order to investigate the effect of pH on the response performance of the optimised membrane electrode, the potential response value of the 2-amino-1-butanol solution (1.0 × 10 −3 mol·L −1 ) was measured at different pH values (pH 2.0-12.0). As shown in Figure 6, the potential response value was stable within a pH range of 5.0-9.0. Furthermore, a large difference between the two enantiomers was observed at pH 9.0. Therefore, pH 9.0 was adopted for measurements using the optimised membrane electrode. was observed at pH 9.0. Therefore, pH 9.0 was adopted for measurements using the optimised membrane electrode.   Table 1.

Interference Ion
Log ,int As shown in Table 1, the membrane electrode exhibited comparable responses to other alkamines with similar configurations to 2-amino-1-butanol. Steric hindrance caused by additional organic groups of other alkamines could impair the recognition performance during ion exchange. The electrode displayed slight enantioselective recognition of enantiomers of 2-amino-3-methyl-1butanol, which have the most similar configuration. However, 2-amino-3-methyl-1-butanol yielded similar potential response values, and caused significant interference.

Recognition of Mixing Samples
To further explore the enantioselectivity of the developed membrane electrode, a mixing sample test was conducted using different molar ratios of S-and R-enantiomers of 2-amino-1-butanol ( Figure   Figure 6. The influence of pH on the response performance of the membrane electrode to S/R-2-amino-1-butanol.  Table 1.

Enantioselectivity Coefficient of the Electrode
As shown in Table 1, the membrane electrode exhibited comparable responses to other alkamines with similar configurations to 2-amino-1-butanol. Steric hindrance caused by additional organic groups of other alkamines could impair the recognition performance during ion exchange. The electrode displayed slight enantioselective recognition of enantiomers of 2-amino-3-methyl-1-butanol, which have the most similar configuration. However, 2-amino-3-methyl-1-butanol yielded similar potential response values, and caused significant interference.

Recognition of Mixing Samples
To further explore the enantioselectivity of the developed membrane electrode, a mixing sample test was conducted using different molar ratios of Sand R-enantiomers of 2-amino-1-butanol (Figure 7). The results showed that the potential response values of the mixing solution increased with increasing proportion of S-2-amino-1-butanol, revealing a clear positive linear correlation between the proportion

Synthesis of CC3-R
CC3-R was synthesised using a previously reported method [29]. Briefly, 20 mL dichloromethane was added dropwise onto 1.0 g 1,3,5-triformylbenzene in a two-necked flask without stirring at room temperature, and 20 μL trifluoroacetic acid was added as a catalyst. Within minutes, 20 mL dichloromethane containing 1.0 g (R,R)-1,2-diaminocyclohexane was dripped slowly into the mixture. After reaction for 72 h at room temperature, white crystals were present on the wall of the flask, which were filtered and rinsed with ethanol/dichloromethane (95:5 v/v).

Preparation of Enantioselective Membrane Electrodes
To prepare the PVC membranes, PVC powder, plasticiser (o-NPOE), and CC3-R were added to 3 mL tetrahydrofuran and stirred to form a transparent solution [40]. This was poured onto a glass sheet and volatilised for 24 h to form a semitransparent film ~0.5 mm thick. The obtained film was incised into an appropriately sized disc and assembled using a PVC tube, which was subsequently filled with 0.1 mol·L −1 KCl as an internal reference solution. A silver chloride electrode was applied as an internal reference electrode, and a saturated calomel electrode was utilised as a reference electrode. For comparison, a CC3-S-modified membrane electrode was prepared in the same way. The overall strategy for enantioselective potentiometric sensor fabrication is depicted in Scheme 1.

Synthesis of CC3-R
CC3-R was synthesised using a previously reported method [29]. Briefly, 20 mL dichloromethane was added dropwise onto 1.0 g 1,3,5-triformylbenzene in a two-necked flask without stirring at room temperature, and 20 µL trifluoroacetic acid was added as a catalyst. Within minutes, 20 mL dichloromethane containing 1.0 g (R,R)-1,2-diaminocyclohexane was dripped slowly into the mixture. After reaction for 72 h at room temperature, white crystals were present on the wall of the flask, which were filtered and rinsed with ethanol/dichloromethane (95:5 v/v).

Preparation of Enantioselective Membrane Electrodes
To prepare the PVC membranes, PVC powder, plasticiser (o-NPOE), and CC3-R were added to Molecules 2018, 23, x FOR PEER REVIEW 6 of 8 Scheme 1. Schematic illustration of enantioselective potentiometric sensor fabrication.
A revised separate solution method was used to calculate the enantioselectivity coefficient (log , ) with the following formula: where ER and ES are the potentials of 0.1 mol·L −1 R-and S-2-amino-1-butanol solutions, respectively, and D is the slope of the response curve of S-2-amino-1-butanol.

Conclusions
The chiral porous organic cage CC3-R proved to be a useful chiral selector for the modification of PVC membrane electrodes to generate enantioselective potentiometric sensors. The optimised membrane electrode containing 3 wt% CC3-R exhibited enantiomeric recognition toward S-2-amino-1-butanol over R-2-amino-1-butanol (log , = −0.98) with acceptable sensitivity, and a near-Nernst response of 25.

Conflicts of Interest:
The authors declare no conflict of interest.
A revised separate solution method was used to calculate the enantioselectivity coefficient (log K Pot S,R ) with the following formula: where E R and E S are the potentials of 0.1 mol·L −1 Rand S-2-amino-1-butanol solutions, respectively, and D is the slope of the response curve of S-2-amino-1-butanol.

Conclusions
The chiral porous organic cage CC3-R proved to be a useful chiral selector for the modification of PVC membrane electrodes to generate enantioselective potentiometric sensors. The optimised membrane electrode containing 3 wt% CC3-R exhibited enantiomeric recognition toward S-2-amino-1-butanol over R-2-amino-1-butanol (log K Pot S,R = −0.98) with acceptable sensitivity, and a near-Nernst response of 25.8 ± 0.3 mV/decade toward S-2-amino-1-butanol at pH 9.0.