One-Pot Solvent-Involved Synthesis of 5-O-Substituted 5H-Chromeno[2,3-b]pyridines

Chromeno[2,3-b]pyridines are substances demanded in medicinal and material chemistry. PASE (pot, atom, and step economy) and in particular one-pot approaches are key green chemistry techniques that are applied for the synthesis of heterocyclic compounds. In this case, the PASE approach was extended with ‘component economy’, as solvent was used also as reactant (solvent-involved reaction). This approach was adopted for the one-pot synthesis of previously unknown O-substituted 5-alkoxy-5H-chromeno[2,3-b]pyridines via two-step transformation, namely the reaction of salicylaldehydes and malononitrile dimer, with the subsequent addition of alcohol. The mechanistic studies revealed the possibility of concurrent reaction. The studies aided in optimizing the reaction conditions for the best yields (77–93%). Thus, the one-pot reaction proceeds efficient and quickly, and the work-up procedure (only simple filtering) is very convenient. The structure of synthesized chromeno[2,3-b]pyridines was confirmed by 2D NMR spectroscopy.


Introduction
Global ecological awareness has made green chemistry a broad, beneficial and promising area of organic chemistry [1,2]. Green chemistry is based on 12 principles [1] that lead to the development of the simplest synthetic approach with the utilization of the least number of components (the less toxic, the better); the result should carry the least number of by-products and environmental threats.
The current level of chemistry does not allow all 12 principles to be applied to resolve all drawbacks and limitations (for example, avoiding solvents or special precursors) in the development of every new process. Thereby, in synthetic chemistry, the PASE approach has become widespread [3][4][5][6][7] and it is currently a key green technique [4]. This approach is focused on the economy (E) of pot (P), atom (A) and steps (S). Atomic economy states that the majority of atoms in a reaction become a component of the final product. This concept prevents waste; however, if the waste is inevitable, according to green chemistry, water is the best by-product.
Step economy is linked to the energy efficiency principle; it is preferable to have the fewest stages (separate transformations, work-ups) [8] and to have all transformations in one pot [8,9].
Among the approaches mentioned above, reactions in which solvents are involved in the transformations [21] have never received special attention. However, it would be a consistent extension of such approaches. More than that, the utilization of solvent as We have accomplished several syntheses of chromeno [2,3-b]pyridines [5,33] and some green transformations with different roles of solvent, among them [12] on-solvent [12], on-water [18], and solid-state ( [15] procedures. In addition to these green methods, we would like to present the first synthesis of O-substituted 5-alkoxy-5H-chromeno [2,3b]pyridines, in which solvent is used as a reactant (solvent-involved- [21]) and plays a role in the final nucleophilic addition, becoming a fragment of the final structure. It delivers atom economy, reduces the number of used components, and thereby makes the whole process greener.

Formation of 5-Alkoxy-5H-Chromeno[2,3-b]pyridine 4
Previously we synthesized various types of 5-C-and 5-P-chromeno [2,3-b]pyridines [5,[33][34][35][36][37][38]. In general, the interaction of a carbonyl group with a CH acid and the subsequent addition of another CH acid are easily feasible. Modifying CH acid to another nucleophilic agent, on the other hand, is a more difficult task. We discovered this for the first time in the synthesis of 5-P-chromeno[2,3-b]pyridines [5]. To complete the synthesis, we had to use an aprotic solvent (CH3CN), which enhances the nucleophilic properties of reactants.
After C-and P-substituted products were synthesized, we concentrated on 5-Osubstituted chromeno [2,3-b]pyridines. It turns out that only a few examples are known, Among these compounds, the structure of 5-phenoxy-5H-chromeno[2,3-b]pyridine ( Figure 1) provides the best potential for modification of the O-substitution. Furthermore, 5-phenoxy-5H-chromeno [2,3-b]pyridine is the only structure known to have an additional oxygen substitution [30]. At the same time, phenol, which is used for the synthesis, is quite a toxic component. It would be better to replace phenol fragments with less toxic and more cheap alcohols. However, the synthesis of 5-phenoxy-5H-chromeno [2,3-b]pyridine is carried out in ethanol [30], and no 5-ethoxy-5H-chromeno[2,3-b]pyridine products were obtained in this procedure.
We have accomplished several syntheses of chromeno [2,3-b]pyridines [5,33] and some green transformations with different roles of solvent, among them [12] on-solvent [12], onwater [18], and solid-state ( [15] procedures. In addition to these green methods, we would like to present the first synthesis of O-substituted 5-alkoxy-5H-chromeno [2,3-b]pyridines, in which solvent is used as a reactant (solvent-involved- [21]) and plays a role in the final nucleophilic addition, becoming a fragment of the final structure. It delivers atom economy, reduces the number of used components, and thereby makes the whole process greener.

