Shape-responsive host–guest chemistry: metal-free tetracationic porphyrin nonplanarity promoted by clay mineral interactions assessed by theoretical simulations

Abstract

Distortions in the porphyrin core from planarity can trigger a unique structure–property relationship, imparting its basicity, chemical stability, redox potential, and excited-state energetics, among other properties. The colour change promoted by such distortion is signed by red shifts in its electronic absorption spectra. The adsorption of guest meso-substituted free-base porphyrin species onto inorganic hosts, such as clay minerals (layered aluminium or magnesium silicates), is known to further promote colour changes. However, the origin of these changes remains a subject of debate without a clear consensus. In this work, an extensive theoretical study was conducted using density functional theory (DFT) to model the interactions between tetracationic porphyrins, specifically meso-substituted groups N-methyl-4-pyridyl (p-TMPyP) and N-methyl-3-pyridyl (m-TMPyP), and montmorillonite (MMT) with the ideal formula [(Al_1.67Mg_0.33)Si₄O₁₀(OH)₂]^−0.33. The following conditions were evaluated: (i) adsorption or intercalation of p-TMPyP into MMT host structure, (ii) intercalation of m-TMPyP into MMT, and (iii) the influence of water on the intercalation process. The electrostatic interactions between the porphyrins and the MMT siloxane surface induced conformational changes in p-TMPyP, characterized by rotation of the substituent groups at the macrocycle periphery and a twist of the porphyrin plane. The nonplanarity of the intercalated p-TMPyP guest produced robust Brønsted basic sites capable of abstracting H⁺ ions from intercalated water molecules, resulting in the formation of a dication. The macrocycle distortion was found to decrease π-conjugation, thereby enhancing the localisation of the lone pair on the imine nitrogen atom. On the other hand, m-TMPyP exhibited slight core macrocycle deformations and minor changes in the dihedral angles of its meso-substituent groups compared to its isomer, with no observed protonation reaction upon intercalation. These findings highlight the clay microenvironment as a promising strategy for inducing conformational alterations in porphyrins, promoting nonplanarity, and exemplifying a shape-responsive system within the framework of guest–host chemistry.

---

Introduction

Porphyrins are natural macrocyclic tetrapyrrolic compounds linked by methine bridges, characterized by an aromatic ring system of conjugated π electrons. This system includes a central cavity capable of coordinating metal cations such as Cu, Fe, Co, Mg, and Zn. The peripheral substituents of the porphyrin ring, which can be located at the β-pyrrolic positions or the meso-positions (Fig. 1a), along with the protonation of the macrocycle (Fig. 1b) and metal coordination (Fig. 1c), significantly influence the conformation and electronic structure of the porphyrins. These structural modifications underscore a remarkable structure–property relationship, a phenomenon also observed in biological systems.¹ The responsiveness of porphyrin properties to such structural changes has driven the exploration of synthetic methods aimed at developing metal- or non-metal macrocyclic systems with distinctive characteristics, offering potential applications across various scientific and technological fields. The versatility of porphyrins has enabled a wide range of studies, including catalysis,² electro-catalysis,³ photocatalysis for H₂ production and CO₂ reduction,⁴ sensor systems,⁵ organic electronics,⁶ electron transfer models,⁷ spintronics,⁸ and the conversion of sunlight into chemical energy.¹

Fig. 1 Schematic of (a) meso-substituted free base porphyrin showing the characteristic α, β, and meso carbon atoms, (b) meso-substituted protonated or dication porphyrin, and (c) metalated porphyrin. The blue circle indicates the N4-core, and R represents para (p-) and meta (m-) methyl-pyridyl substituents.

Structural modifications in porphyrins can lead to significant deviations from the planar conformation of the ring, as observed in the haemoglobin protein. The conformational flexibility of porphyrins, which promotes ring nonplanarity, can arise from various factors, including steric crowding at the periphery of the porphyrin due to β- and/or meso-substituents, steric repulsion within the macrocycle core, metalation, or ligand coordination to the metal centre, among other factors.⁹ The most common conformational modifications involving pyrrole rings alternately tilted out of the plane are shown in Fig. 2.⁹

Fig. 2 Symmetrical non-planar distortions of the porphyrin ring. Orange atoms are above the plane, blue atoms are below the plane, and black atoms are in the plane.

Previous studies on synthesised non-planar porphyrins have highlighted the structure–property correlations and the application potential of this class of macrocycles. Distorted free base porphyrins make available the pyrrole nitrogen lone pair and N–H of the core (N/N–H), which are not practically accessible in the planar framework, opening up the species to hydrogen bond interactions (N–H⋯X, where X = solvent, anion, chiral molecule for instance).^10,11 Hence, the core reactivity allows exploring non-planar porphyrins as base catalysts,¹¹ and sensor devices¹¹ in systems whose planar analogues are otherwise inert.

