<<

International Journal of Molecular Sciences

Article Intramolecular Spodium Bonds in Zn(II) Complexes: Insights from Theory and Experiment

Mainak Karmakar 1, Antonio Frontera 2 , Shouvik Chattopadhyay 1,* , Tiddo J. Mooibroek 3,* and Antonio Bauzá 2,*

1 Department of Chemistry, Inorganic Section, Jadavpur University, Kolkata 700 032, India; [email protected] 2 Departament de Química, Universitat de les Illes Balears, Crta. de Valldemossa km 7.5, 07122 Palma (Baleares), Spain; [email protected] 3 van ‘t Hoff Institute for Molecular Sciences, Universiteit van Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands * Correspondence: [email protected] (S.C.); [email protected] (T.J.M.); [email protected] (A.B.)  Received: 30 August 2020; Accepted: 22 September 2020; Published: 25 September 2020 

Abstract: Two new dinuclear (II) complexes, [Zn (µ -OAc)(L1) ]I MeOH (1) and [Zn (µ - 2 1,3 2 · 2 1,3 OAc)(L2)(NCS)] (2), (where HL1 = 2-(((3-(dimethylamino)propyl)amino)methyl)-6-methoxy-phenol 2 and H2L = 2,20-[(1-Methyl-1,2-ethanediyl)bis(iminomethylene)]bis[6-ethoxyphenol]) have been synthesized and characterized by elemental and spectral analysis. Their X-ray solid state structures have been determined, revealing the existence of intramolecular Zn O spodium bonds in both ··· complexes due to the presence of methoxy (1) or ethoxy (2) substituents adjacent to the coordinated phenolic O-. These noncovalent interactions have been studied using density functional theory (DFT) calculations, the quantum theory of “-in-molecules” and the noncovalent interaction plot. Moreover, a search in the Cambridge structure database (CSD) has been conducted in order to investigate the prevalence of intramolecular spodium bonds in Zn complexes. To our knowledge this is the first investigation dealing with intramolecular spodium bonds.

Keywords: zinc complexes; spodium bonds; σ-hole interactions; CSD analysis; DFT calculations

1. Introduction Noncovalent interactions are very important in many fields of research, including molecular recognition, crystal engineering and catalysis [1–3]. Among the great deal of noncovalent forces, investigations on σ-hole interactions [4,5] are growing very fast and are definitively recognized by the scientific community as an alternative to the ubiquitous bonding [6]. Very recently, the attractive interaction between elements of Group 12 of the and any electron rich ‘accepting’ atom (:A) [7,8] has been termed as a spodium bond (SpB) [9]. Therefore, the SpB has become a new member of the σ-hole family of interactions and it is adequate to differentiate the coordination bond (high covalent character) typical of transition metals from the noncovalent contact. SpBs are directional, the electron rich atom is located at distances that are longer than the sum of covalent radii and they are considerably weaker than coordination bonds. Moreover, the participation of the antibonding σ*(Sp–Y, where Y can be any atom) in the SpB bonding interaction (Y–Sp :A) has been ··· evidenced [9], as is common in σ-hole interactions [4,5]. This manuscript reports the synthesis and X-ray characterization of two new Zn(II) complexes that exhibit intramolecular SpBs. The utilization of two different tridentate and tetradendate ligands allows analyzing the interaction in two different Zn-coordination modes, square-pyramidal in (1)

Int. J. Mol. Sci. 2020, 21, 7091; doi:10.3390/ijms21197091 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2020, 21, 7091 2 of 14 Int. J. Mol. Sci. 2020, 21, 7091 2 of 14 andpseudotetrahedral pseudotetrahedral in (2). in The (2). SpBs The in SpBs both incompounds both compounds have been have studied been using studied density using functional density functionaltheory (DFT) theory calculations (DFT) calculations and characterized and characterized by a combination by a combination of the quantum of the quantumtheory of theoryatoms ofin atomsmolecules in molecules (QTAIM) (QTAIM) [10] and [10 the] and noncovalent the noncovalent interaction interaction (NCI) (NCI) method method [11]. [11Moreover,]. Moreover, the theCambridge Cambridge Structural Structural Database Database (CSD) (CSD) [12] [12 has] has been been examined examined in in order order to to investigate investigate the prevalence ofof intramolecularintramolecular SpBsSpBs inin Zn(II)Zn(II) complexescomplexes andand theirtheir geometric geometric features. features.

