Chemically-driven “molecular logic circuit” based on osmium chromophore with a resettable multiple readout

Abstract

The resettable electro-optical identity of an osmium(II) chromophore has been exploited for integrating miniaturised molecular logic circuits under chemical stimulation. The versatile ‘molecular-probe’ yields multiple outputs using selective stimuli and thus allows precise analysis.

---

Supramolecular receptors,¹ capable of chemical recognition,² have been realised at nano-scale for the meaningful development of molecular gates,³ memory⁴ and devices.⁵ However, a commercial molecular device seems to be an Achilles’ heel without integrated logic circuits.⁶ In this context, the innovative approach of logical transformation of discriminating outputs⁷ has initiated a leeway for molecular circuit engineering. Importantly, coordination based redox active receptors with multiple outputs are of potential interest for the replication of silicon-transistors at molecular level.

Copper, a micronutrient, can cause severe damages, such as Wilson’s disease, anaemia symptoms, neutropenia, impaired growth, overdose imbalance, etc.⁸ Fluoride prevents dental cavities and osteoporosis but at toxic levels could cause hypocalcaemia and dental or skeletal fluorosis.⁹ Importantly, metallic (Cu/Zn) corrosion from fluoridated water during water supply causes the deadly Parkinson and Alzheimer’s diseases.¹⁰ In this viewpoint, a chromophore, 1, (Scheme 1) was designed with redox Os(bpy)₂²⁺ and a conduit imidazole entity for selective multi-dimensional reversible detection of Cu²⁺ and F⁻. Interestingly, discrete chemical information was parallely transduced to multiple spectroscopic and visual outputs to integrate logic circuitry.

Scheme 1 Chemical structure and ORTEP representation of 1 (thermal ellipsoids are drawn at 30% probability level). Hydrogen and water molecules are omitted for clarity. Crystal data for 1: C₃₁H₂₂N₈O₇Os; M = 808.77; monoclinic; space group P21/c; a = 9.5534(2); b = 25.7853(5); c = 12.6272(3) Å; α = 90; β = 93.47(2); γ = 90°; V = 3104.85(12) ų; Z = 4; Dc = 1.730 mg m⁻³; μ = 4.168 mm⁻¹; R₁ = 0.0306; wR₂ = 0.0697.

The chromophore 1 was synthesized in reasonable yield (46%), characterized by a full battery of physico-chemical techniques (Fig. S1–S5‡) and crystallized with a P21/c space group and monoclinic point group. ¹H-NMR of 1 in a DMSO-d₆ solution exhibits a peak at δ = 14.47 ppm due to the single imidazole proton, indicating an imine group binding motif.

The absorption identity of 1 in acetonitrile exhibits an intense peak at λ = 293 nm (ε = 8.8 × 10⁴ M⁻¹ cm⁻¹) due to a π–π* transition of the ligand-centred (LC) band associated with a couple of broad bands at λ = 509 (ε = 1.4 × 10⁴ M⁻¹ cm⁻¹) and 687 nm (ε = 4.1 × 10² M⁻¹ cm⁻¹), which are assigned to singlet and triplet metal-to-ligand charge transfer (¹MLCT and ³MLCT) transitions, respectively.¹¹ Interestingly, 1 displays significant modulation via UV-vis mode on chemical stimulation with Cu²⁺ and F⁻ in acetonitrile at ambient temperature. For instance, the processing via a 50 ppb concentration of Cu²⁺, as the input data, produces a bathochromic shift at λ = 293 nm (Δλ = 18 nm) with emergence of a new shoulder at λ = 337 nm, which is associated with a remarkable “turn-off” modulation at λ = 509 and 687 nm (5.0 and 4.5-fold) and a colourimetric change from light brown to colourless in real time (∼15 s). This logical output could be attributed to the generation of oxidative osmium (Os³⁺) species, as shown from UV-vis and cyclic voltammetry measurements (Fig. 1).¹² The output was input selective, as similar inputs viz. Zn²⁺, Pb²⁺, Ni²⁺, Na⁺, Mn²⁺, Mg²⁺, K⁺, Fe³⁺, Co²⁺, Cd²⁺, Li⁺, Ca²⁺, Hg²⁺ and Fe²⁺ were inactive (5–10% error, Fig. 2a). The simultaneous processing at multiple wavelengths (λ = 293, 509 and 687 nm) allows the potential option for label-free detection. Importantly, the experiment could be replicated in an aqueous medium with similar efficiency (5–8% error). However, applying other input data, comprising of a 30 ppm concentration of F⁻, yields entirely different output signals with a MLCT band at λ = 509 nm as well as a shoulder at λ = 325 nm, undergoing a bathochromic shift (Δλ = 45 and 25 nm) along with a colourimetric change from light brown to red in real time (∼15 s) (Fig. 1).

