Ferrocene-Catalyzed Aromatization and Competitive Oxidative Ring Transformations of 1,2-Dihydro-1-Arylpyridazino[4,5-d]Pyridazines

Abstract

This paper presents the expected and unexpected, but typically substituent-dependent, ferrocene-catalyzed DDQ-mediated oxidative transformations of a series of 5,8-bis(methylthio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines and 8-(3,5-dimethyl-1H-pyrazol-1-yl)-5-(methylthio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines. Under noncatalytic conditions the reactions were sluggish, mainly producing a substantial amount of undefined tarry materials; nevertheless, the ferrocene-catalyzed reactions of the 5,8-bis(methylthio)-substituted precursors gave the aromatic products the expected aromatic products in low yields. Their formation was accompanied by ring transformations proceeding via aryne-generating fragmentation/Diels–Alder (DA)/N₂-releasing retro Diels–Alder (rDA) sequence to construct arene-fused phthalazines. On the other hand, neither the noncatalytic nor the catalytic reactions of the 8-pyrazolyl-5-methylthio-substituted dihydroaromatics yielded the expected aromatic products. Instead, depending on their substitution pattern, the catalytic reactions of these pyrazolyl-substituted precursors also led to the formation of dearylated arene-fused phthalazines competing with an unprecedented multistep fragmentation sequence terminated by the hydrolysis of cationic intermediates to give 4-(methylthio)pyridazino[4,5-d]pyridazin-1(2H)-one and the corresponding 3,5-dimethyl-1-aryl-1H-pyrazole. When 0.6 equivalents of DDQ were applied in freshly absolutized THF, a representative pyrazolyl-substituted model underwent an oxidative coupling to give a dimer formed by the interaction of the cationic intermediate, and a part of the N-nucleophilic precursor remained intact. A systematic computational study was conducted on these intriguing reactions to support their complex mechanisms proposed on the basis of the structures of the isolated products.

1. Introduction

Representing privileged scaffolds among therapeutically active small molecules, structurally diverse pyridazines and their condensed analogs continue to attract considerable interest in pharmaceutical chemistry. Besides the bioactive pyridazine-based compounds found among anti-HIV [1], antiviral [2], antibacterial [3], antihypertensive [4], and anti-inflammatory [5] agents, pyridazine derivatives have also been identified as highly potent anti-cancer drug candidates [6,7,8,9,10]. On the other hand, much less attention has been paid to the pyridazino[4,5-d]pyridazine ring system. Only a limited number of its 1,4-disubstituted derivatives are known in the literature [11,12,13,14,15,16,17].

The introduction of different substituents into positions 5 and 8 seems to be a crucial point in the extension of the group of this class of heterocycles. In one of our previous papers we described the regioselective synthesis of a few 1,2-dihydro-1-arylpyridazino[4,5-d]pyridazines 1a–e and 2a–e (Scheme 1) achieved by substituent-directed addition of polar organometallic reagents on the 1,4-bis-methylthio- and 1-(3,5-dimethyl-1H-pyrazol-1-yl)-4-methylthio derivatives I and II, respectively, used as easily available precursors [18]. As an extension of our research focusing on the synthesis and functionalization of pyridazines, including condensed analogs with potential and proven antiproliferative activity [19,20,21,22,23], we envisaged the dehydrogenation of the aryl adducts 1a–e and 2a–e to access 3a–e and 4a–e, respectively, the aromatic products (Scheme 1) for antiproliferative assays aimed at the extension of SAR established for related heterocycles with fused pyridazine moieties [6,7,8,9,10,20,21,22]. On the other hand, the targeted bicyclic heteroaromatics can be considered as diazadiene components in inverse-electron demand Diels–Alder (DA) reactions followed by an N₂-eliminating retro-DA process (rDA) constructing phthalazines with versatile substituent patterns [13].

A large variety of non-catalytic and catalytic oxidative dehydrogenative aromatization methods have been elaborated for the oxidative dehydrogenation of saturated or partly saturated ring systems to diversely functionalized aromatic scaffolds of potential biological interest, including pyrroles, pyrazoles, imidazoles, oxazoles, thiazoles, quinolones, isoquinolines, pyridines, pyrimidines, triazines, pyridazines, and β-carbolines, as reviewed extensively [24,25]. Most of the reported conventional protocols are based on the use of MnO₂ [26,27], o-iodoxybenzoic acid (IBX), [28,29,30], NBS [31,32], NCS [33], di-tert-butyl peroxide (DTBP) [34], Pb(OAc)₄ [35], PhI(OAc)₂ [36], PhI(OTFA)₂ [37], S₈ [37,38], trichloroisocyanuric acid (TCCA) [39], I₂ [40], benzoyl peroxide (BPO) [41], ceric ammonium nitrate (CAN) [42], chromium(VI) reagents [43], selenium dioxide [44], and DDQ [45,46,47,48,49,50,51] as stoichiometric reagents. Catalytic methods have also been elaborated, exploring iodine [52,53] and a plethora of metal-based systems, nanoparticles, metal–organic frameworks, photoredox-activated complexes, and enzymes as catalysts, most of them combined with molecular oxygen as a terminal oxidant or used under the conditions of acceptorless dehydrogenation [54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74]. However, due to the involvement of highly specialized catalytic systems and experimental setups, these dehydrogenation protocols are often expensive and cannot be considered as scale-up strategies for the batch synthesis of the targeted compounds. On the other hand, DDQ has been identified as one of the most widely applicable stoichiometric and easy-to-handle oxidants in dehydrogenative aromatization of N-heterocycles. In the context of our synthetic goals, it is of pronounced importance that partially saturated pyridazines, such as in situ generated 1,6-dihydropyridazines, were also readily converted into pyridazines via oxidative aromatization by DDQ [50,51]. Accordingly, we first attempted to use this oxidant for converting 1a–e and 2a–e into aromatic pyridazino[4,5-d]pyridazines 3a–e and 4a–e, respectively (Scheme 1).

2. Results

In the first set of experiments, we attempted to perform simple non-catalytic DDQ-mediated aromatization of the bis-methylthio derivatives 1a–e. The reactions conducted for 5 h in the presence of DDQ (1.2 equiv.) in polar solvents (MeCN, THF, and DMSO) at 60 °C (Methods A–C: Scheme 2) allowed the isolation of the expected products 3a–e only in low yields (14–24%: Table 1). The TLC analysis of the reaction mixtures showed a significant reluctance of 1a–e to undergo any appreciable transformation below this temperature. However, under the conditions of Methods A–C, the dehydroaromatization reactions leading to 3a–e were accompanied by the formation of substantial amounts of decomposition products, preventing the recovery even of a small portion of the precursors. Under the same non-catalytic conditions, the attempted dehydrogenations of 2a,b,d,e gave no identifiable products, as these pyrazolyl-substituted precursors underwent practically complete decomposition. On the other hand, an angularly condensed naphtho-fused phthalazine (6c: Scheme 2) could be isolated in low yields (7–12%) from the mixtures obtained by the reactions of 2-naphthyl analog 2c (Table 1). Since the aforementioned noncatalytic reactions proved to be unsatisfactory from the point of view of our synthetic purposes, we attempted to perform the dehydroaromatizations of 1a–e and 2a–e in the presence of a catalytic amount (5%) of ferrocene expected to act as a redox catalyst [75,76,77,78,79,80,81] capable of undergoing fast and kinetically uncomplicated reversible redox change by single electron transfer (SET), which might be beneficial in terms of activating mechanistic pathways leading to the targeted bicyclic aromatic products 3a–e and 4a–e with enhanced selectivity. Accordingly, with the intention of improving the synthetic outcome of the synthetic experiments by circumventing uncontrolled transformations, the reactions of 1a–e were performed under the conditions of Methods D–F using shortened reaction time (3 h) to circumvent or suppress further undesired transformations of the products into tarry materials. The reactions run at 60 °C afforded not only 3a–e but also fused phthalazines (5a and 5bc) along with I derived from a specific dearylation process, as interpreted in detail in the next session. The progress of the reactions was monitored by TLC, which indicated again a definite slowdown of the detectable transformations of the substrates below 60 °C. All products were isolated in low-to-mediocre yields (Scheme 2, Table 1). Due to molecular symmetry, both 1-naphthyl- and 2-naphthyl-substituted precursors 1b and 1c gave identical angularly condensed naphtho[2,1-g]phthalazine (5bc) in yields comparable to those achieved in the isolation of the appropriate aromatic products of type 3. It must be pointed out that under the same conditions 4-methoxyphenyl derivative 1d was converted into 3d as the exclusively isolable product in acceptable-to-good yields (68–80%: Table 1).

