Introduction

Acridine-based compounds have garnered significant attention due to their diverse applications. For instance, a range of substituted 9-aminoacridine derivatives have demonstrated potent chemotherapeutic activity against pancreatic cancer1,2. Acridine-based compounds were also employed as chemical sensors thanks to their fluorescent properties3,4,5,6. While acridine derivatives have been extensively studied in sensing and biological applications, acridine-based metal complexes as catalysts remain relatively less explored. Among these, pincer-type ruthenium complexes have emerged as prominent catalysts in organic transformations. A notable example is the acridine-based PNP-pincer Ru complex, RuHCl(CO)(A-iPr-PNP) [A-iPr-PNP = 4,5-bis-(diiso-propylphosphinomethyl)acridine], which has been shown to catalyze the selective coupling of alcohols with ammonia to yield primary amines, as well as the catalytic conversion of primary alcohols to acetals and esters7,8,9. Furthermore, the acridine-based ruthenium(II) complex was employed in the direct preparation of N-heteroaromatics (i.e., N-substituted pyrroles and pyrazine derivatives) through dehydrogenative coupling of diols10.

Given the importance of acridine compounds in various applications, considerable efforts have been dedicated to developing efficient synthetic methodologies. The Bernthsen reaction, involving the condensation of diphenylamine with a carboxylic acid and zinc chloride (ZnCl2) at elevated temperatures (> 200 °C) over extended periods (up to 24 h), is a straightforward synthesis of acridine compounds11,12,13,14. It has been proposed that the presence of Lewis acid catalysts (e.g., AlCl3, ZnCl2, FeCl3) facilitates the rearrangement of diphenylamine with an aldehyde or imine functional group at the 2-position of the phenyl ring. The compound with either the aldehyde or imine moiety, coordinated to the Lewis acid, undergoes cyclization followed by aromatization to yield the acridine core. While this method is well-established, it often suffers from harsh reaction conditions. Interestingly, a seminal work by Brooker and co-workers demonstrated the synthesis of the first example of tetrahedral acridine-based Schiff base cobalt(II) complexes (Fig. 1c) through a cobalt(II)-mediated rearrangement of diphenylamine-2,2′-dicarboxaldehyde (2,2′-dpadc, Fig. 1a) under mild conditions15. This discovery has opened up new avenues for the exploration of metal-mediated synthetic strategies for acridine derivatives. In our view, this facile method would allow us to further explore this class of ligands, especially in catalytic applications, by incorporation of various transition metals. In addition, this could also be a greener route for synthesis of acridine-based compounds due to its mild condition.

In this study, we investigated the formation of acridine-based metal complexes through the rearrangement of 2,2′-dpadc in the presence of various transition metal ions. This approach led to the successful synthesis and characterization of two novel pentadentate nickel(II) compounds featuring acridine-based Schiff-base ligands, [NiLACR](X)2·CH3CN (X = BF4 (1), ClO4 (2), LACR = (E)-N1-(2-((acridin-4-ylmethylene)amino)ethyl)-N1-(2-aminoethyl)ethane-1,2-diamine, Fig. 1d). To the best of our knowledge, this nickel(II) complex represents the second example of a metal-mediated rearrangement from 2,2′-dpadc to form an acridine-based metal complex. Our work further supports the viability of metal-mediated approaches as promising alternatives for the synthesis of acridine-based compounds. Additionally, the potential application of the nickel(II) complex in electrocatalytic hydrogen evolution reaction (HER) was also demonstrated.

Fig. 1
figure 1

Chemical Structure of (a) 2,2′-dpadc, (b) nickel(II) complexes featuring N4-Schiff base macrocycle ligands16, (c) acridine-based cobalt(II) complexes from diphenylamine-based precursors15, (d) nickel(II) complexes bearing Schiff-base acridine-based ligand in this work.

Full size image

Experimental section

Materials and methods

All reagents were used as received without further purification unless otherwise stated. Solvents used for syntheses were reagent grade except acetonitrile, which was HPLC grade. 2,2′-iminodibenzoic acid (95%), manganese(IV) oxide (activated, 85%), nickel(II) perchlorate hexahydrate (99%), tris(2-aminoethyl)amine (TREN, 96%), tetrabutylammonium hexafluorophosphate (98%), and Celite® filter cel (filter aid, slightly calcined) were obtained from Sigma Aldrich. Lithium aluminum hydride (95%) was purchased from TCI. Nickel(II) tetrafluoroborate hexahydrate (99%) was obtained from ACROS. Glassy carbon electrode and non-aqueous Ag/AgNO3 reference electrode were purchased from CH Instrument Inc.