Potential modifications in the structure of meso-substituted free base porphyrin were suggested in studies about its adsorption on the flat siloxane surfaces of clay minerals, i.e. layered aluminium (or magnesium) silicates, suggesting that varying the microenvironment would be an exciting approach to inducing alterations in the properties of the porphyrin. Macroscopically, a change in colour from red to green (red-shift of the Soret and Q bands) was observed in solid-state samples and aqueous dispersions.¹² Despite the porphyrin/clays system being in the solid-state or aqueous dispersion, the visible spectrum showed absorption in the Soret region compatible with a superposition of two or more bands, whose relative intensities depended on the clay mineral and the amount of porphyrin. Such a change in the electronic states of molecules due to surface adsorption is termed adsorchromism (adsorption plus chromism).¹²

First, it is important to confirm that such a colour alteration is not a consequence of a redox chemical reaction or aggregation phenomenon, and so the main factors to be considered are the protonation (by adsorbed water hydrolysed by the interlayer cation) and the intramolecular structural change of porphyrin in contact with the flat clays surfaces. It was reported that porphyrin protonation promoted the red-shift of the Soret and Q bands and the nonplanar core of the free base porphyrins.¹³ To explain the bathochromic shift of the Soret band, Carrado and Winans¹⁴ and Kuykendall and Thomas¹⁵ suggested the protonation of the core nitrogen atoms of the macrocyclic molecule. However, the formation of a dication by porphyrin protonation (Fig. 1b) is not the main reason for the spectral changes arising from the porphyrin adsorption on clay minerals, because such a red-shift was also observed when metalated porphyrins were used.^16,17

One of the most studied clay-adsorbed porphyrins is the tetracationic meso-substituted porphyrin comprising an N-methyl-4-pyridyl unit (R in Fig. 1), abbreviated p-TMPyP, and trimethylanilinium, in which the aromatic group is almost perpendicular to the porphyrin plane. These species could be immobilised on smectite clays by ion-exchange reactions, establishing electrostatic interactions. Some works proposed the rotation of the methyl-pyridinium unit around the porphyrin core,¹⁸ stating that this could lead to an extension of the conjugate π system and a red-shift of the absorption bands. Chernia and Gill¹⁸ interpreted the spectral changes based on empirical calculations, indicating the methyl-4-pyridyl group twist involved a dihedral angle of 30° or 40°; while the rotation to minor angles was almost impossible due to the increase in the rotational energy barrier. A DFT study about the structure and electronic properties of p-TMPyP and its protonated form was previously reported by Suarez et al.,¹⁹ and also corroborated the energy barrier findings proposed by Chernia and Gill.¹⁸

Dias et al.^16,20,21 used resonance Raman spectroscopy, a technique more sensitive to the molecule structure than UV–Vis electronic spectroscopy, to record the spectra of solid-state samples comprising p-TMPyP-clay. Based on Raman spectroscopic data, they reported that the red-shift of p-TMPyP on LAPONITE® (synthetic hectorite) and vermiculite was mainly due to a non-planar distortion of the porphyrin ring, while core-distorted and the protonated p-TMPyP species were observed in the montmorillonite material. These results evidenced the importance of the clay properties in interacting with porphyrins as it can drive the conformation and characteristics of the adsorbed chromophore.

Although the nature of clay has been recognised as influencing the interaction with p-TMPyP, montmorillonite was chosen for the simulation studies in this work because it is highly water-swellable, a cation exchanger, and a common and valuable material in several fields. Furthermore, previous work showed the presence of two kinds of p-TMPyP in montmorillonite, opening the opportunity to investigate some remaining questions, such as (i) whether adsorption on external basal surfaces or internal (between the layers) surfaces can promote the formation of distinct porphyrin species, (ii) if protonation is facilitated by the core distortion or vice versa, and (iii) the role of the substituent groups in the structural distortion of porphyrins.

In the present work, an extensive theoretical study was reported based on density functional theory (DFT) calculations to evaluate the conformation of p-TMPyP (i) adsorbed or intercalated into montmorillonite (MMT), and (ii) in the absence or presence of water molecules in the system. It was found that the electrostatic interaction between the cationic porphyrin and the negatively charged MMT layers brings about a change in the conformation of p-TMPyP comprising rotation of the methyl pyridine in the periphery of the macrocycle ring and a twist of the pyrrole core in relation to the porphyrin plane. For comparison purposes, DFT simulations were also performed for the isomer N-methyl-3-pyridyl (Fig. 1). It was found that the nonplanarity of the p-TMPyP core on the MMT produced Brønsted basic sites robust enough to abstract H⁺ ions from intercalated water molecules.

Computational details

All calculations were performed in the framework of DFT,²² where the self-consistent Kohn–Sham equations were solved using the pseudopotential and plane-wave method, as implemented in the Quantum ESPRESSO code, version 6.1.^23,24 Vanderbilt's ultrasoft pseudopotentials,²⁵ selected from the PSLibrary, version 1.0.0.,²⁶ were utilised with a kinetic energy threshold of 40 Ry and 400 Ry for the charge density.²⁵ Also, the exchange–correlation functional GGA-PBE²⁷ modified with the van der Waals correction D2 was employed.²⁸ A grid of 9 × 9 × 3 k-points was applied to describe the reciprocal space for the bulk cases according to the Monkhorst–Pack scheme.²⁹ The geometry optimisation was performed using the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm, while Hellmann–Feynman forces smaller than 0.001 eV Å⁻¹ were considered as the convergence criterion.

Na-MMT bulk structure simulation

Since montmorillonite materials present a representable variety of compositions,³⁰ the theoretical montmorillonite model of Pirillo and co-workers was chosen as the starting point for the simulation of the clay structure.³¹ The ideal montmorillonite formula Na_0.33(Al_1.67Mg_0.33)Si₄O₁₀(OH)₂ was considered, in which the cation isomorphic substitution was assumed to occur only in the octahedral sheet of Al³⁺ with Mg²⁺, and Na⁺ ions were hydrated by one-layer of four water molecules in the interlayer region. The model considered a monoclinic unit cell reported in the literature.³¹

