Copper(II) Prevents the Saccarine-Dialkylcyanamide Coupling by Forming Mononuclear (Saccharinate) (Dialkylcyanamide)copper(II) Complexes

: The reaction in the system Cu II /sacNa(H)/NCNR 2 (sacNa(H) = sodium saccharinate (saccharin); R = Me, Et) results in the formation of the complexes [Cu(sac) 2 (NCNR 2 )(H 2 O) 2 ] (R = Me 1 , Et 2 ) instead of the expected products derived from the saccharin–cyanamide coupling. Complexes 1 , 2 , and hydrate 1 · 2H 2 O were characterized by IR, AAS (Cu%), TGA, and also by single-crystal X-ray diffraction for 1 and 1 · 2H 2 O. An integrated computational study of model structure 1 in the gas phase demonstrates that the Cu–N cyanamide and Cu–N sac coordination bonds exhibited a single bond character, polarized toward the N atom and almost purely electrostatic, with the calculated vertical total energies for the Cu–N cyanamide and Cu–N sac of 43.6 and 156.4 kcal/mol, respectively. These data conﬁrmed that the copper(II) completely blocks the nucleophilic centers of ligands via coordination, thus preventing the saccharin–cyanamide coupling.


Introduction
Saccharin (sacH), which is a common and widely used artificial sweetener [1], also finds application in diverse areas of chemistry, particularly being applied as a catalyst of a number of organic reactions (leading to, e.g., compounds of biological importance [2][3][4][5]) and a functional group protector [2,6]. SacH is also applied in the preparation of drugbased co-crystals exhibiting an improved solubility and bioavailability compared to the poorly soluble parent drugs [7].
Our recent finding in the saccharin chemistry was the observation of unusual metalfree "two saccharin-one cyanamide" coupling between sacH and NCNR 2 to grant guanidinated saccharin derivatives [8]. The distinctive feature of this transformation is its formal three-component character, while other known examples of reactivity of sacH toward multiple bond substrates includes the additions to amino alkynes [9,10], isocyanide [11], isocyanates [12], and carbodiimides [12,13]; all these examples include two-component 1:1 additions.
In light of the general interest in coordination chemistry of NCNR 2 (R 2 = Alk 2 , Ar 2 , H/Alk, H/Ar, H/CN, etc.) and their metal-mediated and metal-catalyzed reactions (for our and other group reviews on metal-involving reactivity and coordination chemistry of cyanamides see [14][15][16][17], respectively), we now studied the reaction between sacH and NCNR 2 in the presence of a metal center, which as we assumed, should change its directionality. For these purposes, we addressed copper(II), as a kinetically labile metal center, which is commonly used in various catalytic reactions including coupling-based transformations. We observed that the introduction of copper(II) completely changed the directionality of the reaction by blocking the nucleophilic centers of both reactants Inorganics 2021, 9,  by the ligation to give (sac)(NCNR 2 )Cu II species instead of the guanidinated saccharin derivatives.

