Azides and Porphyrinoids: Synthetic Approaches and Applications. Part 1-Azides, Porphyrins and Corroles.

Azides and porphyrinoids (such as porphyrin and corrole macrocycles) can give rise to new derivatives with significant biological properties and as new materials’ components. Significant synthetic approaches have been studied. A wide range of products (e.g., microporous organic networks, rotaxane and dendritic motifs, dendrimers as liquid crystals, as blood substitutes for transfusions and many others) can now be available and used for several medicinal and industrial purposes.


Introduction
Azide derivatives have been used in synthetic methodologies leading to a wide range of acyclic and cyclic nitrogen compounds. Such a "chemical avenue" started near the end of the 19th century. Phenyl azide was discovered by Griess in 1864 by diazotization of phenyl hydrazine with nitrous acid [1] and in 1890, Curtius, using hydrazine and nitrous acid, discovered hydrazoic acid under an identical diazotization approach [2]. At that time, sodium azide and potassium azide were also obtained from hydrazine and an alkyl nitrite respectively in NaOH or KOH [3].
Nature is really impressive with this use of such small molecules in methodologies that lead to a wide range of compounds, some of them with highly important applications. Such interest and development by scientists took place mainly after the 1950s. Since then, many publications have been taking place, particularly related with the synthesis of several organic azide derivatives and the study of their potential applications (e.g., vaccine preservatives, pesticides and AIDS treatments). A wide range of publications are available for azide synthesis and azide derivatizations [4,5].
Some products of such derivatization methods have demonstrated significant properties and applications. Zidovudin (AZT), an azide derivative displaying anti-AIDS application, is an example [6]. Other azide derivatives are involved in the synthesis of Paclitaxel (Taxol), an anticancer drug and Oseltamivir (Tamiflu), an anti-influenza A/B drug [7,8].
As already mentioned, a significant number of publications on the synthesis and reactions of azides have been put forward, mainly in the recent decades [9]. However, the cycloaddition reactions with azides is a topic that deserves to be highlighted in this introduction. This is due to the fact that depending on the dipolarophiles being used, several biologically active 1,2,3-triazole derivatives can be obtained by 1,3-dipolar cycloadditions [10,11]. In particular, when an organic azide and an alk-1-yne are being used there is the formation of 1,4-and 1,5-disubstituted 1,2,3-triazole isomers.
This review will consider the scientific information published in recent years, involving chemical transformations not only of porphyrinoid and azide species or azide-porphyrinoid derivatives and their applications but also the potential biological products obtained from them. Porphyrinoid macrocycles are known to be of great significance to life. Natural porphyrins and related compounds play vital functions (e.g., respiration, photosynthesis, drug detoxification) and others can be used in very important medicinal applications [16]. Certain formulations have been already approved for the diagnosis of neoplastic situations and in the photodynamic therapy of certain cancer types and in the age-related macular degeneration. The photoinactivation of antibiotic resistant bacteria and viruses is also an important application coming out in the present times. Under such a line of work it can be stated that transformations involving both porphyrinoids with azides or azide-substituted porphyrinoids (porphyrin and corrole types) can lead to new derivatives that might become of great significance as new drugs, as pesticides, and even as catalysts in a wide range of reactions and in the synthesis of certain materials. Newly obtained products include the coupling of species (porphyrinoid-type and other heterocycles), which are themselves biologically active. The biological properties of the new derivatives might be increased and in such case that would presumably be due to any synergistic effect involving the coupled units. A wide variety of information about the azides/porphyrinoids chemical transformations and properties is available, in reviews and books/book chapters published until five years ago. This review is then focused on the scientific information that became available, mainly since 2015. Part one of this review is mainly focused on azides, porphyrins and corroles. Other macrocycles like phthalocyanines, subphthalocyanines and porphyrazines, and their transformations involving azides, are available in part two of the review, also published in this Molecules special issue [17].

Porphyrins
Porphyrins are a group of widely studied tetrapyrrolic macrocycles. This is not only due to their natural biological functions but also to the applications man can get from such a type of macrocycle. Synthetic methodologies based mainly on the azide chemical behavior leading to new porphyrin derivatives and their potential applications will be considered in this review.

