Kinetic Investigation on Tetrakis(4-Sulfonatophenyl)Porphyrin J-Aggregates Formation Catalyzed by Cationic Metallo-Porphyrins

Under mild acidic conditions, various metal derivatives of tetrakis(4-N-methylpyridinium)porphyrin (gold(III), AuT4; cobalt(III), CoT4; manganese(III), MnT4 and zinc(II), ZnT4) catalytically promote the supramolecular assembling process of the diacid 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin (H2TPPS4) into J-aggregates. The aggregation kinetics have been treated according to a well-established model that involves the initial formation of a critical nucleus containing m porphyrin units, followed by autocatalytic growth, in which the rate evolves as a power of time. An analysis of the extinction time traces allows to obtain the rate constants for the auto-catalyzed pathway, kc, and the number of porphyrins involved in the initial seeding. The aggregation kinetics have been investigated at fixed H2TPPS4 concentration as a function of the added metal derivatives MT4. The derived rate constants, kc, obey a rate law that is first order in [MT4] and depend on the specific nature of the catalyst in the order AuT4 > CoT4 > MnT4 > ZnT4. Both resonance light scattering (RLS) intensity and extinction in the aggregated samples increase on increasing [MT4]. With the exception of AuT4, the final aggregated samples obtained at the highest catalyst concentration exhibit a negative Cotton effect in the J-band region, evidencing the occurrence of spontaneous symmetry breaking. The role of the nature of the metal derivative in terms of overall charge and presence of axial groups will be discussed.


Introduction
Spontaneous self-assembling of molecular units into ordered and larger stable arrangements through non-covalent interactions is a topic that in the last few years has received significant attention from the research community due to its large impact on many aspects of chemistry [1,2]. Porphyrins certainly represent a class of compounds that are extremely interesting in this regard because, depending on both their electronic and steric characteristics, they are susceptible to aggregation forming dimers and/or larger molecular oligomers [3][4][5][6]. In this respect, water-soluble porphyrins are excellent candidates to explore the aggregation process, which can be suitably modulated by varying the medium properties, such as pH and ionic strength, or by taking advantage of the coordination chemistry properties of the central macrocycle. The latter may be indeed readily metallated with a wide variety of metal ions, thereby giving access to peculiar photophysical and photochemical properties [7][8][9][10]. However, despite many reports on the formation of porphyrin homo-aggregates, there are relatively few literature investigations on hetero-aggregated systems. Self-assembly between cationic and anionic meso-substituted porphyrins, mainly driven by electrostatic attraction of peripheral substituent groups, hydrophobic interactions and axial coordination, represents an easy way to design hetero-aggregated supramolecular species [11][12][13][14][15][16][17], even if the formation of discrete hetero-dimers has been reported only in water/organic solvents mixtures [18,19]. In this context, the targeted use of anionic porphyrins such as 5,10,15,20-tetrakis(4-sulfonatophenyl) porphyrin TPPS 4 , whose diacid form is able to give both J-(edge-to-edge) and H-type (face-to-face) self-aggregates, is particularly attractive since these structures exhibit peculiar emergent photophysical properties tunable through aggregate topology and molecular arrangement [20][21][22][23][24][25][26][27][28][29][30]. In the literature, some examples concerning the use of TPPS 4 in the formation of hetero-aggregated systems have been reported [31][32][33][34][35][36][37] and, among them, some focus on the possibilities of using TPPS 4 in applicative fields. Shelnutt et al. [35] described the formation of nanotubular J-aggregates whose structure retains own individual porphyrin properties as photocatalytic reduction of metals, while Oztek et al. reported on the potential use of hetero-aggregates as hydrogen storage materials [36].
