New insights into the electrochemical behavior of acid orange 7: Convergent paired electrochemical synthesis of new aminonaphthol derivatives

Electrochemical behavior of acid orange 7 has been exhaustively studied in aqueous solutions with different pH values, using cyclic voltammetry and constant current coulometry. This study has provided new insights into the mechanistic details, pH dependence and intermediate structure of both electrochemical oxidation and reduction of acid orange 7. Surprisingly, the results indicate that a same redox couple (1-iminonaphthalen-2(1H)-one/1-aminonaphthalen-2-ol) is formed from both oxidation and reduction of acid orange 7. Also, an additional purpose of this work is electrochemical synthesis of three new derivatives of 1-amino-4-(phenylsulfonyl)naphthalen-2-ol (3a–3c) under constant current electrolysis via electrochemical oxidation (and reduction) of acid orange 7 in the presence of arylsulfinic acids as nucleophiles. The results indicate that the electrogenerated 1-iminonaphthalen-2(1 H)-one participates in Michael addition reaction with arylsulfinic acids to form the 1-amino-3-(phenylsulfonyl)naphthalen-2-ol derivatives. The synthesis was carried out in an undivided cell equipped with carbon rods as an anode and cathode.

2-Naphthol orange (acid orange 7), C 16 H 11 N 2 NaO 4 S, is a mono-azo water-soluble dye that extensively used for dyeing paper, leather and textiles 1,2 . The structure of acid orange 7 involves a hydroxyl group in the ortho-position to the azo group. This resulted an azo-hydrazone tautomerism, and the formation of two tautomers, which each show an acid− base equilibrium [3][4][5][6][7][8][9][10][11][12] . Despite the number of articles dealing with acid− base properties of the acid orange 7, this topic is not yet well known, and only one pK a , (pK a = 11.4) is reported 6 . On the other hand, azo dyes have been widely used for developing and testing theories of color and constitution, tautomerism, indicator action, and acid-base equilibria 5 . Therefore, detailed mechanistic information is important in understanding of the stability and in identifying of the intermediates structure resulting from the oxidative or reductive decomposition of dye. Consequently, detailed mechanistic information is particularly attractive from the point of view of environmental pollution because of residual dye and the commercial applications 13,14 . Additionally, green/ sustainable synthesis is much more important than conventional synthetic methods. The concept and significance of green sustainable chemistry (GSC), has been recognized throughout the world, and nowadays new processes cannot be developed without consideration of GSC. In recent years, much attention has been paid to electroorganic synthesis as a typical environmentally friendly process [15][16][17][18][19][20][21][22][23] . This method contains the simultaneous incidence of both oxidation (at the anode) and reduction (at the cathode). In conventional electroorganic synthesis, the synthesis of the desired products is done either by the anodic or by the cathodic reaction and so; the reaction product at the counter electrode is undesirable. The simultaneous use of both oxidation and reduction reactions to synthesis of a product is the dream of an organic electrochemist and is a wonderful strategy [24][25][26][27][28] . At ideal conditions, a 200% current efficiency is achievable for paired electrosynthesis when both anodic and cathodic reactions to provide the similar product (convergent strategy) 15 . Furthermore, sulfone compounds and naphthalene derivatives are found in antibiotic drugs such as nafacillin and 4,4-diaminodiphenylsulfone (dapsone) and antifungal drugs such as naftifine, tolnaftate and terbinafine [29][30][31] . These compounds have an effective inhibitory effect against the bacteria 30,31 and antimicrobials effect against wide range of human pathogens 29 . Based on these advances, we anticipate that naphthalene derivatives containing sulfone groups reveals such properties.
The results discussed above prompted us to investigate the electrochemical oxidation and reduction of acid orange 7 in aqueous solutions with different pH values to achieve the following goals: (i) new insights into the electrochemical oxidation and reduction of acid orange 7, (ii) definitive detection of intermediates formed during the oxidative and reductive degradation of acid orange 7, and (iii) convergent paired electrochemical synthesis of new 1-amino-2-naphthol derivatives by constant current electrolysis of acid orange 7 in the presence of arylsulfinic acids as nucleophiles.