Modification at Peripheral Substituents
The iron(III) complex of 5,10,15,20-tetrakis(4-ethynylphenyl)porphyrin (P1-Fe) was used by Son and co-workers to prepare hollow microporous porphyrinic-based organic networks [18]. The authors used Cu 2 O nanocubes as templates and networking catalysts, and a click-chemistry approach. This reaction involved the addition of P1-Fe and 1,4-diazidobenzene to a suspension of Cu 2 O nanocubes in a DMSO/H 2 O mixture. The reaction mixture was heated for 20 h at 85 • C leading to the formation of the microporous Fe(III) porphyrin network on the surface of the Cu 2 O nanocubes. The Cu 2 O core was removed by acid treatment and the Fe(III) porphyrin network linked by bis-triazolylbenzene bridges (Scheme 1) was obtained. The Fe(III) porphyrin-based organic network (FePON) prepared was explored as a catalyst for carbene insertion into N-H bonds; the obtained results revealed excellent catalytic activity, selectivity and recyclability [18]. Wang and co-workers reported the reaction of the Zn(II) complex of 5,10,15,20-tetrakis(4-ethynylphenyl)porphyrin P1-Zn with 5´-azide terminated DNA oligonucleotide (Scheme 2) [19]. The reaction was performed under classical CuAAC conditions, in the presence of bathophenanthroline sulfonate as a Cu(I) stabilizing ligand, yielding a mixture of the corresponding porphyrin adducts containing one, two (cis and trans), three and four (P2 in Scheme 2) triazolyl-DNA oligonucleotides.
The nucleophilic substitution of the para fluorine atom at the pentafluorophenyl ring in porphyrin P3 by propargylamine afforded the alkynyl A 3 B porphyrin P4 that showed to be a useful platform for further functionalization at the meso-substituents of the macrocycle via CuAAC reactions (Scheme 3). Using that strategy, Wiehe and co-workers prepared a set of porphyrin-triazole derivatives P5a-d containing several functions such as hydroxyl, mannosyl and azido groups, respectively P5a, P5b and P5c, and the porphyrinic dimeric species P5d. Macrocycle P5c was further modified with propargyl α-d-mannopyranoside affording derivative P6 at a 95% yield (Scheme 3) [20]. Porphyrin conjugates P5a, P5c, and P6 exhibited photocytotoxicity against human epidermoid carcinoma A-253 and squamous carcinoma CAL-27 cells [20].
The access to porphyrin-fullerene dyads prepared by a CuAAC reaction using a fullerene functionalized with an azido group directly attached to the central bridging subunit was reported by Nierengarten and co-workers [21]. The reaction of the alkynylated Zn(II) porphyrins P7a-c bearing conjugated spacers (m-phenylene, p-phenylene, di-p-phenylene-ethynylene) with the azide moiety 1 was achieved under microwave irradiation (MW) at 60 • C and using CuSO 4 ·5H 2 O and sodium ascorbate in a THF/H 2 O mixture (Scheme 4). Compounds P8a-c were obtained in moderate yields after 3 h (Scheme 4). The authors highlighted that when the reactions were performed at room temperature or under classical heating conditions a slow consumption of the starting material was observed and the desired compounds were obtained in low yields, this being due to the degradation of compound 1. Compounds 8a,b have shown relatively fast, efficient and long-lived photoinduced electron transfer (PET). Other related synthetic approaches involving the preparation of porphyrin-fullerene dyads via CuAAC have been extensively reviewed by Hanh and Nierengarten [22] and Coutsolelos and co-workers [23], and will not be dealt with here.
Nierengarten and co-workers also developed a synthetic approach to prepare pillar [5]arene derivatives peripherally decorated with porphyrin units [24]. The synthetic route started by the preparation of the alkyne-substituted A 3 B-type porphyrin derivatives P10 and P13 (Scheme 5). The preparation of the Zn(II) porphyrin-alkyne complexes began with the preparation of the aldehyde 3 by tosylation of 5-(trimethylsilyl)pent-4-yn-1-ol 2 followed the Williamson etherification of the tosylate with 4-hydroxybenzaldehyde affording 3 in an 85% yield. The next step involved the condensation of pyrrole, mesitaldehyde and 3 in the presence of catalytic amounts of BF 3 .Et 2 O under the conditions described by Lindsey [25]. The porphyrinic core P9 was metalated with Zn(II) and the obtained product was reacted with tetrabutylammonium fluoride (TBAF) in THF to give the porphyrin derivative P10 with a terminal alkyne moiety (Scheme 5) [24]. Scheme 5. Synthetic approaches to prepare mono-substituted alkyne meso-tetraarylporphyrins.
Photophysical studies with the former products have revealed that derivative P14b showed a temperature-dependent intramolecular binding between the 1,2,3-triazole moieties and Zn(II)-porphyrin core. These intramolecular interactions are favored at low temperature; however, at higher temperatures, the Zn(II) porphyrin-triazole coordination is broken and the molecules adopt an open flower-type conformation. The intramolecular coordination can be avoided by the addition of an external chemical stimulus, such as 1-phenylimidazole, a stronger ligand when compared with 1,2,3-triazole moieties [24]. Scheme 6. Synthesis of pillar [5]arene P14 containing porphyrin-triazolyl motifs.
The search for light-harvesting devices led Nierengarten and co-workers [26] to the synthesis of rotaxanes bearing porphyrin moieties by using click chemistry and supramolecular self-assembly approaches (Scheme 7). The authors performed the statistical condensation of pyrrole, mesitaldehyde and aldehyde 4, catalyzed by BF 3, followed by oxidation of the porphyrinogen intermediate, affording the corresponding porphyrin-decyloxy acetate derivative P15a. Hydrolysis of P15a with KOH provided the hydroxydecyloxy porphyrin derivative P15b.
The target azide derivative P16b was obtained at a 90% yield by treating compound P16a with NaN 3 . The clickable [2]rotaxane scaffold P16b was then reacted with Zn(II)-porphyrin-alkyne P10 (see structure in Scheme 5) in the presence of [Cu(phen)(PPh 3 ) 2 )](BF 4 ) in toluene affording derivative P18 at an 18% yield (Scheme 7). The obtained [2]rotaxane peripherally decorated with porphyrin units presents adequate features to act as an antenna for a light-harvesting device, able to transfer energy from the Zn(II) porphyrin moieties to the core [26]. Scheme 7. Synthetic approach to porphyrin [2]rotaxane derivative P18.
More recently, Tuncel and co-workers reported the synthesis of a cucurbit [6]uril (CB6) [5]rotaxane P20 bearing a tetracationic porphyrin decorated with triazole units [27]. The [5]rotaxane P20 was synthesized as depicted in Scheme 8; the tetracationic porphyrin derivative P19 was reacted with tert-butyl azidoethylammonium chloride and CB6 in water for 24 h at 40 • C (Scheme 8). In this reaction the CB6 has a dual function; it is the catalyst for the 1,3-dipolar cycloaddition between the azido and the alkynyl groups, leading to the formation of the 1,2,3-triazole ring, and it is the macrocycle for the inclusion of the four triazole rings. The porphyrin-core [5]rotaxane P20 was obtained at an 87% yield [27].
The photosensitizing properties of the water-soluble compound P20 was assessed on the photoinactivation of Escherichia coli (E. coli), a Gram-negative bacterial strain, Bacillus subtilis (B. subtilis), a Gram-positive bacterium, and breast cancer cell line MCF7. It displayed no cytotoxicity in the dark but under light P20 showed to be an efficient photosensitizer, even at low concentrations, demonstrating the potential for being used in both cancer photodynamic therapy (PDT) and photodynamic inactivation of microorganisms (PDI) [27]. Scheme 8. Preparation of porphyrin-based cucurbit [6]uril [5]rotaxane P20.
Kozaki and co-workers reported the synthesis of a trimeric assembly of dendritic porphyrins with two kinds of porphyrin cores [28]. Compound P21 (Figure 1) was prepared using a CuAAC reaction between a dendrimer having a Zn(II) diethynyldiphenylporphyrin core with two azide terminals and two equivalents of a dendrimer having a Zn(II) tetraphenylporphyrin core with one ethynyl terminal. This reaction was performed using copper sulfate (10 equiv.) and sodium ascorbate (45 equiv.) in dry DMF at room temperature for two days to afford trimer P21 as a black solid at a 66% yield. Absorption and fluorescence studies revealed that trimer P21 has a solvent-dependent conformation; while extended forms are favored in 1,1,2,2-tetrachloroethane (polar), the folded conformation is predominant in toluene (nonpolar). These different conformations explain why the singlet energy transfer from the tetraphenylporphyrins to the central diphenylporphyrin takes place with higher quantum efficiency in toluene than in 1,1,2,2-tetrachloroethane.
Acherar and co-workers reported the preparation of porphyrin-based building blocks using a click chemistry synthetic strategy [29]. The authors started by the preparation of Zn(II) porphyrin complex P23b (Scheme 9). For that the 5-(4-carboxyphenyl)-10,15,20-triphenylporphyrin P22 was converted into the propargyl-amide derivative P23a (96% yield) by reaction with an excess of propargylamine after the in situ activation of the carboxyl group of P22 with DCC and NHS. Compound P23a was metalated with the Zn(II) ion affording derivative P23b in quantitative yield. The reaction of porphyrin P23b and the mono-, di-and tri-substituted azides 6a-c was carried out in the presence of catalytic amounts of CuSO 4 and sodium ascorbate, 0.1 and 0.5 equiv., respectively. During the optimization of the reactional conditions the authors observed that for compound P24a the better conditions involved microwave (MW) irradiation. For compounds P24b,c the best conditions involved stirring at rt for three to five days. The described conditions allowed to obtain derivatives P24a-c in yields ranging from 77% to 97%. Photophysical studies with P24a-c carried out by the same group have revealed that the number of porphyrin units influences their photophysical properties [29]. Leroy-Lhez and co-workers followed a similar synthetic approach to attach a fluorescein-azide moiety to a porphyrin macrocycle bearing a propargyl moiety. The reaction was carried out under CuAAC conditions, affording the porphyrin-fluorescein dyad P25 at a 91% yield ( Figure 2). Photophysical studies supported by density functional theory (DFT) calculations showed strong evidence of efficient energy transfer (~40%) between the porphyrin and fluorescein moieties [30]. The research groups of Cheng [31] and Serrano [32] used the porphyrin-propargyl derivative P26 to prepare porphyrin-based dendrimers peripherally functionalized with alkyl chains linked by triazolyl rings and exhibiting liquid crystal behavior. Both groups developed synthetic routes using porphyrin P26 as the starting reagent for the CuAAC reaction with azide derivatives 7a-d (Scheme 10).
Cheng [31] prepared porphyrin derivatives P27a,b bearing four 1,2,3-triazole moieties with three terminal alkyl chains (n = 16 or 18) in each one of the meso positions that self-assembled into a columnar hexagonal packaging. When the chemosensorial ability of derivative P27b (n = 18) was evaluated in solution towards a series of metal(II) cations, that compound showed a selective emission quenching in the presence of Cu 2+ ions. Based on these results the authors prepared a porphyrin-based organogel in 1,4-dioxane due to the formation of J-aggregated entities assembled in a flower-like sphere by conjugation of hydrogen bonds, van der Waals and π-π-interactions. When this gel was explored as a chemosensor towards Cu 2+ ions, a quenching in the emission was observed due to the interaction of Cu 2+ ions with triazole moieties, which results in the breakdown of gelator molecules.
All systems obtained showed mesomorphic properties and a sponge-like morphology. However, the films obtained with 4,7-di-4-pyridyl-2,1,3-benzothiadiazole present an unordered morphology, while with C 70 a uniform alignment of columns in a columnar hexagonal grid was attained that might lead to anisotropic charge carrier transport along the molecular stacking. Due to the properties presented by compound P27b, it can be investigated for different applications in organic electronic devices or as a chemosensor [31]. Serrano and co-workers prepared porphyrin-core dendrimer derivatives P27c,d with three or six alkyl chains with a terminal coumarin moiety at each one of the meso positions. These compounds were synthesized by a CuAAC reaction between porphyrin-alkyne P26 and the appropriate azido dendrons derivatives 7c,d [32]. The tetra-substituted porphyrin-based functionalized P27c and second-generation dendrimers P27d were isolated in 91% and 71% yields, respectively, and their liquid crystalline behavior explored (Scheme 10). The two derivatives displayed mesogenic properties and the authors found that the coumarin moieties play a key role for the liquid crystal performance with a high contribution to the discotic nematic mesophases showing high hole mobility values when compared with an analogous derivative peripherally functionalized with dodecyloxy alkyl chains. Additionally, the coumarin moieties have an antenna effect, inducing an energy transfer process after excitation, presenting appropriate features to be used in optoelectronic applications [32].
Rajakumar described the preparation of porphyrin-based triazole-bridged fluorenodendrimers P28a,b, P29a,b and P30a,b [33]. The porphyrin-fluorenodendrimers conjugates P28-P30a,b were prepared by reaction of the porphyrin-propargyl derivative P26 with a series of dialkyl-fluorene azido dendrons 8-10a,b through a CuAAC reaction. The reaction was carried out in the presence of CuSO 4 ·5H 2 O (5 mol%) and sodium ascorbate (10 mol%) in a mixture of THF/H 2 O (3:1) for 12 h at room temperature. Compounds P28-P30a,b were isolated in yields ranging from 71% to 82% (Scheme 11). An increase in the number of fluorene moieties around the porphyrin core induced the enhancement of the absorption and fluorescence intensities. The antiproliferative activity of porphyrin-fluorenodendrimers against the PA-1 human ovarian teratocarcinoma cell line was accessed; the second-generation porphyrin-based dendrimers P30a,b showed a more efficient inhibition of cancer cell growth [33].
More recently, Prabakaran and co-workers reported the synthesis of porphyrin derivative P31 bearing four H-cardanol units (Figure 3), a waste by-product of the cashew industry, linked by triazolyl bridges prepared via a CuAAC reaction [34]. J-type aggregates formation is induced when compound P31 was studied in polar protic and aprotic solvents, due to the long alkyl chains of the H-cardanol moieties. However, compound P31 is soluble in non-polar solvents, namely hexane, due to its fat-like behavior when compared with other analogous compounds with smaller alkyl chains (n = 2). This compound showed adequate photophysical properties for a potential medicinal application as a photosensitizer in PDT. Ye and co-workers prepared porphyrin-based porous polymers by the reaction of porphyrin P26 with 1,4-bis(azidomethyl)benzene via CuAAC reaction in the presence of N,N´,N ,N",N"-pentamethyl-diethylenetriamine (PMEDTA) and CuBr in DMF at 100 • C for 48 h. The obtained porous Zn(II) porphyrin polymer P32 ( Figure 3) is able to interact and to act as protein adsorbent, showing potentiality for protein immobilization and other applications in pharmaceutical or food industries [35].
The synthesis of the Pd(II) porphyrin complexes P35 and P36 was described by Bretonniére and co-workers [36]. The synthetic strategy involved the preparation of the Pd(II) complexes P33 and P34 from the hydroxylated porphyrin precursors, followed by the functionalization of the metalloporphyrins with alkyne moieties by a nucleophilic substitution reaction with propargyl bromide in the presence of K 2 CO 3 . The next step involved the CuAAC reaction of P33 and the octa-substituted P34 alkynyl porphyrins with the azido-9,9-diethylfluorene derivative in the presence of CuSO 4 .5H 2 O and sodium ascorbate at 40 • C for 48 h. This allowed the preparation of the porphyrin derivatives P35 and P36 at a 71% and 65% yield, respectively (Scheme 12). The 9,9-diethylfluorene moiety grafted to the Pd(II) porphyrin acts as an artificial antenna enhancing the two-photon absorption when excited at 800 nm and induces an energy transfer to the porphyrin core through a Förster Resonance Energy Transfer (FRET) process [36].
Satake and co-workers reported the preparation of the Gable-type porphyrin P39 bearing two imidazole-alkynyl moieties, thus becoming available to be modified via a CuAAC functionalization [37]. The A 2 BC porphyrin P37 was prepared by the condensation of dipyrromethane 13 with aldehydes 11 and 12 in the presence of trifluoroacetic acid (TFA), followed by acidic deprotection of the acetal protecting group thus affording the corresponding formyl derivative P38. Then, compound P38 was submitted to a further TFA catalyzed condensation with dipyrromethane 13 and aldehyde 11 under the same conditions affording the alkynyl porphyrin P39. Its Zn(II) complex was prepared by treatment with Zn(OAc) 2 salt and functionalized through a CuAAC reaction with an alkyl, nonfluorinated, oligoether group and pyrene derivative moieties by the reaction of the appropriate azide in the presence of CuI and 2,6-lutidine (Scheme 13). Gable-type porphyrins P40 with a triazolyl bridge were obtained in high yields (>88%) after stirring for three days at room temperature. The authors emphasized the utility of the obtained porphyrin derivatives for the preparation of dynamic combinatorial chemistry leading to supramolecular systems by a self-assembly approach, capable of detecting interactions between a target biomaterial (e.g., proteins or nucleic acids) and a ligand [37].
Methyl pheophorbide a P41 was used as the precursor for the synthesis of chlorin derivatives bearing one, two or four quinazoline moieties linked by triazolyl rings via CuAAC reactions (Scheme 14). In the first step, methyl pheophorbide a P41 reacted with propargylamine or a series of diamines affording the corresponding chlorin-amide derivatives P42 and P43 by exocyclic ring-opening; in the second step the free-base amide derivative was metalated with Zn(II). Finally, the Zn(II) chlorin-amide complexes P43a-c were submitted to an acylation reaction at the terminal free amino groups under Steligkich conditions that involved a carboxyl group (pent-4-ynoic acid or 3,5-bis(propargyloxy)benzoic acid), EDC, DMAP and HOBt. After 15 h at rt the corresponding amides P44 and P45 were obtained in yields ranging from 88% to 93% [38]. Scheme 12. Synthesis of porphyrin derivatives P35 and P36 bearing four or eight 9,9-diethylfluorene groups, respectively. The azide-functionalized (arylamino)quinazoline 14 was then connected to the chlorin-alkyne derivatives P42, P44 and P45a-c through click chemistry cycloaddition reactions in the presence of catalytic amounts of CuI after magnetic stirring for 15 h at room temperature, affording the corresponding chlorin bearing one (P46), two (P47 and P48a,b) or four (P48c) triazole-quinazoline units in yields ranging from 40% to 99% (Scheme 15).
The quinazoline moiety attached to the chlorin derivatives is analogous to the unit present in the structure of Vandetanib, a drug approved for the treatment of metastatic medullary thyroid cancer and is known to be a selective epidermal growth factor receptor and vascular endothelial growth factor receptor ligand. The authors expected to find a synergic effect between the photosensitizer properties of the chlorin unit and the selectively cytotoxic agent against tumor cells due to the quinazoline moiety. The chlorin-triazole-quinazoline conjugates presented suitable photophysical properties for the target application; however, their low water solubility precludes the study of their biological activity [38]. Scheme 15. Synthesis of chlorin-triazole-quinazoline conjugates.
The same group described the synthesis of the bis(chlorin)-quinazoline derivative P49 (Figure 4, 57%) by reaction of the chlorin-azide derivative, prepared by reaction of chlorin P43a with 6-azidohexanoic acid, and a quinazoline di-alkyne derivative in the presence of catalytic amounts of CuI in a DMF/water (40:1) mixture [38].