Fig. 1 Representative plot of the processing of two different inputs: Cu²⁺ (50 ppb, CH₃CN, 1 + Cu²⁺) and F⁻ (30 ppm, CH₃CN, 1 + F⁻) via 1 (0.98 × 10⁻⁵ M, CH₃CN, 1) in UV-vis mode. The insets show Benesi–Hildebrand plots at λ = 509 and 554 nm with the addition of Cu²⁺ (a) and F⁻ (b), respectively.

Fig. 2 (a) The UV-vis spectra of 1 (0.98 × 10⁻⁵ M, CH₃CN) with Cu²⁺ and other tested metal ions (50 ppb in acetonitrile); (b) the UV-vis spectra of 1 (0.98 × 10⁻⁵ M, CH₃CN) with F⁻ and other tested anions (30 ppm in acetonitrile).

This could be possibly due to the interaction of F⁻ with the acidic protons of the imidazole unit.¹³ No other tested inputs, viz. I⁻, Br⁻, Cl⁻, NO₂⁻, ClO₄⁻, H₃PO₄⁻, AcO⁻, and CN⁻, could produce a similar output (Fig. 2b). Note that, probe 1 is only suitable for the sensory action under physiological conditions (in a neutral to slightly acidic medium), as 1 loses selectivity with the OH⁻ ion, and hence can be effectively deployed for biological specimens. Thus, both inputs (Cu²⁺ and F⁻) were exclusive and reveal different outputs. Moreover, the parallel addition of a variety of chemical inputs (in the matrix with Cu²⁺ and/or F⁻) could not produce any significant change on the obtained output (Fig. S6‡). Importantly, Cu²⁺ was the predominant input even in case of parallel processing of both Cu²⁺ and F⁻ via 1. Moreover, the output signals were stationary (Benesi–Hildebrand constant: K_1+Cu²⁺ = 2.74 × 10⁵ M⁻¹ and K_1+F⁻ = 4.07 × 10⁴ M⁻¹) and showed no deviation on increasing the concentration/reaction time by up to four times.¹⁴ Notably, the evaluated detection limits (DL_Cu²⁺ = 1.2 × 10⁻⁹ M and DL_F⁻ = 2.8 × 10⁻⁶ M) are superior so far.¹⁵

The repetitive action of a receptor is highly sought after for molecular logic functions. The chemical information written on probe 1 by Cu²⁺ and F⁻ could be erased (>95% reversibility) by H₂O (5 μl in 3 ml) or H⁺ (5 μl, 10⁻⁴ M, 3 ml), respectively (Fig. 3). The reversible tuning through 1 was carried out for 3 cycles and showed significant overall regeneration of up to ∼87%. However, even this minor loss of reversibility could be improved by immobilizing this multitasking receptor on the solid support. Interestingly, the processing of chemical inputs can integrate various complex logic-circuits operating with different outputs.

Fig. 3 (a) Pictorial representation of repetitive processing of Cu²⁺/H₂O and F⁻/H⁺ inputs via 1. (b) Graphical representation of monitoring ΔA (%) at λ = 509 and 554 nm as a function of no. of switching turns with F⁻ (pink/light brown pillar) and Cu²⁺ (white/light brown pillar), respectively.

As shown in Fig. 4, information processing with Cu²⁺ (input 1) and H₂O (input 2) yields an INH (inhibit) logic gate at λ = 509 nm, using a threshold value of A = 0.1 (an absorbance higher/lower than 0.1 will be considered as 0/1, respectively), and an IMP (implication) gate at λ = 293 nm (Fig. S7a and b‡), using a threshold value of A = 0.5 (an absorbance higher/lower than 0.5 will be considered as 0/1, respectively). However, monitoring the information at λ = 554 and 325 nm on processing of F⁻ (input 1) and H⁺ (input 2) yields coupled INH gates, using A = 0.1 (lower “0” and higher “1”) and 0.4 (lower “1” and higher “0”) as threshold values (Fig. S8a and b‡).

Fig. 4 Logic circuit and truth table for the Boolean inputs ((a) Cu²⁺ and H₂O in acetonitrile and (b) F⁻ and H⁺ in acetonitrile) for 1.