Synthesis and Characterization
For the study of the Cu II -involving reactions between cyanamides and saccharin, we addressed the copper salts CuCl 2 ·2H 2 O or CuBr 2 , N,N-disubstituted cyanamides NCNR 2 (R = Me, Et), and sacH or its sodium salt (sacNa). These reactions were performed in a neat NCNR 2 or NCNR 2 /solvent mixture (solvent = H 2 O, MeOH, EtOH) where the copper salts and saccharin(ate) are well soluble. In the systems, Cu II /NCNR 2 /SacNa, [Cu(sac) 2 (NCNR 2 )(H 2 O) 2 ] (R = Me 1, Et 2) complexes were obtained and isolated as crystalline solids. The results previously reported for metal-free process "two saccharinone cyanamide" addition products were neither isolated, nor identified by HRESI + -MS in the reaction mixtures, although peaks from the 1:1 addition product were detected in the HRESI + -MS spectra (see ESI, Supplementary Materials Figures S1 and S2). Furthermore, we optimized the synthesis conditions in order to increase the yield of complexes 1 and 2.
Copper(II) saccharinate/dimethyl cyanamide complexes 1 and 1·2H 2 O (50-75%) were obtained by the dissolution of CuCl 2 ·2H 2 O or CuBr 2 in NCNR 2 (R = Me, Et) at 60 • C followed by the addition of a solution of sacNa in MeOH, EtOH, or H 2 O (Scheme 1). Notably, the choice of solvent affects the release of either hydrated, or anhydrous form of the complex. When the reaction proceeds in dried EtOH, complex 1 was obtained, while from undried MeOH, EtOH, or in H 2 O, hydrate 1·2H 2 O was isolated. In this reaction, sacH can be used instead of sacNa, however, this treatment gives a lower yield of the product (up to 35% for 1·2H 2 O). The yield reduction can be explained by a lower reactivity of the protonated form (pK a of sacH = 1.6 [18] in H 2 O). The reason of the moderate yield of 1 and 1·2H 2 O is the formation of various unidentified by-products. IR monitoring of the filtrate from the reaction mixture verified broad bands at 1641s and 1620 m-s sh cm -1 and band 2251 cm -1 s, which were attributed to ν(C=O) from the saccharinate moiety of the 1:1 addition product, and Me 2 NC(O)NH 2 , and ν(CN) cyanamide from the coordinated NCNMe 2 , respectively. nics 2021, 9, 69 2 of 11 ligation to give (sac)(NCNR2)Cu II species instead of the guanidinated saccharin derivatives.