Other relevant examples in the field involve the additional use of suitable chiral templating units aimed to obtain aggregates exhibiting a well-defined stoichiometric ratio that are able to memorize and amplify the chiral information [33,34,[38][39][40][41]. Despite the many reports on thermodynamic and spectroscopic characterization of these supramolecular species, kinetic studies are rather scarce due to the difficulty in controlling the various parameters that influence the aggregation pathways and eventually the final structures at the nano-or mesoscale [42,43]. Notwithstanding, the knowledge of the kinetic behavior is a prerequisite to understanding the factors controlling the rate determining step and the early stages in these complex phenomena. On the basis of our previous spectroscopic investigations on the formation of hetero-oligomers and the subsequent J-aggregates induction [31], here, we describe a detailed kinetic study on the role of cationic metallo-porphyrins MT 4 in the supramolecular assembling process leading to J-aggregates of TPPS 4 (Scheme 1). Both MT 4 and TPPS 4 porphyrins do not show any tendency towards self-aggregation in the adopted experimental conditions and we anticipate that: (i) the addition of MT 4 to solutions of TPPS 4 easily promotes J-aggregation, even under mild acidic conditions; (ii) the kinetic rates and the extent of aggregation depend linearly on the cationic porphyrin concentration and on the overall characteristics of these metal derivatives (charge and axial substituent groups), following the order AuT 4 > CoT 4 > MnT 4 > ZnT 4 ; (iii) in most cases, the number of porphyrins in the critical nucleus is slightly less than 3-4, as previously reported for the uncatalyzed process; and iv) except for the most effective metal derivative (AuT 4 ), small CD spectra are always detectable, thus adding to the growing number of examples of spontaneous symmetry breaking in J-aggregates [44][45][46][47][48].
Molecules 2020, 25, x FOR PEER REVIEW 2 of 12 varying the medium properties, such as pH and ionic strength, or by taking advantage of the coordination chemistry properties of the central macrocycle. The latter may be indeed readily metallated with a wide variety of metal ions, thereby giving access to peculiar photophysical and photochemical properties [7][8][9][10]. However, despite many reports on the formation of porphyrin homo-aggregates, there are relatively few literature investigations on hetero-aggregated systems. Self-assembly between cationic and anionic meso-substituted porphyrins, mainly driven by electrostatic attraction of peripheral substituent groups, hydrophobic interactions and axial coordination, represents an easy way to design hetero-aggregated supramolecular species [11][12][13][14][15][16][17], even if the formation of discrete hetero-dimers has been reported only in water/organic solvents mixtures [18,19]. In this context, the targeted use of anionic porphyrins such as 5,10,15,20-tetrakis(4sulfonatophenyl) porphyrin TPPS4, whose diacid form is able to give both J-(edge-to-edge) and Htype (face-to-face) self-aggregates, is particularly attractive since these structures exhibit peculiar emergent photophysical properties tunable through aggregate topology and molecular arrangement [20][21][22][23][24][25][26][27][28][29][30]. In the literature, some examples concerning the use of TPPS4 in the formation of heteroaggregated systems have been reported [31][32][33][34][35][36][37] and, among them, some focus on the possibilities of using TPPS4 in applicative fields. Shelnutt et al. [35] described the formation of nanotubular Jaggregates whose structure retains own individual porphyrin properties as photocatalytic reduction of metals, while Oztek et al. reported on the potential use of hetero-aggregates as hydrogen storage materials [36].
Other relevant examples in the field involve the additional use of suitable chiral templating units aimed to obtain aggregates exhibiting a well-defined stoichiometric ratio that are able to memorize and amplify the chiral information [33,34,[38][39][40][41]. Despite the many reports on thermodynamic and spectroscopic characterization of these supramolecular species, kinetic studies are rather scarce due to the difficulty in controlling the various parameters that influence the aggregation pathways and eventually the final structures at the nano-or mesoscale [42,43]. Notwithstanding, the knowledge of the kinetic behavior is a prerequisite to understanding the factors controlling the rate determining step and the early stages in these complex phenomena. On the basis of our previous spectroscopic investigations on the formation of hetero-oligomers and the subsequent J-aggregates induction [31], here, we describe a detailed kinetic study on the role of cationic metallo-porphyrins MT4 in the supramolecular assembling process leading to J-aggregates of TPPS4 (Scheme 1). Both MT4 and TPPS4 porphyrins do not show any tendency towards self-aggregation in the adopted experimental conditions and we anticipate that: i) the addition of MT4 to solutions of TPPS4 easily promotes Jaggregation, even under mild acidic conditions; ii) the kinetic rates and the extent of aggregation depend linearly on the cationic porphyrin concentration and on the overall characteristics of these metal derivatives (charge and axial substituent groups), following the order AuT4 > CoT4 > MnT4 > ZnT4; iii) in most cases, the number of porphyrins in the critical nucleus is slightly less than 3-4, as previously reported for the uncatalyzed process; and iv) except for the most effective metal derivative (AuT4), small CD spectra are always detectable, thus adding to the growing number of examples of spontaneous symmetry breaking in J-aggregates [44][45][46][47][48]. Scheme 1. Pictorial sketch of TPPS4 protonation and subsequent J-aggregation in the presence of cationic metallo-porphyrins MT4.