Nuclear magnetic resonance (NMR) spectra were recorded on a 500 MHz JEOL at 298 K. UV-vis spectra were recorded on a Varian Cary 50 probe UV-vis spectrophotometer. Elemental analyses (C, H, N) were determined by THERMO FLASH 2000 CHNS/O analyzers. Thermogravimetric analyses (TGA) were carried out with a TGA 55 TA Instrument in the temperature range of 30–800 °C under a nitrogen atmosphere flow with a heating rate of 10 °C min− 1. Cyclic voltametric measurements were conducted using Metrohm-Autolab (Model PGSTAT101).

Acridine-based Schiff-base metal complexes from metal-mediated rearrangement of 2,2′-dpadc

In a typical reaction, to a bright-yellow, refluxing solution of 2,2′-dpadc (135 mg, 0.60 mmol) in CH3CN (10 mL) was added metal salts (0.60 mmol) dissolved in CH3CN (10 mL) (metal salts = Ni(BF4)2·6H2O, Ni(ClO4)2·6H2O, Fe(acac)3, Co(BF4)2·6H2O, Cu(BF4)2·6H2O, or Cu(ClO4)2·6H2O (Caution: perchlorate is potentially explosive, so the experiment should be carefully handled.). Then, a solution of TREN (88 mg, 0.60 mmol) in CH3CN (10 mL) was added dropwise over 20 min. Thanks to a characteristic absorption band of the acridine moiety around 360 nm17, formation of the acridine-based ligand during the reaction was monitored by UV-vis spectroscopy. After refluxing for a certain period of time (3–24 h), the solvent was removed under reduced pressure. If applicable, the product was then crystallized by liquid-liquid diffusion (CH3CN/diethyl ether).

[NiLACR](BF4)2 (1) was obtained as dark brown block-shaped crystals (56%). Anal. Calc. for [C20H25B2F8N5Ni]: C 42.31, H 4.44, N 12.34%. Found: C 42.32, H 4.59, N 12.56%. λmax/nm (εmax/ M− 1 cm− 1) = 291 (8800), 344 (6800), 360 (11000), 387 (6800), 403 (7250). ESI-MS (m/z) of [C20H25N5Ni]2+ for calculated: 196.57290; found: 196.57335.

[NiLACR](ClO4)2 (2) was obtained as dark brown block-shaped crystals (61%). Anal. Calc. for [C20H25Cl2N5NiO8]: C 40.51, H 4.25, N 11.81%. Found: C 40.55, H 4.33, N 12.12%. λmax/nm (εmax/ M− 1 cm− 1) = 291 (9100), 344 (7000), 360 (11500), 387 (7100), 403 (7500).

X-ray crystallography

Suitable crystals of complexes 1 and 2 were mounted on MiTeGen micromounts using paratone oil. X-ray diffraction data were collected using a Bruker D8 Quest Cmos Photon II operating at T = 296(2) K. Data were collected using ω and ϕ scans and using Mo-Kα radiation (λ = 0.71073 Å). The total number of runs and images was based on the strategy calculation from the program APEX3 and unit cell indexing was refined using SAINT18. Data reduction was performed using SAINT and SADABS were used for absorption correction. The integrity of the symmetry was checked by using PLATON19. The structure was solved with the ShelXT structure solution program using combined Patterson and dual-space recycling methods20. The structure was refined with a full-matrix least squares on F2 using ShelXL21 and OLEX222. In the final refinement cycles, all non-hydrogen atoms were refined anisotropically. The hydrogen atoms were introduced in calculated positions and refined with fixed geometry and riding thermal parameters with respect to the carrier atoms. A summary of the crystal data and relevant refinement parameters is presented in Table 1. CCDC 2,130,819 (1) and 2,130,820 (2), contain the supplementary crystallographic data for this study. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Homogeneous electrochemical studies of [NiLACR](BF4)2 (1)

The redox property of complex 1 (1 mM) in dimethylformamide (DMF) was examined by cyclic voltammetry (CV) using tetrabutylammonium hexafluorophosphate (TBAPF6) (0.1 M) as a supporting electrolyte. The measurements were performed using a conventional three-electrode one-component configuration in which the working electrode and the reference electrode was a glassy carbon electrode (diameter = 0.3 cm, A = 0.71 cm2) and an Ag/AgNO3 (10 mM) electrode, respectively, and a platinum wire was used as a counter electrode. The measured potentials were quoted with respect to the AgNO3/Ag electrode and internally calibrated to Fc/Fc+ redox couple. The experiments were carried out at 25 °C under N2 atmosphere. The current (i) at a glassy carbon working electrode was recorded at various scan rates of 0.05–1.00 V/s after applying voltage (E).