Results and Discussion
Electrochemical Study of Acid Orange 7. Cyclic voltammograms of acid orange 7 (AO7) in aqueous phosphate buffer solution (c = 0.2 M, pH = 2.0) in two different potential regions (+ 0.5 to − 0.4 and 0.0 to + 1.2 V vs. Ag/AgCl) is shown in Fig. 1. When the electrode potential was scanned from + 0.5 V versus Ag/AgCl to a sufficiently negative voltage (− 0.4 V versus Ag/AgCl), the cyclic voltammogram exhibits a large cathodic peak (C 0 ) at − 0.19 with an anodic peak (A 1 ) at + 0.30 V versus Ag/AgCl at 50 mV s −1 (Fig. 1I). Under these conditions, in the second cycle a new cathodic peak (C 1 ), which is the counterpart of anodic peak (A 1 ) appears with an E p value of + 0.18 V versus Ag/AgCl. Moreover, when the electrode potential was scanned from 0.0 V vs. Ag/AgCl to the positive potentials (+ 1.2 V vs. Ag/AgCl), the voltammogram shows an anodic peak (A 2 ) at + 0.80 V vs. Ag/AgCl with a cathodic peak (C 1 ) (Fig. 1II). As in the previous experiment, a quasi-reversible couple, A 1 /C 1 , appears in the second cycle of the voltammogram. A very important point in this study is the presence of the same redox couple in both cyclic voltammograms. This confirms the generation of the same intermediates from both oxidation and reduction of AO7.
When the potential scan rate increases from 250 to 8000 mV s −1 , the cyclic voltammograms of AO7 in anodic region (Fig. S1) show the following changes: (1) the appearance of peak C 2 , which is the counterpart of peak A 2 .
(2) The increase of the anodic and cathodic peak current ratios (I pA2 /I pA1 ) and (I pC2 /I pCA1 ) and (3) the decrease of peak current function for anodic peak A 2 (I pA2 /v 1/2 ). Increase in the potential scan rate causes a decrease in the CV time-scale and therefore a decrease in the progress of the following chemical reaction. These data confirm the occurrence of a following chemical reaction and generation of a quasi-reversible system after oxidation of AO7 32 . In addition, the effect of potential scan rate, v, (250 to 8000 mV s −1 ) on the cyclic voltammetric response of AO7 in cathodic region which confirms the Irreversibility of the reduction process corresponding to peak C 0 (Fig. S2).
The oxidative and reductive controlled-potential coulometry of AO7 was performed by applying potentials + 0.90 and − 0.20 V vs. Ag/AgCl, respectively. The solutions after coulometry are shown in Supporting Information (Fig. S3). The monitoring of the electrolysis progress was carried out by cyclic voltammetry (Fig. 2). This Figure show that, in both experiments, proportional to the advancement of coulometry, parallel to the decrease in the current of peaks A 2 (Fig. 2I) and C 0 (Fig. 2II), the peaks A 1 and C 1 , increases. In these conditions, the number of transferred electrons in oxidative controlled potential coulometry was obtained 2.9 electrons pre each AO7 molecule. On the other hand, the number of transferred electrons in reductive controlled potential coulometry was obtained 4.1 electrons pre each AO7 molecule. An important point on the oxidation behaviour of AO7 can be seen in Fig. 2Ia. This Figure (and also Fig. 1II) represents a typical behaviour of an ECE pathway in the kinetic region 32 . However, the comparison of I pA2 at the start of coulometry with I pC1 at the end of coulometry shows that, I pC1 at the end of coulometry is equal to I pA2 at the start of coulometry = ( ) . It should be noted that, in an ECE mechanism, the peak current ratio of the starting compound to the product (I st /I pr ) in the kinetic region is {(n 1 + n 2 )/n 2 } 3/2 . Where, n 1 and n 2 are the number of electrons involved in the oxidation of starting compound and product, respectively 32 . This discrepancy implying that oxidation pathway of AO7, is not an ECE and confirms the reaction pathway presented in Fig. 3, for oxidative behaviour of AO7. Contrary to Fig. 2I,II shows that, I pC0 at the start of coulometry, I pC0 start , is more than 2.5 times than that of I pA1 at the end of coulometry ( ) I pA1 end which is near to theoretical value of 2.8 when n 1 = n 2 33 . Diagnostic criteria of cyclic voltammetry accompanied by previously published data on oxidation of AO7 [34][35][36][37][38][39][40][41][42][43] allow us to propose the mechanism presented in Fig. 3 for the electrochemical oxidation and reduction of AO7. In oxidation pathway, generation of AO7 ox is followed by the addition of H 2 O and formation of AO7 OH . At the final step, AO7 OH degraded into 4-nitrosobenzenesulfonate (4NB) and 1-iminonaphthalen-2(1 H)-one (INO). It should be noted that, according to the Fig. 3, the number of transferred electron in oxidative pathway is two electrons, while, the obtained number of transferred electron (from controlled potential electrolysis) was 2.9 electrons. This discrepancy can be related to partially oxidation of 4NB 19 .