Results and Discussion
Scheme 1. Pictorial sketch of TPPS 4 protonation and subsequent J-aggregation in the presence of cationic metallo-porphyrins MT 4 .

Results and Discussion
At neutral or alkaline pH, aqueous solutions of the parent free-base porphyrin at micro-molar concentration stabilize toward self-aggregation. The nitrogen atoms at the core of the tetra-anionic porphyrin TPPS 4 are protonated (pKa = 4.9 [49]) and this species is fully converted to its diacid dianionic form H 2 TPPS 4 at pH 2.8 (B-band at 434 nm, see black line in Figure 1). In the presence of only HCl at this pH, the formation of J-aggregates of H 2 TPPS 4 is kinetically very slow and a tiny concentration can be detected after at least 24 h. In order to investigate the catalytic role of the various metal derivatives of the cationic tetrakis(4-N-methylpyridinium)porphyrin (MT 4 ), these latter species have been pre-incubated at different molar ratios R with TPPS 4 in water (R = [TPPS 4 ]/[MT 4 ], with R spanning in the range 2 ÷ 30) and then aggregation has been triggered by lowering the pH to 2.8 through addition of HCl. All the investigated MT 4 derivatives form hetero-aggregates with the anionic TPPS 4 and their spectroscopic features and stoichiometry have been reported in literature [31]. At neutral or alkaline pH, aqueous solutions of the parent free-base porphyrin at micro-molar concentration stabilize toward self-aggregation. The nitrogen atoms at the core of the tetra-anionic porphyrin TPPS4 are protonated (pKa = 4.9 [49]) and this species is fully converted to its diacid dianionic form H2TPPS4 at pH 2.8 (B-band at 434 nm, see black line in Figure 1). In the presence of only HCl at this pH, the formation of J-aggregates of H2TPPS4 is kinetically very slow and a tiny concentration can be detected after at least 24 h. In order to investigate the catalytic role of the various metal derivatives of the cationic tetrakis(4-N-methylpyridinium)porphyrin (MT4), these latter species have been pre-incubated at different molar ratios R with TPPS4 in water (R = [TPPS4]/[MT4], with R spanning in the range 2 ÷ 30) and then aggregation has been triggered by lowering the pH to 2.8 through addition of HCl. All the investigated MT4 derivatives form hetero-aggregates with the anionic TPPS4 and their spectroscopic features and stoichiometry have been reported in literature [31]. Figure 1 displays the typical time evolution of the UV/Vis spectra observed for both CoT4 (Figure 1, left) and AuT4 (Figure 1, right) at the highest ratio R = 2. The gold(III) metal complex interacting with H2TPPS4 exhibits an initial B-band at 405 nm that decreases in intensity and undergoes a further bathochromic shift to 411 nm, matched by the decrease of the B-and Q-bands of the diacid H2TPPS4 at 434 and 644 nm, respectively, and the concomitant increase of the J-aggregate B-band at 491 nm, together with its Q-band at 704 nm. MnT4 display a B-band at 462 nm and it undergoes a bathochromic shift of 6 nm upon aggregation (see Supporting Information, Figure S7). In the case of CoT4, the B-band of this species (434 nm) is under the envelope of the diacid H2TPPS4 and consequently, not detectable. ZnT4 exhibits a B-band at 436 nm and follows the same pattern of CoT4. Upon addition of HCl, a decrease of the B-band of the diacid species also occurs, while the J-band at 490 nm increases in intensity. The kinetics of growth of J-aggregates can be easily followed by monitoring the extinction of the samples at 490 nm. Figure 2 shows a typical kinetic profile for the aggregation induced by CoT4. As already reported for acid-induced aggregation of H2TPPS4, a sigmoidal behavior is clearly observed with a quite short initial incubation or lag-time followed by an exponential growth. These data have been treated by a well-established autocatalytic model, in which two independent pathways are operative: (i) an uncatalyzed one, controlled by a rate k0 and (ii) a catalyzed one, dominated by a stretched exponential form with kc as the rate constant and n as the time exponent. The rate determining step of the second pathway is the initial formation of a nucleus containing m monomeric porphyrin units [50,51]. All the kinetic parameters have been obtained by a non-linear best-fitting procedure that was applied to the experimental extinction data and collected in Table S1 (see The kinetics of growth of J-aggregates can be easily followed by monitoring the extinction of the samples at 490 nm. Figure 2 shows a typical kinetic profile for the aggregation induced by CoT 4 . As already reported for acid-induced aggregation of H 2 TPPS 4 , a sigmoidal behavior is clearly observed with a quite short initial incubation or lag-time followed by an exponential growth. These data have been treated by a well-established autocatalytic model, in which two independent pathways are operative: (i) an uncatalyzed one, controlled by a rate k 0 and (ii) a catalyzed one, dominated by a stretched exponential form with k c as the rate constant and n as the time exponent. The rate determining step of the second pathway is the initial formation of a nucleus containing m monomeric porphyrin units [50,51]. All the kinetic parameters have been obtained by a non-linear best-fitting procedure that Molecules 2020, 25, 5742 4 of 12 was applied to the experimental extinction data and collected in Table S1 (see Supporting Information), with the exception of k 0 , which makes no or only a minor contribution to the overall fit of the data.