Scheme 2. Non-catalytic and catalytic aromatization reactions of the 1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines and competitive oxidative ring transformations.

Scheme 2. Non-catalytic and catalytic aromatization reactions of the 1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines and competitive oxidative ring transformations.

Table 1. Methods and product distributions of the DDQ-mediated oxidative transformations of 5,8-bis(methyl-thio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines 1a–e and 8-(3,5-dimethyl-1H-pyrazol-1-yl)-5-(methylthio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines 2a–e.

Entry	Precursor	Product(s)	Yields a by A (%/%)	Yields a by B (%/%)	Yields a by C (%/%)	Yields a by D (%/%)	Yields a by E (%/%)	Yields a by F (%/%)
1	1a	3a/5a/Ia	14/–/–	16/–/–	18/–/–	37/24/5	45/22/6	43/30/6
2	1b	3b/5bc/I	17/–/–	21/–/–	19/–/–	35/34/7	41/35/6	39/32/7
3	1c	3c/5bc/I	16/–/–	16/–/–	19/–/–	31/24/5	37/34/11	32/40/8
4	1d	3d	17	20	24	74	68	80
5	1e	3e/I	17/–	20/–	22/–	59/18	47/20	64/24
6	2a	6a/II/7/8a	–/–/–/–	–/–/–/–	–/–/–/–	19/20/7/23	26/13/9/19	30/19/8/24
7	2b	6c/II/7/8a	–/–/–/–	–/–/–/–	–/–/–/–	20/10/17/12	20/8/24/18	26/9/32/25
8	2c	6c/II/7/8a	8/–/–/–	12/–/–/–	7/–/–/–	25/8/36/21	22/10/41/27	25/9/43/32
9	2d	7/8d	–/–	–/–	–/–	44/37	37/44	46/49
10	2e	II	–	–	–	52	59	64

^a the yields of product pairs 5/I, 6/II, and 7/8 listed in table cannot be summarized, as their members originated from common precursor molecules (cf. Scheme 3).

Table 1. Methods and product distributions of the DDQ-mediated oxidative transformations of 5,8-bis(methyl-thio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines 1a–e and 8-(3,5-dimethyl-1H-pyrazol-1-yl)-5-(methylthio)-1-aryl-1,2-dihydropyridazino[4,5-d]pyridazines 2a–e.

Entry	Precursor	Product(s)	Yields a by A (%/%)	Yields a by B (%/%)	Yields a by C (%/%)	Yields a by D (%/%)	Yields a by E (%/%)	Yields a by F (%/%)
1	1a	3a/5a/Ia	14/–/–	16/–/–	18/–/–	37/24/5	45/22/6	43/30/6
2	1b	3b/5bc/I	17/–/–	21/–/–	19/–/–	35/34/7	41/35/6	39/32/7
3	1c	3c/5bc/I	16/–/–	16/–/–	19/–/–	31/24/5	37/34/11	32/40/8
4	1d	3d	17	20	24	74	68	80
5	1e	3e/I	17/–	20/–	22/–	59/18	47/20	64/24
6	2a	6a/II/7/8a	–/–/–/–	–/–/–/–	–/–/–/–	19/20/7/23	26/13/9/19	30/19/8/24
7	2b	6c/II/7/8a	–/–/–/–	–/–/–/–	–/–/–/–	20/10/17/12	20/8/24/18	26/9/32/25
8	2c	6c/II/7/8a	8/–/–/–	12/–/–/–	7/–/–/–	25/8/36/21	22/10/41/27	25/9/43/32
9	2d	7/8d	–/–	–/–	–/–	44/37	37/44	46/49
10	2e	II	–	–	–	52	59	64

^a the yields of product pairs 5/I, 6/II, and 7/8 listed in table cannot be summarized, as their members originated from common precursor molecules (cf. Scheme 3).

Scheme 3. Summary of the modeling-supported pathways that might lead to the versatile products formed in the dehydroaromatization reactions of selected models 1a–c and 2a–c. The ΔG values calculated for the energetics of the elementary steps considered most relevant to estimating the relative feasibility of the competitive pathways are given in [kcal/mol].

Scheme 3. Summary of the modeling-supported pathways that might lead to the versatile products formed in the dehydroaromatization reactions of selected models 1a–c and 2a–c. The ΔG values calculated for the energetics of the elementary steps considered most relevant to estimating the relative feasibility of the competitive pathways are given in [kcal/mol].

Irrespective of the nature of the aryl group bonded to the C1 position, the reactions of 2a–e carried out by Methods D–F did not provide the expected bicyclic heteroaromatics 4a–e. Instead, besides fused phthalazines 6a,c, two dearylated pyridazino[4,5-d]pyridazine derivatives 7 and II, along with N-arypyrazoles of type 8a–d could be isolated from the reaction mixtures in low-to-mediocre yields (Scheme 2 and Table 1).

It was intriguing that both 2b and its isomer 2c were converted into identical angular naphtho-fused pyrazolyl-phthalazine (6c), of which isomer 6b could not be isolated from the mixtures obtained by the reactions of 2b. While phthalazines were formed only in the reactions of 2a–c, under the applied conditions each member of the investigated group of precursor of type 2 gave dearylated product 7. It must be pointed out here that the definite complementarity clearly reflected from the structures of 7 and the isolated N-arylpyrazoles 8a–d suggests that these products are generated via connected fragmentation pathways associated with intramolecular N-arylation and hydrolytic fission of the pyrazolyl substituent on the pyridazino[4,5-d]pyridazine scaffold, as will also be discussed in the next session. On the other hand, we also observed another intriguing difference in the outcomes of the experiments performed with 2d and 2e. While the reactions of 2d performed by Methods D–F afforded only 7 and 8d in mediocre but comparable yields, the reactions of thienyl-substituted precursor 2e gave dearylated aromatic heterocycle II as the exclusively isolable product (Table 1). It must be mentioned here that we could not register analysable NMR spectra of the multicomponent mixtures of the crude products contaminated by a substantial amount of tarry materials with paramagnetic substances destroying observable evolution of NMR signals.

3. Discussion

Apart from the realization of the expected dehydroaromatizations 1a–e → 3a–e, the intriguing outcome of our synthetic experiments obviously needs to be rationalized by such mechanistic pathways that can provide reasonable rationale for the formation of a number of unexpected products. However, first it is necessary to declare that besides 1a–e → 3a–e, the analogous dehydroaromatizations 2a–e → 4a–e are also feasible from the point of view of their thermodynamics, reflected from the change in free energy associated with these transformations, which is almost invariant to the substituents of the bicyclic heteroaromatic framework (ΔG in [kcal/mol] for the overall reactions 1 + DDQ → 3 + DDQH₂: a −33.6; b −33.2; c −33.3; d −36.6; e −34.5./ΔG in [kcal/mol] for the overall reactions 2 + DDQ → 4 + DDQH₂: a −32.9; b −32.2; c −33.8; d −34.8; e −33.9). Thus, it was necessary to generate mechanistic pathways comprising such specific elementary steps that they were suitable to account for the intriguing formation of the unexpected products isolated and identified in our experiments (Scheme 3). Since the catalytic reactions of 1a–c and 2a–c allowed the isolation of all types of products first we focused on the transformations of these precursors to support our view about the complex mechanisms (Scheme 3).

In order to reveal the feasibility of the competing pathways assumed to construct the different structures, we performed a series of comparative DFT modeling studies on the most relevant elementary steps expected to be involved in branched pathways (Scheme 3). In the course of the DFT calculations, the optimization and subsequent frequency calculation on molecular structures were carried out with M06-2X global functional [82] using the 6-31 G(d,p) basis set [83] supplemented by the IEFPCM solvent model [84]. For these computations dielectric constant was set to 47 to represent the polarity of DMSO used as a solvent by Method F, which—in most cases—provided the highest yields of the isolable products.

