A fluorescent naphthalenediimide AND logic gate: a Pourbaix sensor for two protons and one electron

Abstract

A core-substituted naphthalenediimide is demonstrated as a fluorescent H⁺, S₂O₈²⁻-driven AND logic gate in aqueous methanol.

---

Naphthalenediimides (NDIs) are used as building blocks in functional materials, artificial photosynthetic models and sensor devices.¹ Takenaka has used core-unsubstituted ferrocenyl NDIs as electrochemical sensors for studying intercalation with double stranded DNA.² Core-substituted NDIs have far greater potential utility in supramolecular chemistry and molecular recognition.³ They have tuneable optical and electrochemical properties.⁴ For example, primary (NH₂) and secondary amino (RNH) substituents in proximity to a carbonyl contribute towards longer absorbance and emission wavelengths besides higher fluorescence quantum yields (Φ_F).⁵ Despite these notable advantages, molecular logic gates based on NDIs are rare.^6,7 The few examples are a multifunctional redox switch by Thakur,⁸ anion-induced colorimetric logic gates by Mukhopadhyay,⁹ a NAND gate and molecular keypad lock by Liu,¹⁰ and multi-configurable acid–base logic gates by Kalita.¹¹

Previously we reported 2 (Fig. 1).¹² We hypothesised based on precedence with naphthalimide¹³ and perylenediimide,¹⁴ that 2 would function as a Pourbaix sensor,¹⁵ an AND logic gate for acidity and oxidizability based on a photoinduced electron transfer (PET) mechanism.^16,17 The molecule was designed with an electron-donor–spacer₁–fluorophore–spacer₂–receptor format with ferrocene as the electron donor, dimethylamino as the proton receptor and methylene and ethylene spacers. Unexpectedly, we discovered that 2 does not function as a H⁺, oxidant-driven AND logic gate, but rather as a two-input PASS 0 logic gate.^12,18 It remained ‘off’ with no observable fluorescence, impartial to the presence of high H⁺ and high oxidant levels. In hindsight, we are now aware that core-unsubstituted NDIs are non-emissive from the S1 excited state (ππ*) state due to rapid intersystem crossing (ISC) to an upper T4 (nπ*) state.¹⁹ Moreover, core-substituted NDIs such as 3 with a secondary amine are emissive, despite having a heavy Br atom that contributes towards a fast (<200 fs) ISC ππ* → nπ* transition due to strong spin–orbit coupling. With this knowledge at hand, we rationalised that a core-substituted Pourbaix sensor is possible.

Fig. 1 The emissive core-substituted Pourbaix sensor 1, non-emissive model 2,¹² and emissive model 3.¹⁹

Pourbaix sensors with ferrocene have typically been designed for stoichiometric amounts of 1 H⁺ and 1 e⁻.^13,14,20,21 Our anthracene-based Pourbaix sensor²¹ has since been used as a reductive photoredox catalyst.²² An earlier tetrathiafulvalene prototype, was most emissive on receiving 1 H⁺ and donating 2 e⁻,²³ which is equivalent to hydride (H⁻). Now we present a Pourbaix sensor that communicates a fluorescence signal after accepting 2 equivalents of H⁺ and donating 1 equivalent of e⁻. Photosystem II with tyrosine and histidine is a biologically relevant system with a one-electron, two-proton coupled transfer.²⁴

In this study, we present NDI-based Pourbaix sensor 1 as a H⁺, S₂O₈²⁻-driven AND logic gate in aqueous methanol and methanol. The molecular device is constructed in an electron-donor–spacer₁–fluorophore–(spacer₂–receptor)₂ configuration (Fig. 1, see graphical abstract). The NDI fluorophore is centrally positioned between two morpholino receptors attached at the imide termini by an ethylene spacer (spacer₂) and a ferrocene moiety at the core position via a methylene spacer (spacer₁) to a secondary nitrogen atom. Within 1 three intramolecular PET processes^20,25 compete to turn ‘off’ the fluorescence. Turning ‘on’ the logic gate requires two protons and the loss of one electron (2H⁺, e⁻).