Carbohydrate-Porphyrin Conjugates
The click approach was also used to prepare porphyrin-carbohydrate conjugates. In most of the strategies reported, the alkynyl function is present in the porphyrin while the azide group appears in the carbohydrate component.
Other research groups searching for bio-based molecules for supramolecular self-assembly into functional materials for electro-optical applications were able to conjugate, through the CuAAC approach, glycolipid surfactants' moieties such as 15 and 16 with 5,10,15,20-tetrakis(4-ethynylphenyl)porphyrin P1-Zn. These azide glycolipid surfactants (15 and 16) were prepared by ring-opening of lactonic sophorolipids followed by reaction with 3-azidopropylamine under neat reaction conditions (15) (Scheme 17). Since the double bond confers a certain degree of rigidity that influences the self-assembly of the corresponding functional material, the catalytic hydrogenation step was performed before the reaction with 3-azidopropylamine affording derivative 16 with a good yield (>60%) [41]. Senge and co-workers used the bis(4-ethynylphenyl)porphyrin derivative P56 to prepare amphiphilic target molecules P58 to be used in PDT (Scheme 18) [42]. In this study, the authors used the 3-azidopropyl α-d-mannopyranoside as the carbohydrate component. Starting from 5,15-dibromo-10,20-diphenylporphyrin P55, the ethynyl derivative P56 was prepared with a 72% yield via a Suzuki cross-coupling reaction using 4-[(trimethylsilyl)ethynyl]phenylboronic acid pinacol ester, K 3 PO 4 , Pd(PPh 3 ) 4 in THF at 65 • C for 16 h (Scheme 18). The click reaction between a large excess of the ethynyl porphyrin P56 and the azide sugar compound afforded the monosubstituted derivative P57 after 10 min at 115 • C under microwave irradiation. Following a similar approach, the second ethynyl group was then grafted with different azidoalkyl chains N 3 CH 2 (CH 2 ) n CH 3 (n = 7-20). Two of the prepared compound P58 (n = 7 and 16) exhibited singlet oxygen production similar to Foscan. Scheme 18. Synthetic approach to prepare the amphiphilic porphyrin derivatives P58 to be used in PDT.
The CuAAC approach also allowed the efficient preparation (90% yield) of other amphiphilic porphyrin conjugates such as the porphyrin-cyclodextrin conjugate P59 ( Figure 5) [43]. The conjugation between the 6-deoxy-6-azidopermethyl-β-cyclodextrin and the meso-tris(ethynylphenyl)porphyrin derivative was performed in THF under stirring in the presence of copper(II) sulfate and sodium ascorbate in water for 48 h at 60 • C under an argon atmosphere. The obtained amphiphilic porphyrin-cyclodextrin conjugate P59 was used to prepare supramolecular nanoarchitectures to control drug delivery. Drain and co-workers also reported the synthesis of several porphyrin derivatives bearing carbohydrate moieties [44]. These authors prepared carbon-1 and carbon-3-galactosyl porphyrin conjugates P60 and P61 (M = 2H or Zn) under CuAAC conditions and tested them towards four tumoral cell lines (monolayers or spheroids) containing high expression of galactin-1 protein ( Figure 6). The results pointed out that the uptake of the cabon-3 galactose porphyrin derivatives (P61) is higher than the carbon-1 galactose porphyrin derivatives (P60) probably due to the higher hydrophilicity and the binding recognition domain of galectin-1 protein by the available hydroxyl group at carbon-1 [44]. Using the same alkynyl porphyrinic precursor P26, Rajakumar and co-workers [45] and Rivera and co-workers [46] prepared different amphiphilic dendrimers by a convergent synthetic strategy also using click chemistry. The antitumoral activity of some dendrimers was evaluated towards the MIA PaCa-2 cell line and showed that cell growth inhibition increased along the dendritic generation [45]. Other dendrimers exhibited ability to form J-aggregates by increasing the polarity of the solvent but such aggregation can be destroyed by enhancing the temperature [46].
Nilsson and co-workers selected the strategy shown in Scheme 19 to prepare the tri-glucosylated-porphyrin conjugate P65 and analogues (the click step was extended also to galactose and N-acetyl glucosamine azides) to be used as theranostic agents [47,48]. For that, the asymmetric porphyrin precursor P62 was treated successively with chlorosulfonic acid, propargylamine in the presence of N,N-diisopropylethylamine (DIPEA) and Zn(II) acetate in order to gain porphyrin P63 at a total yield of 40%. The click reaction occurred under microwave irradiation after 5 min at 85 • C in a sealed vessel at a 59% yield. After the transformation of the chlorosulfonic groups into the corresponding alkynyl sulfonamide function a new click approach was used to react the product with 2-azidoethyl 2-fluoro-2-deoxy-β-glucopyranoside (structure not shown) to prepare a multimodal imaging and PDT agent. All glycosylated derivatives were deacetylated yielding the hydrophilic glycosyl-porphyrin conjugates in very good yields after treatment with sodium hydroxide, methanol and water at room temperature. The target product has demonstrated significant selectivity towards melanoma cells [48]. Scheme 19. Synthetic strategy used in the preparation of P65.