Prior to analysis of the pathways assumed to be operative in the ferrocene-catalyzed conversions, we calculated the energetics of the non-catalyzed versions of the reactions of 1a–c, which produced isolable amounts of 3a–c. Accordingly, we supposed that the DDQ-mediated hydrogen abstractions proceed via two-step pathways through radical intermediates 9a–c and 13a–c, respectively. The non-catalyzed oxidation can also take place via cationic intermediates 10a–c formed by hydride transfer from 1a–c to DDQ (Scheme 3). Although the overall transformations proved to be favored in terms of thermodynamics, the formation of the highly reactive intermediates is endoergic irrespective of their substitution pattern. We assumed that in the presence of ferrocene the reactions of 1a–c and 2a–c are initiated by a catalyzed single electron transfer (SET) from the substrate to DDQ with the involvement of the Fc/Fc⁺ redox pair in a catalytic cycle resulting in intermediate ion pairs 11a–c and 12a–c composed of a radical cation and radical anion DDQ^•−. Although the calculations indicated that for each of the studied models, this type of SET is unfavored in terms of thermodynamics but can probably be considered as kinetically favored due to ferrocene-catalysis that produces 11a–c and 12a–c in an amount sufficient to maintain a steady state, further advancing the overall reactions towards completion. The formation of the variety of isolable products was interpreted by branched mechanistic pathways starting from different types of thermodynamically favored exoergic specific collisions of the appropriate radical cation and DDQ^•− inside the ion pairs of 11a–c and 12a–c (Scheme 3). Two accessible variants for the recombination of 11a–c generating polycyclic and bicyclic sulfonium phenolate intermediates 19a–c and 27a–c, respectively, were analyzed by theoretical modeling. The results indicated that the elementary steps taking place by recombination with concomitant intramolecular cyclizations (11a–c → 19a–c) are exoergic, but based on the calculated energetic data, the formation of 27a–c seems even more favored in terms of thermodynamics. On the other hand, both the polycyclic and bicyclic sulfonium intermediates are supposed to be converted into the same ion pair of type 23 by the 1,2-elimination-tautomerization sequence 19a–c → 21a–c → 23a–c and sulfonium-mediated intramolecular aromatic electrophilic substitution 27a–c → 23a–c [85], respectively. Advancing further along the proposed mechanistic pathway, the polycyclic sulfonium intermediates 23a–c undergo deprotonation-initiated 1,4-elimination accompanied by concomitant skeletal fragmentation (23a–c → DDQH₂ + I + III or IV). Our calculations identified non-covalent pre-Diels–Alder complexes [86] (25a and 25bc) representing local minima on the potential energy surface as formed by exoergic associations from pyridazino[4,5-d]pyridazine I and the appropriate aryne intermediate III or IV. Finally, 25a and 25bc undergo Diels–Alder (DA) addition followed by N₂-releasing retro Diels–Alder (rDA) of the resulting bridged intermediates 17a and 17bc to afford benzo- and angularly fused naphthophthalazines 5a and 5bc, respectively. Minor portions of these phthalazines can also be formed along the multistep pathway comprising slightly exoergic trans-annular cyclization of the primarily generated radicals 13a–c followed by the H-abstraction-rDA sequence proceeding via intermediates of type 15 and 17. However, besides phthalazines, dearylated pyridazino[4,5-d]pyridazine I was also isolated from the reaction mixtures in yields lower relative to those of phtalazine products (Table 1), pointing to a competition between the formation of pre-Diels–Alder complexes 25a and 25bc and other aryne-consuming conversions [87]. An analogous sequence, which might lead to pyrazolyl-containing fused phthalazines 6a–c, was also analyzed by theoretical modeling. The calculated energetic values disclose a slightly endoergic character of the trans-annular cyclizations of the primarily generated radicals 14a–c, referring to a decreased feasibility of this type of multistep transformation.

We also carried out modeling studies on two versions of the recombination of appropriate radical cations and DDQ^•− inside ion pairs 12a–c, accompanied by simultaneous intramolecular cyclizations with the involvement of the proximal pyrazolyl and aryl groups, generating 1,2-diazepine-fused polyheterocycles 20a–c and spiroheterocycles 29a–c. The calculated data shows that both types of cyclization-assisted radical-radical recombination are markedly exoergic processes characterized by comparable changes in energetics. According to our proposed mechanism, polyheterocycles 20a–c undergo elimination of DDQH⁻ followed by facile tautomerization in the resulting cations associated with rearomatization of the fused carbocyclic ring (20a–c → 24a–c). In the subsequent steps, 24a–c can also be assumed to undergo fragmentation (24a–c → DDQH₂ + II + III or IV) followed by the formation of pre-Diels–Alder complexes and subsequent DA and rDA reactions finally affording benzo- and naphtho-condensed pyrazolylphthalazines of type 6. It is of particular interest that—according to the changes in the energetics accompanying aryne complexation—the association of II and IV leading to 26c is markedly favored over the formation of isomeric complex 26b. The relative energetics calculated for the alternative association modes of II and IV provides rationale for the outcome of one of the intriguing synthetic experiments reported in this contribution, disclosing that both 2b and 2c were converted into 6c, while 6b could not be isolated from the mixtures obtained by the reactions of 2b.

Our modeling studies also found that in a markedly exothermic manner, the ion pairs 12a–c are ready to be converted into spirocyclic intermediates 29a–c, which can be further transformed into 30a–c by 1,4-eliminiation of DDQH⁻ taking place by the cleavage of the C1-C1′ bond accompanied by simultaneous aromatization of the heterocycle and the aryl group. In the pathway-terminating step, 30a–c undergo hydrolysis with the water contamination of the solvents used to give heterocyclic lactam 7 and the corresponding 3,5-dimethyl-1-aryl-1H-pyrazole of type 8. Due to avoiding the formation of a carbon-sulfur bond in a strained spirocyclic framework with a five-membered ring, the structures of 28a–c could not be identified as local minima on the potential energy surface (PES), ruling out the collision-assisted spirocyclization of bis-methylthio-substituted radical ion pairs 11a–c and—in keeping with our experimental findings—the final generation of lactam 7 and thioethers 31a–c.

Finally, in the context of mechanistic pathways summarized in Scheme 3, we assume that the deprotonation of radical cations of type 12⁺ can preferably occur on the N-1 atom (cf. the slightly endoergic conversions of 12a–c → 14a–c) rather than from position 1, as deprotonation of C-1 is assumed to be sterically hindered by the rapidly double-twisting proximal pyrazolyl ring (according to molecular modeling, the full rotation of this bulky substituent with two pending methyl groups is prevented by the sterically congested molecular architecture). Accordingly, the abstraction of a hydrogen atom from the C-1 position of 2a–e is also regarded as a sterically hindered, kinetically unfavored elementary step. This view is supported by the H-bond transmitted correlation detected in the ¹H-¹⁵N-HMBC spectra of compounds of type 2 as we reported earlier [18].

Since 30a–c were obviously consumed by lactame-generating hydrolysis under the conditions of Methods D–F using solvents without prior absolutization, with the intention of trapping at least one representative of these key intermediates in a related molecular architecture, we reacted 2c with a sub-equimolar amount of DDQ at 60 °C in absolutized THF (Method G) to facilitate the construction of dimer 32c (Scheme 4) by a nucleophilic–electrophilic interaction between the portion of 2c that remained intact and cation 30. In accord with our expectation, 32c could also be isolated in not-negligible yield (16%) from the complex reaction mixture. We also managed to separate naphthophthalazine 6c, pyrazole 8c, and a THF-derived butoxy-substituted aromatic heterocycle 33 in moderate yields (Scheme 4a). Besides these products, a small amount (10%) of unreacted 2c was also recovered from the crude reaction mixture. With this experience in hand, disclosing the feasibility of the formation of a coupled heterocycle such as 32c, using equimolar amounts of 1c and 2c, we also performed a cross experiment under the conditions of Method G (Scheme 4b) to check indirectly whether the DDQ-mediated ferrocene-catalyzed primary oxidation of 2c, taking place via SET, thermodynamically is favored over that of 1c, as suggested by the changes in the free energy (ΔG = +22.1 kcal/mol and +17.0 kcal/mol calculated for elementary steps 1c → 11c and 2c → 12c, respectively (Scheme 3). It must be pointed out here that in spite of their endoergic character, the overall feasibility of these SET processes is presumably significantly increased by the involvement of ferrocene as an efficient redox catalyst. Thus, we hypothesized that the aforementioned difference in thermodynamics would eventually facilitate the selective formation of cation 30c, which can react with the unchanged portion of 1c to give the coupled product 34c (Scheme 3). This hypothesis was supported by the outcome of the experiment: besides 6c, 8c, and 33 derived from 2c and a small amount of 3c derived from a partial dehydroaromatization of 1c, 34c could also be isolated in moderate yield (19%).

