SnAr Reactions of 2,4-Diazidopyrido[3,2-d]pyrimidine and Azide-Tetrazole Equilibrium Studies of the Obtained 5-Substituted Tetrazolo[1,5-a]pyrido[2,3-e]pyrimidines

A straightforward method for the synthesis of 5-substituted tetrazolo[1,5-a]pyrido[2,3-e]pyrimidines from 2,4-diazidopyrido[3,2-d]pyrimidine in SnAr reactions with N-, O-, and S- nucleophiles has been developed. The various N- and S-substituted products were obtained with yields from 47% to 98%, but the substitution with O-nucleophiles gave lower yields (20–32%). Furthermore, the fused tetrazolo[1,5-a]pyrimidine derivatives can be regarded as 2-azidopyrimidines and functionalized in copper(I)-catalyzed azide-alkyne dipolar cycloaddition (CuAAC) and Staudinger reactions due to the presence of a sufficient concentration of the reactive azide tautomer in solution. In total, seven products were fully characterized by their single crystal X-ray studies, while five of them were representatives of the tetrazolo[1,5-a]pyrido[2,3-e]pyrimidine heterocyclic system. Equilibrium constants and thermodynamic values were determined using variable temperature 1H NMR and are in agreement of favoring the tetrazole tautomeric form (ΔG298 = −3.33 to −7.52 (kJ/mol), ΔH = −19.92 to −48.02 (kJ/mol) and ΔS = −43.74 to −143.27 (J/mol·K)). The key starting material 2,4-diazidopyrido[3,2-d]pyrimidine presents a high degree of tautomerization in different solvents.


Introduction
Fused-pyrimidine heterocycles are privileged scaffolds that have attracted great interest due to their biological properties [1]. The modification and refinement of such scaffolds are a promising strategy for the development of novel drugs. Among them, pyrido [3,2d]pyrimidine motif as a purine and pteridine analogue is a commonly used building block in drug discovery [2][3][4][5][6][7].

Synthesis
First, we acquired our key starting material, 2,4-diazidopyrido [3,2-d]pyrimidine 2, in excellent yield from commercially available dichloride 1 with sodium azide (Scheme 1). Here and further, the name diazide and structure 2 are used as formal simplification, as it does not exist in pure diazide form, but rather as a mixture of azide-tetrazole tautomeric forms. cannot be substituted and the substitution takes place at the C-2 position; and (3) addition can be completely omitted (Figure 1b, IV). Indeed, SNAr reactions in 2,4-diazidopurines V [21][22][23] and deazapurines VI [24,25] take place at the C-2 position (Figure 1c). However, this is not the case with quinazoline VII [26,27] and pyrido [2,3-d]pyrimidine VIII [28], where a conventional C-4 addition is observed.