Results and Discussion
Previously we synthesized various types of 5-C-and 5-P-chromeno [2,3-b]pyridines [5,[33][34][35][36][37][38]. In general, the interaction of a carbonyl group with a CH acid and the subsequent addition of another CH acid are easily feasible. Modifying CH acid to another nucleophilic agent, on the other hand, is a more difficult task. We discovered this for the first time in the synthesis of 5-P-chromeno[2,3-b]pyridines [5]. To complete the synthesis, we had to use an aprotic solvent (CH 3 CN), which enhances the nucleophilic properties of reactants.
After Cand P-substituted products were synthesized, we concentrated on 5-Osubstituted chromeno [2,3-b]pyridines. It turns out that only a few examples are known, and only a few of them have been synthesized with one-pot or multicomponent methods [39][40][41].
One such example is the multicomponent synthesis of 2,4-diamino-5-phenoxy-5Hchromeno [2,3-b]pyridine-3-carbonitrile (Scheme 1) [30]. Among other things, it is difficult to assess the generality of the reaction, since the authors presented only one such compound. The mechanism of the reaction was also not described thoroughly. and only a few of them have been synthesized with one-pot or multicomponent methods [39][40][41]. One such example is the multicomponent synthesis of 2,4-diamino-5-phenoxy-5Hchromeno [2,3-b]pyridine-3-carbonitrile (Scheme 1) [30]. Among other things, it is difficult to assess the generality of the reaction, since the authors presented only one such compound. The mechanism of the reaction was also not described thoroughly. We were encouraged to deeply investigate such processes in case of malononitrile dimer. First of all, we synthesized chromene 5a for the mechanism investigation. However, during the recrystallization of chromene 5a after the reaction (Scheme 2), some byproducts were found. Thus, in this article, we present a mechanistic investigation of the formation of 5alkoxy-5H-chromeno[2,3-b]pyridine with proven structure (2D NMR) as well as a convenient approach to the new substituted 5-O-chromeno[2,3-b]pyridines.
A two-step transformation results in 5-O-chromeno[2,3-b]pyridine 4a after the complete formation of intermediate 5a (it precipitates, then redissolves) (Scheme 3). The intermediate 5a was detected in the reaction mixture (NMR). The steps of the two-step transformation were controlled by temperature. The final cyclization to 4a (step II, Scheme 3) demands heating at reflux.
If the heating is provided immediately without complete formation (precipitation) of intermediate 5a, the process affords a mixture of compounds 8 and 4a. Thus, in this case We were encouraged to deeply investigate such processes in case of malononitrile dimer. First of all, we synthesized chromene 5a for the mechanism investigation. However, during the recrystallization of chromene 5a after the reaction (Scheme 2), some byproducts were found. and only a few of them have been synthesized with one-pot or multicomponent methods [39][40][41]. One such example is the multicomponent synthesis of 2,4-diamino-5-phenoxy-5Hchromeno [2,3-b]pyridine-3-carbonitrile (Scheme 1) [30]. Among other things, it is difficult to assess the generality of the reaction, since the authors presented only one such compound. The mechanism of the reaction was also not described thoroughly. We were encouraged to deeply investigate such processes in case of malononitrile dimer. First of all, we synthesized chromene 5a for the mechanism investigation. However, during the recrystallization of chromene 5a after the reaction (Scheme 2), some byproducts were found. Thus, in this article, we present a mechanistic investigation of the formation of 5alkoxy-5H-chromeno[2,3-b]pyridine with proven structure (2D NMR) as well as a convenient approach to the new substituted 5-O-chromeno[2,3-b]pyridines.
A two-step transformation results in 5-O-chromeno[2,3-b]pyridine 4a after the complete formation of intermediate 5a (it precipitates, then redissolves) (Scheme 3). The intermediate 5a was detected in the reaction mixture (NMR). The steps of the two-step transformation were controlled by temperature. The final cyclization to 4a (step II, Scheme 3) demands heating at reflux.
If the heating is provided immediately without complete formation (precipitation) of intermediate 5a, the process affords a mixture of compounds 8 and 4a. Thus, in this case and only a few of them have been synthesized with one-pot or multicomponent methods [39][40][41].
One such example is the multicomponent synthesis of 2,4-diamino-5-phenoxy-5Hchromeno[2,3-b]pyridine-3-carbonitrile (Scheme 1) [30]. Among other things, it is difficult to assess the generality of the reaction, since the authors presented only one such compound. The mechanism of the reaction was also not described thoroughly.
We were encouraged to deeply investigate such processes in case of malononitrile dimer. First of all, we synthesized chromene 5a for the mechanism investigation. However, during the recrystallization of chromene 5a after the reaction (Scheme 2), some byproducts were found. Thus, in this article, we present a mechanistic investigation of the formation of 5alkoxy-5H-chromeno[2,3-b]pyridine with proven structure (2D NMR) as well as a convenient approach to the new substituted 5-O-chromeno[2,3-b]pyridines.
A two-step transformation results in 5-O-chromeno[2,3-b]pyridine 4a after the complete formation of intermediate 5a (it precipitates, then redissolves) (Scheme 3). The intermediate 5a was detected in the reaction mixture (NMR). The steps of the two-step transformation were controlled by temperature. The final cyclization to 4a (step II, Scheme 3) demands heating at reflux.
If the heating is provided immediately without complete formation (precipitation) of intermediate 5a, the process affords a mixture of compounds 8 and 4a. Thus, in this case Thus, in this article, we present a mechanistic investigation of the formation of 5-alkoxy-5H-chromeno[2,3-b]pyridine with proven structure (2D NMR) as well as a convenient approach to the new substituted 5-O-chromeno[2,3-b]pyridines.