Synthesis and Characterization
For the study of the Cu II -involving reactions between cyanamides and saccharin, we addressed the copper salts CuCl2·2H2O or CuBr2, N,N-disubstituted cyanamides NCNR2 (R = Me, Et), and sacH or its sodium salt (sacNa). These reactions were performed in a neat NCNR2 or NCNR2/solvent mixture (solvent = H2O, MeOH, EtOH) where the copper salts and saccharin(ate) are well soluble. In the systems, Cu II /NCNR2/SacNa, [Cu(sac)2(NCNR2)(H2O)2] (R = Me 1, Et 2) complexes were obtained and isolated as crystalline solids. The results previously reported for metal-free process "two saccharin-one cyanamide" addition products were neither isolated, nor identified by HRESI + -MS in the reaction mixtures, although peaks from the 1:1 addition product were detected in the HRESI + -MS spectra (see ESI, Supplementary Materials Figures S1 and S2). Furthermore, we optimized the synthesis conditions in order to increase the yield of complexes 1 and 2.
Copper(II) saccharinate/dimethyl cyanamide complexes 1 and 1·2H2O (50-75%) were obtained by the dissolution of CuCl2·2H2O or CuBr2 in NCNR2 (R = Me, Et) at 60 °С followed by the addition of a solution of sacNa in MeOH, EtOH, or H2O (Scheme 1). Notably, the choice of solvent affects the release of either hydrated, or anhydrous form of the complex. When the reaction proceeds in dried EtOH, complex 1 was obtained, while from undried MeOH, EtOH, or in H2O, hydrate 1·2H2O was isolated. In this reaction, sacH can be used instead of sacNa, however, this treatment gives a lower yield of the product (up to 35% for 1·2H2O). The yield reduction can be explained by a lower reactivity of the protonated form (pKa of sacH = 1.6 [18] in H2O). The reason of the moderate yield of 1 and 1·2H2O is the formation of various unidentified by-products. IR monitoring of the filtrate from the reaction mixture verified broad bands at 1641s and 1620 m-s sh cm -1 and band 2251 cm -1 s, which were attributed to ν(C=O) from the saccharinate moiety of the 1:1 addition product, and Me2NC(O)NH2, and ν(CN)cyanamide from the coordinated NCNMe2, respectively. Upon the extension of the complexation to the other cyanamides NCNR2 (R2 = Et2, C4H8, C5H10, C4H8O), we found that CuCl2·2H2O and CuBr2 were almost insoluble in most of these cyanamides (R2 = C4H8, C5H10, C4H8O). Therefore, we modified the reaction conditions, namely CuCl2·2H2O (or CuBr2) was dissolved in THF and then a solution of both sacNa and NCNR2 in MeOH (or EtOH) was added to a THF solution. Under these conditions, a lantern-like complex [Cu2(sac)4(THF)2]·2THF (3·2THF) was obtained. When we attempted to replace THF in this reaction with another solvent ( Complexes 1, 1·2H 2 O, and 3·2THF were characterized by IR, AAS (Cu%), and TGA methods and also by single-crystal X-ray diffraction. Complex 2 was characterized by IR and AAS (Cu%). The AAS (Cu%) data agree with the calculated values for the proposed formulas. In the IR spectra of 1, 1·2H 2 O, and 2, the ν(C≡N) of the NCNR 2 ligand was observed in the range 2226-2245 cm -1 ; these values are comparable with those observed for the homoleptic complexes [Cu(NCNR 2 ) 4 ](BF 4 ) (R = Me, Et; ca. 2240 cm -1 ) [20] and the mixed-ligand copper(I) complexes [Cu(tpm)(NCNR 2 )](BF 4 ) (R = Me, Et; ca. 2250 cm -1 ) [21] and the clusters [Cu 4 X 6 O(NCNMe 2 ) 4 ] (2255-2261 cm -1 ) [22]. In the IR spectra of all complexes, the two strong bands in the ranges 1306-1330 and 1165-1177 cm -1 were attributed to ν sym (SO 2 ) and ν asym (SO 2 ) of the saccharinate ligands, respectively. We also attempted mass-spectrometric characterization of the obtained species, but the HRMS + (ESI) spectra (in MeCN) did not display molecular ions for all complexes and only products of deep fragmentation of complexes (e.g., ions Cu(NCMe) 2 + ) were observed due to the lability of copper(II) complexes in a solution.
Complexes 1 and 2 are stable until ca. 70 • C and then decompose with the loss of the NCNR 2 ligand in the interval ca. 70-150 • C and two H 2 O ligands at ca. 150-250 • C; after 250 • C, the residue undergoes decomposition of sacligands forming yet unidentified species. Hydrate 1·2H 2 O somehow demonstrates greater stability and its decomposition starts at ca. 100 • C with loss of the crystallization water and the NCNMe 2 ligand at 100-150 • C; after 250 • C, the decomposition to yet unidentified species occurs. Complex 3·2THF is stable until ca. 120 • C and then starts losing the solvated THF (125-260 • C), whereupon the coordinated THF (260-326 • C) is lost. After 326 • C, non-stochiometric decomposition of the sacligands occurs.
Complexes 1 and 2 are stable until ca. 70 °C and then decompose with the loss of the NCNR2 ligand in the interval ca. 70-150 °C and two H2O ligands at ca. 150-250 °C; after 250 °C, the residue undergoes decomposition of sacligands forming yet unidentified species. Hydrate 1·2H2O somehow demonstrates greater stability and its decomposition starts at ca. 100 °C with loss of the crystallization water and the NCNMe2 ligand at 100-150 °C; after 250 °C, the decomposition to yet unidentified species occurs. Complex 3·2THF is stable until ca. 120 °C and then starts losing the solvated THF (125-260 °C), whereupon the coordinated THF (260-326 °C) is lost. After 326 °C, non-stochiometric decomposition of the sacligands occurs.
On the other hand, these complexes represent examples of the (N-sac) 2 Cu II -type complexes. Metal complexes of saccharin(ate) exhibit versatile coordination chemistry due to the existence of different coordination sites [34]. For copper(II), several types of complexes are reported, demonstrating monodentate N-, or O-coordination, and N,O-bidentate coordination modes of the sacligand [34]. The (N-sac) 2 (Figure 4).