The synthetic approach for 1 is more convenient than for 2.¹² Intermediate 5 was synthesised by condensation of 2,6-dibromonaphthalene-1,4,5,8-tetracarboxylic dianhydride 4 with excess 4-(2-aminoethyl)morpholine refluxed in glacial acetic acid (Fig. 2). An orange powder, obtained in 79% yield, was subsequently reacted with aminomethylferrocene in DMF and triethylamine (TEA) at 80 °C. Purification by column chromatography on silica gel provided a thermally stable purple solid 1 with a melting point >400 °C in 18% yield. ¹H, ¹³C & ¹³C DEPT NMR, FTIR and HRMS spectroscopic data for 1 and 5 are available in the SI (Fig. S1–S11).

Fig. 2 The synthesis of 1 incorporating PET and ICT mechanisms.

The ¹H NMR spectrum of 1 (Fig. S2) has resonances at 2.59 ppm and 3.67 ppm (morpholine protons) and 2.71 ppm and 4.40 ppm (ethylene protons). A singlet at 4.47 ppm is diagnostic for the CH₂ spacer adjacent to ferrocene. The latter integrates to a 2 : 5 : 2 ratio at 4.25 ppm, 4.29 ppm and 4.32 ppm. Two singlets at 8.34 ppm and 8.87 ppm are assigned to the aromatic core protons. The signal at 10.32 ppm is attributed to the N–H hydrogen atom.

The ¹³C NMR DEPT spectrum provides further evidence for the conversion of 5 to 1. The DEPT spectra of 5 (Fig. S4) has 5 resonances: an up signal at 139.07 ppm for the aromatic hydrogen atoms (CH), and downs signals at 38.16 ppm, 53.81 ppm, 55.90 ppm and 67.00 pm for the CH₂ of the ethylene spacers and morpholine units. The DEPT spectra of 1 (Fig. S6) clearly highlights the loss of C₂ symmetry with two up signals at 138.38 ppm and 120.68 ppm for the aromatic CHs. A down signal at 42.86 ppm is diagnostic for the CH₂ spacer adjacent to ferrocene.

Spectroscopic studies of 1 were conducted in 1 : 1 (v/v) methanol/water and methanol (Tables 1 and 2). The UV-vis absorbance spectrum of 1 reveal peaks at 350 nm, 368 nm and 533 nm (Fig. 3a). The molar absorption coefficients (log ε) are 3.77, 3.84 and 3.82 indicative of allowed π → π* transitions. Isosbestic points at 327 nm, 377 nm, 461 nm and 569 nm are indicative of strong ICT character associated with the conjugated nitrogen atom. With increasing basicity, the absorbance peaks at 350 nm, 371 nm and 533 nm are observed to decrease with notable loss of fine vibrational structure at pH 9.8. The baseline begins to increase, suggestive of aggregation as the morpholine receptors are deprotonated leading to greater lipophilicity. By comparison, the λ_abs of 2 are restricted to the shorter UV wavelengths of 341 nm, 358 nm and 378 nm with log ε of 3.98, 4.20 and 4.25. Dibrominated 5 has peaks at 366 nm, 391 nm and 411 nm with log ε of 4.11, 3.80 and 3.89.

Table 1 Optical properties of 1 in 1 : 1 (v/v) MeOH/H₂O and PASS 0 gate 2 in MeOH^a

	1	2
a 10^−5 M. b Acidic pH adjusted with 0.1 M CH3SO3H for 1 and 0.1 M HCl for 2. c pKa* determined by log[(Imax − I)/(I − Imin)] = −log[H^+] + logβH+ from emission spectra excited at λIsos = 461 nm. d Could not be determined. e Quantum yields relative to fluorescein in 1 M NaOH (ΦF = 0.95). f 10^−4.0 M H^+ & 100 µM Na2S2O8. g Inputs 10^−4.0 M H^+ & 100 µM Fe^3+. h H^+-induced fluorescence enhancement (FE) IFhighest/IF2nd highest. i ΔGPET = Eox − Ered − E0,0 − e^2/εr where Eox (amine) = 1.15 eV, Eox (ferrocene) = 0.45 eV, Ered (1) = −1.00 eV, Ered (2) = −1.11 eV, E0,0 (1) = −2.25 eV, E0,0 (2) = −3.20 eV, e^2/εr = 0.10 eV.^4,12
λ Abs pH 3.6/nm^b	350, 368, 533	341, 358, 378
log ε (cm^−1 mol^−1 L)	3.77, 3.84, 3.82	3.98, 4.20, 4.25
λ Abs Isos/nm	327, 377, 461, 569	—
λ Flu/nm	570	393, 413
pKa*^c	5.4	—^d
Φ f	0.020^f	<0.001^g
FE^h	6.0	<1
ΔGPET/eV^i	−0.20, −0.90	−1.04, −1.74