Based on the results of further modeling studies, we suggest that under the conditions of Method G an endoergic solvolysis of cation 30c, effected by the excess of the solvent molecules, generates cationic intermediate 35, of which highly exoergic displacement reactions with the intact portions of 2c and 1c give 32c and 34c, respectively (Scheme 5). The formation of 33 can be interpreted by an exoergic hydride-transfer from a THF molecule opening the five-membered oxonium ring in 35.

As exemplified by the regioselective transformations of radical ion pairs 11a and 12a (Scheme 3), we have also performed further modeling studies to reveal the spatial distribution of the relevant delocalized molecular orbitals (MOs) of representative radical cations 11a⁺ and 12a⁺, determining the pathways to the isolated products (Scheme 6).

The singly occupied MO of 11a⁺ was found to be delocalized along the bis-methylthio-substituted aromatic pyridazine ring with substantial but comparable shares on the two sulfur centers attached to C-8 and C-5, suggesting at first sight a balanced competition under an assumed kinetic control between the two modes of recombination constructing sulphonium phenolates 28a and 28a*, respectively. However, the calculated energetic values identified 11a → 28a as the process that is thermodynamically favored over the alternative coupling 11a → 28a*. It is of note that in cation 11a⁺, the bonding MOs 88 and 89 indicate a marked phenyl–sulfur interaction, which can be considered as a contributing factor increasing the feasibility of the sulphonium-generation in the S-methyl group on position 8. In this context the MO analysis of 28 disclosed a substantial share of the LUMO on the sulphonium center implicated in an interaction with the proximal π-donor phenyl group (cf. MO-132) preformed for the cyclization yielding 23, of which subsequent multistep conversion finally leads to the formation of benzophthalazine 5a (Scheme 3 and Scheme 6). On the other hand, MO analysis of 28 also identified two bonding orbitals (MO-131 and MO-134) that represent a substantial π–π stacking interaction between the nearly parallel aromatic carbocyclic and heterocyclic fragments, contributing to the overall stability of the molecular architecture.

Underlying the proposed mechanistic pathways starting from radical ion pair 12 (Scheme 3), the singly occupied MO of cation 12a⁺ was found to be delocalized over the proximal interacting pyrazolyl and phenyl substituents with larger spin density on the phenyl ring. On the other hand, the electron delocalization between the interacting rings visualized by MOs 88 and 89 is assumed to assist the alternative modes of cyclization that necessarily accompany the colligation of the radical ion pair taking place on the phenyl ring (Scheme 3 and Scheme 6). It must also be emphasized here that in satisfactory agreement with the ratio of the yields of benzophthalazine 6a and lactam 7 (Table 1, entry 6), the two readily isolable representative products formed by two distinct mechanisms, the competing cyclization-assisted exoergic colligations 12a → 20a and 12a → 29a, can be characterized by comparable thermodynamics as reflected from the calculated data (Scheme 3 and Scheme 6).

Since the protocols applied for the dehydroaromatization of 1d,e and 2d,e provided product distributions markedly different from those obtained by the analogous reactions of 1a–c and 2a–c (Table 1), the energetic profile of the possible competitive pathways of their multistep transformations was also analyzed, focusing on the cascade-initiating steps that determine the direction of the overall transformations (Scheme 7). The exclusive formation of 3d from 1d can be reasoned by comparison of the energetics of the highly favored hydrogen abstraction 11d → 10d and the much less exoergic S-O colligation 11d → 27d. It is obvious that due to a conjugation effect, the 4-methoxyphenyl group substantially enhances the stability of cation 10d⁺. On the other hand, in line with the dominant formation of 3e and the low but reproducible yields of I (Table 1), we calculated somewhat more balanced energetics for the competitive transformations 11e → 10e and 11e → 27e. It is important to mention that competitive couplings 11a–c → 10a–c and 11a–c → 27a–c can be characterized by relatively balanced energetics starting branched pathways, finally constructing products 3a–c and 5a,bc (Scheme 3) in comparable isolated yields (Table 1). However, the aryne intermediates V and VI are assumed to undergo uncontrolled transformations rather than being involved in the DA-rDA sequence terminating the pathway (cf. Scheme 3) started by S-O colligation 11e → 27e.

The pyrazole-containing radical ion pair 12d undergoes a highly exoergic colligation assisted by the formation of spirocyclic intermediate 29d, of which hydrolysis-terminated multistep conversion leads to lactame 7 and pyrazole 8d, the products isolated in mediocre but comparable yields (Table 1). Probably due to the significant difference in the energetics of the alternative cascade-initiating steps 12d → 29d and 12d → 20d, none of the reactions of 2d led to the formation of II. On the other hand, the highly exoergic colligation 12e → 20e is efficiently assisted by the formation of a polyheterocyclic skeleton comprising a thieno [3,2-c][1,2]diazepine subunit with a combination of fused five- and seven-membered rings, which presumably accumulates less inherent strain compared to benzo[c][1,2]diazepines [in 20a (Scheme 3), 20d (Scheme 7)], naphtho [2,1-c][1,2]diazepine, and naphtho [1,2-c][1,2]diazepine [in 20b and 20c, respectively (Scheme 3)] generated by a combination of fused six- and seven-membered rings [88]. Based on the mechanistic considerations outlined in Scheme 3, the multistep transformation of 20e is expected to be terminated by a fragmentation resulting in II and the five-membered heteroarene VI, which, avoiding the formation of a pre-Diels–Alder complex, presumably undergoes fast aryne-consuming conversions [87]. Thus, the formation of II in the reactions of 2d as the exclusive product isolable in relatively high yields (Table 1) is in accord with the relative energetics calculated for the cyclization-assisted colligation 12e → 20e and the less favored spirocyclization formation of the spirocyclic intermediate 12e → 29e (Scheme 7).

The ¹H- and ¹³C NMR data of the novel 1,4-bis(methylthio)-5-aryl-pyridazino[4,5-d]pyridazines 3a–e, benzo- and naphtho-condensed phthalazines of types 5 and 6, lactame 7 pyrazoles of type 8, butyl ether 33, and coupled products 32c and 34c are consistent with their structures, but the following remarks are necessary to make.

In the ¹H-¹³C-HMBC spectrum of 6c, the H-12 singlet shows a three-bond correlation with the C-11 signal identified through its interaction with the protons of the methylthio group. Finally, besides the coupling pattern of the protons on the aromatic ring, the characteristic NOE interaction detected between protons H-1 and H-12 proves the angular arrangement of the tetracyclic skeleton of this naphtho-condensed phthalazine.

In the ¹H-NMR spectrum of 32c, the H-1′ signal is highly downfield-shifted (to 8.06 ppm) due to the cooperative anisotropic deshielding effects exerted by the proximal N-2 and N-p2 atoms situated in the bicyclic heteroaromatic aromatic skeleton and the pyrazole ring, respectively. The H-1′ signal of 34c discernible at 7.65 ppm is less, yet significantly downfield-shifted, referring to a decreased degree of anisotropic effect exerted by the N-2 atom. On the other hand, also indicating the connection of the two heterocyclic units, the highly downfield shifted H-1′′ and H-3′′ signals of 34c discernible at 7.81 ppm and 7.61 ppm, respectively, reflect the strong anisotropic deshielding effect of the proximal N-2 atom. In comparison, the H-1′′ and H-3′′ signals of the 2-naphthyl ring in the ¹H-NMR spectrum of 1c can be found at 6.97 ppm and 6.91 ppm, respectively [18]. However, it is the diagnostic three-bond correlation between H-1′ and C-1 atoms detected by the ¹H-¹³C-HMBC spectra of 32c and 34c that provides unequivocal evidence of the connection between the two heterocyclic fragments in their molecular architecture. The constitution of 34c gained additional support from a ¹H-¹⁵N-HMBC measurement [reference: δ(NH₃) = 0 ppm] indirectly detecting N-6 and N-7 resonances as identified at 396 ppm via their unresolved cross-peaks with H-5 and H-8 signals, N-3′ resonance at 335 ppm, as identified via its cross-peak with the H-1′ signal, and N-2′ resonance at 155 ppm, as identified via its cross-peak also with the H-1′ signal.

The expanded ¹H-¹³C-HMBC spectra of the unexpected products of enhanced structural complexity with graphically highlighted and tabulated correlations are found in the Supplementary Materials (3b: S.1.5, 3c: S.1.8, 5bc: S.1.16, 6a: S.1.21, 6c: S.1.26, 32c: S.1.44, 34c: S.1.51).