Mechanism of Formation of 5-Methoxy
In this case, the addition of MeOH to chromene 5a appears to allow the formation of 5-O-chromeno[2,3-b]pyridine. We have carried out several one-pot transformations to prove this (Scheme 3).
A two-step transformation results in 5-O-chromeno[2,3-b]pyridine 4a after the complete formation of intermediate 5a (it precipitates, then redissolves) (Scheme 3). The intermediate 5a was detected in the reaction mixture (NMR). The steps of the two-step transformation were controlled by temperature. The final cyclization to 4a (step II, Scheme 3) demands heating at reflux.
If the heating is provided immediately without complete formation (precipitation) of intermediate 5a, the process affords a mixture of compounds 8 and 4a. Thus, in this case 5-O-chromeno[2,3-b]pyridine 4a is formed in a multicomponent reaction, but concurrent cascade formation of chromeno[2,3-b]pyridine derivative 8 [42] also occurs (Scheme 3). Based on these results, we decided to accomplish the process in the one-pot two-step format (Scheme 3). In this case, stage I was completed in one hour without heating to avoid the formation of unsubstituted chromeno[2,3-b]pyridine derivative 8 [42]. Stage II also took half an hour, but with heating at reflux. At the end of the two-step transformation, Based on our earlier results [5,33,38] and data from the literature [43], we propose the following one-pot two-step cascade transformation mechanism (Scheme 4). The first stage of the one-pot process was a Knoevenagel condensation with the formation of intermediate   Table 1 in Section 2.3, Entry 1.
Based on these results, we decided to accomplish the process in the one-pot two-step format (Scheme 3). In this case, stage I was completed in one hour without heating to avoid the formation of unsubstituted chromeno[2,3-b]pyridine derivative 8 [42]. Stage II also took half an hour, but with heating at reflux. At the end of the two-step transformation, only 5-O-chromeno[2,3-b]pyridine 4a was detected.
Based on our earlier results [5,33,38] and data from the literature [43], we propose the following one-pot two-step cascade transformation mechanism (Scheme 4). The first stage of the one-pot process was a Knoevenagel condensation with the formation of intermediate It should be noted that obtained 5-O-chromeno[2,3-b]pyridines 4 are unstable in DMSO solution. In publication [45], the authors claimed to have synthesized 2,4-diamino-5-ethoxy-9-methoxy-5H-chromeno[2,3-b]pyridine-3-carbonitrile. Those data, however, are not available, neither in the text of the article nor in the supplementary materials [45]. The authors claim that this is due to the fact that the compound 4a is unstable in DMSO- It should be noted that obtained 5-O-chromeno[2,3-b]pyridines 4 are unstable in DMSO solution. In publication [45], the authors claimed to have synthesized 2,4-diamino-5-ethoxy-9-methoxy-5H-chromeno[2,3-b]pyridine-3-carbonitrile. Those data, however, are not available, neither in the text of the article nor in the Supplementary Materials [45]. The authors claim that this is due to the fact that the compound 4a is unstable in DMSO-d 6 and in solid form. Indeed, we observed the decomposition of 5-O-substituted chromeno[2,3b]pyridines 4 upon standing in DMSO-d 6 in a NMR tube at room temperature, while recording NMR spectra. We solved this problem by preparing a fresh solution. In solid form, these compounds 4 turned out to be stable.

