Transalkylidation reaction: green, catalyst-free synthesis of thiosemicarbazones and solving the NMR conflict between their acyclic structure and intramolecular cycloaddition products

ABSTRACT A series of arylidene thiosemicarbazides have been prepared by a new method, which was titled as transalkylidation. This method is an effective, fast, green and clean method. The mechanism of this reaction has been discussed. Moreover, we clarified the divergences of the structural assignments reported in the literature for the reaction of thiosemicarbazide and aldehydes or ketones in the presence of different catalysts. Where 1,2,4-triazolidine-3-thiones were incorrectly reported as sole product of such reaction, based on occurance of intramolecular cycloaddition of thiosemicarbazones formed in situ. Our findings proved that the reaction stops at earlier stage of thiosemicarbazone and neither cyclization to 1,2,4-triazolidine-3-thiones nor 2-amino-1,3,4-thiadiazoline take place. We have removed the confusion of the NMR interpretation of thiosemicarbazone and their cycloaddition product by carrying out 1H-NMR, 13C-NMR, 15N-NMR and 1H-15N HSQC experiments with temperature gradient. Furthermore, DFT-NMR calculations have been done to make structural distinguish between the three possible structural isomers for this reaction, namely, 1,2,4-triazolidene-3-thione, 2-amino-1,3,4-thiadiazoline and thiosemicarbazone. GRAPHICAL ABSTRACT

Thiosemicarbazones play a significant role in the regulation of plant growth because of their capability for diffusion through the semipermeable membrane of cell lines (22)(23)(24). As well as other industrially important activities, including antifouling and anticorrosion effects (25,26) have also been reported for these compounds. Moreover, The imine bond (C = N) in these compounds are useful in the organic synthesis, in particular for the synthesis of many of pharmaceutical and bioactive heterocycles such as, thiazoles (27)(28)(29)(30)(31), 1,3,4-thiadiazoles (32), thiazolidinones (33,34), quinolines (35), naphthyridines (36), oxadiazoles (37,38), pyrimidines (39). Water plays an important role in our life. Although, the fact that water as a solvent is the cheapest and most non-toxic solvent, there are many limitations of water use as a solvent in organic synthesis. Additionally, its use as a solvent in organic synthesis avoided sophisticated drying of solvents and substrates before use in many reactions Therefore, one of the recent challenges, due to the environmental and health concerns, is to use green synthesis in aqueous medium (40)(41)(42)(43)(44).

Results and discussion
Conventional and unconventional methods for the synthesis of alkylidene thiosemicarbazide had been extensively reported (58)(59)(60)(61)(62)(63). Most of these methods involve condensation reaction of carbonyl compounds and thiosemicarbazide in alcohol with catalytic amounts of acetic acid or mineral acids for about 3-72 h at room temperature or under reflux (59)(60)(61)(62)(63)(64). Moreover, recently a more facile synthesis under ultrasound or microwave (MW)assisted methods were reported (58). Herein, we would like to report a catalyst-free, simpler and cleaner (water as solvent) synthesis of alkylidene thiosemicarbazide in only few minutes reflux.
A reaction of acetone with thiosemicarbazide in water as a solvent proceeded effectively to afford isopropylidene thiosemicarbazide (1) in 95% yield. Subsequent reaction of (1) with different aldehydes 2a-i in aqueous medium gave arylidene thiosemicarbazide (4) instead of the expected arylidene isopropylidene thiosemicarbazide 3. We explained this reaction in the term of transalkylidation in which the arylidene group replaces the isopropylidene in good to excellent yields (72-95%) (Scheme 1). To the best of our knowledge, there are no reports published dealing with transalkylidation reaction either for semicarbazones or thiosemicarbazones. Aromatic aldehydes were added to a hot aqueous solution of acetone thiosemicarbazone 1 to afford the corresponding arylidene which filtered while hot as a pure precipitate. The insolubility of the arylidene thiosemicarbazide in hot water makes its isolation as pure product is so easy and gives this method an advantage over the previously reported direct condensation methods of the aromatic aldehydes with thiosemicarbazide that need an extra step for purification of the products.
To evaluate the solvent effect, the reaction between benzaldehyde and acetone thiosemicarbazone 1 was selected as a model reaction and was conducted using various type of solvents (non-polar, polar protic and polar aprotic solvents) such as toluene, H 2 O, EtOH, DMF and CH 3 CN.
When the reaction was carried out in aprotic solvent either polar or non-polar solvents, DMF, CH 3 CN and toluene, no product was detected (Table 1, entries [3][4][5]. Whereas in ethanol, the reaction occurred with good yield (80%) but it requires longer reaction time (4 h) to go to completion (Table 1, entry 2). Interestingly, best results obtained in case of employing water as a solvent ( Table 1, entry 1). based on the criteria such as reaction time (5 min), green nature, cost-effective and excellent yield (95%), water as solvent proved to be Scheme 1. Transalkylidation reaction.
supreme solvent for the present procedure. Given this excellent result, we decided not to try additional optimization for the experimental conditions and used these to investigate the substrate scope of the reaction ( Table 2).