4. Materials and Methods

All fine chemicals were obtained from commercially available sources (Merck, Budapest, Hungary; Molar Chemicals, Budapest, Hungary; VWR, Budapest, Hungary) and were used without further purification. Tetrahydrofuran (THF) was freshly distilled under N₂ from sodium/benzophenone immediately prior to use under the conditions of Method G. Merck Kieselgel (230–400 mesh, 60 Å) was used for flash column chromatography. To achieve sufficient separation of the multiple products, the ratio of the separable mixture to silica was set at least to 1 g:100 g. Melting points (uncorrected) were determined with an M-560 instrument (Büchi, Essen, Germany). Elemental analyses were performed by a PerkinElmer 2400 CHNS elemental analyzer. Merck Kieselgel 60F254 plates were used for TLC monitoring the reaction mixtures. The NMR spectra were recorded in DMSO-d₆ or CDCl₃ solution in 5 mm tubes at RT on a Bruker DRX-500 spectrometer (Bruker Biospin, Karlsruhe, Baden Württemberg, Germany) at 500 (¹H), 125 (¹³C), and 50 (¹⁵N) MHz, with the deuterium signal of the solvent as the lock and TMS as the internal standard (¹H, ¹³C) and NH₃ (liq.) as the external reference (¹⁵N). The ¹⁵N NMR chemical shifts in compound 34c were assigned on the basis of the correlations revealed by its ¹H-¹⁵N HMBC spectrum. The ¹H-¹H-COSY, ¹H-¹³C-HSQC, ¹H-¹³C-HMBC, and ¹H-¹⁵N-HMBC spectra were registered by using the standard Bruker pulse programs. The energetic profile of the transformations was given by the changes in Gibbs free energy (∆G). The free energy values of the optimized structures were obtained by correcting the computed total energy with zero-point vibrational energy (ZPE) and thermal corrections. All calculations were carried out using the Gaussian 16 (revision c.01) software (Gaussian Incorporation, Pittsburgh, PA, USA) package [89]. The optimized structures are available from the authors.

4.1. General Procedure for the Dehydroaromatization of 1a–e and 2a–e Effected by DDQ Under the Conditions of Methods A–F

At room temperature DDQ (545 mg, 2.4 mmol, 1.2 equiv.) was added to the solution of 1-aryl-1,2-dihydropyridazino[4,5-d]pyridazine (2 mmol) in 20 mL of the solvent (MeCN by Methods A and D; THF by Methods B and E; DMSO by Methods C and F). When the reactions were conducted under the conditions of Methods D–E, prior to addition of DDQ. Ferrocene (18.6 mg, 0.1 mmol) was also dissolved in the reaction mixture. All reaction mixtures were stirred under argon at 60 °C for 5 h (Methods A–C) or 3 h (Methods D–F) (cf. Scheme 2). After the stirring was stopped, the mixture was allowed to cool down to room temperature and quenched with saturated NaHCO₃ solution (10 mL) and then diluted with water (100 mL). The resulting suspension was extracted with CH₂Cl₂ (5 × 40 mL). The combined organic layers were washed with brine, dried over Na₂SO₄, passed through a Celite pad, and concentrated to dryness. The thick brown oily residue was triturated with water. The resulting solid was filtered off, dried, and subjected to flash column chromatography on silica gel using petroleum ether/EtOAc (80:1–20:1, v/v) as eluent. The separated products were further purified by trituration with hexane/Et₂O (3:1–1:1, v/v). The yields of the products are listed in Table 1.

4.2. Procedure for the Reaction of 2c Effected by a Sub-Equimolar Amount of DDQ Under the Conditions of Method G

At room temperature ferrocene (9 mg, 5 mol%) and DDQ (136 mg, 0.6 mmol, 0.6 equiv.) were sequentially added to the solution of 2c (401 mg, 1 mmol, 1 equiv.) in THF (10 mL) distilled freshly under N₂ from sodium/benzophenone. The reaction mixture was stirred under argon at 60 °C for 3 h. After the stirring was stopped, the mixture was allowed to cool down to room temperature and quenched with saturated NaHCO₃ solution (5 mL) then diluted with water (50 mL). The resulting suspension was extracted with CH₂Cl₂ (5 × 20 mL). The combined organic layers were washed with brine, dried over Na₂SO₄, passed through a Celite pad, and concentrated to dryness. The resulting thick brown oil was triturated with water to obtain the mixture of the crude products, which were separated and further purified using the procedure described in Section 4.1. The yields of the products are presented in Scheme 4a.

4.3. Procedure for the Reaction of a 1:1 Mixture of 1c and 2c Effected by a Sub-Equimolar Amount of DDQ Under the Conditions of Method G

At room temperature ferrocene (9 mg, 5 mol%) and DDQ (136 mg, 0.6 mmol, 0.6 equiv.) were sequentially added to the mixture of 1c (176 mg, 0.5 mmol, 0.5 equiv.) and 2c (200 mg, 0.5 mmol, 0.5 equiv.) dissolved in THF (10 mL) distilled freshly under N₂ from sodium/benzophenone. The reaction mixture was stirred under argon at 60 °C for 3 h. After the stirring was stopped, the mixture was allowed to cool down to room temperature and quenched with saturated NaHCO₃ solution (5 mL) then diluted with water (50 mL). The resulting suspension was extracted with CH₂Cl₂ (5 × 20 mL). The combined organic layers were washed with brine, dried over Na₂SO₄, passed through a Celite pad, and concentrated to dryness. The resulting thick brown oil was triturated with water to obtain the mixture of the crude products, which were separated and further purified using the procedure described in Section 4.1. The yields of the products are presented in Scheme 4b.

4.4. Characterization of the Products

4.4.1. 1,4-Bis(methylthio)-5-phenylpyridazino[4,5-d]pyridazine (3a)

White powder. M.p. 133–134 °C. ¹H-NMR (CDCl₃): 9.85 (s, 1H, H-8); 7.63 (m, 1H, H-4′); 7.60–7.52 (overlapping m’s, 4H, H-2′,6′ and H-3′,5′); 2.90 (s, 3H, SCH₃ on C-1); 2.54 (s, 3H, SCH₃ on C-4). ¹³C-NMR (CDCl₃): 158.2 (C-5); 157.3 (C-4); 156.4 (C-1); 145.7 (C-8); 136.4 (C-1′); 130.8 (C-4′); 130.5 (C-3′,5′); 128.9 (C-2′,6′); 119.0 (C-4a); 117.0 (C-8a); 15.9 (SCH₃ on C-4); 13.6 (SCH₃ on C-1). Anal. calcd. for C₁₄H₁₂N₄S₂: C 55.98%; H, 4.03%; N, 18.65%; S, 21.35%. Found: C, 55.83%; H, 4.11%; N, 18.81%; S, 21.28%.

4.4.2. 1,4-Bis(methylthio)-5-(naphthalen-1-yl)pyridazino[4,5-d]pyridazine (3b)

White powder. M.p. 177–180 °C. ¹H-NMR (CDCl₃): 9.97 (s, 1H, H-8); 8.14 (d, J = 8.1 Hz, 1H, H-4′); 8.00 (d, J = 8.1 Hz, 1H, H-5′); 7.66 (dd, J = 8.1 Hz and 7.7 Hz, 1H, H-3′); 7.56 (dd, J = 8.4 Hz and 7.7 Hz, 1H, H-6′); 7.52 (d, J = 7.4 Hz, 1H, H-2′); 7.41 (dd, J = 8.4 Hz and 7.7 Hz, 1H, H-7′); 7.20 (d, J = 8.4 Hz, 1H, H-8′); 2.91 (s, 3H, SCH₃ on C-1); 2.36 (s, 3H, SCH₃ on C-4). ¹³C-NMR (CDCl₃): 157.4 (C-5); 157.0 (C-4); 156.1 (C-1); 145.9 (C-8); 133.5 (C-4′a); 133.2 (C-1′); 132.5 (C-8′a); 130.9 (C-4′); 129.0 (C-2′); 128.5 (C-5′); 127.2 (C-7′); 126.5 (C-6′); 125.3 (C-3′); 124.7 (C-8′); 118.4 (C-4a); 117.9 (C-8a); 15.6 (SCH₃ on C-4); 13.3 (SCH₃ on C-1). Anal. calcd. for C₁₈H₁₄N₄S₂: C 61.69%; H, 4.03%; N, 15.99%; S, 18.30%. Found: C, 61.50%; H, 4.15%; N, 15.80%; S, 18.42%.