Potential energy study of p-TMPyP on the MMT surface

The molecular structure of the non-protonated p-TMPyP (para-isomer) tetracation (C₄₄H₃₈N₈) was simulated, to optimise its geometry, with a total charge of +4, isolated in vacuum.¹⁹ This structure was transferred to the (001) surface of MMT (adsorption) or the internal surfaces (intercalation) to simulate the MMT-clay model, according to the following steps: (1) intercalated Na⁺ ions were removed from the structure; (2) the interlayer distance between adjacent layers (crystallographic c-axis) was increased; (3) the crystallographic a and b dimensions of the unit cell were duplicated to accommodate p-TMPyP. Thus, a 4 × 2 × 1 supercell of MMT with 336 atoms was generated with the lattice parameters: a = 21.03 Å; b = 17.54 Å; c = 47.49 Å and β = 95.34°. Including the p-TMPyP atoms, the TMPyP-MMT surface model had 426 atoms. Due to the system's size, a fixed molecule scan on the clay surface on possible adsorption sites on the MMT (001) surface was performed. Only the gamma point was considered in the reciprocal space.

p-TMPyP (m-TMPyP) intercalated into MMT

After the study of p-TMPyP adsorbed on an anhydrous MMT surface, the lattice parameter c was reduced to 15.24 Å, for simulating the intercalation process, and tests for optimisation of the geometry were then performed.²⁰ Here, 28 water molecules were randomly added, using Visual Molecular Dynamics (VMD) software,³² and the solvation properties of the intercalated p-TMPyP-MMT system, at 1000 kg m⁻³ density, were investigated in triplicate.³⁰ The optimisation geometry step was repeated afterwards. The m-TMPyP intercalation study started from the anhydrous p-TMPyP structure, thus modifying the para-substituent, for carrying out geometry optimisation, as well as for the solvation step.

UV–Vis spectroscopy modelling

The optical transitions were carried out using the time-dependent density functional theory (TD-DFT), as implemented in the Gaussian 09 computational package.³³ For these calculations, the 6-311G(d,p) basis set was employed, using the CAM-B3LYP exchange–correlation functional.³⁴ To mimic the MMT's influence on p-TMPyP, first, the geometric structure of the molecule in the +4 charge state after the optimisation on the clay was extracted. Then, the p-TMPyP was submitted to the Polarizable Continuum Model (PCM)³⁵ environment to mimic the MMT, using the high-frequency dielectric constant (11.21) and static dielectric constant (12.75), obtained using the method reported by Pasquarello et al.³⁶ Indeed, the use of PCM to simulate the environment was successfully employed before providing the water effects for this molecule.¹⁹

Results and discussion

Na-MMT (anhydrous and hydrated) model

The layer of montmorillonite, a smectite clay, was composed of two sheets of linked [SiO₄] tetrahedra sandwiching one sheet of [AlO₄(OH)₂] or [MgO₄(OH)₂] octahedra (a 2 : 1 dioctahedral layer). The layers had a negative electric charge, mainly due to the isomorphic substitutions in the octahedral sheets that were neutralised by cations intercalated between an array of face-to-face stacked layers (Fig. 3a). The results of simulating the Na-MMT structure in the anhydrous and hydrated states are presented in Fig. 3b and c.

Fig. 3 (a) Schematic of a smectite clay; (b) anhydrous Na-MMT; (c) hydrated Na-MMT. Atom/geometry (colour): oxygen (red), hydrogen (white), silicon tetrahedra (blue), magnesium octahedra (orange), aluminium octahedra (silver), and sodium (purple).

The calculated lattice parameters for the dehydrated Na-MMT were a = 5.05 Å, b = 8.58 Å, c = 10.41 Å, and β = 97.80°, showing deviations below 5% compared to the unit cell parameters obtained for a dehydrated Na-MMT using the exchange–correlation PBE-D2 and similar plane-wave cutoff:³¹a = 5.20 Å; b = 9.20 Å, c = 10.13 Å, and β = 98.99°. The structure refinement of a dioctahedral smectite with the chemical composition Na_0.46[(Al_1.68Mg_0.32)(Si_3.86Al_0.14)O₁₀(OH)₂], i.e. close to that one simulated in this work, gave the following results: a = 5.18 Å, b = 8.98 Å, c = 10.10 Å and β = 99.5°.³⁰ The bond lengths are presented in Table S1 in the ESI.†

The simulation of hydrated Na-MMT was achieved by adding four water molecules around the intercalated cation. The calculated lattice parameters for the hydrated Na-MMT were a = 5.24 Å, b = 8.87 Å, c = 12.73 Å, and β = 95.11°. Significant differences included an increase in the lattice parameters b and c, an angular deformation (β) adjustment of less than 5°, and an approximately 2.3 Å increase in interplanar distance.

p-TMPyP model

A systematic theoretical investigation was performed to establish a model describing the structure of p-TMPyP and its dication form (H₂TMPyP²⁺).¹⁹ The best result was obtained considering the +4 charge of p-TMPyP (without the neutralising chloride ions). The calculated average value of the methyl-pyridyl dihedral angles, i.e. the angle between the porphyrin core ring and the phenyl substituent, was 64.04°, while the mean C_α–C_meso–C_α–C_β dihedral angles were 175.23° (a value of 180° corresponds to a planar geometry). The pyrrole N/N–H moieties were practically screened within the core in such a conformation. These results, obtained with the Quantum ESPRESSO, were in good agreement with previous work¹⁹ that employed the Gaussian 09 software.

p-TMPyP on the MMT surface model

A model comprising p-TMPyP on the MMT surface was created using the optimised anhydrous Na-MMT structure described above, removing sodium ions and increasing the space between the layers, i.e. the crystallographic c-direction. Due to the dimensions of the p-TMPyP being about 220 Ų, a 4 × 2 × 1 supercell was constructed with a basal space to include the p-TMPyP, as represented in Fig. 4a posteriorly. The system had a total of 426 atoms (90 atoms from p-TMPyP) and the crystallographic parameters a = 21.03 Å, b = 17.53 Å, c = 47.49 Å, and β = 95.34°.

