Bond Trading: Intramolecular Metal and Ligand Exchange within a NO/Ni/Co Complex

Abstract With the goal of generating hetero‐redox levels on metals as well as on nitric oxide (NO), metallodithiolate (N2S2)CoIII(NO−), N2S2 = N,N‐ dibenzyl‐3,7‐diazanonane‐1,9‐dithiolate, is introduced as ligand to a well‐characterized labile [Ni0(NO)+] synthon. The reaction between [Ni0(NO+)] and [CoIII(NO−)] has led to a remarkable electronic and ligand redistribution to form a heterobimetallic dinitrosyl cobalt [(N2S2)NiII∙Co(NO)2]+ complex with formal two electron oxidation state switches concomitant with the nickel extraction or transfer as NiII into the N2S2 ligand binding site. To date, this is the first reported heterobimetallic cobalt dinitrosyl complex.


Introduction
The extreme reactivity of the nitric oxide (NO) molecule that leads to its usefulness in human physiology requires transfer between various stabilizing entities. [1]Both thiols and metals serve in this capacity. [2]Related to its existence in three redox levels, NO + , NO • and NO − , with  orbitals well matched to frontier orbitals of transition metals, nitric oxide (NO) is a versatile donor ligand in coordination chemistry, well known to form classes of complexes that display its multiple redox states, frequently inextricably entwined with those of the metal to which it binds. [3]Such ambiguity is the genesis of the Enemark-Feltham (E-F) notation in transition metal nitrosyls, {M(NO)} n , that sums the valence d-electrons of M with the frontier -electrons of NO, without commitment to the oxidation state of the metal or the charge on the NO. [4]ynthetic programs that target heterobimetallic nitrosyl complexes bridged by sulfur atoms are of potential importance in creating and understanding redox non-innocence in the M─NO DOI: 10.1002/advs.202307113units. [5]In this regard we have explored metal nitrosyls in combination with exogeneous metals in attempts to unravel redox behaviours in M and in NO.Thus, while developing metal-nitrosyls as synthons in the formation of isolable hetero-bi and trimetallic complexes, we discovered an interesting "bond exchange" reaction involving Co(NO) within a N 2 S 2 ligand field and the molybdenum(II) synthon, [Mo 2 (CH 3 CN) 10 ] 4+ . [6]ursuing the concept of [(N 2 S 2 )Co(NO)] as a NO source in the formation of nitrosylated heterobimetallics, and using the interesting [Ni(NO)(CH 3 NO 2 ) 3 ] + synthon ({Ni 0 (NO + )} 10 ), [7] we set out to explore the possibility of generating hetero-redox levels in {Co(NO)•Ni(NO)} bimetallics containing NO − on Co III and NO + on Ni 0 .These bimetallic systems are of potential significance for understanding the mechanism of coupling between two NO molecules (via NO reduction) that generates the greenhouse gas N 2 O as was seen for the single metal site as well for the [(bipy) 2 Ni(NO)] + , bipy = bipyridine [8] Another interesting aspect of the current work is to study the analogy between bidentate "bipy" and nitrosylated metallodithiolate as ligands for the {Ni(NO)} 10 synthon.In fact, we observed NO transfer and/or rearrangement in the S-bridged Co─Ni bimetallic, concomitant with bond exchange and considerable electron rearrangement governed by thermodynamic preferences.