4.4.3. 1,4-Bis(methylthio)-5-(naphthalen-2-yl)pyridazino[4,5-d]pyridazine (3c)

White powder. M.p. 199–201 °C. ¹H-NMR (CDCl₃): 9.85 (s, 1H, H-8); 8.03 (overlapping br s and d, J = 8.1 Hz, 2H, H-1′ and H-4′, respectively); 7.96 (d, J = 8.2 Hz, 1H, H-5′); 7.93 (d, J = 8.4 Hz, 1H, H-8′); 7.66 (dd, J = 8.1 Hz and 1.5 Hz, 1H, H-3′); 7.61 (dd, J = 8.2 Hz and 7.7 Hz, 1H, H-6′); 7.58 (dd, J = 8.4 Hz and 7.7 Hz, 1H, H-7′); 2.88 (s, 3H, SCH₃ on C-1); 2.45 (s, 3H, SCH₃ on C-4). ¹³C-NMR (CDCl₃): 158.3 (C-5); 157.5 (C-4); 156.4 (C-1); 145.6 (C-8); 134.5 (C-4′a); 133.8 (C-2′); 133.2 (C-8′a); 131.2 (C-1′); 129.2 (C-8′); 128.8 (C-4′); 128.4 (C-5′); 128.0 (C-6′); 127.3 (C-7′); 127.1 (C-3′); 119.2 (C-4a); 117.2 (C-8a); 15.9 (SCH₃ on C-4); 13.6 (SCH₃ on C-1). Anal. calcd. for C₁₈H₁₄N₄S₂: C, 61.69%; H, 4.03%; N, 15.99%; S, 18.30%. Found: C, 61.48%; H, 3.94%; N, 16.10%; S, 18.37%.

4.4.4. 5-(4-Methoxyphenyl)-1,4-bis(methylthio)pyridazino[4,5-d]pyridazine (5d)

Light yellow powder. M.p. 182–184 °C. ¹H-NMR (CDCl₃): 9.78 (s, 1H, H-8); 7.49 (d, J = 8.1 Hz, 2H, H-2′,4′); 7.08 (d, J = 8.1 Hz, 2H, H-3′,5′); 3.93 (s, 3H, OCH₃ on C-4′); 2.87 (s, 3H, SCH₃ on C-1); 2.54 (s, 3H, SCH₃ on C-4). ¹³C-NMR (CDCl₃): 161.7 (C-4′); 157.7 (C-5); 157.1 (C-4); 155.9 (C-1); 144.9 (C-8); 131.8 (C-2′,6′); 128.2 (C-3′,5′); 118.8 (C-4a); 114.1 (C-3′,5′); 116.7 (C-8a); 55.5 (OCH₃ on C-4′); 15.5 (SCH₃ on C-4); 13.2 (SCH₃ on C-1). Anal. calcd. for C₁₅H₁₄N₄OS₂: C, 54.53%; H, 4.27%; N, 16.96%; S, 19.41%. Found: C, 54.70%; H, 4.41%; N, 17.13%; S, 19.52%.

4.4.5. 1,4-Bis(methylthio)-5-(thiophen-2-yl)pyridazino[4,5-d]pyridazine (3e)

Light yellowish powder. M.p. 136–138 °C. ¹H-NMR (CDCl₃): 9.80 (s, 1H, H-8); 7.72 (d, J = 4.6 Hz, 1H, H-5′); 7.38 (d, J = 2.7 Hz, 1H, H-3′); 7.24 (dd, J = 4.6 Hz and 2.7 Hz, 1H, H-4′); 2.89 (s, 3H, SCH₃ on C-1); 2.59 (s, 3H, SCH₃ on C-4). ¹³C-NMR (CDCl₃): 157.2 (C-4); 156.2 (C-1); 152.6 (C-5); 145.6 (C-8); 135.9 (C-2′); 133.7 (C-3′); 130.9 (C-5′); 128.1 (C-4′); 119.2 (C4a); 117.7 (C-8a); 15.8 (SCH₃ on C-4); 13.6 (SCH₃ on C-1). Anal. calcd. for C₁₂H₁₀N₄S₃: C, 47.04%; H, 3.29%; N, 18.28%; S, 31.39%. Found: C, 47.21%; H, 3.32%; N, 18.09%; S, 31.23%.

4.4.6. 1,4-Bis(methylthio)benzo[g]phthalazine (5a)

Light yellowish powder. M.p. 184–186 °C. ¹H-NMR (CDCl₃): 8.65 (s, 2H, H-5,10); 8.12 (m, 2H, H-6,9); 7.69 (m, 2H, H-7,8); 2.83 (s, 6H, SCH₃). ¹³C-NMR (CDCl₃): 158.6 (C-1,4); 135.1 (C-5a, 9a); 129.5 (C-6,9); 128.9 (C-7,8); 124.7 (C-5,10); 122.2 (C-4a,10a); 13.5 (SCH₃). Anal. calcd. for C₁₄H₁₂N₂S₂: C, 61.73%; H, 4.44%; N, 10.28%; S, 23.54%. Found: C, 61.52%; H, 4.60%; N, 10.17%; S, 23.60%.

4.4.7. 8,11-Bis(methylthio)naphtho [1,2-g]phthalazine (5bc)

Light yellowish powder. M.p. 198–201 °C. ¹H-NMR (CDCl₃): 9.19 (s, 1H, H-12); 8.72 (br d, J = 7.7 Hz, 1H, H-1); 8.41 (s, 1H, H-7); 7.88 (br d, J = 7.7 Hz, 1H, H-4); 7.79 and 7.71 (A and B part of an AB spin system, J_AB = 9.1 Hz, 2x1H, H-6 and H-5, respectively); 7.74 (td, J = 7.7 Hz and 1.5 Hz, 1H, H-2); 7.70 (td, J = 7.7 Hz and 1.5 Hz, 1H, H-3); 2.85 (s, 3H, SCH₃ on C-8); 2.87 (s, 3H, SCH₃ on C-11). ¹³C-NMR (CDCl₃): 157.9 (C-11); 157.6 (C-8); 134.3 (C-6a); 132.9 (C-12a); 132.4 (C-4a); 130.6 (C-5); 129.6 (C-12b); 129.0 (C-4); 128.6 (C-3); 127.7 (C-2); 126.5 (C-6); 125.6 (C-7); 123.4 (C-1); 122.1 (C-7a); 122.0 (C-11a); 118.3 (C-12); 13.2 (s, 3H, SCH₃ on C-8); 13.1 (s, 3H, SCH₃ on C-11). Anal. calcd. for C₁₈H₁₄N₂S₂: C, 67.05%; H, 4.38%; N, 8.69%; S, 19.89%. Found: C, 66.90%; H, 4.46%; N, 8.77%; S, 19.92%.

4.4.8. 1-(3,5-Dimethyl-1H-pyrazol-1-yl)-4-(methylthio)benzo[g]phthalazine (6a)

Light yellow powder. M.p. 206–207 °C. ¹H-NMR (CDCl₃): 8.85 (s, 1H, H-10); 8.77 (s, 1H, H-5); 8.17 (d, J = 7.8 Hz, 1H, H-6); 8.12 (d, J = 7.8 Hz, 1H, H-9); 7.73 (t, J = 7.8 Hz, 1H, H-7); 7.69 (t, J = 7.8 Hz, 1H, H-8); 6.19 (s, 1H, H-4′); 2.93 (s, 3H, SCH₃ on C-3′); 2.52 (s, 3H, CH₃ on C-5′); 2.43 (s, 3H, CH₃ on C-3′). ¹³C-NMR (CDCl₃): 161.5 (C-4); 151.13 (C-3′); 151.08 (C-1); 143.3 (C-5′); 135.5 (C-9a); 135.0 (C-5a); 130.0 (C-9); 129.3 (C-6); 129.2 (C-7); 128.8 (C-8); 128.0 (C-10); 124.8 (C-4a); 124.0 (C-5); 119.7 (C-10a); 108.1 (C-4′); 14.2 (CH₃ on C-3′); 13.6 (SCH₃), 12.9 (CH₃ on C-5′). Anal. calcd. for C₁₈H₁₆N₄S: C, 67.47%; H, 5.03%; N, 17.49%; S, 10.01%. Found: C, 67.65%; H, 5.12%; N, 17.55%; S, 9.94%.