Immobilization of Porphyrins in Different Matrixes
Due to the interesting photophysical and biological properties displayed by porphyrin derivatives, they have been incorporated/embedded in or covalently linked to several organic and inorganic supports. Among several coupling approaches, the CuAAC reaction conditions were used successfully to link natural or synthetic porphyrin derivatives to cellulose fibers, dextran, mesoporous organosilicas, SiO 2 -magnetic nanoparticles or functionalized glass.
For instance, in 2015 Topka and Dinolfo explored the CuAAC approach in combination with the layer-by-layer (LbL) process to assemble over an azide-rich functionalized glass or quartz the alkynyl porphyrin P1-Zn, the azide-BODIPY 17 and the tris-azido linker 18 to obtain the multilayer film GQ-MP1-Zn (Scheme 20) [49]. The LbL process started with the conjugation of the ethynyl P1-Zn with the azide functionalized glass or quartz surface (prepared by anchoring 11-azidoundecyltrimethoxysilane to glass/quartz slides surface exposed to a piranha solution) followed by reaction of the remaining ethynyl moieties with a mixture of azide-BODIPY 17 and the tris-azido linker 18 in order to allow the formation of another porphyrinic layer. In this way, it was possible to add multiple layers and consequently to modulate the optical density through their thickness. In fact, this strategy is very convenient to generate multichromogenic (donor-acceptor) energy transfer assemblies to develop full spectral light harvesting materials for DSSC devices [49]. The same research group, by following the LbL process and through a click approach, prepared other porphyrin-based molecular multilayer films (ITO-MP1-Zn) on indium tin oxide (ITO) electrodes for photovoltaic and photogalvanic devices [50]. Scheme 20. Strategy used to immobilize P1-Zn, 17 and/or 18 in glass, quartz or ITO surface by LbL.
Zhang and co-workers described the preparation of organic-inorganic hybrid porphyrin derivatives with a polyhedral oligomeric silsesquioxane (POSS) [51,52]. The authors followed two different synthetic approaches. Firstly, POSS functionalized with an azido group (POSS-N 3 ) [53] was directly attached to the porphyrin-alkyne Zn(II) complex P67, under classical click chemistry conditions, with the formation of the expected triazole ring. Zn(II) porphyrin-POSS hybrid P68a was obtained at a 49% yield after 24 h at 40 • C (Scheme 21) [51].
The corresponding free-base derivative P68b (96% yield) was obtained after the treatment of hybrid 34a with conc. HCl in chloroform and their self-assembly behavior was explored. The authors found that different morphologies, such as core-shell spherical micelles and ordered square sheets, can be obtained depending on the polarity of the solvent or solvent mixtures used [51]. The same group prepared the photosensitive organic-inorganic porphyrin-POSS hybrid P69 with an azobenzene bridge (Figure 7) [52]. The synthetic route involved the reaction of porphyrin derivative P67 with 4-azido-4´-hydroxyazobenzene affording the formation of the porphyrin-azobenzene derivative linked by 1,2,3-triazole rings (50% yield). Through a substitution reaction with 3-bromoprop-1-yne in the presence of potassium carbonate, this compound afforded the corresponding alkyne derivative that was submitted to a second click chemistry reaction with POSS-N 3 to obtain the Zn(II) complex of P69-trans at a 51% yield. The treatment with HCl afforded the free-base derivative P69-trans in quantitative yield [52]. The azo bridge allowed the change from a trans to a cis configuration and, in such way, there is a significant reduction in the molecule size. This was performed by irradiation with UV light (350 nm), with visible light (450 nm) or with heating ( Figure 7). The supramolecular self-assembly of derivative P69-trans, as observed for compound P68, has a strong solvent dependence but additionally it also depends on the configuration. The possibility to modulate the structure by using light presents potentialities for applications in bio-imaging, drug release and catalysis [52].
Aiming to obtain nanocatalysts able to regulate polymerization kinetics in response to dual external stimuli (light and magnetism), Cai and co-workers have reported the covalent anchorage of the alkynyl Zn(II) complex P67 into the Fe 3 O 4 @aSiO 2 Janus type nanoparticles (magnetic nanoparticles with the magnetic off-centered core) bearing azide units (Scheme 22). The coupling was performed in the presence of PMEDTA, Cu(II) acetate and sodium ascorbate in DMF at 50 • C for 12 h under an argon atmosphere [54]. The catalyst efficacy of Fe 3 O 4 @aSiO 2 -P67 in photo-induced electron transfer reversible addition-fragmentation chain transfer (PET-RAFT) processes was tested using vinylic monomers such as styrene, acrylates or acrylamides and led to the preparation of several homopolymers after green light activation (520 nm, 5 W, 1.3 mW cm −2 ). The results showed that the nanocatalyst was regenerated without significant photodegradation.  The Fe 3 O 4 magnetic core can also be capped with other shell types such as the polysaccharide dextran, a biocompatible polymer, with high affinity to iron oxide. Sol and co-workers selected Fe 3 O 4 @Dextran nanoparticles to prepare nanoplatforms for delivering photosensitizers to tumoral cells for the PDT approach [56]. The synthesis of the conjugates involved first the functionalization of the dextran surface with epichlorohydrin followed by reaction with sodium azide, in order to functionalize the nanoparticle surface with the required azido functions (Scheme 23). The conjugation with the water-soluble alkynyl porphyrin derivatives P71-P73 took place in aqueous media at room temperature for 24 h in the presence of Cu(II) acetate and sodium ascorbate; the water solubility was conferred by the presence of glucosyl, sulfonate, or pyridinium groups. The graft yields obtained after dialysis (72 h against pure water) varied between 24% and 49%, the best values being found for the anionic and cationic derivatives P72 (43%) and P73 (49%). The results showed that the cationic conjugate Fe 3 O 4 @Dextran-P73 displayed magnetic properties to be used not only as contrast agent for MRI and drug delivery but also in hyperthermia therapy [56].