To investigate HER activity of 1, the electrocatalytic performance of complex 1 was measured in N2-saturated DMF/0.1 M TBAPF6 with various concentrations of acetic acid (0–30 mM) as a proton source. The measurements were carried out in a three-electrode system in a H-type cell with the anodic chamber (containing working and counter electrodes) and cathodic chamber (containing reference electrode) as illustrated in Figure S7.

The generated current corresponding to the catalytic HER activity of complex 1 was also recorded and plotted against the applied potential. Furthermore, controlled-potential electrolysis (CPE) was performed in the same electrochemical apparatus with the continuous stirring in both chambers during electrolysis. At the end of the electrolysis, aliquots of headspace gas (4 mL) from the cathodic chamber were collected and analyzed by gas chromatograph (Agilent Technologies 8890 GC instrument coupled with TCD detector, He carrier gas).

Results and discussion

Formation of pentadentate nickel(II) complexes bearing acridine-based Schiff-base ligand through rearrangement of 2,2′-dpadc

It has been previously reported that the metal complexes with diphenylamine-based Schiff-base metal macrocycles could be synthesized through a one-pot reaction of metal-templated [1 + 1] Schiff-base condensation23,24. Therefore, along with our study of nickel(II) complexes featuring N4 Schiff-base macrocycles (Fig. 1b)16, we initially intended to synthesize penta-coordinated Schiff-base nickel(II) macrocycles (NiLN5MCC)+ through this synthetic route using the same head unit (i.e., 2,2′-dpadc) to react with TREN in the presence of nickel(II) salts (i.e., Ni(BF4)2, or Ni(ClO4)2) (Fig. 2). To our surprise, the crystallographic data, however, revealed an unexpected formation of acridine-based nickel(II) Schiff-base complexes, [NiLACR](BF4)2 (1) and [NiLACR](ClO4)2 (2) (Fig. 1d).

Fig. 2
figure 2

Possible reaction between 2,2′-dpadc and TREN in the presence of metal ions.

Full size image

Structural descriptions

X-ray quality crystals were obtained by slow diffusion of diethyl ether into the solution of complexes in acetonitrile. The single crystal X-ray diffraction analyses reveal that complexes 1 and 2 are isostructural and crystallize in the centrosymmetric triclinic system with space group P-1 (Table 1). The asymmetric unit consists of one crystallographic unique nickel(II) ion, two counter anions, and one lattice acetonitrile molecule. The molecular structure of cationic species of the complex is shown in Fig. 3. The central nickel(II) ion is five-coordinated with a distorted trigonal bipyramidal geometry surrounded by five nitrogen atoms from the ligand, in which N1 and N3 atoms occupy the axial positions, while three basal positions being taken up by N2, N4, and N5 atoms. The Ni − N bond lengths are in the range of 1.957(2)-2.110(2) for 1 and 1.967(2)-2.115(2) Å for 2 (Table S1), which are comparable to those reported in other related nickel(II) complexes25,26,27.

Table 1 Crystal data and structural refinement for 1 and 2.
Full size table
Fig. 3
figure 3

ORTEP diagram of [NiLACR]2+ showing the coordination environment for nickel(II) atom in 1. The atoms are represented by 50% probability thermal ellipsoids.

Full size image

A close inspection of the crystal structure discloses that a pair of symmetry-related [NiLACR]2+ complexes are stacked through face-to-face π − π interactions exist between acridine moieties of the LACR ligands to from a supramolecular dimeric structure (centroid-to-centroid distances = 3.865(3) − 3.924(3) Å for 1, 3.866(2) − 3.921(2) Å for 2). The dimers are assembled into a one-dimensional column structure by N − H···F/O hydrogen bonds between the amine groups of the ligands and the counter anions, Fig. 4. Ultimately, the columns are interconnected with counter anions and lattice acetonitrile molecules through complementary N − H···F/O and C − H···F/O hydrogen bonding interactions, giving rise to a three-dimensional supramolecular architecture. Details of the hydrogen bonding geometry for the complexes 1 and 2 are given in Table S2.