e]pyrimidines
3a-c,f were obtained in moderate yields with substitution proceeding at the expected C-4 position. These conditions were found to be most suitable in our previous work on pyrido [2,3-d]pyrimidines [28]. However, we discovered that the reaction could be undertaken in DCM using NEt 3 as a base (conditions b). In these conditions, the work-up was easier and products 3d-f were obtained in higher yields. Scheme 1. Synthesis of 2,4-diazidopyrido [3,2-d]pyrimidine (2).
As the initial conditions for the substitution of diazide 2 with thiols, we chose th K2CO3/DMF system (Scheme 2, conditions a) 5-Thiotetrazolo[1,5-a]pyrido [2,3-e]pyrimidines 3a-c,f were obtained in moderate yield with substitution proceeding at the expected C-4 position. These conditions were found to be most suitable in our previous work on pyrido [2,3-d]pyrimidines [28]. However, w discovered that the reaction could be undertaken in DCM using NEt3 as a base (condi tions b). In these conditions, the work-up was easier and products 3d-f were obtained in higher yields. Scheme 2. SNAr reaction of diazide 2 with thiols.
Next, we explored the SNAr reaction between diazide 2 and the amines. As with th thiols, we adopted the previously used reaction conditions [28] and an addition o p-methoxybenzylamine to diazide 2 in DMSO provided product 4a in 49% yield withou an additional base. At this point, we decided to investigate the solvent effect on tautom erization and thus manipulate the site of nucleophile attack. To do this, we carried ou SNAr reactions of diazide 2 with p-methoxybenzylamine in various solvents: toluene benzene, DCM, EtOH, CHCl3, DMSO, and MeCN. In all cases, the same product 4a wa obtained. This means that 5-azidotetrazolo[1,5-a]pyrimidine tautomer 2AT is alway predominant, despite the selected solvents. The highest yield with the easiest work-up procedure was obtained in DCM, and it was the solvent of choice in further research. To explore the scope of the reaction, we used optimized conditions for the synthesis of dif ferent amino derivatives 4a-g in good yields (Scheme 3). Products bearing the benzyli 4a, aliphatic primary 4b, and secondary 4c, 4d, 4g amine moieties were obtained. In ad dition, ammonia and hydrazine showed good reactivity and provided products 4h and 4f. However, substitution of diazide 2 with aromatic amine (anisidine) to 4h was unsuc cessful and only the starting material was recovered after 3 days of stirring. Scheme 2. SNAr reaction of diazide 2 with thiols.
Next, we explored the SNAr reaction between diazide 2 and the amines. As with the thiols, we adopted the previously used reaction conditions [28] and an addition of p-methoxybenzylamine to diazide 2 in DMSO provided product 4a in 49% yield without an additional base. At this point, we decided to investigate the solvent effect on tautomerization and thus manipulate the site of nucleophile attack. To do this, we carried out SNAr reactions of diazide 2 with p-methoxybenzylamine in various solvents: toluene, benzene, DCM, EtOH, CHCl 3 , DMSO, and MeCN. In all cases, the same product 4a was obtained. This means that 5-azidotetrazolo[1,5-a]pyrimidine tautomer 2AT is always predominant, despite the selected solvents. The highest yield with the easiest work-up procedure was obtained in DCM, and it was the solvent of choice in further research. To explore the scope of the reaction, we used optimized conditions for the synthesis of different amino derivatives 4a-g in good yields (Scheme 3). Products bearing the benzylic 4a, aliphatic primary 4b, and secondary 4c, 4d, 4g amine moieties were obtained. In addition, ammonia and hydrazine showed good reactivity and provided products 4h and 4f. However, substitution of diazide 2 with aromatic amine (anisidine) to 4h was unsuccessful and only the starting material was recovered after 3 days of stirring. Substitution of diazide 2 with simple alcohols proceeded in the presence of a base (K2CO3) in dry MeCN, yielding products 5a and 5b (Scheme 4), although the products were obtained in low yields, mainly due to partial hydrolysis in the basic reaction medium as a side reaction. A complex mixture of unidentified products was obtained in the reaction of diazide 2 with phenol, most probably due to further substitution reactions of the phenyloxy moiety as a leaving group. Given that the compounds 2-5 persist in equilibrium between tetrazolo[1,5-a]pyrimidines and 2-azidopyrimidines, it should be possible to functionalize them as hetarylazides [12,34] and tetrazoles [35]. Indeed, a series of 1,2,3-triazole-substituted pyrido [3,2-d]pyrimidines 6a-e were obtained from tetrazolo [1,5-a]pyrido [2,3-e]pyrimidine 4a in the copper(I)-catalyzed azide-alkyne cycloaddition (CuAAC) reaction (Scheme 5). Substitution of diazide 2 with simple alcohols proceeded in the presence of a base (K 2 CO 3 ) in dry MeCN, yielding products 5a and 5b (Scheme 4), although the products were obtained in low yields, mainly due to partial hydrolysis in the basic reaction medium as a side reaction. A complex mixture of unidentified products was obtained in the reaction of diazide 2 with phenol, most probably due to further substitution reactions of the phenyloxy moiety as a leaving group. Substitution of diazide 2 with simple alcohols proceeded in the presence of a base (K2CO3) in dry MeCN, yielding products 5a and 5b (Scheme 4), although the products were obtained in low yields, mainly due to partial hydrolysis in the basic reaction medium as a side reaction. A complex mixture of unidentified products was obtained in the reaction of diazide 2 with phenol, most probably due to further substitution reactions of the phenyloxy moiety as a leaving group. Given that the compounds 2-5 persist in equilibrium between tetrazolo[1,5-a]pyrimidines and 2-azidopyrimidines, it should be possible to functionalize them as hetarylazides [12,34] and tetrazoles [35]. Indeed, a series of 1,2,3-triazole-substituted pyrido [3,2-d]  Given that the compounds 2-5 persist in equilibrium between tetrazolo[1,5-a]pyrimidines and 2-azidopyrimidines, it should be possible to functionalize them as hetarylazides [12,34] and tetrazoles [35]. Indeed, a series of 1,2,3-triazole-substituted pyrido [3,2-d]pyrimidines 6a-e were obtained from tetrazolo[1,5-a]pyrido[2,3-e]pyrimidine 4a in the copper(I)-catalyzed azide-alkyne cycloaddition (CuAAC) reaction (Scheme 5).