2.2. Results and Discussion

2.1.2.1. Synthesis InIn this this work, work, two two acetate acetate bridged bridged dinuclear dinuclear zinc complexes zinc complexes have been have synthesized been synthesized using reduced using 1 2 1 reducedSchiff bases Schi ff HLbases1 and HL Hand2L2 Has2 L ligands.as ligands. HL1 HLwas wasprepared prepared by by1:1 1:1 condensation condensation of of 3-3‐ methoxysalicylaldehydemethoxysalicylaldehyde and andN, NN-dimethyl-1,3-diaminopropane,N‐dimethyl‐1,3‐diaminopropane in methanol in methanol and subsequent and subsequent reduction 2 usingreduction NaBH using4 following NaBH4 following a literature a literature method [ 13method–15]. In[13–15]. the same In the way, same H2 way,L was H2L prepared2 was prepared by 2:1 condensationby 2:1 condensation of 3-ethoxysalicylaldehyde of 3‐ethoxysalicylaldehyde and rac and-1,2-diaminopropane rac‐1,2‐diaminopropane in methanol in methanol followed followed by the 1 reductionby the reduction with NaBH with4. NaBH Methanol4. Methanol solution solution of HL on of reaction HL1 on with reaction zinc acetatewith zinc dihydrate acetate anddihydrate and 2 iodidepotassium in a iodide 2:2:1 molar in a 2:2:1 ratio molar produced ratio complexproduced1 .complex On the other1. On hand,the other methanol hand, methanol solution of solution H2L on of reactionH2L2 on withreaction zinc with acetate zinc dihydrate acetate dihydrate and and thiocyanate sodium thiocyanate in a 1:2:1 molar in a 1:2:1 ratio molar produced ratio complex produced2. Thecomplex synthetic 2. The routes synthetic to the routes complexes to the1 complexesand 2 are shown 1 and 2 in are Figure shown1. in Figure 1.

FigureFigure 1.1. Synthetic routesroutes toto complexescomplexes 11 andand 22..

Int. J. Mol. Sci. 2020, 21, 7091 3 of 14

2.2. Description of Compounds 1 and 2 A perspective view of complexes 1 and 2 along with selective atom numbering scheme is depicted in Figure 2a,b respectively. 1 crystallizes in the orthorhombic space group P212121 and its asymmetric unit consists of an acetato bridged dinuclear zinc(II) L12 complex, one I– counter anion and a methanol solvent molecule. Each zinc is coordinated to the tridentate L1 via two amine N‐ atoms, one bridging acetate ligand and two bridging phenolate atoms (see Supplementary Figure S1). Both penta‐coordinated zinc(II) centers are in a distorted square pyramidal geometry (see SupplementaryInt. J. Mol. Sci. 2020 ,Figures21, 7091 S2 and S3). The asymmetric unit contains one lattice methanol molecule3 that of 14 is H‐bonded to the I− counter anion. The counter ion is also H‐bonded to N4–H4 of the reduced Schiff base ligand. 2.2. Description of Compounds 1 and 2 Compound 2 (Figure 2b and Supplementary Figure S4) crystallizes in the triclinic space group Pī andA perspectivedoes not contain view of additional complexes 1ionsand or2 alongsolvent with molecules selective atomin its numbering asymmetric scheme unit. isZn2 depicted is N‐ coordinatedin Figure2a,b to respectively. the thiocyanate,1 crystallizes connected in theto O6 orthorhombic of the bridging space acetate group andP21 2bound121 and to its two asymmetric bridging 1 – phenolateunit consists O‐ ofatoms an acetato to form bridged a distorted dinuclear tetrahedron zinc(II) (see L 2 complex, Supplementary one I counter Figure S5). anion Zn1 and is asituated methanol in asolvent distorted molecule. square Each pyramidal zinc is environment coordinated to where the tridentate the ligand L1 [O1,N1,N2,O3]via two amine N-atoms,donor atoms one form bridging the baseacetate of ligandthe pyramid and two and bridging O5 of the phenolate bridging oxygen acetate atomsis at the (see apex. Supplementary The coordination Figure distances S1). Both of penta-coordinatedcompounds 1 and 2 zinc(II) are given centers in Table are in1 and a distorted range from square 1.92 pyramidal to 2.21 Å. geometryThese distances (see Supplementary are similar or slightlyFigures S2longer and S3).that The the asymmetric sum of covalent unit contains radii ( oneΣRcov lattice = 1.93 methanol and 1.86 molecule Å for thatZn+N is H-bonded and Zn+O, to respectively)the I− counter and anion. can Thebe considered counter ion as is normal also H-bonded [16]. to N4–H4 of the reduced Schiff base ligand.