In the reduction pathway, the first two-electron reduction converts AO7 to AO7 R . In the next step, AO7 R via an irreversible two-electron degradation process converts to 4-aminobenzenesulfonate (4AB) and 1-aminonaphthalen-2-ol (ANO).
According to the above data, the cathodic peak C 0 corresponds to reduction of AO7 to its reduced form (AO7 R ). The anodic peak A 2 corresponds to the two-electron oxidation of AO7 to its oxidized form (AO7ox). Obviously, the cathodic peak C 2 corresponds to the reduction of AO7 ox into AO7 and the redox couple A 1 /C 1 , are related to the oxidation of ANO to INO and reduction of INO to ANO within a quasi-reversible two-electron, two-proton process.
Adsorption Study. Fig. S4 shows the normalized oxidative cyclic voltammograms of AO7 (the data of  S1). The normalization was performed by dividing the current by the square root of the potential scan rate (I/v 1/2 ). According to the proposed pathway for the electrochemical oxidation of AO7, the increasing of normalized A 2 peak current (I pA2 /v 1/2 ), was unexpected. One possibility for such inconsistency is adsorption of AO7 on the electrode surface. To confirm this finding, the plot of log I PA2 vs. log v at pH values 2.4, 7.2 and 10.3 is shown in Fig S5. It was reported that when the slope log I PA2 vs. log v is 0.5, the electrochemical reaction is a diffusion controlled process, while when the slope increases to 1, the electrochemical reaction occurs via an adsorption-controlled process 32 . It is clear that, in all pHs, the slope is more than 0.5 and increases with increasing pH from 0.58 to 0.74. These values are greater than 0.5 for the diffusion controlled process and are less than one, which is theoretical value for the adsorption-controlled electrode process. Therefore, it is clear that the electrochemical oxidation of AO7 at glassy carbon electrode in aqueous media is adsorption/diffusion process. These results show that the interaction between anionic forms of AO7 and the electrode surface is stronger than neutral form. The Effect of pH. The electrochemical behavior of AO7 has been studied in different pH values. The oxidative and reductive cyclic voltammograms of AO7 in aqueous solution with various pHs are shown in Figs S6 and S7, respectively. As seen in Fig. S6, with increasing pH, E pA2 shifts to negative values. This confirms the participation of proton(s) in the oxidation of AO7. The variation of E pA2 with pH is given by: where m is the number of protons involved in the reaction, = E pA2 pH 0 is the anodic peak potential at pH = 0.0 and R, T, and F have their usual meaning. The E pA2 -pH diagram comprise three linear segments with different equations and slopes at pH values 7.4 and 11.4 (Fig. 4I). This diagram indicates that in the aqueous solutions, AO7 is in different reduced and oxidized forms, that their relative amounts are dependent on the pH and electrode potential. At pHs lower than 7.4, the E pA2 value shifts by − 29 mV/pH indicating that the redox reaction is two-electron/ one-proton process involving the oxidation protonated AO7 (HAO7) to protonated AO7 ox (Fig. 5, Eq. 1) [44][45][46] . Moreover, at pH range 7.4-11.4, the E pA2 value shifts by − 61 mV/pH. In this range of pH, the redox reaction is two-electron/two-proton process involving the oxidation of HAO7 to AO7 ox (Fig. 5, Eq. 2).
Finally, at pHs > 11.4, the E pA2 value is independent of pH, showing that the redox reaction involves a two-electron process without participation of any proton including the oxidation of AO7 anion (AO7 − ) to AO7 ox (Fig. 5, Eq. 3). The important point of Fig. 5 is the absence of neutral AO7, which can be related to the high tendency of AO7 to keep the proton on the nitrogen atoms due to the intramolecular hydrogen bonding (Fig. S8). In addition, the calculated pK a for HAO7 ox /AO7 ox and AO7/AO7 − equilibria which is also shown in Fig. 5, Eqs 4 and 5, are: 7.4 and 11.4, respectively.