Substrate scope
To examine the scope of the transalkylidation reaction in an aqueous medium for the preparation of thiosemicarbazones, a series of aromatic aldehydes or ketones with structurally divergent functional groups were examined ( Table 2).
It was found that transalkylidation method is tolerant for both electron-rich (Me, OH, Me 2 N) and electrondeficient (NO 2 , Cl) aromatic aldehydes. But less tolerant for aromatic ketone such as acetophenone. This could be attributed to lower reactivity of ketones compared to aldehydes towards nucleophilic attack. Moreover, the reaction proceeded effectively with unsaturated aldehyde (cinnamaldehyde) to afford the corresponding thiosemicarbazone (allylidenethiosemicarbazide). Interestingly, different positions (o, m, p) on the phenyl group did not show any significant effect on the reaction time, purity and yield percent. Unfortunately, when aliphatic aldehydes were employed the reaction gives the corresponding thiosemicarbazones in poor yield. (Table 2, entries 12,13)

Reaction mechanism
The findings that the transalkylidation reaction only takes place in polar protic solvent and doesn't proceed in aprotic polar or non-polar solvents, suggests an important role for the solvent. So, we proposed a mechanism of the transalkylidation reaction as shown in Scheme 2. Nucleophilic attack of water molecule on the carbon atom of imine group (C = N) of acetone thiosemicarbazone induces the nucleophilic attack of the π-electrons of the imine double bond on the carbonyl group (C = O) of the aromatic aldehyde to afford the intermediate I. then Proton transfer occur in the second step to give the intermediate II, subsequent loss of water and acetone molecules afford the final product. The nucleophilicity of the solvent (water or ethanol) plays a vital part in this mechanism as it initiates the reaction. Furthermore, its polar character stabilizes the ionic intermediates I and II which was formed during the reaction pathway. That was confirmed when, no product detected either in polar aprotic (DMF, CH 3 CN) or nonpolar aprotic solvent (toluene) as they are considered to be non-nucleophilic solvents.