---

Table 2 Truth table for AND gate 1 in 1 : 1 (v/v) methanol/water and methanol^a

Input1 (H^+)^b	Input2 (S2O8^2−)^c	Output 1 (MeOH/H2O,ΦF)^d	Output 1 (MeOH,ΦF)^d
a 12 µM 1 excited at 461 nm. Abs at 0.15. b High input1 10^−4.0 M H^+ and low input1 H^+ 10^−10 M adjusted with CH3SO3H and 1 M NaOH in 1 : 1 (v/v) MeOH/H2O. High input1 10^−4.0 M H^+ and low input1 with no acid added in MeOH. c High input1 100 µM Na2S2O8 and low input2 no Na2S2O8 added. d Relative ΦF of 1versus 10^−6 M fluorescein in aqueous 1 M NaOH (ΦF = 0.95). The high threshold output level is ΦF > 0.01 from duplicate measurements.
0 (low)	0 (low)	0 (low, 0.0010)	0 (low, 0.0017)
0 (low)	1 (high)	0 (low, 0.0016)	0 (low, 0.0044)
1 (high)	0 (low)	0 (low, 0.0035)	0 (low, 0.0073)
1 (high)	1 (high)	1 (high, 0.020)	1 (high, 0.049)

---

Fig. 3 (a) UV-vis absorbance spectra of 13 µM 1 in 1 : 1 (v/v) MeOH/H₂O with [H⁺] from 10^−10.0 M to 10^−4.0 M; (b) Emission spectra of 1 in methanol, λ_ex = 505 nm with four input conditions (A–D); (c) pH titration of 1 in 1 : 1 (v/v) MeOH/H₂O with 100 µM Na₂S₂O₈, λ_ex = 461 nm; (d) solutions of 1 in MeOH irradiated with 365 nm UV light: (A) no inputs, (B) 100 µM Na₂S₂O₈, (C) 10^−4.0 M H⁺ and (D) 100 µM Na₂S₂O₈ and 10^−4.0 M H⁺.

Molecule 1 is the first fluorescent NDI Pourbaix sensor. It has a longer emission wavelength than previous naphthalimide¹³ and perylenediimide prototypes, limited to a yellow fluorescence at 522 nm.¹⁴ Pourbaix sensor 1 has a maximum absorbance wavelength of 533 nm with a broad tail (Fig. 3a) and emits orange fluorescence (Fig. 3d). The Φ_F of 0.020 is similar in intensity to blue-emitting anthracene Pourbaix sensors.^21,23 The fluorescence enhancement (FE) ratio of 6.0 in 1 : 1 (v/v) H₂O/MeOH exceeds that of many Pourbaix sensors studied in methanol and THF.^13,14 The longer emission wavelength is due to the strengthened intramolecular hydrogen bond between the N–H and the C=O group in the excited state.⁵

Fluorescence titrations were carried out by titrating 1 with one input as the titrant while the other input was held constant (Fig. 3c). On titration with acid, a sigmoidal-shaped trendline was observed between pH 4.6 and pH 6.5. A pK_a* of 5.4 was evaluated by fitting the pH-intensity data to the Henderson–Hasselbalch equation (see footnotes Table 1, SI). The spatially separated tertiary amines are expected to have near identical pK_a* values. Titration with persulfate (E° = 2.5 V) oxidises ferrocene (+2) to ferrocenium (+3) generating two mole equivalents of sulfate.²⁶ Subsequent addition of thiosulfate reduces the ferrocenium to ferrocene; and hydroxide deprotonates the protonated amines, which in both cases turns off the emission (Fig. S12–S14). Addition of more CH₃SO₃H and Na₂S₂O₈ restores the emission output (Fig. S15).