Additionally, we were able to functionalize 5-aminotetrazolo[1,5-a]pyrimidine 4b in the Staudinger reaction to iminophosphorane 8 (Scheme 7). Interestingly, its NMR and single crystal X-ray analysis revealed a protonated form 8′, which was obtained by the precipitation of compound 8 with anhydrous HCl in the DCM/MTBE system. The protonation occurred at the N(1) position of the molecule. Synthesis of 2,4-bistriazole from diazide 2 was not successful due to the formation of multiple side products. The major component was found to be partially reduced 2-triazolylpyrido[3,2-d]pyrimidine 7 (Scheme 6). We [36][37][38] and others [39] have previously observed that azido groups can be selectively reduced to their respective amino derivatives by Cu(I), which is generated by the CuSO 4 /ascorbate system. Synthesis of 2,4-bistriazole from diazide 2 was not successful due to the formation of multiple side products. The major component was found to be partially reduced 2-triazolylpyrido[3,2-d]pyrimidine 7 (Scheme 6). We [36][37][38] and others [39] have previously observed that azido groups can be selectively reduced to their respective amino derivatives by Cu(I), which is generated by the CuSO4/ascorbate system. Scheme 6. Partial reduction of compound 2 during its CuAAC reaction.
Additionally, we were able to functionalize 5-aminotetrazolo[1,5-a]pyrimidine 4b in the Staudinger reaction to iminophosphorane 8 (Scheme 7). Interestingly, its NMR and single crystal X-ray analysis revealed a protonated form 8′, which was obtained by the precipitation of compound 8 with anhydrous HCl in the DCM/MTBE system. The protonation occurred at the N(1) position of the molecule. Scheme 6. Partial reduction of compound 2 during its CuAAC reaction.
Additionally, we were able to functionalize 5-aminotetrazolo[1,5-a]pyrimidine 4b in the Staudinger reaction to iminophosphorane 8 (Scheme 7). Interestingly, its NMR and single crystal X-ray analysis revealed a protonated form 8 , which was obtained by the precipitation of compound 8 with anhydrous HCl in the DCM/MTBE system. The protonation occurred at the N(1) position of the molecule. Synthesis of 2,4-bistriazole from diazide 2 was not successful due to the form multiple side products. The major component was found to be partially 2-triazolylpyrido[3,2-d]pyrimidine 7 (Scheme 6). We [36][37][38] and others [39] ha ously observed that azido groups can be selectively reduced to their respectiv derivatives by Cu(I), which is generated by the CuSO4/ascorbate system. Scheme 6. Partial reduction of compound 2 during its CuAAC reaction.
Additionally, we were able to functionalize 5-aminotetrazolo[1,5-a]pyrimid the Staudinger reaction to iminophosphorane 8 (Scheme 7). Interestingly, its N single crystal X-ray analysis revealed a protonated form 8′, which was obtaine precipitation of compound 8 with anhydrous HCl in the DCM/MTBE system. tonation occurred at the N(1) position of the molecule.  To confirm that the C-4 position is the more reactive site in pyrido [3,2-d]pyrimidines with identical leaving groups in positions C-2 and C-4 (substrate 1), we switched the order of nucleophile addition. Indeed, the addition of amine first to the 2,4-dichloropyrido [3,2d]pyrimidine (1), followed by sodium azide, afforded the expected 4b (Scheme 8). However, it should be mentioned that the addition rate of the second nucleophile-azide was rather slow. It took 3 days to achieve near complete conversion of intermediate 9. Previously, Boyomi et al. [29] reported failed attempts of 4-amino and 4-benzyloxy substituted 2-chloropyrido [3,2-d]pyrimidine substitution with sodium azide in refluxed ethanol. Moreover, in our case, the amino product 9 was obtained in high yield without the formation of the diamino product. The electron-donating effect of the amino group slowed or even inhibited further SNAr process. To confirm that the C-4 position is the more reactive site in pyrido [3,2-d]pyrimidine with identical leaving groups in positions C-2 and C-4 (substrate 1), we switched the o der of nucleophile addition. Indeed, the addition of amine first to th 2,4-dichloropyrido [3,2-d]pyrimidine (1), followed by sodium azide, afforded the ex pected 4b (Scheme 8). However, it should be mentioned that the addition rate of th second nucleophile-azide was rather slow. It took 3 days to achieve near complete con version of intermediate 9. Previously, Boyomi et al. [29] reported failed attempts o 4-amino and 4-benzyloxy substituted 2-chloropyrido [3,2-d]pyrimidine substitution wit sodium azide in refluxed ethanol. Moreover, in our case, the amino product 9 was ob tained in high yield without the formation of the diamino product. The electron-donatin effect of the amino group slowed or even inhibited further SNAr process.