Fig. 4 (a) Representation of the 4 × 2 × 1 supercell, unit cell and initial position z₀ for scanning the surface potential in the crystallographic c-direction; (b) energy variation of the p-TMPyP-MMT system scanning in the z- and y-axes. Inset: energy variation around z = 2 and y = 1.6. Atom/geometry (colour): oxygen (red), hydrogen (white), silicon tetrahedra (blue), magnesium octahedra (orange), and aluminium octahedra (silver).

The initial step involved a single-point calculation scan to model the adsorption of p-TMPyP on the MMT layer. This entailed placing the molecule on the surface and seeking a local energy minimum before proceeding to the geometry optimisation step. Subsequently, the optimised geometry of p-TMPyP in a vacuum was utilised for the surface scan. Fig. 4a shows the initial configuration for the surface scanning, where p-TMPyP was situated 3.8 Å from the surface, and a 0.1 Å variation was applied along the z-axis. In total, 18 single-point calculations were conducted. The scan in the y-axes started from the point of the minimum energy obtained in the z position, i.e. 2 Å. Due to the large dimensions and symmetry of p-TMPyP, variations in the x-axis were treated equivalently to those in the y-axis. The in-plane scan was accomplished with a translation of 0.05 Å, performing 30 simulations. Fig. 4b shows the overlap of the scans in the z- and y-axes and the region of simultaneous minimum energy for both directions after creating a grid of the potential energy surface with 48 different positions. Subsequent calculations for the geometry optimisation of p-TMPyP on the MMT surface were performed using the GGA-PBE functional.

The conformations of p-TMPyP calculated in vacuum conditions and after atomic relaxation under interaction with the (001) MMT surface (basal siloxane plane) are shown in Fig. 5a, b and c, d, respectively. The atomic bond lengths of both simulated structures were found to be similar (see Table S2†), but the dihedral angles presented a significant change (Table S3†). The p-TMPyP mean dihedral angles calculated in a vacuum were 64.04° and 175.23° for the methyl-pyridyl and the porphyrin core, respectively. After optimisation on the MMT surface, the following values were obtained: 44.12° for the methyl-pyridyl twist and 165.46° for the macrocycle core distortion. Hence, the new p-TMPyP conformation on the anhydrous MMT (001) surface included a rotation of the methyl pyridine in the periphery of the ring and a twist of the pyrrole core (wave-type distortion, Fig. 2). The calculation assumed interactions between p-TMPyP and the MMT surface comprising electrostatic (positive guest and negative host) and van der Waals interactions. The system exhibiting p-TMPyP on the MMT surface simulated in this work could be related to the situation of p-TMPyP at low concentration in a flat orientation on the external surface of this clay mineral.

Fig. 5 Conformations of p-TMPyP calculated in (a, b) vacuum conditions, showing the methyl-pyridyl dihedral angles (in red); (c, d) after atomic relaxation under interaction with the (001) MMT surface, showing the methyl-pyridyl dihedral angles (in red). Atom (colour): C (grey), N (blue), and H (white).

p-TMPyP on the MMT surface: modelling the UV–Vis spectrum

Calculations were performed to simulate the UV–vis spectrum of p-TMPyP on the MMT surface, as shown in Fig. S1.† A good match between the experimental (451 nm) and the calculated UV–Vis spectra (448 nm) was observed. The system presented a red-shift, from 425 nm to 448 nm, of the Soret band compared to the free p-TMPyP. Moreover, since no relevant rotation of the methyl-pyridyl substituent was noticed, it could be concluded that the Soret band red-shift was due to p-TMPyP electrostatic interactions with MMT, and slight macrocycle plane deformation, with no structural distortion that would flatten the molecule. The electronic properties of porphyrin adsorbed on MMT were mainly dictated by the dielectric constant of the solid surface rather than by the solvent. This conclusion aligns with studies about the p-TMPyP/synthetic saponite clay system in water–acetone mixtures (10%–90% v/v), which reported observing a small red-shift in the Soret band, and interpreted this as a predictable solvent effect.³⁷ On the other hand, the UV–Vis spectra of mixtures of p-TMPyP and aqueous dispersions of hectorite clays showed that increasing the layer charge density of the clay (isomorphic substitution) promoted higher shifts of the Soret and Q bands to a low energy region (from 421 to 450–455 nm).³⁸ In another work, p-TMPyP was mixed with aqueous dispersions of montmorillonite clays having different layer charge densities (1.35 to 0.49 mmol of negative charge per g).³⁹ It was observed that the Soret band shifted regularly from 421 (without clay) to approximately 450 nm as the layer charge density was increased. In the UV–Vis spectra of samples with a high layer charge, a shoulder was observed around 485 nm, which was not assigned to an adsorbed porphyrin but rather to an intercalated one.