The Gibbs energy for the two PET pathways is predicted from the Weller equation using available thermodynamic data (Table 1).^12,27 The ΔG_PET from either of the morpholines or ferrocene to the excited state NDI is favourable with ΔG_PET = −0.20 eV and −0.90 eV, respectively. These values for 1 are less exothermic than for 2 due to the considerably lower HOMO–LUMO gap from core substitution.

Compound 1 functions as a H⁺, S₂O₈²⁻-driven AND logic gate. The truth tables for the four input-output situations are summarised in Table 2. Emission spectra are provided in Fig. 3b and Fig. S12. In the absence of both inputs, or in the absence of only one input, the fluorescence is ‘off’ due to PET from either ferrocene, or from one of the two morpholino nitrogen lone pairs to the excited state NDI fluorophore. However, when both inputs are present in excess, the two tertiary amines are protonated and the ferrocene is oxidised, resulting in an orange fluorescence (Fig. 3d). In the ‘on’ state: in 1 : 1 (v/v) MeOH/H₂O, the Φ_Fmax is 0.020 with a FE of 5.7; in MeOH the Φ_Fmax is 0.049 with a FE of 6.7.

In summary, we presented the first fluorescent NDI-based logic AND gate for acidity and oxidizability. An orange fluorescence is observed above high H⁺ and high S₂O₈²⁻ threshold levels. Appending a secondary amino substituent on the NDI aromatic core induces favourable fluorescence properties consistent with a modular PET design. The bromine atom at the core position provides a site for future synthetic modification such as: next-in-line AND gate prototypes with improved photophysical properties (higher Φ_F), other two-input logic gate types (INHIBIT),²⁸ multi-diagnostic three-input logic gates,^20,29 and covalent attachment to a solid support.³⁰ New research endeavours are also envisioned in corrosion detection,¹⁴ and fluorescent cellular imaging³¹ among others.¹⁵

Author contributions

D.C.M.: conceptualization, supervision, writing – original draft, writing – review & editing; J.V., M.C. investigation.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

Supplementary information (SI): experimental details, ¹H, ¹³C & ¹³C DEPT NMR, FTIR, mass spectrum with accurate mass and emission spectra. See DOI: https://doi.org/10.1039/d6nj00080k.

Acknowledgements

The University of Malta is thanked for financial support.