Results and Discussion
A stoichiometric amount of [Ni(NO)(CH 3 NO 2 ) 3 ][PF 6 ] [7] was added to a CH 3 CN solution of (dadt Bz )Co(NO) (dadt Bz is N,N-dibenzyl-3,7-diazanonane-1,9-dithiolate) producing a color change within time of mixing at ca. 22 °C from dark purple to dark brown, concomitant with the (NO) IR changes that are displayed in Figure 1.The absorption at 1603 cm −1 for (NO) of the precursor (dadt Bz )Co(NO) complex (Figure S2, Supporting Information) in CH 3 CN diminished as new bands at ca. 1652, 1780, and 1842 cm −1 appeared.Over the course of 4 h, the absorption bands at 1780 and 1842 cm −1 continued to increase while the absorption at 1652 cm −1 diminished (Figure 1).Upon addition of diethyl ether (Et 2 O), a brown crystalline product was isolated in ≈60% yield.While mass spectral results gave an elemental composition consistent with the targeted 1:1 adduct of [(dadt Bz )Co(NO)•Ni(NO)] + (Figure S5, Supporting Information), the X-ray analysis of crystals showed a rearranged Co-Ni product corresponding to the exchange of nickel We propose this trimetallic species to be a side product, rather than a reactive intermediate essential to the formation of the  S2 (Supporting Information).a)  5 value for Co in complex A, b)  4 value for Ni 2+ complex B and C, c)  4 value for {Co(NO) 2 } [10] in complexes B and C.  S2 (Supporting Information).Complexes A and B crystallize in the same monoclinic P2 1 /n space group, while complex C crystallized in the monoclinic C2/c space group.The cobalt within the N 2 S 2 binding site of (dadt Bz )Co(NO), adopts an almost ideal square pyramidal geometry with  5 = 0.04.The displacement of the cobalt from the mean N 2 S 2 plane is 0.298 Å.The apical NO molecule is severely bent, with ∠Co1-N1-O1 = 127.88˚.These parameters are consistent with those of several reported [(N 2 S 2 )Co(NO)] complexes with various hydrocarbon connectors between the heteroatoms. [6,10]he nickel center within the N 2 S 2 pocket of [(dadt Bz  direction, likely reflecting a packing difference between complex B and C (vide infra).The dihedral angle between the benzylic rings in complex B is found to be 7.61˚, whereas for complex C it is found to be 62.67˚(Figure S12 − anion (green in Figure 3), is found to be lower in energy compared to complex C (blue in Figure 3) having the BPh 4 − anion by 2.5 kcal mol −1 (negligible).This supports the conclusion that packing differences in the two structures (complex B and C) give rise to differences in geometry and orientation of the benzylic phenyl rings.
The nitrosyl IR stretching frequencies as well as the M─N─O bond angle, and the M─N(O) and N─O bond lengths of reported cationic N donor and P donor cobalt dinitrosyl complexes are tabulated in Table S4 (Supporting Information).All are similar to the [(dadt Bz )Ni•Co(NO) 2 ] + species.Despite the significant differences in geometrical orientation and Co-N-O angles in {Co(NO) 2 } 10 , the two NO units in Co(NO) 2 are strongly coupled (2D IR, see Figure S15 Supporting Information) and best described as a unit, having symmetric and asymmetric stretching vibrations, similar to that of Fe(NO) 2 in the [(N 2 S 2 )Ni•Fe(NO) 2 ] + analogue. [11]he fast, time of mixing, reaction of (dadt Bz )Co(NO) with [(CH 3 NO 2 ) 3 Ni(NO)] + suggest formation of an adduct, presumably the [(dadt Bz )Co(NO)•Ni(NO)] + analogous to Hayton's [(bipy)•Ni(NO)] + derivative, [8a] expressed in Figure 4  + was imported to use as the starting coordinates for energy calculations, performed using TPSSTPSS [12]  functional and triple- basis set 6-311++G (d, p) [13] in Gaussian 16 Revision C.01. [14] All species were confirmed to be minimum energy structures by the absence of imaginary frequencies.The optimized structures of the expected/hypothetical product, and the isolated/rearranged product, in gas phase were then calculated using MeCN in the smd solvent model.The ground state energy difference, between these two structures (Figure 4) is found to be 14.5 kcal mol −1 in favour of the rearranged product.
To account for the reactivity observations, we pursue the hypothesis that the initial mixing of substrates generates a Co-Ni species, presumably {Co(NO)} 8 {Ni(NO)} 10 (Scheme 1).A rapid electron transfer is expected to occur from {Ni(NO)} 10 to {Co(NO)} 8 to generate a {Co(NO)} 9 {Ni(NO)} 9 species followed by NO release.The release of NO gas was detected by headspace analysis of the reaction mixture by gas chromatography at retention time of 2.56 min (Figure S25, Supporting Information).A similar observation was made by the Hayton group for the {Ni(NO)} 10 unit in the [(bipy)Ni(NO)] + or {Ni(NO)} 10 complex which was found to release NO upon oxidation to {Ni(NO)} 9 . [15]he control experiment found that separately the {Co(NO)} 8 and {Ni(NO)} 10 synthons did not show any NO release in MeCN, under similar reaction times (Figure S26, Supporting Information).The release of NO upon stirring and its attack on the {Co(NO)} 9 would account for generation of {Co(NO) 2 } 10 as it comes out of the N 2 S 2 pocket.The resultant empty binding site is rapidly replaced by Ni(II) to form the stable square planar NiN 2 S 2 complex.
The role of released NO during the reaction is supported by the significant increase in the isolated yield of the final product (from 15% to 60%), when the reaction is carried out in a septumsealed vial in order to inhibit NO loss.However, there is an optimal NO concentration as performing the reaction under excess pressure of exogeneous or added NO gas considerably decreased the rate of the reaction as well as the amount of product formation (see Figure S22, Supporting Information).Based on these observations, NO release during the reaction is likely involved in the rate determining step for the rearrangement that forms the final product.The mechanism expressed in Scheme 1 is consistent with these results, including the increased yield in the closed system.However, coupling between the NO molecules has not been observed or indicated during the course of the reaction.Scheme 2. Analogy between bidentate bipyridine (reported by Hayton's group [7,8,15] ) and nitrosylated metallodithiolate as ligands for the {Ni(NO)} 10 synthon.
The oxidation state ambiguity of the {Ni(NO)} 10 -unit (Enemark-Feltham notation) as expressed within the Hayton's bipyridine complexes, [7,8,15] indicates a two-electron redistribution of oxidation state when a second bipyridine is added to the tri-coordinated, [(bipy)Ni(NO)] + , as shown below in Scheme 2. In our case, the four donors within the N 2 S 2 tetradentate ligand are well known to stabilize Ni II .The thermodynamic downhill rearrangement reaction promotes, in a bimetallic fashion, the two-electron oxidation state change for nickel.The transfer of •NO and an electron (overall NO − ) to {Co(NO)} 8 , generating the {Co(NO) 2 } 10 , maintains the overall electron balance, but requires an impressive amount of oxidation state redistribution.That is, the chemical non-innocence of the N 2 S 2 Co(NO) metallodithiolate ligand supports the rearrangement of Ni 0 /Ni 2+ by an intramolecular bimetallic process.Further computational studies using the CASSCF valance bond analysis method is required to estimate the amount of NO +/•/− character in each complex of Scheme 2. [16]