Figure 2. Perspective ball ball and stick view of the complexes 1 (a) and 2 (b) along with selective atom numbering scheme (only the coordinating atoms are labeled forfor clarity).clarity).

TableCompound 1 also 2gathers(Figure longer2b and Zn Supplementary∙∙∙O distances Figure (values S4) in crystallizesitalics) where in thethe triclinic O‐atom space belongs group to the P ¯ı methoxyand does not(in L contain1) or ethoxy additional (in L2 ions) substituent or solvent of molecules the aromatic in its ring. asymmetric These distances unit. Zn2 are is N-coordinated around 0.8 Å longerto the thiocyanate,than ΣRcov and, connected consequently, to O6 can of the be bridgingconsidered acetate as intramolecular and bound tospodium two bridging bonds, phenolateas further analyzedO-atoms to below. form These a distorted interactions tetrahedron are depicted (see Supplementary in Figure 3 as Figure black S5).dashed Zn1 lines. is situated In compound in a distorted 1 the SpBssquare are pyramidal located opposite environment to the Zn–O(acetate) where the ligand coordination [O1,N1,N2,O3] bonds donorat an angle atoms of form~170°. the In basecompound of the pyramid2, the ethoxyde and O5 O of‐atoms the bridging are located acetate opposite is at the the phenoxy apex. The O’s coordination at an angle of distances ~145°. These of compounds differences1 ˚ inand O–Zn2 are∙∙∙ givenO angles in Table likely1 and originate range fromfrom 1.92 the todifferent 2.21 A. geometries These distances involved; are similar the square or slightly pyramidal longer ˚ geometrythat the sum in 1 of leaves covalent ample radii space (ΣRcov for= an1.93 additional and 1.86 Adonor for Zn atom+N and(SpB) Zn while+O, respectively) the accessible and space can beat theconsidered tetrahedral as normal Zn in 2 [ 16is ].more restricted. Table1 also gathers longer Zn O distances (values in italics) where the O-atom belongs to the ··· methoxy (in L1) or ethoxy (in L2) substituent of the aromatic ring. These distances are around 0.8 A˚ longer than ΣRcov and, consequently, can be considered as intramolecular spodium bonds, as further analyzed below. These interactions are depicted in Figure3 as black dashed lines. In compound 1 the SpBs are located opposite to the Zn–O(acetate) coordination bonds at an angle of ~170◦. In compound 2, the ethoxyde O-atoms are located opposite the phenoxy O’s at an angle of ~145◦. These differences in O–Zn O angles likely originate from the different geometries involved; the square pyramidal ··· geometry in 1 leaves ample space for an additional donor atom (SpB) while the accessible space at the tetrahedral Zn in 2 is more restricted. The supporting information (Supplementary Figures S6–S10) contains a more detailed structural description complexes 1 and 2 (Section1), the Hirshfeld surface analysis (Section2) and additional spectral characterizations (Section3; IR, UV-vis and XRPD spectra). Int. J. Mol. Sci. 2020, 21, 7091 4 of 14

Int. J. Mol. Sci. 2020, 21, 7091 4 of 14

Table 1. Selected bond lengths (A)˚ in compounds 1 and 2. Table 1. Selected bond lengths (Å) in compounds 1 and 2.