NMR argument for thiosemicarbazones and their cycloaddition products
All prepared products were fully identified and characterized by analytical, spectroscopic techniques (see supplementary information).
For all of the prepared thiosemicarbazones, the NH 2 group appears as two peaks in proton NMR with one proton for each. This is due to the partial double bond character induced by the thioamide moiety Figure 1, which restricts the free rotation, making the two protons of the NH 2 group diastereotopic protons or electronically nonequivalent. One is trans to the NH and the other is cis. This is a very common characteristic of many amide or thioamide-containing molecules.
To confirm this assumption, the temperature effect on proton NMR have been investigated for compound 4c. At low temperature, the energy barrier for rotation is high to prevent averaging (fast rotation), the two NH 2 protons are non-equivalent and showed two peaks. Rising up the temperature, signal averaging takes place, and the two protons begin to exchange on the NMR timescale. As this rate of exchange becomes faster, the signal coalescing and finally complete signal averaging to the point where the two protons are equivalent Figure 2.
Additionally, 15 N-NMR and 1 H-15 N Heteronuclear Single Quantum Coherence (HSQC) experiments were carried out for compounds 4c and 4d. As shown in Figures 3 and 4. Three and four 15 N-NMR peaks were observed for 4c and 4d respectively. The four nitrogen peaks of compound 4d were assigned to the three nitrogens of the thiosemicarbazone and the fourth nitrogen corresponds to the nitro group. The cross-peaks due to correlating the 15 N chemical shifts with H chemical shifts which are spin-coupled showed that the nitrogen appeared at highest field correlates to two protons in 1 H-NMR, which clearly proves its chemical nature as NH 2 group. Moreover, cross-peak has been observed between the next nitrogen and the most deshielded  proton indicating NH group. While no cross peak relates the third nitrogen to any proton which corresponds to imine nitrogen (C = N). For compound 4d the most downfield nitrogen appeared is due to the nitro group of the arylidene moiety. As mentioned before, the reaction of aldehydes or ketones with thiosemicarbazide could afford either condensated product (thiosemicarbazone) or continued further to afford 1,2,4-triazolidene-3-thione or 2amnio-1,3,4-thiadiazoline as cycloaddition product of thiosemicarbazone intermediate.Herein, we present examples in which incorrect interpretation of NMR was reported. In all of these examples, the products were incorrectly assigned as 1,2,4-triazolidene-3thione (B) instead of thiosemicarbazones (A) Figure 5. This confusion is due to the appearance of three exchangeable peaks in 1 H NMR, which wrongly interpreted to three NH peaks correspond to 1,2,4-triazolidene-3-thione instead of NH and NH 2 peaks corresponding to thiosemicarbazones. Furthermore, in order to make structural distinguish between the three structural isomers namely: triazolidenethione, thiadiazole and thiosemicarbazone. DFT-NMR calculations were done for the previously optimized possible structures using gauge-including atomic orbital GIAO/ B3LYP density functional method with 6-31G basis set by Gaussian 09 software. The atom labelling for the three isomers are shown in Figure 6, and the calculated chemical shifts are listed in Table 3. Notes: a reaction conditions: 2 mmol of Aldehyde or ketones, 2 mmol of Acetone Thiosemicarbazone 1, 20 ml water, 5-10 min. b nd, not detected. Scheme 2: Transalkylidation reaction mechanism. Figure 1. Resonance of thioamode group in thiosemicarbazones. Table 3 showed that besides the presence of the thiocarbonyl group (C = S) which appears at ≈170-180 ppm, thiosemicarbazone is characterized by its azomethine proton and carbon at chemical shift ≈7-8 and 140 ppm in 1 H NMR and 13 C-NMR respectively. While cycloaddition of thiosemicarbazone to afford triazolidene-3-thione or thiadiazoline led to disappearance of the azomethine group and presence of saturated carbon involved in the ring formation. The saturated carbon and its proton should appear at chemical shift ≈70-80 and 4-5 ppm in 13 C-NMR and 1 H-NMR respectively. Furthermore, the thiocarbonyl moiety still exists in triazolidine but absent in case of thiadiazoline. To confirm the above data, acetylation of 1 with acetic hydride in pyridine to afford thiadiazoline 5 has been done (Scheme 3).

Inspection of
Thiadiazoline 5 showed a peak at 80 ppm in 13 C-NMR corresponding to the saturated carbon of the thiadiazoline ring and disappearance of peak at 179.17 ppm corresponding to thiocarbonyl moiety, which agrees to the theoretical calculation.
Based on these findings, cyclocondensation of thiosemicarbazide and aldehydes or ketones affording 1,2,4-triazolidine-3-thiones has been incorrectly reported several times in the literature, either using uncatalyzed reaction (56,65) or using catalyst such as DMAP (46) (45). In all these reports' authors based their structural assignment on the presence of three exchangeable protons for three NH groups but both peaks at ≈80 and 4 ppm were missing in 13 C-NMR and 1 H-NMR respectively. Instead, peaks at 140 and 8 ppm were present, which are characteristic for the imine functionality; Moreover, the three exchangeable protons could also correspond to the NH and NH 2 of the acyclic isomers as proven before. This indicates that the reaction stopped at the earlier stage of condensation affording thiosemicarbazone and no cyclization to triazolidine had been taking place. The presence of three exchangeable protons accounts for the confusion between the acyclic and the cyclized form in these reports.
To support our idea, the spectral data ( 1 H and 13 C) for previously reported 5-phenyl-1,2,4-triazolidine-3-thione or 5-(4-methoxyphenyl)-1,2,4-triazolidine-3-thione (45-48,56,69) together with the spectral data of their acyclic forms (compounds 4a and 4j respectively) obtained from this work are listed in Tables 4 and 5 respectively. Inspection of Table 4 or 5 showed that the reported spectral data of the cyclic forms are identical to that of the acyclic forms. This shows the misleading in the assignment of the structure in these previous studies.
Furthermore, beside this method, there is another reaction titled Criss-Cross Cycloadditions of ketazines led to formation of 3,7-diphenyltetrahydro- [