Molecules 2020, 25, x FOR PEER REVIEW 4 of 12 Supporting Information), with the exception of k0, which makes no or only a minor contribution to the overall fit of the data. As an example, Figure 3 (left) displays the typical linear dependence of the values of the rate constant kc as a function of the increasing concentration of [CoT4]. The related values for the time exponent n range between 3 and 4, while the size of the initial nucleus m ≈ 2-3 (dimer or trimer) is slightly lower with respect to what is reported in the literature for acid-induced aggregation of the parent porphyrin. In these latter cases, m values suggested the involvement of a trimer or a tetramer in the rate determining step [47,[52][53][54]. This difference could be ascribed to the different nature of the starting building block for the aggregation process, where a heterodimer stabilized by electrostatic interactions could be responsible for the seeding of the growth. As already reported in the literature [31], the cationic porphyrins investigated here form such kinds of supramolecular species with variable stoichiometry. This observation, together with a certain degree of variation on m values measured for ZnT4 and AuT4 (see Table S1), suggests the potential formation of concentration dependent hetero-oligomers as initiators for the catalytic growth of the H2TPPS4 J-aggregates.  is slightly lower with respect to what is reported in the literature for acid-induced aggregation of the parent porphyrin. In these latter cases, m values suggested the involvement of a trimer or a tetramer in the rate determining step [47,[52][53][54]. This difference could be ascribed to the different nature of the starting building block for the aggregation process, where a heterodimer stabilized by electrostatic interactions could be responsible for the seeding of the growth. As already reported in the literature [31], the cationic porphyrins investigated here form such kinds of supramolecular species with variable stoichiometry. This observation, together with a certain degree of variation on m values measured for ZnT 4 and AuT 4 (see Table S1), suggests the potential formation of concentration dependent hetero-oligomers as initiators for the catalytic growth of the H 2 TPPS 4 J-aggregates. Supporting Information), with the exception of k0, which makes no or only a minor contribution to the overall fit of the data. As an example, Figure 3 (left) displays the typical linear dependence of the values of the rate constant kc as a function of the increasing concentration of [CoT4]. The related values for the time exponent n range between 3 and 4, while the size of the initial nucleus m ≈ 2-3 (dimer or trimer) is slightly lower with respect to what is reported in the literature for acid-induced aggregation of the parent porphyrin. In these latter cases, m values suggested the involvement of a trimer or a tetramer in the rate determining step [47,[52][53][54]. This difference could be ascribed to the different nature of the starting building block for the aggregation process, where a heterodimer stabilized by electrostatic interactions could be responsible for the seeding of the growth. As already reported in the literature [31], the cationic porphyrins investigated here form such kinds of supramolecular species with variable stoichiometry. This observation, together with a certain degree of variation on m values measured for ZnT4 and AuT4 (see Table S1), suggests the potential formation of concentration dependent hetero-oligomers as initiators for the catalytic growth of the H2TPPS4 J-aggregates.  Resonance light scattering from aggregated samples confirms the presence of quite large arrays of electronically coupled porphyrin units (N > 25) [55]. The intensity of a resonance light scattering (RLS) peak depends on the scattering and the absorption cross-sections of the aggregates and therefore on their size and concentration [56]. Figure 4 shows the RLS spectra of aggregated samples for the case of the CoT 4 catalyzed process. The intensities of the peaks centered at about 490 nm increase on increasing [CoT 4 ].