According to Eguchi,⁴⁰ synthetic smectite clay turned green in contact with a red non-charged meso-tetraphenylporphyrin (TPP) solution. The adsorption was irreversible, and a bathochromic shift of the Soret band was also observed, indicating that the spectral profile of the porphyrin/clay was unrelated to the solvent. In that case, the interaction between the non-charged TPP and the clay surface was not as strong as that of charged p-TMPyP and the clay. Still, the Soret red-shift in water or hexane indicated that the dielectric property of the inorganic layers must be considered in the adsorption process. The influence of the dielectric property of the medium on the Soret band absorption was also observed in a previously reported DFT simulation.¹⁹

Although the experimental studies assigned the red-shift to a flattening of the meso-methyl-pyridyl group, the computational data obtained in the present work indicated that the bathochromic shift could be assigned to a structural change mainly in the aromatic macrocycle, producing non-planar species.

p-TMPyP intercalated into the MMT model

The next step involved the simulation of p-TMPyP intercalated between MMT layers (i) in an anhydrous environment and (ii) in the presence of intercalated water molecules. The unit cell in which the p-TMPyP was intercalated into MMT was designed considering the optimised p-TMPyP on MMT shown in Fig. 4a. As mentioned in the computational details, a reduced c-crystallographic parameter was chosen to simulate an intercalated anhydrous condition. The results of the optimised p-TMPyP intercalated system are shown in Fig. 6a. Then, water molecules were randomly added in to the interlayer. The system was subjected to newer geometry relaxation. Such optimisation distorted the p-TMPyP porphyrin ring, as shown in Fig. 6b. As previously noticed in this work for the adsorbed porphyrin on MMT, the atomic bond lengths of the simulated p-TMPyP structure adsorbed or intercalated into anhydrous or hydrated MMT were similar (Table S4†). Table S5† shows the dihedral angles of the methyl-pyridyl substituent concerning the porphyrin core and the C_α–C_meso–C_α–C_β dihedral angles. The average dihedral angle between the meso-substituent and the plane of the porphyrin ring decreased from 40.65° to 38.63° when intercalated water molecules were introduced in the simulation. At the same time, the core distortion changed from 159.43° to 150.68°.

Fig. 6 Simulated unit cells of p-TMPyP intercalated into MMT (a) after structural optimisation in an anhydrous environment, and (b) after structural optimisation with water molecules randomly located between the layers. Atom/geometry (colour): oxygen (red), hydrogen (white), silicon tetrahedron (blue), magnesium octahedra (orange), and aluminium octahedra (silver).

Next, X-ray diffraction patterns were simulated to analyse the optimised unit cells of intercalated p-TMPyP-MMT systems under both anhydrous and hydrated conditions (Fig. S2†). The basal spacing values (d₀₀₁) decreased from 15.1 to 14.8 Å in the presence of water molecules. These theoretical results suggest the smallest thickness of p-TMPyP in the sandwiched condition between the layers. Hydrated solid samples of p-TMPyP intercalated into SWy-2 MMT (or into LAPONITE® XLS) showed a d₀₀₁ value of about 14.3 Å after intercalation.²⁰ Hence, the developed model fitted well with the experimental results.

The relaxation of the TMPyP-MMT intercalated system following the hydration process indicated a spontaneous proton transfer from both hydrogen atoms of a water molecule in the upper part of the p-TMPyP plane to two non-protonated nitrogen atoms of the pyrrole groups (video available in the ESI†). This transfer led to a doubly protonated species, as highlighted in Fig. 7a and b. Furthermore, three protonated nitrogen atoms of the p-TMPyP core exhibited displacement in the same direction relative to the macrocycle ring. In contrast, one protonated nitrogen atom was observed to be tilted in the opposite position, displaying a C1 symmetry. The distorted conformation observed did not conform to the typical structure depicted in Fig. 2, nor did it resemble the saddle-type non-planar distortion typically associated with dicationic tetraphenylporphyrins.⁴¹ Three N–H units in the core were directed up while one unit was down from the macrocycle plane (Fig. 7b).

Fig. 7 Structure of p-TMPyP intercalated into hydrated MMT (a) before structural optimisation, showing the hydrogen bond interactions (dashed lines) between a water molecule and nitrogen atoms from the porphyrin core, and (b) after structural optimisation, showing the protonation of two N atoms in the porphyrin core; (c) isosurfaces depicting the electronic charge density difference for the MMT-TMPyP-Water-MMT system: yellow and green regions represent negative (gain of charge density) and positive (loss of charge density) isosurfaces, respectively.

Such a structural change in p-TMPyP was not observed in the optimised hydrated p-TMPyP on the MMT surface and in the non-optimised-intercalated-TMPyP-MMT system in the anhydrous condition, indicating that confinement and hydration conditions are needed to generate highly distorted p-TMPyP. Hydrated clay minerals show Brønsted acid properties because of the higher hydrolysis degree of intercalated water than bulk water molecules.⁴² The macrocycle distortion can decrease the π-conjugation, enhancing the localisation of the lone pair in the imine nitrogen atom and, consequently, increasing the intrinsic basicity of the macrocycle (i.e. p-TMPyP protonation).¹³ An increase in water acidity combined with the rise in basicity of p-TMPyP due to the porphyrin distortion could promote the formation of species with the conformation exhibited in Fig. 7b.