Scheme 8. Conventional synthesis route for compound 4b.
It is interesting to note that the addition of two azido groups in the synthesis of dia zide 2 was relatively fast (<1 h). This suggests that the first azido group after the additio to C-4 tautomerizes to tetrazole 10T, where tetrazole, as an electron-withdrawing group makes the pyrimidine system more reactive toward a second nucleophilic addition, an the final 2,4-disubstituted system is formed (Scheme 9). Scheme 9. Synthesis pathway of diazide 2.

Single Crystal X-Ray Analysis
Compounds 2, 3b, 3f, 4a, 4d, 5a, and the product 8 protonated form 8′ were obtaine in crystalline form and their chemical structures were confirmed by single crystal X-ra analysis. Crystal data and refinement details for the studied crystals are presented i Table 1. Search of the Cambridge structure database (CSD, version 5.43, November 2021 for synthesized pyrido[2,3-e]tetrazolo[1,5-a]pyrimidine heterosystem did not reveal an hits and, thus, gave evidence that it had not been studied by single crystal X-ray diffrac tion yet. Below, we discuss the geometry of this new tricyclic heterosystem in detail. Th pyrido [3,2-d]pyrimidine heterosystem search gave five hits [40][41][42]. Comparison o compound 6c with structures from CSD showed that their geometry fit very well. In the crystal structure 2, the tricyclic heterosystem was planar within ±0.021(1) Å Atoms N12 and N13 of the azide group deviated from this plane by 0.0736(9) Å an 0.1356(11) Å, respectively. Thus, the azide group is involved in a common conjugat system of the molecule, and the C5-N11 single bond, equal to 1.391(2) Å, was shortene when compared to a standard single C-N bond [43]. The azide group was not exactl linear and the valence angle N11-N12-N13 was 171.88(12)°.

Scheme 8. Conventional synthesis route for compound 4b.
It is interesting to note that the addition of two azido groups in the synthesis of diazide 2 was relatively fast (<1 h). This suggests that the first azido group after the addition to C-4 tautomerizes to tetrazole 10T, where tetrazole, as an electron-withdrawing group, makes the pyrimidine system more reactive toward a second nucleophilic addition, and the final 2,4-disubstituted system is formed (Scheme 9). Scheme 7. Synthesis of iminophosphorane 8 and its HCl salt 8′.
To confirm that the C-4 position is the more reactive site in pyrido [3,2-d]pyrimidin with identical leaving groups in positions C-2 and C-4 (substrate 1), we switched the der of nucleophile addition. Indeed, the addition of amine first to 2,4-dichloropyrido [3,2-d]pyrimidine (1), followed by sodium azide, afforded the pected 4b (Scheme 8). However, it should be mentioned that the addition rate of second nucleophile-azide was rather slow. It took 3 days to achieve near complete co version of intermediate 9. Previously, Boyomi et al. [29] reported failed attempts 4-amino and 4-benzyloxy substituted 2-chloropyrido [3,2-d]pyrimidine substitution w sodium azide in refluxed ethanol. Moreover, in our case, the amino product 9 was o tained in high yield without the formation of the diamino product. The electron-donati effect of the amino group slowed or even inhibited further SNAr process.

Scheme 8. Conventional synthesis route for compound 4b.
It is interesting to note that the addition of two azido groups in the synthesis of d zide 2 was relatively fast (<1 h). This suggests that the first azido group after the additi to C-4 tautomerizes to tetrazole 10T, where tetrazole, as an electron-withdrawing grou makes the pyrimidine system more reactive toward a second nucleophilic addition, a the final 2,4-disubstituted system is formed (Scheme 9). Scheme 9. Synthesis pathway of diazide 2.