Scheme 23.
Magnetic nanoparticle coated with native dextran and linked to water-soluble photosensitizers.
The alkynyl-porphyrin derivative P4 was immobilized in hyperbranched polyglycerol (hPG) functionalized with azide units affording the corresponding porphyrin-hPG conjugates under CuAAC conditions. To improve the solubility in water, some of the azide units at the hPG core were then functionalized with methoxy poly(ethylene glycol) (mPEG)-propargyl affording the porphyrin-hPG-PEG conjugates P74a,b (Figure 9) as biocompatible drug nanocarrier systems [20].
The authors extended this approach to obtain porphyrin-hPG conjugates P75, P76 and P77 with different linkers and mannose content and demonstrated that some of them exhibit strong phototoxicity against S. aureus and several tumor cell lines at micromolar concentration range [57,58]. Durand and co-workers also used porphyrin P78 to synthesize the derivative P79, which was used to produce mesoporous organosilica nanoparticles containing Zn(II) porphyrin moieties [59]. The authors obtained the sol-gel precursor P79, featuring eight triethoxysilyl groups, through the CuAAC reaction of P78 with N,N-bis(3-triethoxysilylpropyl)prop-2-ynylamine (19) in dry THF in the presence of CuBr(PPh 3 ) 3 and under MW irradiation (Scheme 24). The nanoparticles were prepared by co-condensation of porphyrin P79 with bis(triethoxysilyl)ethene in water at 80 • C, under basic catalysis, and using cetyltrimethylammonium bromide as the structure-directing agent. The nanoparticles revealed to be two-photon-sensitive and their two-photon imaging capacity was assessed in MCF-7 breast cancer cells. Also, owing to their high porosity, the nanoparticles were applied as drug nanocarriers. Nanoparticles were loaded with doxorubicin (drug loading of 20 wt%) and it was observed that the release of doxorubicin was pH dependent. The authors concluded that these doxorubicin-loaded organosilica nanoparticles were very efficient at inducing MCF-7 breast cancer cell killing, showing promising potential for theranostic applications. Salvatini and co-workers also selected the click approach to link covalently the alkynyl porphyrin P80 with SiC/SiO x nanowires functionalized with azide units (Scheme 25) [60]. The experimental design was based on the condensation of the nanowires with 3-azidopropyltrimethoxysilane to provide the required azido groups (N 3 -SiC/SiO x ) at their surface. The alkynyl derivative P80 was prepared at a 60% yield by reacting 5,10,15,20-tetrakis(4-carboxyphenyl)porphyrin with propargylamine in the presence of EDC and HOBt in DMF for 24 h (Scheme 25). In order to avoid the coordination of the free-base porphyrin with copper(II), the cycloaddition was performed in DMSO using thermal conditions (160 • C for 24 h). The P80-SiC/SiO x nanowires linked covalently to the porphyrin produced singlet oxygen efficiently after a low dose (0.4-2 Gy) of X-ray irradiation; also, a reduction in 75% of lung adenocarcinoma cells viability was observed after 12 days. In 2019, lyocell fibers (a cellulose II matrix), after being submitted to a silanization process with (3-azidopropyl)triethoxysilane, were used to immobilize the alkynyl-derived protoporphyrin IX (P81) in order to produce photo-bactericidal materials C-P81 [61]. The coupling was performed in DMF in the presence of copper(II) sulphate and ascorbic acid for 72 h at room temperature (Scheme 26). The photobactericidal properties of the new materials were evaluated towards two Gram-positive bacterial strains (S. aureus and B. subtilis) and have demonstrated significant activity against these two strains upon light activation [61]. Scheme 26. Immobilization of alkynyl-derived protoporphyrin IX P81 on modified cellulose II matrix.
Other azide cellulose-derived fibers (kraft pulp) were used to covalently link cationic porphyrin derivatives to be tested as photobactericidal materials (Scheme 27) [62]. It is well-known that cationic porphyrin derivatives exhibit a great antibacterial activity in Gram-positive and Gram-negative bacteria strains independently of their antibiotic resistance and without development of resistance [63][64][65][66]. Thus, pulp kraft was previously tosylated by reaction with an excess of tosyl chloride in triethylamine at room temperature for 96 h; the obtained material was then treated with sodium azide in DMF for 96 h at 50 • C, thus being transformed into the azide kraft pulp (N 3 -kraft pulp) derivative [62]. The propargylated porphyrin ZnTPPyP was obtained by treating the neutral TPyP with an excess of propargyl bromide at room temperature for nine days, followed by the complexation step with
The 5-(4-azidophenyl)porphyrin P82 was also used to form porphyrin-DNA conjugates [68]. The conjugates were prepared by solid-phase click reaction between the azide P82 and oligodeoxynucleotides with an ethynyl group on controlled pore glass support (Scheme 29). Porphyrin-DNA duplexes and a dimer of porphyrin were created by hybridization with the appropriate complementary DNA strands. Porphyrin-DNA modified electrodes constructed by a self-assembled DNA monolayer with the porphyrin units on the gold electrode showed a photocurrent response upon light irradiation. Other examples of click chemistry using 5-(4-azidophenyl)porphyrins to synthesize new sensitizers for dye-sensitized solar cells (DSSCs) include the synthesis of the porphyrin-triazole conjugates P87 and P89 that incorporate carboxyl groups as anchoring groups (Scheme 30) [69]. DSSCs fabricated with dyes P87 and P89 showed better efficiency (eight and four times better, respectively) when compared to reference compounds without the triazole ring. It was concluded that the carboxylate unit, for this type of compounds, is a more efficient anchoring group than the cyanoacrylate unit.

Scheme 30. Synthesis of porphyrin-triazole conjugates incorporating carboxyl groups.
Durantini and co-workers reported the synthesis of a cruciform dendrimer, which bears a nucleus of Zn(II) porphyrin substituted at the meso positions by four bis(carbazolyl)triphenylamine units linked by triazole rings [70]. The new compound P91 was obtained from the reaction of the meso-tetrakis(azidophenyl)porphyrin P90 and the acetylenic dendron 23 using the CuAAC method (Scheme 31). The electrochemical oxidation of the carbazole groups in P91 was used to obtain stable and reproducible fully π-conjugate photoactive polymeric films that are highly stable to photobleaching and produce singlet oxygen in both DMF and water. These films revealed to be very efficient in the photodynamic inactivation of Staphylococcus aureus and Escherichia coli in both planktonic and biofilm forms.
Boyle and co-workers reported the synthesis of the first water-soluble porphyrin radiolabeled with fluorine-18 [71]. The tricationic porphyrin P93 was obtained through click conjugation of the cationic azidophenylporphyrin P92 with the tetraethoxy PEG derivative 24 (Scheme 32). Biological studies revealed that this porphyrin derivative showed cellular uptake, good photocytotoxicity, and minimal dark toxicity in human adenocarcinoma (HT-29) cells and demonstrated potential properties as a radiotracer in vivo. The new theranostic agent P93 may be considered a clinically relevant theranostic agent since it integrates the therapeutic selectivity of PDT with the imaging efficacy of positron emission tomography (PET). Scheme 32. Synthesis of a water-soluble porphyrin radiolabeled with fluorine-18.

meso-(p-Azidomethylphenyl)porphyrins
The covalent cage P97 consisting of two porphyrins connected by four flexible spacers each incorporating two 1,2,3-triazolyl ligands has been synthesized starting from the meso-tetrakis(4-azidomethylphenyl)porphyrin P78 and the TIPS-monoprotected alkyne 25 (Scheme 33) [72]. The first CuAAC reaction gave rise to compound P94 at a 89% yield. Deprotection of the TIPS-protected alkyne using TBAF led to porphyrin P95, the second precursor of the cage. The final step involved the formation of the DABCO-templated hetero dimer P96 (in a dynamic equilibrium mixture with two homo dimers) and the construction of the last four triazolyl groups. The last cyclisation reaction was performed in CH 2 Cl 2 at room temperature using [Cu(tren')]Br as catalyst and Na 2 CO 3 as a base. The reaction was completed after five days and the cage P97 was obtained at a 25% yield. Due to the flexibility of the ether spacers between the triazole moieties this cage adopts a compact conformation in solution stabilized by π-π interactions. However, the coordination of the triazole moieties with four Ag(I) ions locks the porphyrin macrocycles in a face-to-face disposition, leading to the formation of a rigid cage. Tuncel and co-workers reported the synthesis of a photoactive glycosylated porphyrin covalently attached to a monofunctionalized cucurbit [7]uril unit [73]. The synthesis of the target compound involved an initial CuAAC reaction involving the tetraazidoporphyrin P78 with the propargylated tetraacetylmannose 26 (Scheme 34). The resulting monoazidoporphyrin P98 was then deacetylated using sodium methoxide in methanol and, finally, a new CuAAC reaction between azide P99 and the monopropargyloxycucurbit [7]uril 27 afforded the multifunctional porphyrin derivative P100.
The host-guest chemistry of conjugate P100 was investigated by 1 H NMR experiments using a bisimidazolium guest and the formation of an inclusion complex with the cucurbit [7]uril was observed. It was also found that the 1 O 2 generation efficiency of the synthesized compound was significantly higher than that of the unfunctionalized porphyrin, indicating that this supramolecular assembly can serve as an efficient photosensitizer for biological applications. In 2019, Tuncel and co-workers reported the use of the glycosylated porphyrin-cucurbituril conjugate P100 as a photosensitizer in photodynamic antibacterial and cancer therapy and as a drug carrier [74]. The results of the study showed that this photosensitizer efficiently inactivated both Gram-negative (Escherichia coli) and Gram-positive bacteria (Bacillus subtilis) when exposed to white light for 1 min (20 mW/cm 2 ). The in vitro citotoxicity and photocytotoxicity of conjugate P100 on the MCF7 breast cancer cell line revealed that it did not show any cytotoxic effect in the absence of light but caused a significant decrease in the viability of the cancer cells when exposed to white light for 5 min (20 mW/cm 2 ). It was also demonstrated that conjugate P100 served as a doxorubicin carrier for chemo-photodynamic dual cancer therapy.
Gros, Harvey and co-workers reported the synthesis of cyclotriveratrylene derivatives covalently bonded to 1, 2, 3, or 6 Zn(II) porphyrin units via click chemistry (Schemes 35 and 36) [75]. The new compounds were used as hosts for C 60 but the binding constants were considered to be not good (K a between 400 and 4000 M −1 ). The formation of host-guest assemblies with C 60 was studied using absorption and fluorescence spectroscopy, including fluorescence quenching by C 60 . Surprisingly, the largest K a value was obtained for the monoporphyrinyl derivative P102, which exhibits the least sterically demanding structure. Computer modelling studies revealed that the flexibility of the anchoring chains between the cyclotriveratrylene bowl and the porphyrin units prevented favorable conformation to capture C 60 . Instead, pincer conformations were found to be responsible for the host-guest associations.