The multi-output generation by a single input could be potentially used for precise and defect-free detection. The recognition propensity of receptor 1 can be monitored via optical as well as electrochemical modes. The receptor 1 exhibits dual fluorescence at λ = 371 and 463 nm on excitation at λ = 293 nm.¹¹ Therefore, the addition of chemical stimuli to 1 triggers dual modulation in fluorescence identity. The processing of 50 ppb of Cu²⁺ via a fluorogenic mode demonstrates complete quenching of the signal in real time (109-fold at λ = 463 nm and 8-fold at λ = 371 nm). The differential quenching extent with processing of 30 ppm of F⁻ in quick time (6-fold at λ = 463 nm and 371 nm) potentially leads to the selective estimation of each stimulus (Fig. 5). The processing of Cu²⁺ was predominant in the mixture of analytes with an equal concentration and the written information could be erased (>92% reversibility) by H₂O (Cu²⁺) or H⁺ (F⁻) as discussed in the case of the UV-vis mode. The binding constant (Benesi–Hildebrand constant: K_1+Cu²⁺ = 2.32 × 10⁵ M⁻¹ and K_1+F⁻ = 3.56 × 10⁴ M⁻¹) and detection limit (DL_Cu²⁺ = 1.2 × 10⁻⁹ M and DL_F⁻ = 2.8 × 10⁻⁶ M) evaluated were in agreement with the values obtained by the UV-vis data.

Fig. 5 Fluorogenic processing via 1 (0.98 × 10⁻⁵ M, CH₃CN) of the chemical inputs of 50 ppb Cu²⁺ and 30 ppm of F⁻ in acetonitrile in real time. The insets show the monitoring of the emission intensity at λ = 463 nm with the addition of Cu²⁺ (a) and F⁻ (b).

Assuming Cu²⁺ (In 1), F⁻ (In 2), H₂O (In 3), and H⁺ (In 4) as the four inputs and monitoring the output at λ = 463 nm (threshold value = 50 (a.u), intensity higher/lower than 50 (a.u) will be considered as 1/0, respectively) provide a competent molecular system for the exclusive processing of four distinct chemical informations (Fig. S9‡). Importantly, chromophore 1 yields a distinct response for the exclusive and combinatorial information input following the Boolean logic circuit¹⁶ (Fig. 6). The 4-input example, based on fluorescence modulation, is likely to be the first example of this type of integrated logic circuit mimic.

Fig. 6 Molecular logic circuit constructed using four (Cu²⁺, F⁻, H₂O and H⁺) inputs and an exclusive output, and the corresponding truth table is shown alongside.

Similar to the optical mode, the sensing mechanism of 1 could be monitored in electrochemical mode also with significant reversibility. The receptor 1 shows one electron transfer reversible redox wave at half wave potential (E_1/2) of 0.35 V with peak to peak separation (ΔE) of 75 mV at 300 mV s⁻¹ vs. Ag/AgCl.¹¹ Importantly, the addition of chemical inputs exhibits two-way tuning i.e., anodic shift via Cu²⁺ (ΔE_1/2 = + 0.14 V) and cathodic shift via F⁻ (ΔE_1/2 = −0.15 V), supporting the proposed sensory action of 1 (Fig. 7). This discriminating behaviour is exclusive and finds eminent applications in low voltage electrochromic devices.

Fig. 7 Cyclic voltammograms of 1 (0.97 × 10⁻³ M, 0.1 mM TBAP, acetonitrile) on processing 50 ppb of Cu²⁺ (a) and 30 ppb of F⁻ (b) in real time.

Conclusions

In summary, stable duo-optical (chromogenic and fluorogenic) and electrochemical responses of chromophore 1 were exploited for the dual recognition of a cation (Cu²⁺, 0–50 ppb) and an anion (F⁻, 0–30 ppm). Moreover, due to its reversible and rapid output/processing under chemical stimulation, an array of logic circuits, viz. two-input/two-output and four-input/one-output, was integrated. The monitoring of the outputs via multiple modes (optical, electrochemical, and visual) provides accurate quantification, widespread utility, and multi-bit information processing at different wavelengths. Importantly, our processor provides label-free detection in a matrix of analytes and allows the retention of unique information with no scope of mixing.

Acknowledgements

RDG thanks DST (SERB/F/1424/2013-14) and South Asian University, New Delhi, India for financial assistance. AK and MC thanks UGC and CSIR for SRF. University of Delhi is greatly acknowledged for the technical support.