Molecules 2020, 25, x FOR PEER REVIEW 5 of 12 Resonance light scattering from aggregated samples confirms the presence of quite large arrays of electronically coupled porphyrin units (N > 25) [55]. The intensity of a resonance light scattering (RLS) peak depends on the scattering and the absorption cross-sections of the aggregates and therefore on their size and concentration [56]. Figure 4 shows the RLS spectra of aggregated samples for the case of the CoT4 catalyzed process. The intensities of the peaks centered at about 490 nm increase on increasing [CoT4]. Analogously, both the intensity of RLS corrected for the extinction of the solutions and the extinction increase linearly on increasing [CoT4] (Figure 5), in line with an increment in size and concentration of the formed aggregates. Since the enhancement of the scattered light at resonance with the absorption feature reflects the strong electronic coupling among the porphyrins and increases with their numbers in the aggregate [55], this evidence points to a progressive stabilization of the nano-assemblies mediated by the cationic porphyrins.  Resonance light scattering from aggregated samples confirms the presence of quite large arrays of electronically coupled porphyrin units (N > 25) [55]. The intensity of a resonance light scattering (RLS) peak depends on the scattering and the absorption cross-sections of the aggregates and therefore on their size and concentration [56]. Figure 4 shows the RLS spectra of aggregated samples for the case of the CoT4 catalyzed process. The intensities of the peaks centered at about 490 nm increase on increasing [CoT4]. Analogously, both the intensity of RLS corrected for the extinction of the solutions and the extinction increase linearly on increasing [CoT4] (Figure 5), in line with an increment in size and concentration of the formed aggregates. Since the enhancement of the scattered light at resonance with the absorption feature reflects the strong electronic coupling among the porphyrins and increases with their numbers in the aggregate [55], this evidence points to a progressive stabilization of the nano-assemblies mediated by the cationic porphyrins.  Interestingly, the intensity of RLS increases on increasing the aggregation rate constant k c ( Figure 6). The data have been analyzed through a linear best fitting, obtaining I RLS corr = (−912 ± 905) + 10 5 × (7.53 ± 0.56) × k c . On one hand, this behavior is in contrast with that observed for the acid-induced aggregation of TPPS 4 , where an inverse dependence has been reported for the growth of porphyrin nanotubes [47]. On the other hand, when these nanoassemblies are formed in the presence of a high concentration of Zn 2+ , again a linear dependence between these two parameters has been reported [52]. These findings suggest that the structural arrangement of the chromophores in these specific supramolecular species are quite different with respect to that involved in nanotubes. Further evidence for a difference in the aggregate architecture stems from the slight broadening of the J-band for all the final aggregated samples. These spectral features appear generally larger than those reported for systems where the Frenkel exciton model for the electronic coupling among chromophores is operative [57].