2-Azido-or 2-Azidomethylporphyrins
Singh and Nath reported the synthesis of β-triazole bridged coumarin-porphyrin conjugates by the CuAAC reaction between 2-azido-5,10,15,20-tetraphenylporphyrinatoCu(II) (P106) with various alkyne-substituted coumarins (29-31) (Scheme 37) [76]. The corresponding free-bases were obtained in a good yield (71%-80%) by demetalation of the Cu(II) derivatives. Metalation of the free-bases with Zn(II) acetate afforded the corresponding Zn(II) complexes. The photophysical characterization of these conjugates revealed, for some of them, a considerable electronic communication between both units. Additionally, in some of these dyads, a significant intramolecular energy transfer between both chromophores was observed. Using a similar strategy, the same authors also synthesized the Zn(II) β-triazolylmethyl-bridged coumarin-porphyrin dyads P111-P113 (Scheme 38) [77]. Their synthesis involved the 1,3-dipolar cycloaddition between the 2-azidomethyl-5,10,15,20-tetraphenylporphyrinatozinc(II) (P110) and the alkyne-substituted coumarins 29-31. The reported yields for the dyads are in the range of 84%-92%. Demetalation of the Zn(II) complexes afforded the corresponding dyads with free-base porphyrins. The photophysical characterization of the new compounds showed the occurrence of an intramolecular energy transfer between the units of some conjugates. The same authors also reported the use of porphyrins P106 and P110 for the synthesis of β-triazolyland β-triazolylmethyl-bridged porphyrin-xanthone conjugates [78]. For both porphyrin derivatives, 1:1 (P114) and 2:1 (P115) porphyrin-xanthone conjugates were prepared (Scheme 39). The photophysical evaluation of these conjugates revealed a bathochromic shift in their electronic absorption and fluorescence spectra when compared to the spectra of the meso-tetraarylporphyrins.

Other Azido-Substituted Porphyrins
Xu and co-workers reported the synthesis of a series of fluoride functionalized conjugated microporous polymers (CMPs) based on Zn(II) porphyrin building blocks [79]. The synthetic route involved the Suzuki-coupling reaction of meso-tetrakis(p-bromophenyl)porphyrinatozinc(II) (P116) with a mixture of 4,4'-biphenyldiboronic acid and (2,5-bis(azidomethyl)-1,4-phenylene)diboronic acid at varying molar ratios (Scheme 40). The azide groups incorporated within the pores of the resulting CMPs P117 were then subjected to a CuAAC reaction with pentafluorophenylethyne. The resulting porous frameworks P118 exhibited enhanced CO 2 sorption properties. Drašar and co-workers reported the synthesis of the two trilobolide-porphyrin conjugates P119 and P120 (Figure 10) [80]. Trilobolide is a natural sesquiterpene lactone isolated from horse caraway (Laser trilobum, L. Borkh) that exerts significant pharmacological properties. The trilobolide unit was linked to porphyrin derivatives via CuAAC reactions and the in vitro cytotoxicity of the resulting conjugates was tested against rat peritoneal cells and four human tumor cell lines: LNCaP (prostate carcinoma), U-2 OS (osteosarcoma), MCF-7 (breast carcinoma), and MiaPaCa-2 (pancreatic carcinoma). The intracellular localization of the conjugates was investigated using live-cell fluorescence microscopy. Both conjugates were localized in mitochondria and lysosomes of HeLa and LNCaP cells at 5.0 µM concentration after 2 h. Eggleston and co-workers reported the synthesis of several cell-penetrating peptide-porphyrin conjugates using a diversity of bioconjugation reactions, including the CuAAC approach [81]. That work aimed to convert classical hydrophobic PDT agents into amphiphilic conjugates suitable for targeted photochemical internalization and PDT. Examples of the synthesized conjugates are illustrated in Scheme 41. Coutsolelos and co-workers explored the potential application as sensitizers for DSSCs of two porphyrin-based dyads linked by a variable-length triazole bridge [82]. The A 2 BC porphyrin derivative P125 bearing an ester and an amino group at para positions of two opposite meso-phenyl rings was functionalized with an azide group by following two different synthetic pathways. One of them involved the insertion of an alkyl chain as a spacer between the meso-phenyl and the azido group by reaction of the amino-porphyrin P125 with 5-bromopentanoyl chloride, which yielded the corresponding bromoalkyl porphyrin amide derivative P126a, this was then converted into the azide derivative P126b by treatment with NaN 3 . A second strategy involved the treatment of porphyrin P125 with NaNO 2 and NaN 3 in acidic medium allowing the preparation of derivative P127 with the azide group directly linked at the meso-phenyl group (Scheme 42). Scheme 42. Synthetic approach to porphyrin-based dyads P128 and P129 bearing triazolyl bridge.
Porphyrin-azide adduct P126b was reacted with porphyrin-alkyne P67 via a click CuAAC reaction in the presence of CuSO 4 /ascorbate catalytic system affording the triazolyl bridged porphyrin dyad P128a at a 42% yield. Dyad P129a was synthesized (41% yield) by following an analogous synthetic approach; however, since the azide is directly linked to a phenyl ring, the replacement of the CuSO 4 /ascorbate catalytic system by catalytic amounts of CuI in the presence of DIPEA was required [82,83]. Basic hydrolysis of the methyl ester groups of precursors P128a and P109a gave rise to target porphyrinic dyads P128b and P129b at a 91% yield. Compound P129b, with a smaller triazolyl bridge, displayed a higher photovoltaic performance than the analogous derivative P128b (5.1% vs. 3.8%) when attached to titanium oxide via the carboxyl group [82].
Boyle, Chudasama and co-workers developed a method for the site-selective modification of a full antibody to provide a defined multi-porphyrin antibody conjugate, which showed to be active in vitro [84].
The experimental procedure involved the synthesis of the dibromopyridazinedione-strained alkyne 35 that was used for the functional re-bridging of the disulfides of trastuzumab (34, a clinically approved antibody for the treatment of breast cancer) (Scheme 43). The insertion of the dibromopyridazinedione 35 into the disulfide bonds was carried out by adding tris(2-carboxyethyl)phosphine (a reducing agent) to a solution of trastuzumab in borate buffered saline solution (pH 8.0), containing an excess of 35, and the mixture was incubated at 4 • C for 16 h. The resulting conjugate 36 was then coupled with the water-soluble porphyrin azide P130 to afford the trastuzumab-porphyrin conjugate P131. This conjugate P131 was formed in near quantitative yield in the reaction of conjugate 36 with five equivalents of the porphyrin azide for 4 h at 37 • C. Conjugate P131 exhibited remarkable abilities to eradicate HER2+ cells (ca. 90% kill) while, at the same concentration, HER2-cells were unaffected. In contrast, trastuzumab alone showed, on irradiation, no cytotoxicity and minimal cytotoxicity was observed for the unconjugated porphyrin P130 with both cell lines.

Scheme 43. Synthesis of a trastuzumab-porphyrin conjugate.
Boyle and co-workers also used the tricationic porphyrin P130 to produce a molecular theranostic agent suitable for use as a PET radiotracer and as a photosensitizer for PDT [85]. These authors developed a procedure to prepare 69/71 Ga and 68 Ga radiolabeled azide-functionalized porphyrins P132, and the metalated porphyrins were then bioconjugated to the alkyne-functionalized dodecapeptide TWYKIAFQRNRK 37 (Scheme 44). This peptide exhibits a good affinity for the α6β1-integrin, which is involved in cellular migration and is also upregulated in multiple cancers, including breast carcinomas and glioblastomas. The radiolabeled peptide-porphyrin conjugate P133 exhibited excellent ability to eradicate the high integrin expressing HeLa cell line (ca. 80% kill), while at the same concentration showed considerably lower cell killing in a cell line displaying minimal integrin expression (U87). The CuAAC reaction was used to synthesize different generations (from 1 to 3) of porphyrin-cored dendrimers consisting of siloxane-poly(amidoamine) dendron-like arms (Si-PAMAM) as illustrated in Scheme 45 for G-1 [86]. The encapsulation of the porphyrin unit inside the dendritic Si-PAMAM shell (P135) allowed to increase the solubility of porphyrins and to prevent their aggregation in aqueous solutions. The porphyrin-cored dendrimers showed strong fluorescence emission and the intensity increased significantly as the generation is growing. The dendrimers also showed to be pH-responsive in fluorescence emission, indicating that they may be able to act as pH-responsive probes suitable for biomedical imaging, diagnosis and treatment.
More recently, the same research group, and using the same approach, reported the synthesis of glycosylated porphyrin-cored Si-PAMAM dendrimers [87]. The new glycoporphyrin dendrimers exhibited high fluorescence quantum yields and singlet oxygen production efficiency, but also showed specific recognition of lectin and temperature-responsive property (20-80 • C). In 2019, Zhang and co-workers reported the synthesis of an amphiphilic polymer P136 (Figure 11) that can be self-assembled into micelles with excellent stability, ultra-fast sensitivity of redox-triggered porphyrin release, and significant photodynamic anticancer performance [88]. The in vitro biological results revealed that these micelles can effectively enhance the cellular uptake and cellular internalization of porphyrin, have an extremely low dark toxicity, and are efficient towards A549 cells upon light irradiation with visible LED lamp. A successful conjugation of hemin to the G-quadruplex DNA was reported in 2017 [89]. The G-quadruplex-hemin conjugate P138 was prepared from the reaction of the azide-modified hemin P137 with a modified PS2.M oligonucleotide (5'GTG GGT AGG GCG GGT TGG3') 38 with dibenzoazacyclooctynyl groups using the Cu-free strain-promoted alkyne-azide cycloaddition approach (Scheme 46). The PS2.M-hemin conjugate exhibits peroxidase activity and catalyzes the oxidation of fluorogenic substrate 4-(1-methylhydrazineyl)-7-nitrobenzo[c] [1,2,5]oxadiazole but with slightly lower efficiency when compared to PS2.M/hemin complex. In 2018, Tsourkas and co-workers reported a new approach for the site-specific labeling of dye-stabilized nanoemulsions with affibodies for cellular targeting [90]. The method involved the preparation of dye-stabilized nanoemulsions bearing azide-handles on the surface. The azide groups were then conjugated with site-specific affibodies labeled at the C-terminus with a constrained alkyne, namely dibenzoazacyclooctyne. The affibody-conjugated nanoclusters were prepared by the Cu-free strain-promoted alkyne-azide cycloaddition approach by reacting the azide-functionalized nanoclusters (based on P139 and 39) with alkyne-functionalized affibody 40 in PBS for 12 h at room temperature with shaking ( Figure 12).