Fig. 4
figure 4

Views of the one-dimensional column structures forming by the complementary hydrogen bonding and π-π stacking interactions in 1 (a) and 2 (b).

Full size image

Electronic and electrochemical properties of acridine-based nickel(II) complexes

UV-vis spectra of the two nickel(II) complexes showed the identical absorptions (in agreement with their isostructure) at 291, 344, 360, 387, and 403 nm which could be assigned to charge transfer transitions (Fig. 5)6,18. It should be noted that the characteristic absorption band of the acridine moiety in the complexes was observed at ca. 360 nm, consistent with the previous report6. However, no d-d transition band was clearly seen, probably due to its lower absorptivity with respect to that of the ligand.

The cyclic voltametric measurements of nickel(II) complexes were conducted in a N2-saturated DMF solution containing 0.1 M TBAPF6 (Figure S5). The cyclic voltammogram of 1 exhibited two irreversible processes at E = -1.18 V and E = -0.45 V for reduction (Epc,1) and oxidation events (Ep.a.,1), respectively. Moreover, at more negative potentials, a quasi-reversible process was observed at E1/2 = -1.89 V. Likewise, the cyclic voltammogram of 2 showed the redox events in negative potentials at E = -0.48 V and − 1.19 V for irreversible process, and at E1/2 = -1.90 V for a quasi-reversible process. These similar redox potentials and peak shapes indicate their similar electronic properties, which is consistent with the alike structural configuration.

Fig. 5
figure 5

UV-vis spectroscopic characterization of 0.1 mM [NiLACR](BF4)2 (1, red solid line) [NiLACR](ClO4)2 (2, black solid line), and 2,2′-dpadc in acetonitrile solution (blue dashed line).

Full size image

Proposed mechanism of acridine-based complex formation

According to the previous reported cobalt(II) mediated rearrangement of diphenylamine-based starting materials18, a similar mechanism of acridine formation for our nickel(II) complexes via rearrangement of 2,2′-dpadc was proposed as depicted in Fig. 6. Firstly, 2,2′-dpadc was reacted with tris(2-aminoethyl)amine (TREN) via Schiff base condensation, resulting in one imine arm. Next, the imine arm was activated by the nickel(II), providing more electrophilicity of the carbon atom in the imine which hereafter underwent a 6-membered cyclization. Thereafter, a couple of deprotonations took place to rearomatize and formed the acridine with a pendant carbonyl arm meanwhile releasing nickel(II) salt and TREN back into the solution. The carbonyl-dangling acridine and TREN were further subjected to nickel-templated Schiff-base condensation which ultimately formed the acridine-based Schiff-base Ni(II) complex.

Fig. 6
figure 6

Plausible mechanism of nickel-mediated acridine-based ligand formation.

Full size image

Investigation of rearrangement of 2,2′-dpadc mediated by various metal ions

To explore whether this nickel(II)-mediated rearrangement is applicable to other metal ions, a range of transition metal salts was tested. Unfortunately, when monitoring the reaction with Co(BF4)2, Fe(acac)3 or BF3 (non-metal Lewis acid), characteristic absorption of the acridine moiety at 360 nm was not observed (Figure S1). This indicated that these two metal salts as well as BF3 may not be able to induce the rearrangement to form an acridine-based ligand. Also, attempts to identify the product from these reactions resulted in unidentified substances.

Interestingly, when Cu(BF4)2 was employed in the reaction, [CuLN5MCC](BF4) (LN5MCC = N5-donor macrocyclic ligand) was obtained, and no indication of acridine formation was observed. Although attempts to obtain single crystals suitable for X-ray crystallography have not been successful, the copper(II) complex could be confirmed by UV-vis spectroscopy and mass spectrometry (ESI-MS). Markedly, the UV-vis spectrum of [CuLN5MCC](BF4) in CH3CN is similar to that of the previously reported [CuLEt](BF4)28,29 with a small bathochromic shift of 10–20 nm. On the other hand, when Cu(ClO4)2 was used as the metal source, a mixed product of [CuLACR(ClO4)]+ and [CuLN5MCC]+ was evidenced by UV-vis spectroscopy and ESI-MS. Unfortunately, the pure product of [CuLACR(ClO4)]+ could not be obtained for further characterization.