Notes and references

1. (a) M. Al Kobaisi, S. V. Bhosale, K. Latham, A. M. Raynor and S. V. Bhosale, Chem. Rev., 2016, 116, 11685; (b) S. V. Bhosale, M. Al Kobaisi, R. W. Jadhav, P. P. Morajkar, L. A. Jones and S. George, Chem. Soc. Rev., 2021, 50, 9845.
2. (a) S. Sato and S. Takenaka, J. Inorg. Biochem., 2017, 167, 21; (b) S. Sato and S. Takenaka, J. Organomet. Chem., 2008, 693, 1177.
3. (a) N. Sakai, J. Mareda, E. Vauthey and S. Matile, Chem. Commun., 2010, 46, 4225; (b) C. Thalacker, C. Röger and F. Würthner, J. Org. Chem., 2006, 71, 8098; (c) F. Würthner, S. Ahmed, C. Thalacker and T. Debaerdemaeker, Chem. Eur. J., 2002, 8, 4742.
4. J. Shukla and P. Mukhopadhyay, Eur. J. Org. Chem., 2019, 7770.
5. S. Maniam, H. F. Higginbotham, T. D. M. Bell and S. J. Langford, Chem. – Eur. J., 2019, 25, 7044.
6. A. P. de Silva, Molecular Logic-based Computation, The Royal Society of Chemistry, Cambridge, 2013. See ch. 14 for applications.
7. (a) A. Ciupa, RSC Adv., 2025, 15, 10565; (b) L. Liu, P. Liu, L. Ga and J. Ai, ACS Omega, 2021, 6, 30189; (c) C.-Y. Yao, H.-Y. Lin, H. S. N. Crory and A. P. de Silva, Mol. Syst. Des. Eng., 2020, 5, 1325; (d) D. C. Magri, Coord. Chem. Rev., 2021, 426, 213598.
8. S. Jana, S. Kumar Bag, B. Mondal, M. Karmakar and A. Thakur, Organometallics, 2024, 43, 1459.
9. M. R. Ajayakumar, G. Hundal and P. Mukhopadhyay, Chem. Commun., 2013, 49, 7684.
10. W. Jiang, M. Han, H.-Y. Zhang, Z.-J. Zhang and Y. Liu, Chem. – Eur. J., 2009, 15, 9938.
11. H. J. Bora, P. Barman, S. Bordoloi, G. Gogoi, B. Gogoi, N. S. Sarma and A. Kalita, RSC Adv., 2021, 11, 35274.
12. J. Grech, J. C. Spiteri, G. J. Scerri and D. C. Magri, Inorg. Chim. Acta, 2023, 544, 121176.
13. (a) J. C. Spiteri, J. S. Schembri and D. C. Magri, New J. Chem., 2015, 39, 3349; (b) A. D. Johnson, K. A. Paterson, J. C. Spiteri, S. A. Denisov, G. Jonusauskas, A. Tron, N. D. McClenaghan and D. C. Magri, New J. Chem., 2016, 40, 9917; (c) J. C. Spiteri, S. A. Denisov, G. Jonusauskas, S. Klejna, K. Szaciłowski, N. D. McClenaghan and D. C. Magri, Org. Biomol. Chem., 2018, 16, 6195; (d) J. C. Spiteri, A. D. Johnson, S. A. Denisov, G. Jonusauskas, N. D. McClenaghan and D. C. Magri, Dyes Pigm., 2018, 157, 278; (e) G. J. Scerri, M. Caruana, N. Agius, G. Agius, T. J. Farrugia, J. C. Spiteri, A. D. Johnson and D. C. Magri, Molecules, 2022, 27, 5939.
14. G. J. Scerri, J. C. Spiteri and D. C. Magri, Mater. Adv., 2021, 2, 434.
15. D. C. Magri, Supramol. Chem., 2017, 29, 741.
16. A. P. de Silva, T. S. Moody and G. D. Wright, Analyst, 2009, 134, 2385.
17. D. C. Magri, Analyst, 2017, 142, 676.
18. F. Magro, L. Camenzuli and D. C. Magri, Chemosensors, 2023, 11, 1.
19. O. Yushchenko, G. Licari, S. Mosquera-Vazquez, N. Sakai, S. Matile and E. Vauthey, J. Phys. Chem. Lett., 2015, 6, 2096.
20. D. C. Magri, M. Camilleri Fava and C. J. Mallia, Chem. Commun., 2014, 50, 1009.
21. T. J. Farrugia and D. C. Magri, New J. Chem., 2013, 37, 148.
22. P. Garrido-Barros, C. G. Romero, J. R. Winkler and J. C. Peters, J. Am. Chem. Soc., 2024, 146, 12750.
23. D. C. Magri, New J. Chem., 2009, 33, 457.
24. G. Chararalambidis, S. Das, A. Trapali, A. Quaranta, M. Orio, Z. Halime, P. Fertey, R. Guillot, A. Coutsolelos, W. Leibl, A. Aukauloo and M. Sircoglou, Angew. Chem., Int. Ed., 2018, 57, 9013.
25. D. C. Magri, G. D. Coen, R. L. Boyd and A. P. de Silva, Anal. Chim. Acta, 2006, 568, 156.
26. L. B. Feng, Y. Zeng, P. Wang, N. Duan, H. D. Ji and X. Zhao, Processes, 2024, 12, 2361.
27. X. N. Cui, J. Z. Zhao, Z. R. Lou, S. J. Li, H. J. Wu and K.-L. Han, J. Org. Chem., 2015, 80, 568.
28. G. J. Scerri, M. Cini, J. S. Schembri, P. F. da Costa, A. D. Johnson and D. C. Magri, Chem. Phys. Chem., 2017, 18, 1742.
29. G. J. Scerri, J. C. Spiteri, C. J. Mallia and D. C. Magri, Chem. Commun., 2019, 55, 4961.
30. (a) M. Vella Refalo, J. C. Spiteri and D. C. Magri, New J. Chem., 2018, 42, 16474; (b) M. Vella Refalo, N. V. Farrugia, A. D. Johnson, S. Klejna, K. Szaciłowski and D. C. Magri, J. Mater. Chem. C, 2019, 7, 15225.
31. F. Doria, M. Folini, V. Grande, G. Cimino-Reale, N. Zaffaroni and M. Freccero, Org. Biomol. Chem., 2014, 13, 570.

---