Interestingly, the intensity of RLS increases on increasing the aggregation rate constant kc ( Figure  6). The data have been analyzed through a linear best fitting, obtaining IRLS corr = (−912 ± 905) + 10 5 × (7.53 ± 0.56) × kc. On one hand, this behavior is in contrast with that observed for the acid-induced aggregation of TPPS4, where an inverse dependence has been reported for the growth of porphyrin nanotubes [47]. On the other hand, when these nanoassemblies are formed in the presence of a high concentration of Zn 2+ , again a linear dependence between these two parameters has been reported [52]. These findings suggest that the structural arrangement of the chromophores in these specific supramolecular species are quite different with respect to that involved in nanotubes. Further evidence for a difference in the aggregate architecture stems from the slight broadening of the J-band for all the final aggregated samples. These spectral features appear generally larger than those reported for systems where the Frenkel exciton model for the electronic coupling among chromophores is operative [57]. When comparing the values of kc for the various metal derivatives, a monotonic increase is well evident together with a different efficiency in promoting the aggregation of TPPS4. Figure 7 reports the behavior of kc as a function of increasing [MT4] for the investigated cationic porphyrins. All the data have been best-fitted to the equation kc = k'MT4 × [MT4] and the values of the relative slope k'MT4 follow the trend: AuT4 > CoT4 > MnT4 > ZnT4 (Figure 8, left). The ability of the various porphyrins in accelerating the formation of J-aggregates is paralleled by their attitude in stabilizing larger aggregates or higher concentrations, as shown in Figure 8 (right), where the values of IRLS corr and extinction for fully aggregated samples at the highest investigated concentration are displayed. Interestingly, the intensity of RLS increases on increasing the aggregation rate constant kc ( Figure  6). The data have been analyzed through a linear best fitting, obtaining IRLS corr = (−912 ± 905) + 10 5 × (7.53 ± 0.56) × kc. On one hand, this behavior is in contrast with that observed for the acid-induced aggregation of TPPS4, where an inverse dependence has been reported for the growth of porphyrin nanotubes [47]. On the other hand, when these nanoassemblies are formed in the presence of a high concentration of Zn 2+ , again a linear dependence between these two parameters has been reported [52]. These findings suggest that the structural arrangement of the chromophores in these specific supramolecular species are quite different with respect to that involved in nanotubes. Further evidence for a difference in the aggregate architecture stems from the slight broadening of the J-band for all the final aggregated samples. These spectral features appear generally larger than those reported for systems where the Frenkel exciton model for the electronic coupling among chromophores is operative [57]. When comparing the values of kc for the various metal derivatives, a monotonic increase is well evident together with a different efficiency in promoting the aggregation of TPPS4. Figure 7 reports the behavior of kc as a function of increasing [MT4] for the investigated cationic porphyrins. All the data have been best-fitted to the equation kc = k'MT4 × [MT4] and the values of the relative slope k'MT4 follow the trend: AuT4 > CoT4 > MnT4 > ZnT4 (Figure 8, left). The ability of the various porphyrins in accelerating the formation of J-aggregates is paralleled by their attitude in stabilizing larger aggregates or higher concentrations, as shown in Figure 8     An intriguing property of TPPS4 J-aggregates is their propensity to spontaneous symmetry breaking, leading to an unbalance in the mixture of enantiomorphous nanostructures. This occurrence can be detected through circular dichroism spectroscopy and these supramolecular assemblies exhibit exciton coupled CD spectra at the wavelength of their absorption bands. In the case of J-aggregates formed in the presence of inorganic acids, the intensity of the usually observed positive Cotton effect can be quite large and an inverse dependence on the aggregation rate has been reported [46,47]. In the present case, very weak negative exciton split CD spectra are detectable at the J-band for all the samples obtained in the presence of the cationic porphyrins, with the exception of AuT4 (Figure 9). This evidence supports the hypothesis that the presence of MT4 species promotes structurally different aggregates with respect to nanotubes.  An intriguing property of TPPS 4 J-aggregates is their propensity to spontaneous symmetry breaking, leading to an unbalance in the mixture of enantiomorphous nanostructures. This occurrence can be detected through circular dichroism spectroscopy and these supramolecular assemblies exhibit exciton coupled CD spectra at the wavelength of their absorption bands. In the case of J-aggregates formed in the presence of inorganic acids, the intensity of the usually observed positive Cotton effect can be quite large and an inverse dependence on the aggregation rate has been reported [46,47]. In the present case, very weak negative exciton split CD spectra are detectable at the J-band for all the samples obtained in the presence of the cationic porphyrins, with the exception of AuT 4 ( Figure 9). This evidence supports the hypothesis that the presence of MT 4 species promotes structurally different aggregates with respect to nanotubes.