4.4.9. 8-(3,5-Dimethyl-1H-pyrazol-1-yl)-11-(methylthio)naphtho [1,2-g]phthalazine (6c)

Yellowish powder. M.p. 232–235 °C. ¹H-NMR (CDCl₃): 9.49 (s, 1H, H-12); 8.90 (br d, J = 8.1 Hz, 1H, H-1); 8.78 (s, 1H, H-7); 7.93 (br d, J = 8.0 Hz, 1H, H-1); 7.86 and 7.84 (A and B part of an AB spin system, J_AB = 9.2 Hz, 2x1H, H-5 and H-6, respectively); 7.80 (t, J = 8.0 Hz, 1H, H-2); 7.75 (t, J = 8.0 Hz, 1H, H-3); 6.18 (s, 1H, H-4′); 2.96 (s, 3H, SCH₃); 2.52 (s, 3H, CH₃ on C-5′); 2.43 (s, 3H, CH₃ on C-3′). ¹³C-NMR (CDCl₃): 161.4 (C-11); 151.2 (C-3′); 150.9 (C-8); 143.3 (C-5′); 135.4 (C-6a), 133.7 (C-12a); 133.1 (C-4a); 130.6 (C-5); 130.1 (C-12b); 129.5 (C-4); 129.3 (C-3); 128.2 (C-2); 127.6 (C-6); 127.3 (C-7); 125.5 (C-11a); 124.1 (C-1); 120.4 (C-7a); 118.4 (C-12); 108.2 (C-4′); 14.2 (CH₃ on C-3′); 13.8 (SCH₃), 12.9 (CH₃ on C-5′). Anal. calcd. for C₂₂H₁₈N₄S: C, 71.33%; H, 4.90%; N, 15.12%; S, 8.65%. Found: C, 71.60%; H, 4.99%; N, 15.25%; S, 8.71%.

4.4.10. 4-(Methylthio)pyridazino[4,5-d]pyridazin-1(2H)-one (7)

Yellow powder. M.p. 289–293 °C. ¹H-NMR (DMSO-d₆): 13.47 (s, 1H, NH); 9.84 (s, 1H, H-8); 9.71 (s, 1H, H-5); 2.57 (s, 3H, SCH₃). ¹³C-NMR (DMSO-d₆): 157.0 (C-1); 147.9 (C-8); 146.5 (C-5); 142.2 (C-4); 124.3 (C-4a); 121.8 (C-8a); 12.9 (SCH₃). Anal. calcd. for C₇H₆N₄OS: C, 43.29%; H, 3.11%; N, 28.85%; S, 16.51%. Found: C, 43.20%; H, 3.19%; N, 28.61%; S, 16.44%.

4.4.11. 3,5-Dimethyl-1-phenyl-1H-pyrazole (8a)

Thick light yellow oil. ¹H-NMR (DMSO-d₆): 7.47 (d, J = 4.2 Hz, 4H, H-2′,3′,5′,6′); 7.37 (m, 1H, H-4′) 6.06 (s, 1H, H-4); 2.28 (s, 3H, CH₃ on C-3); 2.19 (s, 3H, CH₃ on C-5). ¹³C-NMR ¹³C-NMR (DMSO-d₆): 148.3 (C-3); 140.2 (C-1′); 139.5 C-5); 129.5 (C-3′,5′); 127.3 (C-4′); 124.5 (C-2′,6′); 13.7 (CH₃ on C-5); 12.6 (CH₃ on C-3). Anal. calcd. for C₁₁H₁₂N₂: C, 76.71%; H, 7.02%; N, 16.27%. Found: C, 76.47%; H, 7.29%; N, 16.40%.

4.4.12. 3,5-Dimethyl-1-(naphthalen-1-yl)-1H-pyrazole (8b)

Thick light yellow oil. ¹H-NMR (CDCl₃): 7.95 (d, J = 8.1 Hz, 1H, H-8′); 7.95 (d, J = 8.1 Hz, 1H, H-5′); 7.58–7.48 (overlapping m’s, 4H, H-2′,3′,6′,7′); 7.36 (d, J = 7.9 Hz, 1H, H-4′); 6.10 (s, 1H, H-4); 2.39 (s, 3H, CH₃ on C-3); 2.08 (s, 3H, CH₃ on C-5). ¹³C-NMR (CDCl₃): 149.0 (C-3); 141.5 (C-5); 136.2 (C-1′); 134.2 (C-4′a); 130.9 (C-8′a); 125.4 (C-2′); 125.1 (C-8′); 123.2 (C-4′); 129.3 (C-5′); 127.3 (C-6′); 128.1 (C-3′); 126.6 (C-7′); 105.4 (C-4); 13.7 (CH₃ on C-5); 11.4 (CH₃ on C-3). Anal. calcd. for C₁₅H₁₄N₂: C, 81.05%; H, 6.35%; N, 12.60%. Found: C, 81.38%; H, 6.49%; N, 12.36%.

4.4.13. 3,5-Dimethyl-1-(naphthalen-2-yl)-1H-pyrazole (8c)

Thick yellow oil. ¹H-NMR (DMSO-d₆): 8.05–8.00 (three overlapping signals: two d’s, J = 8.9 Hz and 8.1 Hz, and a br s, 3H, H-4′, H-5′, and H-1′); 7.98 (d, J = 8.1 Hz, 1H, H-8′); 7.69 (dd, J = 8.9 Hz and 1.8 Hz,1H, H-3′); 7.58 (~t, J ~ 8 Hz, 1H, H-7′); 7.55 (~t, J ~ 8 Hz, 1H, H-6′); 6.11 (s, 1H, H-4); 2.38 (s, 3H, CH₃ on C-3); 2.22 (s, 3H, CH₃ on C-5). ¹³C-NMR (DMSO-d₆): 148.6 (C-3); 139.9 (C-5); 137.6 (C-2′); 133.4 (C-8′a); 131.9 (C-4′a); 129.3 (C-4′); 128.5 (C5′); 128.1 (C-8′); 127.3 (C-7′); 126.7 (C-6′); 123.4 (C-3′); 121.9 (C-1′); 107.8 (C-4); 13.8 (CH₃ on C-5); 12.8 (CH₃ on C-3). Anal. calcd. for C₁₅H₁₄N₂: C, 81.05%; H, 6.35%; N, 12.60%. Found: C, 81.68%; H, 6.49%; N, 12.48%.

4.4.14. 1-(4-Methoxyphenyl)-3,5-dimethyl-1H-pyrazole (8d)

Thick yellow oil. ¹H-NMR (CDCl₃): 7.32 (d, J = 8.5 Hz, 2H, H-2′,6′); 6.95 (d, J = 8.5 Hz, 2H, H-3′,5′); 5.97 (s, 1H, H-4); 3.83 (s, 3H, OCH₃); 2.29 (s, 3H, CH₃ on C-3); 2.24 (s, 3H, CH₃ on C-5). ¹³C-NMR (CDCl₃): 158.8 (C-4′); 148.4 (C-3); 139.6 (C-5); 132.9 (C-1′); 126.4 (C-2′,6′); 114.1 (C-3′,5′); 106.7 (C-4); 55.5 (OCH₃); 13.4 (CH₃ on C-5); 12.1 (CH₃ on C-3). Anal. calcd. for C₁₂H₁₄N₂O: C, 71.26%; H, 6.98%; N, 13.85%. Found: C, 71.59%; H, 7.03%; N, 13.71%

4.4.15. 8′-(3,5-Dimethyl-1H-pyrazol-1-yl)-4,5′-bis(methylthio)-1′-(naphthalen-2-yl)-1′H- 1,2′-bipyridazino[4,5-d]pyridazine (32c)

Yellowish powder. M.p. 245–247 °C. ¹H-NMR (CDCl₃): 10.53 (d, J = 1.1 Hz, 1H, H-8); 9.72 (d, J = 1.1 Hz, 1H, H-5); 8.064 and 8.059 (2xs, 2H, H-4′ and H-1′, resp.); 7.62 (m, 1H, H-5″); 7.55 (d, J = 8.6 Hz, 1H, H-4″); 7.53 (m, 1H, H-8″); 7.35–7.30 (m. 2H, H-6″ and H-7″); 7.18 (br s, 1H, H-1″); 6.91 (dd, J = 8.6 Hz and 1.6 Hz, 1H, H-3″); 6.07 (s, 1H, H-p4); 2.77 (s, 3H, SCH₃ on C-5′); 2.70 (s, 3H, SCH₃ on C-4); 2.35 (s, 3H, CH₃ on C-p5); 2.04 (s, 3H, CH₃ on C-p3). ¹³C-NMR (CDCl₃): 156.7 (C-5′); 155.4 (C-4); 152.5 (C-p3); 150.7 (C-8′); 149.7 (C-1); 148.8 (C-8); 146.4 (C-5); 142.8 (C-p5); 136.4 (C-2″); 133.42 and 133.28 (two hardly separated lines, C-4′ and C-4″a); 133.3 (C-8″a); 129.2 (C-4″); 128.4 (C-5″); 127.9 (C-8″); 126.87 and 126.83 (C-6″ and C-7″, interchangeable assignments); 125.4 (C-1″); 125.1 (C-8′a); 123.9 (C-3″); 120.5 (two coalesced lines C-4a and C-4′a); 113.1 (C-8a); 108.7 (C-p4); 53.8 (C-1′); 13.7 (SCH₃ on C-5′); 13.5 (SCH₃ on C-4); 14.4 (CH₃ on C-p5); 12.3 (CH₃ on C-p3). Anal. calcd. for C₂₉H₂₄N₁₀S₂: C, 60.40%; H, 4.19%; N, 24.29%; S, 11.12%. Found: C, 60.62%; H, 4.27%; N, 24.20%; S, 11.03%.