The reductive cyclic voltammograms of AO7 in aqueous solution with different pHs are shown in Fig. S7. As seen, with increasing pH, peak C 0 shifts to negative potentials, indicating the participation of proton(s) in the electrode process. The E PC0 -pH diagram is shown in Fig. 4II. It has two linear segments at pH 7.7. At pH values lower than 7.7, E pC0 value shifts by − 90 mV/pH indicating the redox reaction is two-electron/three-proton process involving the reduction of HAO7 to H2AO7 R (Fig. 5, Eq. 6). However, at pHs > 7.7, the situation is a little complicated, as the E pC0 value shifts by − 40 mV/pH. It probably including unrecognizable reductions HAO7 to HAO7 R (two-electron/two-proton process) and HAO7 to AO7 R (two-electron/one-proton process) (Fig. 5, Eqs.  7 and 8).   The cyclic voltammograms regarding A 1 /C 1 peaks at different pHs and its E 1/2 -pH diagram are shown in Fig. S9. The E 1/2 values were calculated as the average of the anodic and cathodic peak potentials (E pA1 + E pC1 )/2. The E 1/2 -pH diagram displays a simple linear dependence of the E 1/2 on pH, with the slope of − 85 mV/pH, indicating that the redox reaction is two-electron/three-proton process involving the oxidation of HANO to INO in the forward scan and reduction of INO to HANO in the reverse scan (Fig. 5, Eq. 9). The extrapolation of this line to pH = 0.0 provides the E 1/2 = + 0.60 V versus Ag/AgCl for redox couple A 1 /C 1 .

Electrochemical Study of AO7 in the Presence of Arylsulfinic acids. Because of two reasons: a)
confirmation of the proposed mechanism in Fig. 3 and b) electrochemical synthesis of some new organic compounds, in this part, the electrochemical behavior of AO7 in the presence of 4-toluenesulfinic acid (1a) was studied and compared with that of AO7 in the absence of 1a (Fig. 6). The cyclic voltammogram of AO7 in aqueous phosphate buffer (c = 0.2 M, pH = 2.0), in the presence of 1a is shown in Fig. 6b. Comparison of the voltammogram with that of AO7 in the absence of 1a (Fig. 6a), shows two important differences: (a) the appearance of a new redox couple (peaks A 3 and C 3 ) at E 1/2 = 0.39 V vs Ag/AgCl. (b) The disappearance of the cathodic peak C 1 in the reverse scan. Under these conditions, the peak current ratio, I pC1 /I pA1 , depends on the both potential scan rate and 1a concentration, so that, I pC1 /I pA1 increases with increasing scan rate and decreasing 1a concentration. In Fig. 6, curve c is the voltammogram of 1a in the same conditions and in the absence of AO7 that does not show any peak in the working potential range. In addition, cyclic voltammogram d, is belong to the isolated product from the electrolysis of AO7 in the presence of 1a. In this part, the same results were obtained for the other two arylsulfinic acids (1b and 1c). These results and the spectroscopic data of the isolated electrolysis product all point out to compound 3a (final product), which would have been formed according to the pathway shown in Fig. 7.
The first step in the synthesis of 3a is the generation of INO. This compound is directly generated from oxidative cleavage of AO7 (Fig. 3), and/or from the two-electron oxidation of ANO (Fig. 7). In the next step, INO would serve as a Michael acceptor in a reaction with 1a to form the final product 3a. The oxidation of 3a did not occur during the electrolysis due to the insolubility of 3a in electrolysis solution (aqueous phosphate buffer). Based on Fig. 7, the anodic and cathodic peaks A 3 and C 3 pertain to the oxidation of 3a to 4a and vice versa. According to the proposed mechanism in Fig. 3, we have designed a paired electrochemical strategy for the synthesis of the sulfone derivatives 3a-3c. The paired electrochemical synthesis of 3a-3c has been successfully performed in a one-pot process at the current density of 0.32 mA/cm 2 , in an undivided cell equipped with carbon rods as cathode and anode. The electrolysis was terminated when the cathodic peak that corresponds to the reduction of AO7 (C 0 ) disappears. This peak disappears after consumption of 4.0 F/mol electricity. Under these conditions, the monitoring of the electrolysis progress was carried out by cyclic voltammetry and shown in Fig. 8. This Figure shows that, proportional to the advancement of coulometry, I pC0 , I pA1 and I pA2 decreases while I pA3 and I pC3 increases. The variation of I pC0 vs. charge consumed is also shown in Fig. 8 (inset). This curve show that I pC0 decrease exponentially with advancing coulometry (I pC0 = 10.15 e −0.03Q ). The total amount of charge passed for terminating the reaction was determined from the extrapolation to the X-axis. The calculated charge passed confirms consumption of about 4e − per molecule of AO7.