Following the structural optimisation of the hydrated p-TMPyP-MMT system, the simulation revealed the formation of an oxide ion instead of the hydroxide ion from water deprotonation. This observation emphasises the pronounced Brønsted basicity of the non-planar p-TMPyP species. Also, the oxide ion was bonded to the silicon atom of the MMT tetrahedral sheet after breaking the apical Si–O linkage. Despite the modification in the MMT structure after optimisation of the hydrated p-TMPyP intercalated between the layers, the distortion observed in the porphyrin structure agreed with the Raman experimental results reported by Dias et al.^16,20,21 The Raman spectra of p-TMPyP sorbed into four clays, including an MMT, were recorded with laser lines at 457.9, 488.0, and 514.5 nm. The first two lines were in resonance with the Soret band, and were mainly related to electronic transitions, respectively, of the porphyrin dication (protonated species) and the macrocycle distortion from the planarity of non-protonated species. At 457.9 nm, the spectrum of p-TMPyP showed a downshift off the band at 1550 cm⁻¹ (non-protonated species) to 1536 cm⁻¹ (protonated species) after interaction with MMT. When excited at 488 nm, the band at 1550 cm⁻¹ predominated over the one at 1536 cm⁻¹, while the bands assigned to the out-of-plane vibrations below 1000 cm⁻¹ were intensified. To explore the qualitative charge transfer in the p-TMPyP-MMT system, the electronic charge density difference (shown in Fig. 7c) was calculated considering the total charge of the complete system minus the total charge of the isolated components (p-TMPyP, MMT, and water molecules). This revealed a reorganisation of the electronic structure, indicating the interactions between the water molecules and p-TMPyP in the MMT-confined environment. To demonstrate the impact of MMT on the charge transfer (or H⁺) between water and p-TMPyP, a model was constructed excluding MMT and employing a smaller supercell to maintain the anticipated water density. Remarkably, the theoretical analysis did not exhibit the spontaneous protonation of p-TMPyP under these conditions. This result corroborated the proposition regarding the influential role of MMT's dielectric properties in enhancing water acidity within a confined setting. Consequently, MMT is postulated to facilitate the transfer of H⁺ ions between water molecules and p-TMPyP, offering a more efficient pathway for charge transfer.

meta-TMPyP intercalated into the MMT model

To investigate the steric influence of the meso-substituted methyl-pyridyl group into the porphyrin distortion, simulations of the meta-isomer of TMPyP (Fig. 1, m-TMPyP) intercalated into anhydrous and hydrated MMT were also performed. The same methodology previously described for optimisation of the geometry of the p-TMPyP-MMT system was employed. The m-TMPyP atomic bond lengths of simulated anhydrous and hydrated systems were found to be practically identical, as shown in Table S6.† However, the average dihedral angle of meta-methyl-pyridyl was twisted around 10° relative to the porphyrin core when water molecules were inserted between the layers (Table S7†), while the C_α–C_meso–C_α–C_β dihedral angles also decreased ca. 10°, suggesting distortion of the porphyrin core.

For comparison purposes, Table 1 shows the dihedral angles of the two TMPyP isomers intercalated into MMT in the absence and presence of water molecules. The twisting of the meso-substituent and the core non-planar conformation were more pronounced in the para-isomer, and such results could be attributed to the higher energy barrier for m-TMPyP inclination due to steric factors. The lower rotation of the meso-group of m-TMPyP hinders the core distortion of porphyrin as an extension to that observed for the para-isomer.

Table 1 Dihedral angles of optimised isomers of TMPyP intercalated into anhydrous and hydrated MMT

Simulated system	Dihedral angles (°) of porphyrin substituted group intercalated into MMT	Dihedral angles (°) Cα–Cmeso–Cα–Cβ of porphyrin intercalated into MMT
p-TMPyP-MMT anhydrous	40.65	159.30
p-TMPyP-MMT hydrated	38.63	151.38
m-TMPyP-MMT anhydrous	64.01	175.23
m-TMPyP-MMT hydrated	55.01	165.79

---

Fig. 8a and b illustrate the unit cells simulated m-TMPyP-MMT systems carried in the hydrated condition before and after optimisation, respectively. Considering the free m-TMPyP, the decreased dihedral angles of the methyl-pyridyl group related to the core ring when adsorbed on the clay or intercalated without water or in a hydrated environment were about 20°, 23.4°, and 25.4°, respectively. The variations of the dihedral angle C_α–C_meso–C_α–C_β of porphyrin for the three conditions were around 10.8°, 16°, and 23.9°. In terms of the percentage variation of the dihedral angles, the rotation of the methyl-pyridyl group was more pronounced in the adsorption process of m-TMPyP on the MMT external surface. At the same time, the distortion from planarity was more susceptible in the intercalation process. Unlike the previously shown intercalated para-isomer system, no nitrogen core protonation occurred after geometry optimisation of the meta-isomer in the hydrated MMT. Fig. 8c highlights the m-TMPyP conformation in the clay-confined site and shows the hydrogen bonds between the water molecules and non-protonated nitrogen atoms in the core. Hence, the position of the methyl-pyridyl group affected the distortion degree of the macrocycle ring and the basicity of the porphyrin, as shown in Fig. 9. The porphyrin protonation depends on an intrinsic proton affinity and a structural term related to the macrocycle distortion.¹³ Considering the similarity of the substituent groups in the two porphyrins, the basic character can be distinct due to the different degrees of distortion from planarity.

Fig. 8 Unit cells of m-TMPyP intercalated into MMT (a) before structural optimisation in a hydrous environment, and (b) after structural optimisation with water molecules randomly located between the layers; (c) optimised structure of m-TMPyP intercalated into hydrated MMT, showing the hydrogen bond interactions (dashed lines) with water molecules.

Fig. 9 Conformations of m-TMPyP calculated in (a, b) vacuum conditions and (c, d) optimised structure when intercalated into hydrated MMT. Atom (colour): C (grey), N (blue), and H (white).

A work about films of ortho-, meta-, or para-isomers of TMPyP with a synthetic clay, prepared by the Langmuir–Blodgett method, revealed that the Soret band shifts (Δλ) in relation to the free TMPyP were 6, 11, and 28 nm, respectively.⁴³ This corroborates the tendency of the decrease in the energy barrier to twist the meso-substituted group in the periphery of the macrocycle plane. The calculations provided in this study align with those results. Additionally, the minor red-shifts observed for the ortho- and meta-isomers indicated that the core distortion was less pronounced than that of the para-isomer, making protonation improbable.