[MT4] for the various metal derivatives. AuT4 (black squares), CoT4 (red circles), MnT4 (blue triangles), ZnT4 (green triangles).  An intriguing property of TPPS4 J-aggregates is their propensity to spontaneous symmetry breaking, leading to an unbalance in the mixture of enantiomorphous nanostructures. This occurrence can be detected through circular dichroism spectroscopy and these supramolecular assemblies exhibit exciton coupled CD spectra at the wavelength of their absorption bands. In the case of J-aggregates formed in the presence of inorganic acids, the intensity of the usually observed positive Cotton effect can be quite large and an inverse dependence on the aggregation rate has been reported [46,47]. In the present case, very weak negative exciton split CD spectra are detectable at the J-band for all the samples obtained in the presence of the cationic porphyrins, with the exception of AuT4 (Figure 9). This evidence supports the hypothesis that the presence of MT4 species promotes structurally different aggregates with respect to nanotubes.  On the basis of the collected evidence, a mechanistic pathway to aggregation can be proposed (Scheme 2). In water at neutral pH, the TPPS 4 porphyrin is not protonated and therefore it is a tetra-anion able to strongly interact with cationic MT 4 metallo-porphyrins, leading to different hetero-aggregates (MT 4 @TPPS 4 ). These latter species possess an overall charge that depends on the coordinated metal ion (Au(III) +5; Co(III) +5; Mn(III) +5; Zn(II) +4) in the cationic species. Further, the nature of the metal ion imposes different coordination numbers and geometries: (i) Au(III) 4-coordinated and square planar; (ii) Co(III) and Mn(III) 6-coordinated and octahedral; and (iii) Zn(II) 5-coordinated and square pyramidal. The resulting different charge and geometrical structures of the various metal complexes have an impact on the stoichiometry and interaction among porphyrins in the formation of the hetero-aggregated species. Moreover, upon lowering the pH to 2.8, the TPPS 4 porphyrin is further protonated at the macrocyclic core and its overall charge decreases to −2 in the diacid H 2 TPPS 4 . This event determines a rearrangement of the hetero-aggregates (MT 4 @H 2 TPPS 4 ) and their stoichiometry were reported as ranging from 1:3 to 2:3 for the AuT 4 @H 2 TPPS 4 species and 2:3 for MnT 4 @H 2 TPPS 4 and ZnT 4 @H 2 TPPS 4 [31]. The preference for an edge-to-edge interaction (J-type) in these hetero-oligomers was explained (i) in terms of a steric role of the axial substituent groups, as in the penta-(ZnT 4 ) and hexa-coordinated (CoT 4 and MnT 4 ) complexes, thus preventing a potential face-to-face contact (H-type) and (ii) the net positive charge at the core of the macrocycle (AuT 4 ), leading to a repulsive interaction with the diprotonated H 2 TPPS 4 . All these factors determine the observed trend in the efficiency of the MT 4 species to influence both the rates and the eventual extent of aggregation. Indeed, these J-hetero-oligomers can be considered as the real initiators of the assembling process. Alternatively, the further addition of H 2 TPPS 4 monomers determines the formation of the critical nucleus (MT 4 @(H 2 TPPS 4 ) m ). From this point on, the autocatalytic pathway controlled by the rate constants k c leads to the growth of the final J-aggregates. It is interesting to observe that, when detectable, the final position of the B-band related to the MT 4 species in the UV/Vis spectra suggests their involvement into the nano-assemblies ( Figure S7). In addition, slight bathochromic shifts can be measured in the B-and Q-bands of the final J-aggregates (AuT 4 , 492/704 nm; CoT 4 , 492/709 nm; MnT 4 493/709 nm; ZnT 4 , 496/712 nm), supporting the hypothesis of a substantial incorporation of the cationic metallo-porphyrins in the J-aggregates. This evidence is in line with literature data showing the incorporation of Sn(IV) cationic species into nanotubes of H 2 TPPS 4 [35].