Notes and references

1. (a) J. M. Lehn, Science, 1993, 260, 1762–1763; (b) J. M. Lehn and A. V. Eliseev, Science, 2001, 291, 2331–2332; (c) M. Schmittel and H.-W. Lin, Angew. Chem., Int. Ed., 2007, 46, 893–896.
2. (a) J. M. Lehn, Science, 1985, 227, 849–856; (b) S. D. Bull, M. G. Davidson, J. M. H. Van Den Elsen, J. S. Fossey, A. T. A. Jenkins, Y. B. Jiang, Y. Kubo, F. Marken, K. Sakurai, J. Zhao and T. D. James, Acc. Chem. Res., 2013, 46, 312–326; (c) A. Kumar, M. Chhatwal, A. K. Singh, V. Singh and M. Trivedi, Chem. Commun., 2014, 50, 8488–8490; (d) K. Chen and M. Schmittel, Analyst, 2013, 138, 6742–6745; (e) K. Chen and M. Schmittel, Chem. Commun., 2014, 50, 5756–5759.
3. (a) A. P. de silva and S. Uchiyama, Nat. Nanotechnol., 2007, 2, 399–410; (b) T. Gupta and M. E. van der Boom, Angew. Chem., Int. Ed., 2008, 47, 5322–5326; (c) D. C. Magri, T. P. Vance and A. P. de Silva, Inorg. Chim. Acta, 2007, 360, 751–764; (d) A. Credi, Angew. Chem., Int. Ed., 2007, 46, 5472–5475.
4. (a) C. Simão, M. Mas-Torrent, J. C. Montenegro, F. Otón, J. Veciana and C. Rovira, J. Am. Chem. Soc., 2011, 133, 13256–13259; (b) A. Kumar, M. Chhatwal, P. C. Mondal, V. Singh, A. K. Singh, D. A. Cristaldi, R. D. Gupta and A. Gulino, Chem. Commun., 2014, 50, 3783–3785; (c) U. Pischel, Angew. Chem., Int. Ed., 2010, 49, 1356–1358; (d) U. Pischel, Aust. J. Chem., 2010, 63, 148–164.
5. (a) D. Margulies, G. Melman and A. Shanzer, Nat. Mater., 2005, 4, 768–771; (b) K. Chen, Q. Shu and M. Schmittel, Chem. Soc. Rev., 2015, 44, 136–160.
6. (a) A. P. de Silva, Nat. Mater., 2005, 4, 15–16; (b) B. Daly, J. Ling and A. P. de Silva, Chem. Sci. Eng., 2014, 8, 240–251; (c) A. P. de Silva and N. D. McClenaghan, Chem.–Eur. J., 2004, 10, 574–586.
7. (a) J. J. Lavigne and E. V. Anslyn, Angew. Chem., Int. Ed., 2001, 40, 3118–3130; (b) A. Kumar, M. Chhatwal and T. Gupta, Tetrahedron Lett., 2012, 53, 5691–5694; (c) K. Chen, J. W. Bats and M. Schmittel, Inorg. Chem., 2013, 52, 12863–12865.
8. N. W. Solomons, J. Am. Col. Nutr., 1985, 4, 83–105.
9. J. A. Camargo, Chemosphere, 2003, 50, 251–264.
10. (a) O. Fejerskov, J. Ekstrand and A. B. Burt, Fluoride in dentistry, Munksgaard, Copenhagen, 2nd edn, 1996; (b) B. D. Gessner, M. Beller, J. P. Middaugh and G. M. Whitford, N. Engl. J. Med., 1994, 330, 95–99.
11. A. Kumar, A. K. Singh and T. Gupta, Analyst, 2013, 138, 3356–3359.
12. D. Saha, S. Das, S. Karmakar, S. Dutta and S. Baitalik, RSC Adv., 2013, 3, 17314–17334.
13. D. Saha, S. Das, C. Bhaumik, S. Dutta and S. Baitalik, Inorg. Chem., 2010, 49, 2334–2348.
14. H. A. Benesi and J. H. Hildebrand, J. Am. Chem. Soc., 1949, 71, 2703–2707.
15. S. Qiu, S. Gao, L. Xie, H. Chen, Q. Liu, Z. Lin, B. Qiu and G. Chen, Analyst, 2011, 136, 3962–3966.
16. G. de Ruiter and M. E. van der Boom, Angew. Chem., Int. Ed., 2012, 51, 8598–8601.

---

Footnotes

† A dedication to the late Dr Tarkeshwar Gupta.

‡ Electronic supplementary information (ESI) available: X-Ray analysis data for 1, experimental details, characterization, and sensing methodology. CCDC 995811. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra14269a

---