Fig. 10 summarises the structural data of TMPyP in all conditions simulated in this work, including the results from the two isomers. Revisiting the results from the Raman spectroscopy analyses of the solid-state samples of p-TMPyP in terms of its interaction with different clays,¹⁹ the relationship between the porphyrin localisation on the host and the UV–Vis spectral profile could be indicated. MMT has guest cations on the external surface (absorption at about 458 nm) and when intercalated in a hydrous environment (absorption at 490 nm). p-TMPyP on LAPONITE® and vermiculite was located on the external surfaces, which is plausible because the first clay tends to be exfoliated, and the second one is not as expansible as MMT. The larger amount of non-planar macrocycles on vermiculite compared to the other clays (smectites), as indicated by Raman spectroscopy, was due to its higher layer charge density.

Fig. 10 Modifications in the TMPyP conformation when interacting with MMT external or internal (intercalation) surfaces, considering anhydrous (grey shade, left) or hydrous (blue shade, right) environments.

Summing up, the shape-responsive host–guest process can modulate the properties of porphyrins, such as their basicity, chemical stability, redox potential, and nonlinear optical response.⁴⁴ This opens the opportunity to obtain non-planar porphyrins by interaction with inorganic external or internal surfaces without needing to synthesise new molecules.

Conclusions

Employing various levels of approximation, theoretical calculations in the framework of DFT indicated distinct conformations of p-TMPyP porphyrin adsorbed on the external basal surfaces or between the layers of the MMT host structure. Minor changes were noticed in bond lengths while the angles varied significantly, yielding non-planar species due to the pyrrole rings tilting relative to the mean plane of the porphyrin. Modelling of the absorption electronic spectrum of p-TMPyP on the clay surface showed a red-shift of the Soret band when the dielectric constant of MMT was considered. The structural optimisation of intercalated p-TMPyP in a hydrous confined environment indicated a higher level of distortion of the macrocycle core compared to in the non-intercalated species. Such distortion exposed the nitrogen lone pairs and promoted the transfer of H⁺ ions from water to p-TMPyP, indicating the increase in basicity of the porphyrin. Hence, protonation was facilitated by the core distortion conformation. On the other hand, the nonplanarity in intercalated meso-substituted porphyrins is related to the nature of the substituents; for instance, m-TMPyP was not expressively distorted when intercalated. Such a result is associated with the steric constraint of N-methyl-3-pyridyl phenyl rotation. The shape-responsive behaviour of the guest porphyrins in interaction with the clay host evidences new paradigms regarding the charge transfer and chemical reactivity that may be explored to research and develop potential new nanodevices.

Author contributions

Investigation and methodology: E. D. S., F. C. D. A. L. and A. V. G. R.; writing (original draft): V. R. L. C. and F. C. D. A. L.; Revision and editing: E. D. S., F. C. D. A. L., A. V. G. R., V. R. L. C. and H. M. P.; conceptualization: V. R. L. C. and H. M. P.; project administration, supervision and funding acquisition: H. M. P.

Data availability

The datasets generated during the current study are not publicly available but can be obtained from the authors upon reasonable request (hmpetril@if.usp.br, vrlconst@iq.usp.br).

Conflicts of interest

There is no conflict of interest to declare.

Acknowledgements

E. D. S. acknowledges the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for the PhD scholarship. The authors are grateful for the computational resources from high-performance computing (HPC-USP) and the National High-Performance Processing Center in São Paulo (CENAPAD/SP). H. M. P., V. R. L. C. and F. C. D. A. L. acknowledge the financial support from Sao Paulo Foundation (FAPESP, INCT-INEO 2014/50869-6). Additionally, H. M. P. and V. R. L. C. are thankful to CNPq (308438/2022-1 and 314034/2021-8) and F. C. D. A. L. to FAPESP (2023/17506-6) for research grants.