Theoretical Calculations
In order to obtain additional indirect evidence supporting the blocking role of the copper(II) in the saccharine/cyanamide coupling, we studied the nature of Cu-Ncyanamide and Cu-Nsac coordination bonds in 1 and carried out an integrated computational study including the full geometry optimization of the model of structure 1 in the gas phase using the appropriate experimental XRD structure as a starting point, the topological analysis of the electron density distribution (QTAIM) [44], the natural bond orbital and charge decomposition analyses (NBO [45] and CDA [46]), and calculation of the vertical total energies for the Cu-Ncyanamide and Cu-Nsac coordination bond dissociations (see ESI for all details). This approach has already been successfully used by us upon studies of bonding properties in various similar transition metal complexes [20,[47][48][49][50]. The results of our computational study revealed that (i) crystal-packing strongly affects the structural characteristics of 1 in the solid state; (ii) the dialkylcyanamide copper(II) complexes featuring noticeable contribution of the heterocumulene mesomeric form; (iii) Cu-Ncyanamide and Cu-Nsac coordination bonds in 1 exhibit a single bond character, clearly polarized toward the N atom and almost purely electrostatic; (iv) the {M}←L σ-donation substantially prevails  [41]. The bond length and angles for the saccharinate ligand were similar to those reported for the [Cr 2 L 2 (µ-sac) 4 ] complexes (L = THF, Py) [42,43]. In addition, 3·2THF represents a rare reported example of saccarinate-based lantern-like complexes with a core {M 2 (Sac) 4 }. No significant H-bonding was detected in the structures of the complexes, while weak π-π interactions between aryl···aryl rings were identified.

Theoretical Calculations
In order to obtain additional indirect evidence supporting the blocking role of the copper(II) in the saccharine/cyanamide coupling, we studied the nature of Cu-N cyanamide and Cu-N sac coordination bonds in 1 and carried out an integrated computational study including the full geometry optimization of the model of structure 1 in the gas phase using the appropriate experimental XRD structure as a starting point, the topological analysis of the electron density distribution (QTAIM) [44], the natural bond orbital and charge decomposition analyses (NBO [45] and CDA [46]), and calculation of the vertical total energies for the Cu-N cyanamide and Cu-N sac coordination bond dissociations (see ESI for all details). This approach has already been successfully used by us upon studies of bonding properties in various similar transition metal complexes [20,[47][48][49][50]. The results of our computational study revealed that (i) crystal-packing strongly affects the structural characteristics of 1 in the solid state; (ii) the dialkylcyanamide copper(II) complexes featuring noticeable contribution of the heterocumulene mesomeric form; (iii) Cu-N cyanamide and Cu-N sac coordination bonds in 1 exhibit a single bond character, clearly polarized toward the N atom and almost purely electrostatic; (iv) the {M}←L σ-donation substantially prevails over the {M}→L π-back-donation in both Cu-N cyanamide and Cu-N sac coordination bonds in 1; (v) the calculated vertical total energies (E v ) for the Cu-N cyanamide and Cu-N sac coordination bond dissociation in optimized equilibrium model structure 1 were 43.6 and 156.4 kcal/mol, respectively. Overall, one can conclude that the nature of Cu-N coordination bonds in 1 is similar, but the saccharinate is a stronger ligand toward the copper(II) center than the cyanamide.

Materials and Methods
All solvents and reactants were obtained from commercial sources and used as received. Sodium saccarinate was obtained according to the published method [51]. Atomic absorption spectrometry (AAS) was carried out on a Shimadzu AA-7000 spectrometer (Shimadzu, Japan) (spectral range 189-900 nm) using the flame emission spectroscopy method. Standard Cu samples for the calibration solutions were prepared by MERCK standard (Merck KGaA, Darmstadt, Germany) in 0.1 M HNO 3 ; calibration solutions were 0.01-100.0 mg/L. Spectral analysis of the sample solutions was carried out with 100-fold dilution. Infrared spectra (4000-400 cm -1 ) were recorded using a Bruker FTIR TENSOR 27 (Bruker, Germany) instrument in Nujol. The thermogravimetry/differential thermal analysis was performed with a NETZSCH TG 209 F1 Libra thermoanalyzer (NETZSCH Group, Selb, Germany) and MnO 2 powder was used as the standard. The initial weights of the samples were in the range 1.1-1.8 mg. The experiments were run in an open aluminum crucible in a stream of argon at a heating rate of 10 K/min. The final temperature was 530 • C. Processing of the thermal data was performed with Proteus analysis software [52].