11 Zn1–N1 (coord.)Zn1–N1 1 (coord.) 2.169(5)1 2.169(5) Zn2–N3 Zn2–N3 (coord.) 2.160(5) 2.160(5) Zn1–N2 (coord.)Zn1–N2 (coord.)2.074(4) 2.074(4) Zn2–N4 Zn2–N4 (coord.) 2.085(4) 2.085(4) 2 Zn1–O1 (coord.)Zn1–O1 2 (coord.) 2.212(4)2.212(4) Zn2–O1 Zn2–O1 (coord.) 1.983(4) 1.983(4) Zn1–O3 (coord.) 1.977(3) Zn2–O3 (coord.) 2.165(3) Zn1–O3 (coord.) 1.977(3) Zn2–O3 (coord.) 2.165(3) Zn1–O6 (coord.) 2.051(4) Zn2–O5 (coord.) 2.075(4) Zn1–O6 (coord.)Zn1–O4 (SpB)2.051(4) 2.688(5) Zn2–O5 Zn2–O2 (coord.) (SpB) 2.667(4) 2.075(4) Zn1–O4 (SpB)Zn1 Zn22.688(5) 3.047(1) Zn2–O2 (SpB) 2.667(4) ··· Zn1∙∙∙Zn2 3.047(1) 2 2 Zn1–N1 (coord.) 2.091(2) Zn2–N3 (coord.)a 1.922(2) Zn1–N1 (coord.) 2.091(2) Zn2–N3 (coord.)a 1.922(2) Zn1–N2 (coord.) 2.099(2) Zn2–O1 (coord.) 2.017(1) Zn1–N2 (coord.)Zn1–O1 (coord.)2.099(2) 2.058(2) Zn2–O1(coord.) Zn2–O3 (coord.) 2.007(2) 2.017(1) Zn1–O1 (coord.)Zn1–O3 (coord.)2.058(2) 2.042(2) Zn2–O3 Zn2–O6 (coord.) 1.976(2) 2.007(2) Zn1–O3 (coord.)Zn1–O5 (coord.)2.042(2) 1.978(2) Zn2–O6Zn2–O2 (coord.) (SpB) 2.692(2) 1.976(2) Zn1 Zn2 2.9025(5) Zn2–O4 (SpB) 2.664(2) Zn1–O5 (coord.) ··· 1.978(2) Zn2–O2 (SpB) 2.692(2) Zn11 Sum∙∙∙Zn2 of Zn and N covalent2.9025(5) radius: 1.93 A;˚ 2 SumZn2–O4 of Zn and (SpB) O covalent radius: 1.862.664(2)A.˚ 1 Sum of Zn and N covalent radius: 1.93 Å; 2 Sum of Zn and O covalent radius: 1.86 Å.

FigureFigure 3. Spodium3. Spodium bonds bonds (black (black dashed dashed lines) lines) and and interacting interacting anglesangles in compounds 11 ((aa)) and and 22 (b(b).). H-atomsH‐atoms omitted omitted for for clarity. clarity.