Single Crystal X-Ray Analysis
Compounds 2, 3b, 3f, 4a, 4d, 5a, and the product 8 protonated form 8′ were obtain in crystalline form and their chemical structures were confirmed by single crystal X-r analysis. Crystal data and refinement details for the studied crystals are presented Table 1. Search of the Cambridge structure database (CSD, version 5.43, November 20 for synthesized pyrido[2,3-e]tetrazolo[1,5-a]pyrimidine heterosystem did not reveal a hits and, thus, gave evidence that it had not been studied by single crystal X-ray diffr tion yet. Below, we discuss the geometry of this new tricyclic heterosystem in detail. T pyrido [3,2-d]pyrimidine heterosystem search gave five hits [40][41][42]. Comparison compound 6c with structures from CSD showed that their geometry fit very well. In the crystal structure 2, the tricyclic heterosystem was planar within ±0.021(1) Atoms N12 and N13 of the azide group deviated from this plane by 0.0736(9) Å a 0.1356(11) Å, respectively. Thus, the azide group is involved in a common conjug system of the molecule, and the C5-N11 single bond, equal to 1.391(2) Å, was shorten when compared to a standard single C-N bond [43]. The azide group was not exac linear and the valence angle N11-N12-N13 was 171.88(12)°. Scheme 9. Synthesis pathway of diazide 2.

Single Crystal X-ray Analysis
Compounds 2, 3b, 3f, 4a, 4d, 5a, and the product 8 protonated form 8 were obtained in crystalline form and their chemical structures were confirmed by single crystal X-ray analysis. Crystal data and refinement details for the studied crystals are presented in Table 1. Search of the Cambridge structure database (CSD, version 5.43, November 2021) for synthesized pyrido[2,3-e]tetrazolo[1,5-a]pyrimidine heterosystem did not reveal any hits and, thus, gave evidence that it had not been studied by single crystal X-ray diffraction yet. Below, we discuss the geometry of this new tricyclic heterosystem in detail. The pyrido [3,2d]pyrimidine heterosystem search gave five hits [40][41][42]. Comparison of compound 6c with structures from CSD showed that their geometry fit very well. In the crystal structure 2, the tricyclic heterosystem was planar within ±0.021(1) Å. Atoms N12 and N13 of the azide group deviated from this plane by 0.0736(9) Å and 0.1356(11) Å, respectively. Thus, the azide group is involved in a common conjugate system of the molecule, and the C5-N11 single bond, equal to 1.391(2) Å, was shortened when compared to a standard single C-N bond [43]. The azide group was not exactly linear and the valence angle N11-N12-N13 was 171.88 (12) • .
The crystal structure 3b was a dichloromethane solvate. Heterocyclic fragment of the molecule was strictly planar. Deviation of the S11 atom from this plane was 0.121(1) Å. The lone electron pairs of S11 atom were involved in the common conjugate system of the heterosystem, which resulted in shortening of the bond C5-S11 = 1.731 (2) Å compared to a standard single C-S bond [43]. Aromatic fragments of the molecule, forming a dihedral angle of 5.15(9) • , were nearly parallel to each other.
In the crystal structure 3f, some violation of the planarity of the heterocyclic system was observed. The dihedral angle between the tetrazolo-pyrimidine and pyridine fragments was 5.97(5) • . The C5-S11 bond (1.745(1) Å) in the 3f structure was longer than in 3b. The least squares mean planes of tricyclic heterosystem and cyclohexane fragments formed a dihedral angle of 61.40 (7) • .
In the crystal structure 4a, the heterocyclic system was strictly planar (±0.01 Å). Deviation of the N11 atom from this plane was 0.0358(11) Å and the lone electron pair of N11 atom participated in the common conjugate system of a tricycle. The dihedral angle between aromatic fragments of the molecule was 73.98(5) • . Orientation of the methoxy group was characterized by the torsion angle C20-O19-C16-C15 = −6.8(2) • .
In the crystal structure 4d, we again observed minor violation of the planarity of the heterocyclic system. The dihedral angle between the tetrazolo-pyrimidine and pyridine fragments was 3.39 (7) • . The least squares mean planes of the heterocyclic system and morpholine fragment formed a dihedral angle of 19.71 (7) • . The morpholine fragment assumed a chair conformation. Atoms N11 and O14 deviated from the plane formed by four carbon atoms by 0.6142(15) Å and −0.6703(14) Å, respectively.
In the structure 6c, the bicyclic heterosystem was sufficiently planar. The C4-N9 bond length was 1.3366 (15) Å. Dihedral angles of the triazole fragment with mean planes of bicycle and adjacent phenyl ring were 4.38(6) • and 11.16(7) • , respectively. The slope of the mean plane of the second phenyl fragment to the plane of the bicycle was 75.71(5) • . Orientation of the methoxy group was characterized by the torsion angle C18-O17-C14-C13 = 7.2(2) • .
Compound 8 was crystallized in the form of hydrochloride chloroform disolvate. In contrast to the previous structure 6a in 8, atom N1 became protonated and bond C2-N16 [1.328(2) Å] assumed a double bond character. Least squares planes of pyridine and pyrimidine fragments in the heterosystem formed a dihedral angle of 3.87(8) • . Two atoms at the end of an aliphatic chain in the structure were disordered and assumed two positions with an occupancy ratio of 0.7:0.3.
The analysis of Table 2 shows that, overall, the geometry of the tetrazolo[1,5-a]pyrimidine fragment in the studied compounds corresponded to the published data. The geometry of the tetrazole fragment was the most conservative and was practically the same in all structures. The most variable bonds of the heterocyclic system were C3a-N4 and N4-C5. Their length was related to the type of substituent in position 5.