One-Pot Synthesis of 5-Alkoxy-5H-Chromeno[2,3-b]pyridines 4a-i
As the mechanism of the reaction had been established, we were interested in optimizing the reaction conditions. Further, transformations of salicylaldehyde 1a, malononitrile dimer 2 and alcohol 3a were done in the one-pot, two-step format (Scheme 3, Path A, two-step, one-pot) to avoid the concurrent formation of byproduct 8 (Scheme 3, Path B). Namely, salicylaldehyde 1a and malononitrile dimer 2 in alcohol were stirred until the intermediate 5 was precipitated (Stage I), and the reaction mixture was heated at reflux to produce 5-O-chromeno[2,3-b]pyridines 4 (Stage II). The list of conditions is presented in Table 1. The data from Table 1 indicate that amines (Entries 1-4, 8-13) tend to produce higher yields of 4a than inorganic catalysts. Stage I can be accomplished with any type of catalyst in high yields. It is possible that amines form a transitional intermediate that increases the rate of the nucleophilic addition in Stage II.
The reaction in Stage I is easily catalyzed, and the product precipitates in one hour process without heating. In Stage II, there should be a balance between reagent and product solubility, and the amount of alcohol is important (Entries 8-10). Thus, 10 mL of methanol were used in Stage I, and another 20 mL of methanol were used in Stage II to achieve the best results. According to Entries 8, 11, 12, Stage II finishes in a half hour.
Among the amines used (Entries 1-4), triethylamine (Entry 4) is the best catalyst. This is probably due to its strong basicity. 10% of catalyst is enough for the reaction (Entries 4, 8).
To sum up, Entry 11 is the most optimal condition for this new transformation. To isolate the synthesized compound, when the reaction was finished, the reaction mixture was left at room temperature for 3 h to crystallize the final compound 4 in pure form.