4.4.16. 1-Butoxy-4-(methylthio)pyridazino[4,5-d]pyridazine (33)

Light yellow powder. M.p. 96–99 °C. ¹H-NMR (CDCl₃): 9.89 (d, J = 1.2 Hz, 1H, H-8); 9.88 (d, J = 1.2 Hz, 1H, H-5); 4.71 (t, J = 6.7 Hz, 2H, H-1′); 2.84 (s, 3H, SCH₃ on C-4); 1.95 (~qi, J ~ 7 Hz, 2H, H-2′); 1.58 (~sex, J ~ 7 Hz, 2H, H-3′); 1.03 (t, J = 7.4 Hz, 3H, H-4′). ¹³C-NMR (CDCl₃): 158.2 (C-1); 154.1 (C-4); 146.3 (C-8); 146.1 (C-5); 120.7 (C-4a); 112.6 (C-8a); 68.9 (C-1′); 31.1 (C-2′); 19.6 (C-3′); 14.2 (C-4′); 13.3 (SCH₃ on C-4). Anal. calcd. for C₁₁H₁₄N₄OS: C, 52.78%; H, 5.64%; N, 22.38%; S, 12.81%. Found: C, 52.99%; H, 5.69%; N, 22.45%; S, 12.91%.

4.4.17. 4,5′,8′-Tris(methylthio)-1′-(naphthalen-2-yl)-1′H-1,2′-bipyridazino[4,5-d]pyridazine (34c)

Light yellowish powder. M.p. 214–216 °C. ¹H-NMR (CDCl₃): 10.57 (d, J = 1.1 Hz, 1H, H-8); 9.72 (d, J = 1.1 Hz, 1H, H-5); 8.05 (s, 1H, H-4′); 7.81 (br s, 1H. H-1″); 7.71–7.64 and 7.65 (overlapping m’s and br s, 4H, H-4″, H-5″, H-8″ and H-1′, resp.); 7.61 (dd, J = 8.6 Hz and 1.8 Hz, 1H, H-3″); 7.37–7.34 (m, 2H, H-6″ and H-7″); 2.78 (s, 3H, SCH₃ on C-4); 2.76 (s, 3H, SCH₃ on C-5′); 2.66 (s, 3H, SCH₃ on C-8′). ¹³C-NMR (CDCl₃): 156.8 (C-8′); 155.3 (C-4); 153.6 (C-5′); 150.7 (C-8′), 149.6 (C-1); 148.9 (C-8); 146.4 (C-5); 134.7 (C-4′); 134.6 (C-2″); 133.7 (C-4″a); 133.2 (C-8″a); 129.2 (C-4″); 128.8 (C-8″); 128.3 (C-1″); 128.0 (C-5″); 127.2 (C-7″); 126.9 (C-6″); 126.4 (C-4′a); 125.9 (C-3″); 120.6 (C-4a); 117.8 (C-8′a); 113.0 (C-8a); 53.6 (C-1′); 14.1 (SCH₃ on C-8′); 13.58 (SCH₃ on C-4); 13.54 (SCH₃ on C-5′). ¹⁵N-NMR chemical shifts [ref.: δ(NH₃) = 0 ppm] detectable by ¹H-¹⁵N-HMBC (CDCl₃): 396 (not resolved, N-6 and N-7 as identified via cross-peaks with H-5 and H-8 signals); 335 (N-3′, as identified via cross-peak with H-1′ signal); 155 (N-2′, as identified via cross-peak with H-1′ signal). Anal. calcd. for C₂₅H₂₀N₈S₃: C, 56.80%; H, 3.81%; N, 21.20%; S, 18.19%. Found: C, 56.67%; H, 3.92%; N, 21.25%; S, 18.11%.

4.4.18. 1,4-Bis(methylthio)pyridazino[4,5-d]pyridazine (I)

Yellow powder. M.p. 190–193 °C. ¹H-NMR (CDCl₃): 9.76 (s, 2H, H-5,8); 2.86 (s, 6H, SCH₃ on C-1 and C-4). Anal. calcd. for C₈H₈N₄S₂: C, 42.84%; H, 3.60%; N, 24.98%; S, 28.59%. Found: C, 43.02%; H, 3.72%; N, 24.85%; S, 28.67%.

4.4.19. 1-(3,5-Dimethyl-1H-pyrazol-1-yl)-4-(methylthio)pyridazino[4,5-d]pyridazine (II)

Yellow powder. M.p. 223–226 °C. ¹H-NMR (CDCl₃): 10.44 (d, J = 1.2 Hz, 1H, H-8); 9.86 (d, J = 1.2 Hz, 1H, H-5); 6.17 (s, 1H, H-4′); 2.91 (s, 3H, SCH₃); 2.62 (s, 3H, CH₃ on C-5′); 2.36 (s, 3H, CH₃ on C-3′). Anal. calcd. for C₁₂H₁₂N₆S: C, 52.93%; H, 4.44%; N, 30.86%; S, 11.17%. Found: C, 53.06%; H, 4.32%; N, 30.65%; S, 11.22%.

5. Conclusions

This contribution presents an in-depth synthetic and mechanistic study on the complex and intriguing transformations of 1,2-dihydro-1-arylpyridazino[4,5-d]pyridazines 1a–e and 2a–e. Besides the expected aromatic heterocycles 3a–e, the DDQ-mediated reactions catalyzed by ferrocene afforded a variety of unexpected products. The research disclosed novel types of heterocyclic transformations controlled by the substitution pattern of the substrate molecules. The proposed mechanisms were analyzed and confirmed by theoretical modeling of the key elementary steps that initiate the competing pathways leading to the formation of the different types of products. The newly recognized transformations along with the specific structure-reactivity correlations reported in this contribution might also be utilized in other fields of modeling-supported synthetic chemistry. It is also worth noting that most of the rationalized reactions proceed via C–C bond cleavage and recombination in radical cascades, representing an emerging area of intensively researched organic transformations with special regard to designed sequential functionalization [90,91,92,93,94].

Finally, the prepared novel compounds also have potential relevance in the research on the field of bioactive pharmaceutical ingredients. According to our best knowledge, there are no characteristic examples in the available literature for bioactive pyridazino[4,5-d]pyridazines represented by 3a–e; however, these products can be considered precursors that are explorable in an inverse electron-demand DA/rdA sequence to give a variety of highly functionalized, potentially bioactive phthalazine derivatives.

On the other hand, 5a, 5bc, 6a, and 6c are novel and further functionalizable representatives of the large family of phthalazine-based compounds. Phthalazine can be considered as one of the most privileged scaffolds in pharmaceutical chemistry, as demonstrated by the selection of investigational and approved active agents presented in Figure 1. Olaparib (AZD2281) is a poly(adenosine diphosphate-ribose)polymerase (PARP) inhibitor approved by the FDA for first-line treatment of adults with advanced ovarian cancer [95]. Azelastine is also an FDA-approved H₁-receptor-blocking antiallergic agent for the treatment of ocular allergic symptoms [96]. Zopolrestat, an aldose reductase inhibitor originally developed by Pfizer for the treatment of diabetic complications, was later identified as a potent competitive inhibitor of human glyoxalase I (GLOI) associated with a substantial potential in treating various cancers and other diseases [97]. Vatalanib, featuring a remarkably simple structure, is an anti-cancer agent acting as a tyrosine-kinase inhibitor in angiogenesis and used in combination with first- and second-line chemotherapy for the treatment of metastatic colorectal cancer and non-small cell lung cancer [98].