2.3. TheoreticalThe supporting DFT Study information (Supplementary Figures S6–S10) contains a more detailed structural description complexes 1 and 2 (Section 1), the Hirshfeld surface analysis (Section 2) and additional Theoretical models of 1 and 2 (10 and 20, see Figure4, top) have been used to investigate the existencespectral of characterizationsσ-holes at the Zn (Section atoms. 3; IR, The UV utilization‐vis and XRPD of such spectra). models is needed because in the real systems the potential σ-holes are “hidden” as a consequence of their intramolecular interaction with 2.3. Theoretical DFT Study the electron rich O-atoms. In these models the aromatic rings were eliminated to leave hydroxides as bridgingTheoretical ligands instead models of of a phenoxides.1 and 2 (1′and The 2′ geometry, see Figure of 4, the top) models have was been optimized used to investigate while keeping the theexistence zinc ions andof σ‐ theirholes surrounding at the Zn atoms. donor The atoms utilization frozen. Forof such comparison models purposes,is needed thebecause acetate in bridgingthe real systems the potential σ‐holes are “hidden” as a consequence of their intramolecular interaction with ligand in 1 has been replaced by a dianionic carbonate ligand in 10 (i.e., so that both 10 and 20 are charge neutral).the electron rich O‐atoms. In these models the aromatic rings were eliminated to leave hydroxides as bridgingThe MEP ligands surface instead of 1 ofshows a phenoxides. that, as The expected, geometry the of most the models positive was MEP optimized is located while at keeping the NH the zinc ions and their surrounding donor atoms frozen. For comparison purposes, the acetate group. Additionally, two regions of positive potential are present at the extension of the O–Zn bonds bridging ligand in 1 has been replaced by a dianionic carbonate ligand in 1′ (i.e., so that both 1′ and (+23 kcal/mol), which is consistent with σ-holes adequate for interacting with electron rich atoms. The 2′ are charge neutral). MEP value is most negative at the bridging O-atoms ( 55 kcal/mol). The MEP surface of 1 shows that, as expected, the− most positive MEP is located at the NH group. The MEP surface of 2 on a default scale (bottom left in Figure4b) is also most positive at the NH Additionally, two regions0 of positive potential are present at the extension of the O–Zn bonds (+23 groups. Plotting the MEP on a different scale clearly reveals two patches of electropositive potential kcal/mol), which is consistent with σ‐holes adequate for interacting with electron rich atoms. The (+10 kcal/mol), congruent with the anticipated σ-holes. MEP value is most negative at the bridging O‐atoms (−55 kcal/mol).

Int. J. Mol. Sci. 2020, 21, 7091 5 of 14

The MEP surface of 2′ on a default scale (bottom left in Figure 4b) is also most positive at the NH groups. Plotting the MEP on a different scale clearly reveals two patches of electropositive potential Int. J. Mol. Sci. 2020, 21, 7091 5 of 14 (+10 kcal/mol), congruent with the anticipated σ‐holes.