2D-NMR Study of the Structure of Compound 4d
The structure of the obtained compounds 4a-i was confirmed by 1 H and 13 C NMR data, IR spectroscopy, and mass spectrometry. The structure of the compound 4d was confirmed by various NMR correlation spectroscopy techniques (Figures 3 and 4).
The assignment of one-dimensional (1D) 1 H and 13 C-NMR spectra signals was performed using two-dimensional (2D) NMR experiments such as 1 H-13 C HSQC and 1 H-13 C HMBC (Figure 4). 1 H-13 C HSQC showed two NH2-groups which had no cross-peaks in the spectrum. In the HMBC spectrum, the coupling between amino-protons and C 3 was found (Figures 3 and 4). Additionally, 5-OMe gave the only correlation with C 5 through three chemical bonds. This is confirmed by the spin interaction from the HMBC spectrum of H 5 with the carbons of the benzene and pyridine rings (Figures 3 and 4).

2D-NMR Study of the Structure of Compound 4d
The structure of the obtained compounds 4a-i was confirmed by 1 H and 13 C NMR data, IR spectroscopy, and mass spectrometry. The structure of the compound 4d was confirmed by various NMR correlation spectroscopy techniques (Figures 3 and 4).
The assignment of one-dimensional (1D) 1 H and 13 C-NMR spectra signals was performed using two-dimensional (2D) NMR experiments such as 1 H-13 C HSQC and 1 H-13 C HMBC (Figure 4). 1 H-13 C HSQC showed two NH2-groups which had no cross-peaks in the spectrum. In the HMBC spectrum, the coupling between amino-protons and C 3 was found (Figures 3 and 4). Additionally, 5-OMe gave the only correlation with C 5 through three chemical bonds. This is confirmed by the spin interaction from the HMBC spectrum of H 5 with the carbons of the benzene and pyridine rings (Figures 3 and 4).  It should be noted that regardless of the type of alcohol, the highest yields of chromeno[2,3b]pyridine 4a-c were obtained in the reaction with salicylaldehyde 1a (Table 2). When electron-donor or halogen substituted salicylaldehyde 1 is used, the yields decrease slightly but remain comparable. This is preusmably due to the greater solubility of the intermediate 5 in alcohols in these cases. When alcohol was changed, the highest yields of chromeno[2,3b]pyridines 4 were obtained with methanol 3a (Table 2). Most likely, this can be explained by steric factors.