Free Energy Calculation for Azide-Tetrazole Equilibrium of Substituted Tetrazolo[1,5-a]pyrido[2,3-e] Pyrimidines
The system in equilibrium can be quantitatively characterized by thermodynamic values-Gibbs free energy, enthalpy, and entropy. The Gibbs free energy describes the equilibrium at given state of conditions, while the enthalpy defines the absolute stability of the tetrazole system (a higher value means a higher stability of tetrazole).
The Gibbs-Helmholtz equation ∆G = −RTln(K eq ) was used to calculate the Gibbs free energy of tautomerization [50]. 1 Table 3. Very similar results were also obtained by plotting the van't Hoff equation (see Supplementary Materials). Errors were calculated using the mean square error method. As previously mentioned, the azide-tetrazole equilibrium is influenced by the solvent polarity, temperature, substituent electronic effects, and sterics [8]. In our case, the equilibrium of tetrazolo[1,5-a]pyrido[2,3-e]pyrimidines 3-5 was fully shifted toward tetrazole in DMSO-d 6 and the azido tautomer was not observed in this solvent. On the other hand, the equilibrium in less polar CDCl 3 was notable and varied with different substituents.
As expected, the equilibrium shifted toward the azido tautomer at elevated temperatures. The calculated negative enthalpy values confirmed that the tetrazole is an energetically more stable form and the negative Gibbs free energy affirms that tetrazole in pyrido [3,2d]pyrimidines 3-5 is a major tautomer present at 25 • C.
It is well-known that electron donating substituents stabilize the fused tetrazole ring, while electron withdrawing substituents favor the azido tautomer. In our case, the Gibbs free energy values of p-methoxybenzylamino-(4a) and hexylamino-(4b) products were the highest. Therefore, the equilibrium was strongly shifted toward the tetrazole tautomer ( Figure 2). However, the Gibbs free energy values for products containing secondary amine moieties piperidine (4c), morpholine (4d), and N-methylpiperazine (4g) were significantly lower than those of primary amine moieties 4a and 4b. Additionally, alkoxy-substituted 5a and 5b were shifted toward the tetrazole tautomer and the Gibbs free energy values were higher than those of the thiols.