Figure 4. (a) Top: Model of compound 1 (named 1′) used to investigate the existence of σ‐holes. Figure 4. (a) Top: Model of compound 1 (named 10) used to investigate the existence of σ-holes. Bottom: Bottom: MEP surface of 1′; (b) Top: Model of compound 2 (named 2′) used to investigate the existence MEP surface of 10;(b) Top: Model of compound 2 (named 20) used to investigate the existence of σ-holes.of σ‐holes. Bottom: Bottom: MEP MEP surfaces surfaces of of2 2′(using (using two two didifferentfferent MEP energetic energetic scales). scales). All All isosurfaces isosurfaces have have Int. J. Mol. Sci. 2020, 21, 7091 0 6 of 14 beenbeen computed computed at at the the PBE0-D3 PBE0‐D3/def2/def2-TZVP‐TZVP (isosurface (isosurface 0.002 0.002 a.u.). a.u.). The The energies energies at at selected selected points points of of the the isosurfaces are given in kcal/mol. Forisosurfaces example, the are givendensities in kcal of /themol. bond CPs of about 0.02 is in between the 0.01 observed in a tetrel‐ bonding adduct and the 0.03 in a hydrogen‐bonding adduct [17]. InIn order order to characterizeto characterize the SpBsthe SpBs in compounds in compounds1 and 21, weand have 2, we performed have performed combined QTAIMcombined/NCI QTAIM/NCIThe same method behavior analyses is observed of both for complexes, the energy which densities are shown that are in smaller Figure 5.for Some the SpBs surfaces, compared bond methodto coordination analyses bonds. of both Another complexes, interesting which areresult shown is that in the Figure total5 .energy Some surfaces,density H(r), bond is negligible critical points in (CPs)critical and points bond paths(CPs) inand compound bond paths1 corresponding in compound to 1 intramolecular corresponding C–H to intramolecularπ interactions C–H have∙∙∙π been interactionsthe SpBs and have negative been omitted in the coordinationfor clarity. Similarly, bonds, whichfor compound is an indication 2 the critical··· of dominant points and omitted for clarity. Similarly, for compound 2 the critical points and bond paths of some intramolecular pathscharacter of somein the coordinationintramolecular bonds. H‐bonds Finally, involving the delocalization the H‐atoms index of (DI) the values, ethoxy which groups are a andmeasure the H-bonds involving the H-atoms of the ethoxy groups and the thiocyanate ligand have been omitted for thiocyanateof bond order, ligand are alsohave clearly been omitteddifferent for in coordinationclarity. The complete (ranging QTAIM/NCIfrom 0.327 to method0.517) and analyses spodium of clarity. The complete QTAIM/NCI method analyses of compounds 1 and 2 are shown in Supplementary compoundsbonds (ranging 1 and from 2 are 0.059 shown to 0.075).in Supplementary Therefore, theFigure QTAIM S11. parameters is a convenient method to Figure S11. differentiateFigure 5a both shows types that of each bonding. SpB in 1 is characterized by a bond CP and bond path connecting the O‐atom to the Zn atom. Moreover, they are also characterized by blue isosurfaces in the NCI plot, thus evidencing an attractive interaction. The NCI isosurface extends toward the region in between the phenolic and methoxy O‐atoms where the color changes to yellow, thus evidencing some repulsion between these atoms. It is also worth mentioning that similar isosurfaces are not present in the coordination bonds, due to their covalent character (covalent bonding is not revealed by the NCI method using the 0.004 cut‐off for the electron density). In compound 2, each SpB is characterized by bond CPs and bond path interconnecting the O‐atom to the Zn‐atom. The NCI analysis also shows two blue isosurfaces between the Zn and the O‐atoms, evidencing attractive interactions. Similar to compound 1, the isosurface extends toward the region between the O‐atoms where the color is yellow, evidencing a repulsive O∙∙∙O interaction. For both compounds, the bond CPs that characterize the SpBs have been labelled as “a” and “b” and two additional bond CPs that characterize Zn–O and Zn–N coordination bonds have been labelled as “c” and “d”. The QTAIM parameters measured at bond CPs a‐d are summarized in Table 2. The values of ρ(r) and the Laplacian of ρ(r) at the bond CPs that characterize the SpBs are significantly smaller than those at the coordination bonds. In fact, the values at the bond CPs corresponding to SpBs are in the range of typical noncovalent interactions.

FigureFigure 5. 5.Quantum Quantum theory theory of of atomsatoms in molecules (QTAIM) (QTAIM) and and noncovalent noncovalent interaction interaction (NCI) (NCI) method method analysesanalyses combined combined in in the the same same representation representation forfor compounds 1 ((aa)) and and 22 (b(b).). Noncovalent Noncovalent bond bond paths paths areare represented represented as as dashed dashed lines. lines.

Table 2. Electron charge density (ρ), its Laplacian (∇2ρ), kinetic (V), Lagrangian (G) and total (H) energy densities at the bond critical points (CPs) labelled in Figure 5 for compounds 1 and 2 in a.u.

CP ρ(r) ∇2ρ(r) V(r) G(r) H(r) DI 1 a 0.017 0.055 −0.014 0.014 0.000 0.072 b 0.018 0.057 −0.015 0.015 0.000 0.075 c 0.081 0.428 −0.136 0.121 −0.015 0.370 d 0.076 0.313 −0.107 0.093 −0.014 0.394 2 a 0.016 0.055 −0.014 0.014 0.000 0.059 b 0.017 0.059 −0.015 0.015 0.000 0.064 c 0.073 0.388 −0.117 0.107 −0.010 0.327 d 0.099 0.475 −0.171 0.145 −0.026 0.510

Two theoretical models have been used to evaluate energetically the SpBs in compound 2. First, a slightly modified model of 2 (denoted as 2a, see Figure 6) has been constructed where the ethoxy substituent has been changed by a methoxy group in order to eliminate the H‐bonds between the ethoxy and the thiocyanate group that are present in 2 (see Supplementary Figure S11). Secondly, a hypothetical complex was calculated where the methoxy group is located para instead of ortho with respect to the phenolate O‐atom (denoted as 2b). In this complex the spodium bonds cannot be formed, while the basicity of the phenolate atoms is similar to that of 2a. To estimate the intramolecular SpB energy, the formation energies of complexes 2a and 2b from the reaction of the metalloligands La and Lb (see Figure 1) with Zn(CH3COO)(NCS) have been computed. Secondly, the