Synthetic Work
Synthesis of [Cu(sac) 2 (NCNR 2 )(H 2 O) 2 ] (1, 1·2H 2 O, 2). CuCl 2 ·2H 2 O or CuBr 2 (0.25 mmol) was dissolved in 5-fold excess of NCNR 2 (R 2 = Me 2 1, Et 2 2; 0.10 mL, 1.3 mmol) at room temperature (RT), whereupon a solution of sodium saccharinate (0.25 mmol) in certain solvents (2 mL, dried EtOH for 1, MeOH, EtOH, or H 2 O for 1·2H 2 O and MeOH or EtOH for 2) was added. The resulting mixture was left to stand for 3-5 days at RT without stirring and the bright greenish-blue prismatic crystals were precipitated. These crystals were filtered off, washed by methanol, and dried in air at RT. One-or two-fold reprecipitation from the mother liquid allowed the increased yield of 1·2H 2 O (75%). The total isolated yields were 50-75%. Alternatively, 1·2H 2 O was obtained from a CuX 2 /saccharin mixture with excess NCNMe 2 and without solvents by keeping for 1 h at 60 • C and then at RT for 2-3 days, however, the reaction proceeds in lower yields (15-35%). A few crystals of pure 1-2 were mechanistically separated from the reaction mixture and yields were not calculated.   [57] program complex using spherical harmonics, implemented in the SCALE3 ABSPACK scaling algorithm. The crystallographic data and structure refinement parameters are given in Table S1

Computational Details
The full geometry optimization of model structure 1 was carried out at the DFT level of theory using the M06 functional [58] with the help of the Gaussian-09 program package [59]. No symmetry restrictions were applied during the geometry optimization procedure and appropriate experimental X-ray structure 1 was used as a starting point. The calculations were carried out using the multi electron fit fully relativistic energy-consistent pseudopotential MDF10 of the Stuttgart/Cologne group that described 10 core electrons and the appropriate contracted basis set for the copper atom [60]) and the 6-31G(d) basis sets for other atoms. The Hessian matrix was calculated analytically for the optimized model structure 1 in order to prove the location of correct minima on the potential energy surface (no imaginary frequencies). The topological analysis of the electron density distribution with the help of the "atoms in molecules" method developed by Bader (QTAIM) [44] and charge decomposition analysis developed by Dapprich and Frenking (CDA) [46] were carried out by using the Multiwfn program (version 3.7) [61]. The Cartesian atomic coordinates for optimized equilibrium model structure 1 are presented in Table S2, ESI, and as the attached xyz-file.

Conclusions
In this study, we demonstrate that in the system Cu II /SacNa(H)/NCNR 2 , copper(II) forms complexes with saccharinate and disubstituted cyanamides. The directionality of the reaction is completely different from that observed for the metal-free reaction between saccharin and NCNR 2 ; the latter results in the formation of guanidinated saccharins [8].
We succeeded in the isolation and characterization of the [Cu(sac) 2 (NCNR 2 )(H 2 O) 2 ] (R = Me, Et) complexes, which represent the first example of structurally characterized mononuclear (cyanamide)Cu II species. The experimental X-ray structure of 1 was used as a starting point for an integrated computational study of model structure 1 in the gas phase. As can be inferred from inspection of the obtained theoretical data, Cu-N cyanamide and Cu-N sac coordination bonds in 1 exhibited a single bond character, clearly polarized toward the N atom and predominately electrostatic, but the saccharinate was a better ligand toward the copper(II) center than the cyanamide (the calculated vertical total energies for the Cu-N cyanamide and Cu-N sac were 43.6 and 156.4 kcal/mol, respectively). All these data confirmed the role of the copper(II) in the change of the directionality of the reaction between saccharin and NCNR 2 . The copper(II) completely blocks the nucleophilic centers of ligands via coordination, thus preventing the saccharin-cyanamide coupling.