Tautomerism of Diazidopyrido[3,2-d]Pyrimidine
Finally, we looked at the tautomeric equilibrium of diazide 2. The 1 H-NMR spectra of diazide 2 in various solvents are shown in Figure 3. A different number of tautomeric forms were present depending on the solvent polarity. Tetrazole as the electron withdrawing moiety shifted signals downfield, while azido tautomer signals were more upfield. By going up in solvent polarity, the signals appeared more downfield and the ratio of the downfield/upfield signals increased. Thus, as in theory, the tetrazole tautomer becomes more dominant in polar solvents. At the present time, we are unable to undeniably provide the structural identity of each set of signals. For a thorough assignment of tautomeric forms, 15 N labeling is required.
In most cases, three to four tautomeric forms were observed. In TFA, only one tautomeric form was present and two tautomeric forms were observed in D2SO4. It is most likely that these solvents shift the equilibrium to bistetrazole 2TT due to far-out polarity. However, the pyridine ring in such acidic conditions can be protonated, making the ring system extremely electron deficient and shifting the equilibrium toward diazide 2P. It is interesting to note that in AcOD-d4, seven out of nine possible tautomeric forms were present. There are five possible tautomeric structures and three betaine structures for diazide 2 (Figure 4). To prove that these are indeed tautomeric forms, we acquired spectra after prolonged storage and redissolving the stored sample in a different solvent. To our delight, acquiring spectra after 7 days and 30 days of storage at 4 °C in acetic acid solution presented identical spectra to that of the freshly prepared sample ( Figure 5). Furthermore, evaporation of the acetic acid and redissolving the 30 day stored sample in CDCl3 provided identical spectra to one obtained by dissolving diazide 2 in CDCl3.

Tautomerism of Diazidopyrido[3,2-d]Pyrimidine
Finally, we looked at the tautomeric equilibrium of diazide 2. The 1 H-NMR spectra of diazide 2 in various solvents are shown in Figure 3. A different number of tautomeric forms were present depending on the solvent polarity. Tetrazole as the electron withdrawing moiety shifted signals downfield, while azido tautomer signals were more upfield. By going up in solvent polarity, the signals appeared more downfield and the ratio of the downfield/upfield signals increased. Thus, as in theory, the tetrazole tautomer becomes more dominant in polar solvents. At the present time, we are unable to undeniably provide the structural identity of each set of signals. For a thorough assignment of tautomeric forms, 15 N labeling is required.
In most cases, three to four tautomeric forms were observed. In TFA, only one tautomeric form was present and two tautomeric forms were observed in D 2 SO 4 . It is most likely that these solvents shift the equilibrium to bistetrazole 2TT due to far-out polarity. However, the pyridine ring in such acidic conditions can be protonated, making the ring system extremely electron deficient and shifting the equilibrium toward diazide 2P. It is interesting to note that in AcOD-d 4 , seven out of nine possible tautomeric forms were present. There are five possible tautomeric structures and three betaine structures for diazide 2 (Figure 4). To prove that these are indeed tautomeric forms, we acquired spectra after prolonged storage and redissolving the stored sample in a different solvent. To our delight, acquiring spectra after 7 days and 30 days of storage at 4 • C in acetic acid solution presented identical spectra to that of the freshly prepared sample ( Figure 5). Furthermore, evaporation of the acetic acid and redissolving the 30 day stored sample in CDCl 3 provided identical spectra to one obtained by dissolving diazide 2 in CDCl 3 .

General Information
Reagents purchased from Alfa Aesar, Acros Organics, Sigma Aldrich were used as received. All solvents were distilled prior to use. THF and toluene were distilled from Na under an Ar atmosphere. DMF and DMSO were distilled from CaH 2 under reduced pressure. For column chromatography, ROCC silica gel (40-60 µm, 60 Å) was used. Chromatography was monitored by TLC (E. Merck Kieselgel 60 F 254 ) and visualized with UV light.
HPLC analysis was performed using an Agilent Technologies 1200 Series system equipped with an X Bridge C-18 column, 4.6 × 150 mm, particle size 3.5 µm, with a flow rate of 1 mL/min, using 0.1% TFA/H 2 O and MeCN for the mobile phase.
The IR spectra were recorded in KBr with a Perkin-Elmer Spectrum BX FTIR spectrometer (4000−450 cm −1 ).
High-resolution mass (HRMS) (electrospray ionization (ESI)) was recorded with an Agilent 1290 Infinity series ultra-high pressure liquid chromatography connected to an Agilent 6230 time-of-flight mass spectrometer or (atmospheric pressure chemical ionization (APCI)) on a 7 T solariX XR (Bruker Daltonik GmbH) Fourier transform ion cyclotron resonance mass spectrometer equipped with an APCI source.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27227675/s1, Gibbs and Van't plots, complete table of thermodynamic values calculated and 1 H, 13 C and 31 P NMR spectra are available in the supporting information.