Conclusion
In conclusion, the (dadt Bz )Co(NO) metalloligand reacts with a [Ni(NO)] + synthon to form an unstable intermediate within the time of mixing but undergoes rearrangement slowly (ca. 4 h) at room temperature to yield a thermodynamically stable heterobimetallic Ni(N 2 S 2 ) cobalt dinitrosyl species.8a] While there are several cobalt dinitrosyl complexes reported in the literature, [9,17] to the best of our knowledge this is the first reported heterobimetallic cobalt dinitrosyl complex (DNCC).The identity of the original adduct can only be surmised; however, a Co/Ni interchange is most likely resulting from intramolecular process evolving from the first formed adduct.The rearranged product, [Ni•Co(NO) 2 ] + requires a formal overall transfer of an NO − species, or an NOcoupled electron transfer (NOCET).The polarizability of the electron density between metal and NO ligands strengthens the suggested role of metal carriers in NO capture and its transfer, i.e., regulation which is ultimately critical in biological NO chemistry.
, Supporting Information).The crystal packing diagrams of complexes B and C are shown in Figures S13 and S14 (Supporting Information).Overlay of the [(dadt Bz )Ni•Co(NO) 2 ] + structures from crystallization with two different anions PF 6 − and BPh 4 − is shown in Figure 3. From DFT optimized structures of the two isomers using MeCN in the smd model, complex B, that with the PF 6 as a DFT optimized structure.Via an unknown pathway of rearrangement, the adduct then forms the isolated product, in which Ni and Co have exchanged positions, along with NO transfer from Ni to Co, yielding the [(dadt Bz )Ni•Co(NO) 2 ] + thermodynamic product.The redox changes would involve an overall two electron change, i.e., Ni 0 to Ni II , and NO + to NO − .A computational study addressed the driving force for this rearrangement of [(dadt Bz )Co(NO)•Ni(NO)] + to [(dadt Bz )Ni•Co(NO) 2 ] + .The crystal structure of [(dadt Bz )Ni•Co(NO) 2 ]

Figure 4 .
Figure 4. Energy difference between the expected/hypothetical and rearranged product as obtained from DFT calculations.
Scheme 1. Probable steps for the formation of [(dadt Bz )Ni•Co(NO) 2 ][PF 6 ] species.Including the presumed initial adduct, the species shown in grey background are hypothetical intermediates during the transformation.