Figure 12.
Structures of the azide-dyes and alkyne-affibody used for the preparation of affibody-conjugated nanoclusters.
Lay, Gallo and co-workers report the synthesis of iron and ruthenium glycoporphyrins using the CuAAC approach [91]. The new derivatives were used as catalysts in cyclopropanation and aziridination reactions using diazo compounds and aryl azides as carbene and nitrene precursors, respectively (see Section 3.). As illustrated in Scheme 47, glycoporphyrins P141 were obtained from the reaction of tetraazidoporphyrin P140 and the methyl 4-O-propargyl-α-d-glucopyranoside 41.

Scheme 47.
Synthesis of glycoporphyrins used as catalysts in cyclopropanation and aziridination reactions.

Other Reactions Involving Azides
The reactivity of porphyrin derivatives with azide compounds was also explored in other synthetic routes not involving alkyne derivatives. For instance, Vroemans and co-workers described a synthetic method using 2-formyl-meso-tetraphenylporphyrin P142 as starting material in a catalyzed multicomponent synthetic approach [92]. The reaction of P142 with benzyl azide and nitroalkanes was performed under a nitrogen atmosphere in the presence of a mixture of p-toluenesulfonic acid/morpholine salt in catalytic amounts and butylhydroxytoluene as an anti-oxidant. When the reaction was carried out with the non-activated nitromethane the 2-(1,5-disubstituted-1,2,3-triazolyl)porphyrin derivative, P143 (R = H) was obtained at a 22% yield. However, the reaction with ethyl nitroacetate afforded 2-(1,4,5-trisubstituted-1,2,3-triazolyl)porphyrin P143 (R = CO 2 Et) at a 45% yield (Scheme 48). Similar yields were obtained when the reaction was performed with the Zn(II) complex of 2-formylporphyrin P142-Zn. All these reactions were regioselective. The Schmidt reaction conditions, involving the treatment of ketones with sodium azide in an acidic medium, are usually used to prepare lactams from cyclic ketones by ring expansion and is an alternative to the Beckmann rearrangement. However, when Brückner and co-workers treated the octaethylporphyrin derivative P144 under the Schmidt reaction conditions a α-hydroxyoxochlorin derivative was isolated (P145), or meso-chlorinated products (P146), instead of the expected lactams [93]. The reaction with the oxochlorin P144 required the use of a large excess of NaN 3 and conc. H 2 SO 4 at 130 • C. After 2 h of reaction, a green product corresponding to P145 was isolated as a racemic mixture with a 23% yield (Scheme 49). The authors proposed that the mechanistic pathway starts with the formation of an azidohydrin derivative, followed by dehydration, the formation of a nitrene by nitrogen elimination and 1,3-migration of an ethyl group, and finally hydrolysis of the ethylamine derivative. The attempt to perform the Schmidt reaction with conc. hydrochloric acid instead of sulfuric acid led to the formation of meso-chlorinated products P146a-f [93]. The treatment of 10-bromo-5,15-diarylporphyrins P147 with sodium azide allowed for the preparation of the corresponding azidoporphyrin derivative as an intermediate, in high yields, through an aromatic nucleophilic substitution. The quickly in situ reduction of the azide intermediate by sodium ascorbate and K 2 CO 3 afforded the amino derivative P148 in yields ranging from 80% to 90% (Scheme 50) [94]. This procedure shows to be an excellent alternative to the previous method reported, involving the thermal decomposition of the azide (yields up to 44%).

Scheme 50.
Preparation of meso-aminoporphyrin P148 by reduction of the corresponding azide intermediate.
The authors extended this method to the free-base 5,15-dibromodiarylporphyrins, however only the mono-aminated derivative was isolated from the reaction (80% yield). When the S N Ar reaction was performed with the Ni(II) complex of the 5,15-dibromodiarylporphyrin the diaminated P149a was obtained; notwithstanding, this showed to be easily oxidized to the corresponding quinone derivative during purification. However, the crude of the Ni(II) complex of meso-substituted diaminoporphyrin P149a was successfully converted into the stable bis(trifluoroacetamide) P149b analogous ( Figure 13) [94].

Catalytic Applications and Azide Transformations
Amongst several catalytic systems that have been reported for nitrene transfer reactions, some porphyrin metal complexes attracted attention as efficient catalysts. On the other hand, organic azides revealed to be potentially green resources to transfer nitrenes, since the only by-product formed throughout the generation of nitrenes from azides is dinitrogen. Alkene aziridination and C-H amination are among the most common uses for nitrene transfer reactions, although several other very promising applications have been recently explored [95][96][97][98][99][100][101][102][103][104]. Those most recent developments employing porphyrin metal complexes as efficient catalysts are summarized below.
Molecules containing an aziridine functional group are a versatile class of organic synthons due to the presence of a strained three-membered ring, which can be easily involved in ring-opening reactions and the aziridine functionality often shows interesting pharmaceutical and/or biological behaviors. Gallo and collaborators recently reviewed the most important results on the catalytic activity of iron and ruthenium porphyrin complexes in the aziridination of alkenes (Scheme 51) [105]. An interesting and straightforward two-step synthesis of N-arylaziridines starting from azides and anilines was efficiently achieved under continuous flow conditions, which was highlighted in the Gallo´s review [106]. The same group was able to prepare several iron and ruthenium glycoporphyrin complexes P141 (Scheme 47) and reported their uses as catalysts in the aziridination of α-methylstyrene in the presence of several aryl azides. The yields obtained ranged from 10% to 96%, depending on the porphyrin metal complex and also on the substitution pattern at the phenyl ring of the azide [91].
Zhang and collaborators reported that aryloxysulfonyl azides (ArOSO 2 N 3 ) can be successfully activated by the Co(TPP) catalyst, at room temperature, for selective aziridination of alkenes (Scheme 53). This Co(TPP)-based catalytic aziridination showed to be adequate for the reactions of several olefins and aryloxysulfonyl azides, thus generating N-aryloxysulfonyl aziridine derivatives in good to excellent yields [108].
As already stated the activation of organic azides by a metal-based catalyst allows selective amination reactions through the formation of active nitrene intermediate species [114]. The C-H bond amination reactions catalyzed by hemoproteins, iron porphyrins and phthalocyanines, also by non-heme-type iron complexes, as well as metal-organic framework-supported iron complexes, were recently reviewed and will not be explored further ( [98] and references cited therein). Another review dealing with the use of ruthenium porphyrins to catalyze hydrocarbon amination reactions showed that this class of porphyrin catalysts are able to activate organic azides in the synthesis of several aza-derivatives. Various synthetic methodologies have been reported there, concerning ruthenium porphyrins as catalysts for nitrene transfer reactions in C-H bond amination reactions [114]. Intramolecular ring-closing C-H bond amination enables direct synthesis of various N-heterocycles from aliphatic azides. Pyrrolidines (Scheme 58), oxazolidines (Scheme 59), imidazolidines (Scheme 60), isoindolines and tetrahydroisoquinoline (Scheme 61) were obtained by de Bruin and collaborators in good to excellent yields in a single reaction step catalyzed by cobalt(II) porphyrins ( Figure 15). In the absence of an amine trapping agent, complete recovery of the starting material was observed after 16 h at 100 • C. The use of di-tert-butyldicarbonate (Boc 2 O) significantly enhances the reaction rate by preventing competitive binding of the resulting amine and the use of toluene gave better results than benzene. Marked improvements both in yield and TON were observed using Co(TMP) as catalyst in the formation of pyrrolidines, and this was the Co(II)porphyrin complex of choice for the other reactions reported. Furthermore, the enantioselective ring-closing amination reaction was achieved when using a chiral cobalt-porphyrin catalyst with four (1S)-(−)-camphanic ester substituents in the second coordination sphere. Low enantioselectivities (29% ee at 100 • C and 46% ee at 80 • C) were attained for the ring-closure of (4-azidobutyl)benzene [99]. Recently, the same group reported the first example of the use of a cobalt(II) corrole complex in nitrene transfer reactions, more precisely in the ring-closing C-H amination of an aliphatic azide in the presence of Boc 2 O, giving the corresponding Boc-protected pyrrolidine (Scheme 58) [115]. Moreover, Che and collaborators described the synthesis of an iron(III) N-heterocyclic carbene (NHC) porphyrin complex, [Fe(III)(TDCPP)(IMe) 2 ]I (TDCPP = meso-tetrakis(2,6-dichlorophenyl)porphyrinato; IMe = 1,3-dimethylimidazol-2-ylidene), which was used to catalyze the intramolecular amination of C-H bonds with a wide range of alkyl azides in the presence of Boc 2 O and under thermal and microwave-assisted conditions [116].