It can be seen that among the metal salts used in this work, only Ni(BF4)2 and Ni(ClO4)2 can mediate the rearrangement of 2,2′-dpadc in the reaction with TREN to give the purified product of acridine-based metal complexes, i.e. complex 1 and 2. The origin of exclusivity of the Ni2+ ions in this particular reaction may be attributed to the size and Lewis acidity of Ni2+ ions, but it has not been clearly understood. Further studies such as DFT calculations are needed to get deeper insight into this reaction. Thus, the studies regarding their potential application hereafter would focus on the nickel(II) complex. From the literature, a number of nickel(II) complexes with imine derivative ligands have been reported and used in various applications30,31,32,33,34. In particular, several Ni(II) complexes and materials have been demonstrated as potential electrocatalysts for hydrogen evolution reaction (HER)35,36. Due to the importance of HER in addressing energy and environmental challenges, it is of interest to further investigate the potential of the nickel(II) complex (1) in this particular application.

Electrocatalytic hydrogen evolution reaction (HER) of complex 1

Due to their similarity in the structure and properties, further studies regarding the HER activity of acridine-based nickel(II) complexes were concentrated only on complex 1. Firstly, complex 1 was examined whether it could be an active HER electrocatalyst in comparison to Ni(BF4)2. CVs of the complex 1 were performed in a N2-saturated 0.1 M TBAPF6 /DMF solution. Upon addition of acetic acid (0–30 mM) as the proton source, catalytic current enhancement around the reduction peak of Ni2+/ Ni+ in complex 1 was clearly observed and significantly higher than that of Ni(BF4)2. Next, the CPE was conducted at applied potential (Eapplied) of -2.1 V vs. Fc/Fc+ for 1 h in the N2-saturated 0.1 M TBAPF6 /DMF solution with acetic acid (30 mM) as the proton source (Fig. 7a). Then, the H2 gas product was analyzed by GC. It was found that the reaction employing complex 1 as the electrocatalyst could generate significantly higher amount of H2 than that using free Ni2+ ions (Table 2). This result clearly showed that complex 1 was active toward the HER. Moreover, the stable current of -2.11 mA observed throughout the 1-h electrolysis indicated satisfactory stability of complex 1 under this operational condition (Fig. 7b). It should be noted that increasing the applied potential from -2.1 V to -2.3 V did not vastly improve %FE of catalyst 1. Although the HER performance of complex 1 (40% FE) was in the lower end when compared to those of previously reported Ni2+ complexes (46–94%FE)37,38,39,40, the opportunity for other applications and several approaches for further development are still open such as ligand modification as well as molecular-catalyst immobilization onto carbonaceous materials.

Fig. 7
figure 7

Comparison of (a) charge accumulation profile and (b) current profile of the 1-h CPE in the absence of catalyst (blank, black line), and in the presence of Ni(BF4)2 (red line) and complex 1 (blue line) at -2.1 V vs. Fc/Fc+ in a N2-saturated 0.1 M TBAPF6 solution in DMF containing 30 mM acetic acid.

Full size image
Table 2 Summarized catalytic activity of catalysts for HER at -2.1 V vs. Fc/Fc+.
Full size table

Each reaction contains 1 mM catalyst in a N2-saturated 0.1 M TBAPF6 solution in DMF containing 30 mM acetic acid for 1 h electrolysis. %FE is the percentage of the Faradaic efficiency of the catalyst.

Conclusions

We report the synthesis and characterization of novel pentadentate nickel(II) complexes featuring an acridine-based Schiff base ligand. The complexes were prepared via a nickel-promoted one-pot condensation of diphenylamine-2,2′-dicarboxaldehyde and TREN, accompanied by a ligand rearrangement at the head-unit motif. Consistent with the previous report by Brooker and co-workers, the rearrangement was proposed to proceed after Schiff-base condensation and facilitated by the nickel ions as a Lewis acid. The presence of the acridine moiety in the complexes was confirmed by X-ray crystallography with two distinct counterions and corroborated by the characteristic absorption band of acridine in UV-vis spectroscopy. Furthermore, the electrocatalytic hydrogen evolution reaction (HER) activity of a representative nickel(II) complex (1) was investigated. The complex demonstrated moderate HER activity in 0.1 M TBAPF6 solution in DMF with the faradaic efficiency of 40% at -2.1 V vs. Fc/Fc+ in the presence of acetic acid as the proton source.

This facile synthetic approach through the metal-mediated rearrangement of diphenylamine derivative presents a promising avenue for the rational design and facile synthesis of novel acridine-based ligands. Moreover, this strategy facilitates the exploration of these ligands and their corresponding metal complexes in various applications, including organic transformations, electro- and photocatalysis, and biological applications.