(Scheme 2). In water at neutral pH, the TPPS4 porphyrin is not protonated and therefore it is a tetraanion able to strongly interact with cationic MT4 metallo-porphyrins, leading to different heteroaggregates (MT4@TPPS4). These latter species possess an overall charge that depends on the coordinated metal ion (Au(III) +5; Co(III) +5; Mn(III) +5; Zn(II) +4) in the cationic species. Further, the nature of the metal ion imposes different coordination numbers and geometries: (i) Au(III) 4coordinated and square planar; (ii) Co(III) and Mn(III) 6-coordinated and octahedral; and (iii) Zn(II) 5-coordinated and square pyramidal. The resulting different charge and geometrical structures of the various metal complexes have an impact on the stoichiometry and interaction among porphyrins in the formation of the hetero-aggregated species. Moreover, upon lowering the pH to 2.8, the TPPS4 porphyrin is further protonated at the macrocyclic core and its overall charge decreases to −2 in the diacid H2TPPS4. This event determines a rearrangement of the hetero-aggregates (MT4@H2TPPS4) and their stoichiometry were reported as ranging from 1:3 to 2:3 for the AuT4@H2TPPS4 species and 2:3 for MnT4@H2TPPS4 and ZnT4@H2TPPS4 [31]. The preference for an edge-to-edge interaction (J-type) in these hetero-oligomers was explained (i) in terms of a steric role of the axial substituent groups, as in the penta-(ZnT4) and hexa-coordinated (CoT4 and MnT4) complexes, thus preventing a potential face-to-face contact (H-type) and (ii) the net positive charge at the core of the macrocycle (AuT4), leading to a repulsive interaction with the diprotonated H2TPPS4. All these factors determine the observed trend in the efficiency of the MT4 species to influence both the rates and the eventual extent of aggregation. Indeed, these J-hetero-oligomers can be considered as the real initiators of the assembling process. Alternatively, the further addition of H2TPPS4 monomers determines the formation of the critical nucleus (MT4@(H2TPPS4)m). From this point on, the autocatalytic pathway controlled by the rate constants kc leads to the growth of the final J-aggregates. It is interesting to observe that, when detectable, the final position of the B-band related to the MT4 species in the UV/Vis spectra suggests their involvement into the nano-assemblies ( Figure S7). In addition, slight bathochromic shifts can be measured in the B-and Q-bands of the final J-aggregates (AuT4, 492/704 nm; CoT4, 492/709 nm; MnT4 493/709 nm; ZnT4, 496/712 nm), supporting the hypothesis of a substantial incorporation of the cationic metallo-porphyrins in the J-aggregates. This evidence is in line with literature data showing the incorporation of Sn(IV) cationic species into nanotubes of H2TPPS4 [35]. Ext t = Ext ∞ + (Ext 0 − Ext ∞ ) (1 + (m -1){k 0 t + (n + 1) −1 (k c t) n+1 }) −1/(m−1) (1) where Ext 0 , Ext ∞ , k 0 , k c , m, and n are the optimized parameters.

Conclusions
Kinetic investigations offer the opportunity to gain insights on how the growth of supramolecular assemblies is controlled by specific changes of the medium and other external factors. The J-aggregates formed by the water sulfonated porphyrin TPPS 4 have been highly investigated for their ability to self-assemble into a variety of nano-and mesoscopic supramolecular structures. The specific arrangement and electronic coupling of the chromophores inside these species control their spectroscopic features and properties, such as their chirality. Understanding the role of other chemical components able to mediate, catalyze, and eventually to be incorporated into the final aggregates is therefore very important. The present study adds another small piece of information on these systems. Here, we demonstrated that MT 4 porphyrins are effective in accelerating the J-aggregation of H 2 TPPS 4 and that their efficiency is strongly dependent on their electronic and steric characteristics. Cationic metallo-porphyrins enter as part of the nano-architecture and the presence of transition metal ions could potentially introduce novel properties, such as catalytic activity.