Scheme 58.
Catalytic intramolecular ring-closing C-H bond amination starting from (4-azidobutyl)benzene as substrate with formation of the corresponding pyrrolidine. A broad and atom-economical method for the preparation of cyclic sulfoximines in up to 98% yield has been established by Bolm and collaborators via intramolecular imidation reactions of azido-substituted sulfoxides as nitrene precursors (Scheme 62). The catalytic process is based on a commercially available Fe(II) phthalocyanine (FePc) as catalyst for these for ring-closing sulfur imidations and exhibits large functional group tolerance. Toluene was chosen as the solvent and the yields were higher at 100 • C [117]. Under the best conditions studied and using 5 mol% of catalyst, several 3-azidoalkyl sulfoxides underwent intramolecular imidation, giving rise to the corresponding cyclized products with yields up to 98% (Scheme 63) [117].
Starting from aryl sulfoxides with benzylic azido groups, the corresponding intramolecular imidation products were also obtained with yields for benzo[d]isothiazoles ranging from 76%-98% (Scheme 64) [117]. A general method for the catalytic synthesis of o-aminoazobenzenes based on the commercially available Co(II) meso-tetraphenylporphyrin Co(TPP) was recently disclosed, which consists in a formal dimerization of two phenyl azides with simultaneous loss of two molecules of dinitrogen (Scheme 65) [118].
In the pursuit of using 2-OH and 2-NH 2 substituted aryl azides in Co(II) porphyrin catalyzed nitrene transfer reactions, de Bruin and collaborators found that from the ortho-hydroxy aryl azides both phenoxyzinones and benzoxazines could be synthesized in high yields, whereas from the 2-amino aryl azide substrates, azobenzenes were obtained as main products. So, the actually observed resulting compounds contradict the expected cobalt mediated catalytic coupling of nitrenes to alkynes or alkenes (Scheme 66). When the reaction was carried out in neat alkyne (thus using an excess of the alkyne), almost exclusive formation of phenoxyzinone was obtained [101]. Moreover, the aziridination of alkenes such as cyclohexene, styrene and 1-etoxyethene was also tried with o-azidophenol, although only phenoxyzinone was identified again as the major product. The authors were able to run this reaction with 2-azido-5-nitrophenol in the presence of Co(TPFPP) since Co(TPP) was inactive in this case, and the phenoxyzinone was not obtained in any case (Scheme 67) [101].

Corrole Macrocycles
Corroles are aromatic tetrapyrrolic macrocycles related to porphyrins but having only three methine bridges. Since 1999, the availability of these contracted macrocycles through facile synthetic strategies [119][120][121][122] allowed the scientific community to explore them successfully in different fields like catalysis, medicine (biomedical imaging, MRI contrast, cancer PDT, PDI of microorganisms, etc.) and in the design of photovoltaic materials and devices [123]. Among the different strategies used in the functionalization/post-functionalization of easily available corroles to obtain target molecules, the CuAAC approach is being successfully used. New corrole derivatives linked to other corroles [124], to porphyrins [125,126] and to BODIPYs [127] were successfully obtained, as recently reviewed by some of us [123]. In this section, only a few publications are being considered to illustrate the impact of the CuAAC approach in the functionalization of corroles.
For instance, in 2015, Desbois and co-workers selected this approach as a convenient and rapid synthetic access to heterobimetallic complexes based on corroles (Cor 2) and on porphyrins (P152 and P153) for medical imaging studies [128]. In the selection of the metal ions, authors took into account the potentiality of the Mn and Gd corrole complexes in the enhancement of contrast for MRI applications and of the "cold" metals complexes (Cu, Ga, In) for radionuclear imaging applications. The cycloaddition reactions were performed using azide Cor1 or P82 and the propargylamido-DOTA or propargylamido-NOTA in the presence of CuI and DIPEA under microwave irradiation. The expected complexes were obtained in yields ranging from 26% to 80% (Figure 16). Preliminary relaxivity assays seemed to indicate that these bimodal complexes are promising contrast agents in MRI. In 2017, Ngo and co-workers explored the CuAAC approach to develop the rotaxanes based on porphyrins and corroles presented in Scheme 68 [129]. The reaction involved the azide corrole Cor1, the alkynyl porphyrin P154 and the macrocycle 44 in the presence of CuI. The porphyrin-corrole dyad Cor3 was obtained in 98% yield. When the method was extended to the synthesis of the triad [3]rotaxane Cor4 and pentad [5]rotaxane Cor5 the desired derivatives were also isolated in excellent yields (96% and 70% respectively). The study showed that these interlocked scaffolds are sterically hindered although the electronic properties of the axle components were not affected and so such compounds can be explored in the development of new catalysts and materials. Scheme 68. Strategy used to obtain porphyrin-corrole dyad Cor3 and structures of the [3]rotaxane Cor4 and [5]rotaxane Cor5.
In 2019, Cao and co-workers showed that the attachment of the cobalt complex of corrole Cor6 as the azide component on carbon nanotubes functionalized with alkyne units leads to a system able to catalyze the 4e-reduction of O 2 to H 2 O (Scheme 69) [130]. Typically, mononuclear Co complexes mediate the two-electron reduction of O 2 to H 2 O 2 but authors demonstrated that with an adequate covalent bond interaction to carbon nanotubes, mononuclear Co corroles can also become intrinsically active for the 4e-oxygen reduction reaction of O 2 to H 2 O. Scheme 69. Synthetic strategy used to attach Cor6 to carbon nanotubes.

Final Remarks
Azides and pyrrolic derivatives are groups of compounds which play a key position in the synthesis of a large diversity of new derivatives, particularly those containing the triazole and porphyrin/corrole moieties. A wide range of potential applications are being put forward, particularly those related with light-harvesting, photovoltaic and photogalvanic devices, chemosensing, catalysis, and significantly on the medical side in MRI, drug delivery, hyperthermia therapy, in photodynamic therapy of cancer cells and in the photoinactivation of microorganisms.
The adaptation of the precursors to the final application requires some engineering in each component. In fact, different combinations with the alkynyl and azide components is in line with the excellence of the organic synthetic work developed under the context of CuAAC methods selected in this review. In most of the reported works the protection of the inner core of the macrocycle with a metal ion, mainly Zn(II) (which was easily removed by acid treatment), was required in order to avoid the coordination with Cu(II) ions. The used solvents comprised of general aqueous organic solvents but the use of only organic solvents does not seem to limit the applicability of the approach. Having in mind the significance of 1,2,3-triazoles in the medicinal and agricultural areas, it can be anticipated that the triazole unit might bring a synergic effect in the biological properties of the final target derivatives.
Besides the tremendous positive impact of the CuAAC reactions in organic synthesis, it must be stressed that they have some limitations. The main one is that CuAAC reactions are limited to terminal alkynes. These reactions lead to 1,4-disubstituted 1,2,3-triazoles but the regioisomeric 1,5-disubstituted 1,2,3-triazoles may be obtained selectively using the ruthenium(II)-catalyzed azide-alkyne cycloaddition. Other metal complexes (of silver, iridium, lutetium, scandium, and samarium) are also selective catalysts for the azide-alkyne cycloaddition reaction but they are rarely used [131]. Other methods for the synthesis of 1,2,3-triazole derivatives are also available in the literature. In some cases, they require very mild reaction conditions, including room temperature and aqueous media, and without the need to have alkyne derivatives.
Some metalloporphyrins are also excellent catalysts for nitrene transfer reactions leading to important transformations such as alkene aziridination and C-H amination suggesting that this partnership could be explored in more sophisticated systems.
Certainly, this scientific area in which this review is included will continue to be a "great target" in everyday objectives, mainly medicinal and industrial ones, of many researchers.

Conflicts of Interest:
The authors declare no conflict of interest.