2D-NMR Study of the Structure of Compound 4d
The structure of the obtained compounds 4a-i was confirmed by 1 H and 13 C NMR data, IR spectroscopy, and mass spectrometry. The structure of the compound 4d was confirmed by various NMR correlation spectroscopy techniques (Figures 3 and 4).
The assignment of one-dimensional (1D) 1 H and 13 C-NMR spectra signals was performed using two-dimensional (2D) NMR experiments such as 1 H-13 C HSQC and 1 H-13 C HMBC (Figure 4). 1 H-13 C HSQC showed two NH 2 -groups which had no cross-peaks in the spectrum. In the HMBC spectrum, the coupling between amino-protons and C 3 was found (Figures 3 and 4). Additionally, 5-OMe gave the only correlation with C 5 through three chemical bonds. This is confirmed by the spin interaction from the HMBC spectrum of H 5 with the carbons of the benzene and pyridine rings (Figures 3 and 4). performed using two-dimensional (2D) NMR experiments such as 1 H-13 C HSQC and 1 H-13 C HMBC (Figure 4). 1 H-13 C HSQC showed two NH2-groups which had no cross-peaks in the spectrum. In the HMBC spectrum, the coupling between amino-protons and C 3 was found (Figures 3 and 4). Additionally, 5-OMe gave the only correlation with C 5 through three chemical bonds. This is confirmed by the spin interaction from the HMBC spectrum of H 5 with the carbons of the benzene and pyridine rings (Figures 3 and 4).  Detailed 1D 1 H and 13 C-NMR spectra and 2D NMR spectra of the compound 4d are presented in the Supplementary Materials (Figures S19-S22).
All melting points were measured with a Gallenkamp melting-point apparatus (Gallenkamp & Co., Ltd., London, UK) and were uncorrected. 1 H and 13 C-NMR spectra were recorded in DMSO-d6 with Bruker AM300 and Bruker AV500 spectrometers (Bruker Corporation, Billerica, MA, USA) at ambient temperature. Chemical shift values are relative to Me4Si. Some 1 H-NMR spectra have underestimated NH2 signals integrals. These protons were exchanged with D2O (it is present as an impurity in DMSO-d6). Two- Detailed 1D 1 H and 13 C-NMR spectra and 2D NMR spectra of the compound 4d are presented in the Supplementary Materials (Figures S19-S22).
All melting points were measured with a Gallenkamp melting-point apparatus (Gallenkamp & Co., Ltd., London, UK) and were uncorrected. 1 H and 13 C-NMR spectra were recorded in DMSO-d 6 with Bruker AM300 and Bruker AV500 spectrometers (Bruker Corporation, Billerica, MA, USA) at ambient temperature. Chemical shift values are relative to Me 4 Si. Some 1 H-NMR spectra have underestimated NH 2 signals integrals. These protons were exchanged with D 2 O (it is present as an impurity in DMSO-d 6 ). Two-dimensional (2D) NMR spectra were registered with a Bruker AV400 spectrometer (Bruker Corporation, Billerica, MA, USA) at ambient temperature. The IR spectrum was recorded with a Bruker ALPHA-T FT-IR spectrometer (Bruker Corporation, Billerica, MA, USA) in a KBr pellet. MS spectra (EI = 70 eV) were obtained directly with a Kratos MS-30 spectrometer (Kratos Analytical Ltd., Manchester, UK). High-resolution mass spectra (HRMS) were measured on a Bruker micrOTOF II (Bruker Corporation, Billerica, MA, USA) instrument using electrospray ionization (ESI).

One-Pot Synthesis of 5-Alkoxy-5H-Chromeno[2,3-b]pyridines 4a-i
Salicylaldehyde 1a-c (1 mmol) and malononitrile dimer 2 (1 mmol, 132 mg) were stirred in alcohol 3a-c (10 mL) for 1 h at room temperature. The formation of a thick yellowish precipitate was observed. Another portion of alcohol 3a-c (20 mL) was added to the precipitate with stirring, and the reaction mixture was refluxed for an additional 30 min. After the reaction was completed, the flask was left at room temperature for 3 h. The solid was filtered, washed with well-chilled ethanol/water mixture (1:1, 2 × 2 mL), and dried to isolate pure 5-alkoxy-5H-chromeno [

Conclusions
To conclude, the synthesis of 5-O-substituted 5H-chromeno [2,3-b]pyridines differs from that of 5-P-and 5-C-substituted derivatives. The nucleophile, which is supposed to interact with the intermediate, determines the final structure.
In this case, the solvent (alcohol) acted as a nucleophile; it interacted with the intermediate formed from salicylaldehyde and malononitrile dimer to form 5-O-substituted 5H-chromeno [2,3-b]pyridines. This two-step, one-pot transformation extends the PASE approach with 'component economy' as alcohol is used both as a solvent and a reactant (solvent-involved reaction).
The mechanistic studies revealed the possibility of concurrent processes. These processes tend to form more thermodynamically stable products. Alcohol, as a weak nucleophile, forms the kinetic product, which is unstable in solution.
The mechanistic studies aided in optimizing reaction conditions. Thus, the one-pot, two-step transformation of salicylaldehydes, malononitrile dimer and alcohol proceeds efficiently and quickly with the formation of 5-O-substituted 5H-chromeno [2,3-b]pyridines in high yields of 77-93%. It is easy to isolate the final compounds directly from the reaction mixture.