References

1. W. Auwärter, D. Écija, F. Klappenberger and J. V. Barth, Nat. Chem., 2015, 7, 105–120.
2. A. Rezaeifard and M. Jafarpour, Catal. Sci. Technol., 2014, 4, 1960.
3. J. Qiao, Y. Liu, F. Hong and J. Zhang, Chem. Soc. Rev., 2014, 43, 631–675.
4. E. Nikoloudakis, I. López-Duarte, G. Charalambidis, K. Ladomenou, M. Ince and A. G. Coutsolelos, Chem. Soc. Rev., 2022, 51, 6965–7045.
5. S. Jiang, R. Cheng, X. Wang, T. Xue, Y. Liu, A. Nel, Y. Huang and X. Duan, Nat. Commun., 2013, 4, 2225.
6. R. E. Andernach, S. Rossbauer, R. S. Ashraf, H. Faber, T. D. Anthopoulos, I. McCulloch, M. Heeney and H. A. Bronstein, ChemPhysChem, 2015, 16, 1223–1230.
7. L. Ji, S. Devaramani, X. Mao, Z. Zhang, D. Qin, D. Shan, S. Zhang and X. Lu, J. Phys. Chem. B, 2017, 121, 9045–9051.
8. W. Kuch and M. Bernien, J. Phys.: Condens. Matter, 2017, 29, 023001.
9. C. J. Kingsbury and M. O. Senge, Coord. Chem. Rev., 2021, 431, 213760.
10. M. Kielmann and M. O. Senge, Angew. Chem., Int. Ed., 2019, 58, 418–441.
11. M. Roucan, M. Kielmann, S. J. Connon, S. S. R. Bernhard and M. O. Senge, Chem. Commun., 2018, 54, 26–29.
12. M. Eguchi, A. S. Nugraha, A. E. Rowan, J. Shapter and Y. Yamauchi, Adv. Sci., 2021, 8, 2100539.
13. O. S. Finikova, A. V. Cheprakov, P. J. Carroll, S. Dalosto and S. A. Vinogradov, Inorg. Chem., 2002, 41, 6944–6946.
14. K. A. Carrado and R. E. Winans, Chem. Mater., 1990, 2, 328–335.
15. V. G. Kuykendall and J. K. Thomas, Langmuir, 1990, 6, 1350–1356.
16. P. M. Dias, D. L. A. De Faria and V. R. L. Constantino, J. Inclusion Phenom. Macrocyclic Chem., 2000, 38, 251–266.
17. S. Banik, J. Bhattacharjee, S. A. Hussain and D. Bhattacharjee, J. Phys. Chem. Solids, 2015, 87, 128–135.
18. Z. Chernia and D. Gill, Langmuir, 1999, 15, 1625–1633.
19. E. D. Suarez, F. C. D. A. Lima, P. M. Dias, V. R. L. Constantino and H. M. Petrilli, J. Mol. Model, 2019, 25, 264.
20. P. M. Dias, D. L. A. de Faria and V. R. Leopoldo Constantino, Clays Clay Miner., 2005, 53, 361–371.
21. V. R. L. Constantino, C. A. S. Barbosa, M. A. Bizeto and P. M. Dias, An. Acad. Bras. Cienc., 2000, 72, 45–50.
22. W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133–A1138.
23. P. Giannozzi, O. Andreussi, T. Brumme, O. Bunau, M. Buongiorno Nardelli, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, M. Cococcioni, N. Colonna, I. Carnimeo, A. Dal Corso, S. de Gironcoli, P. Delugas, R. A. DiStasio, A. Ferretti, A. Floris, G. Fratesi, G. Fugallo, R. Gebauer, U. Gerstmann, F. Giustino, T. Gorni, J. Jia, M. Kawamura, H.-Y. Ko, A. Kokalj, E. Küçükbenli, M. Lazzeri, M. Marsili, N. Marzari, F. Mauri, N. L. Nguyen, H.-V. Nguyen, A. Otero-de-la-Roza, L. Paulatto, S. Poncé, D. Rocca, R. Sabatini, B. Santra, M. Schlipf, A. P. Seitsonen, A. Smogunov, I. Timrov, T. Thonhauser, P. Umari, N. Vast, X. Wu and S. Baroni, J. Phys.: Condens. Matter, 2017, 29, 465901.
24. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari and R. M. Wentzcovitch, J. Phys.: Condens. Matter, 2009, 21, 395502.
25. D. Vanderbilt, Phys. Rev. B: Condens. Matter Mater. Phys., 1990, 41, 7892–7895.
26. A. Dal Corso, Comput. Mater. Sci., 2014, 95, 337–350.
27. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
28. S. Grimme, A. Hansen, J. G. Brandenburg and C. Bannwarth, Chem. Rev., 2016, 116, 5105–5154.
29. H. J. Monkhorst and J. D. Pack, Phys. Rev. B, 1976, 13, 5188–5192.
30. S. I. Tsipursky and V. A. Drits, Clay Miner., 1984, 19, 177–193.
31. S. Pirillo, C. R. Luna, I. López-Corral, A. Juan and M. J. Avena, J. Phys. Chem. C, 2015, 119, 16082–16088.
32. W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14, 33–38.
33. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, 2009, preprint, https://www.gaussian.com.
34. T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51–57.
35. J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005, 105, 2999–3094.
36. P. Umari and A. Pasquarello, Phys. Rev. Lett., 2002, 89, 157602.
37. M. Eguchi, Y. Watanabe, Y. Ohtani, T. Shimada and S. Takagi, Tetrahedron Lett., 2014, 55, 2662–2666.
38. A. Čeklovský, A. Czímerová, K. Lang and J. Bujdák, J. Lumin., 2009, 129, 912–918.
39. A. Čeklovský, P. Boháč and A. Czímerová, Appl. Clay Sci., 2016, 126, 68–71.
40. M. Eguchi, Bull. Chem. Soc. Jpn., 2020, 93, 898–903.
41. C. J. Kingsbury, M. Kielmann, B. Twamley and M. O. Senge, Molecules, 2022, 27, 7060.
42. S. Yariv, Int. Rev. Phys. Chem., 1992, 11, 345–375.
43. Y. Hirade, T. Ishida, T. Shimada and S. Takagi, Colloids Surf., A, 2021, 624, 126747.
44. M. O. Senge, ECS Trans., 2015, 66, 1–10.

---

Footnote

† Electronic supplementary information (ESI) available: Bond lengths between atoms in the simulated hydrated Na-MMT structure. Optimised bond lengths (Å) of p-TMPyP in vacuum and adsorbed on the (001) MMT surface. Optimised dihedral angles (°) of p-TMPyP in vacuum and adsorbed on the (001) MMT surface. Simulated UV-vis electronic spectra of p-TMPyP under different conditions. Optimised bond lengths (Å) of p-TMPyP on MMT, p-TMPyP intercalated into anhydrous MMT, and p-TMPyP intercalated into hydrated MMT. Optimised dihedral angles (°) of p-TMPyP on MMT, p-TMPyP intercalated into anhydrous MMT, and p-TMPyP intercalated into hydrated MMT. Simulated X-ray diffraction pattern of p-TMPyP intercalated into MMT. Optimised bond lengths (Å) of m-TMPyP intercalated into anhydrous MMT, and m-TMPyP intercalated into hydrated MMT. Optimised dihedral angles (°) of m-TMPyP intercalated into anhydrous MMT and m-TMPyP intercalated into hydrated MMT. See DOI: https://doi.org/10.1039/d4dt03437f

---