Materials and methods
Unless otherwise stated, CH 3 CN, DMF and toluene were purified and dried by distillation over 4A molecular sieves. Thiosemicarbazide, aldehydes and ketones were purchased from Sigma-Aldrich and used without further purification. 1 H NMR, 13 C NMR, and 15 N NMR spectra were recorded at room temperature on a Bruker AC 300 FT, an Avance Bruker DPX 300, or on a Bruker Avance DRX 400 spectrometers.

Synthesis of 2-(propan-2ylidene)hydrazinecarbothioamide
A hot solution of thiosemicarbazide (0.91 g, 10 mmol) in 20 mL of water was added to Acetone (10 mmol), Then the mixture was heated under reflux for 2 h. After cooling, the reaction mixture was filtered, and the obtained white solid was washed with water. The crude product was purified by crystallization from hot water.

General procedure for synthesis of alkylidene thiosemicarbazide
Aldehyde or ketones (2 mmol) were added to hot solution of Acetone Thiosemicarbazone 1 (2 mmol) in 20 ml water then the suspension was heated and stirred for few minutes until starting material 1 is consumed as detected from TLC. The precipitate formed was collected by filtration while hot.

Conclusions
In conclusion, we introduced a new, green and a more facile reaction titled transalkylidation which can be used for replacing isopropylidene moiety by arylidene group affording a green synthesis of a variety of arylidene thiosemicarbazide. Based on the presence of three exchangeable protons for thiosemicarbazones and their cycloaddition products, it was not so easy to distinguish between them. This confusion promotes us to make full identification for each isomer using spectral data and DFT calculations. Furthermore, simple distinguish between the three isomers could be done using 13 C-NMR. Where the appearance of peaks at ≈80 and 180 ppm in 13 C-NMR indicates the existence of triazolidine-3-thione ring while lacking a peak at ≈80 ppm and presence of peak at 180 ppm only proves the presence of the acyclic form. Furthermore, appearance of a peak at 80 ppm and absence of peak at 180 ppm verifies the presence of thiadiazoline ring Table 6.   Table 6. Summary for the most characterstic peaks 13 C-NMR for the three structural isomers. 13 C-NMR (ppm) triazolidene-3thione thiosemicarbazone thiadiazoline ring

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributors
Mohamed A. El-Atawy, an Assistant Professor of Organic Chemistry, his research interest is focused on synthestic methods and structure elucidation of heterocyclic compounds. Moreover, experimental and theoretical studies of the reaction mechanism of certain reactions. Among his recent research "new synthetic methodolgy for pyrroles from nitrodienes".
Alaa Z. Omar, an Assistant Professor of Organic Chemistry. He currently is the principal investigator for a research group that has ten postgraduate students. His current research interests include the synthesis of unique polymers for solar cell applications, synthesis of novel dyes for industrial applications, design and specified synthesis for antiviral agents, and the synthesis of anticorrosive agents.
Mohamed Hagar, an Associate Professor of Organic Chemistry. His current research interests include the synthesis and structure elucidation of heterocyclic compounds, synthesis of variety of materials used for liquid crystal applications.
Essam M. Shashira, a Professor of Organic Chemistry. His research interest is focused on both photochemistry and synthesis of heterocyclic compounds with evaluation of their biological activities.