In addition, the UV-visible spectra of AO7 in the presence of 4-toluenesulfinic acid (1a) were collected during a constant current coulometry in the same conditions as before (Fig. S11). Under these conditions, the absorption spectrum of AO7 consists of three absorption bands at 312, 410 and 488 nm. Our data show as the coulometry is carried out, the height of all three peaks decrease and a new peak appears at 352 nm and grows in intensity.
INO is an asymmetric Michael acceptor and can be attacked by 1a via 1,4 (site A) or 1,6 (site B) Michael addition reaction to yield two types of products (3a and 3a') ( Fig. 9). However, the recorded NMR spectrum shows a singlet peak at 8.04 ppm belongs to aromatic proton. This result confirms that the 1,4-Michael addition is more probable reaction and compound 3a is the final product of the electrochemical oxidation of AO7 in the presence of 1a. In order to increase the yield of 3a-3c, some affecting factors must be optimized. Therefore, the effects of two of the most important factors, applied current density and charge were investigated by setting all parameters to be constant and optimizing one each time. The effect of charge passed was studied in the range of 1 to 6 F mol −1 , while the other parameters are as follows: temperature = 298 K, current density = 0.32 mA/cm 2 , electrode surface = 31.2 cm 2 , AO7 = 0.1 mmol and 1a = 0.1 mmol are kept constant. As is shown in Fig. S10a, the maximum product yield appears in 4.1 F mol −1 charge consumed. The product yield decreases with increasing charge passed from 4.1 F mol −1 probably due to the occurrence of side reaction(s) such as over-oxidation. Furthermore, the effect of applied current density on product yield was studied in the range 0.16-1.6 mA cm −2 , while the other parameters (temperature = 298 K, charge consumed = 40 C, electrode surface = 31.2 cm 2 , AO7 = 0.1 mmol, and of 1a = 0.1 mmol) are kept constant. The results show that, with increasing the current density from 0.32 mA cm −2 , the product yield decreases (Fig. S10b). The product yield decreasing in current densities greater than 0.32 mA cm −2 , can be related to some side reactions such as oxidation of solvent, nucleophile or over-oxidation of 3a and/or INO. To evaluate the usefulness of the pair strategy in the synthesis of 3a-3c, electrochemical synthesis of 3a was performed in a divided cell in both oxidative and reductive conditions. Our data confirms that in a divided cell (in both cases) (unpaired condition), (a) the yield of 3a is lower and (b) the charge consumption is greater than that of in undivided cell.

Conclusions
This work provides new insights into the electrochemical behavior of AO7 in aqueous solutions in both oxidative and reductive regions and shows that both oxidation and reduction of AO7 leading to the formation of a redox couple (ANO/INO) (Fig. 3). In addition, the pH dependence of AO7 and other intermediates was studied in order to understand the predominant species, oxidation and reduction pathways and adsorption study. For example, our data shows that, the interaction between anionic forms of AO7 and the electrode surface is stronger than neutral form. Furthermore, in this work, the electrochemical oxidation/reduction of AO7 has been investigated in the presence of arylsulfinic acids (1a-1c) as nucleophiles, in acidic solutions. Our data display that the intermediate (INO) is attacked by the nucleophile, 1a-1c, to give the final product 3a-3c (Fig. 7). Clean synthesis, technical feasibility (using galvanostatic method and simple cell), use of electricity instead of oxidative or reductive reagents, one-step process, work in room temperature and pressure and use of aqueous solution instead of organic solvents, are the advantages of this method.

Materials and Methods
Apparatus and Reagents. Cyclic voltammetry, controlled-potential coulometry and preparative electrolysis were performed using an Autolab model PGSTAT 30 and a Behpazho potentiostat/galvanostat. The working and counter electrode used in macro-scale electrolysis and coulometry was an assembly of four ordinary soft carbon rods (6 mm diameter and 4 cm length). Working electrode used in the cyclic voltammetry experiments was a glassy carbon disc (1.8 mm diameter) and a platinum rod was used as a counter electrode. The electrosynthesis were performed under constant-current condition in an undivided cell. The glassy carbon electrode was polished using alumina slurry (from Iran Alumina Co.). More details are described in our previous paper 47 . Acid orange 7, arylsulfinic acids and phosphate salts were obtained from commercial sources. These chemicals were used without further purification.