Int. J. Mol. Sci. 2020, 21, 7091 6 of 14

Figure5a shows that each SpB in 1 is characterized by a bond CP and bond path connecting the O-atom to the Zn atom. Moreover, they are also characterized by blue isosurfaces in the NCI plot, thus evidencing an attractive interaction. The NCI isosurface extends toward the region in between the phenolic and methoxy O-atoms where the color changes to yellow, thus evidencing some repulsion between these atoms. It is also worth mentioning that similar isosurfaces are not present in the coordination bonds, due to their covalent character (covalent bonding is not revealed by the NCI method using the 0.004 cut-off for the electron density). In compound 2, each SpB is characterized by bond CPs and bond path interconnecting the O-atom to the Zn-atom. The NCI analysis also shows two blue isosurfaces between the Zn and the O-atoms, evidencing attractive interactions. Similar to compound 1, the isosurface extends toward the region between the O-atoms where the color is yellow, evidencing a repulsive O O interaction. For both compounds, the bond CPs that characterize the SpBs ··· have been labelled as “a” and “b” and two additional bond CPs that characterize Zn–O and Zn–N coordination bonds have been labelled as “c” and “d”. The QTAIM parameters measured at bond CPs a-d are summarized in Table2. The values of ρ(r) and the Laplacian of ρ(r) at the bond CPs that characterize the SpBs are significantly smaller than those at the coordination bonds. In fact, the values at the bond CPs corresponding to SpBs are in the range of typical noncovalent interactions. For example, the densities of the bond CPs of about 0.02 is in between the 0.01 observed in a tetrel-bonding adduct and the 0.03 in a hydrogen-bonding adduct [17].

Table 2. Electron charge density (ρ), its Laplacian ( 2ρ), kinetic (V), Lagrangian (G) and total (H) energy ∇ densities at the bond critical points (CPs) labelled in Figure5 for compounds 1 and 2 in a.u.

CP ρ(r) 2ρ(r) V(r) G(r) H(r) DI ∇ 1 a 0.017 0.055 0.014 0.014 0.000 0.072 − b 0.018 0.057 0.015 0.015 0.000 0.075 − c 0.081 0.428 0.136 0.121 0.015 0.370 − − d 0.076 0.313 0.107 0.093 0.014 0.394 − − 2 a 0.016 0.055 0.014 0.014 0.000 0.059 − b 0.017 0.059 0.015 0.015 0.000 0.064 − c 0.073 0.388 0.117 0.107 0.010 0.327 − − d 0.099 0.475 0.171 0.145 0.026 0.510 − −

The same behavior is observed for the energy densities that are smaller for the SpBs compared to coordination bonds. Another interesting result is that the total energy density H(r), is negligible in the SpBs and negative in the coordination bonds, which is an indication of dominant covalent character in the coordination bonds. Finally, the delocalization index (DI) values, which are a measure of bond order, are also clearly different in coordination (ranging from 0.327 to 0.517) and spodium bonds (ranging from 0.059 to 0.075). Therefore, the QTAIM parameters is a convenient method to differentiate both types of bonding. Two theoretical models have been used to evaluate energetically the SpBs in compound 2. First, a slightly modified model of 2 (denoted as 2a, see Figure6) has been constructed where the ethoxy substituent has been changed by a methoxy group in order to eliminate the H-bonds between the ethoxy and the thiocyanate group that are present in 2 (see Supplementary Figure S11). Secondly, a hypothetical complex was calculated where the methoxy group is located para instead of ortho with respect to the phenolate O-atom (denoted as 2b). In this complex the spodium bonds cannot be formed, while the basicity of the phenolate atoms is similar to that of 2a. To estimate the intramolecular SpB energy, the formation energies of complexes 2a and 2b from the reaction of the metalloligands La and Lb (see Figure1) with Zn(CH COO)(NCS) have been computed. Secondly, the difference ∆E ∆E 3 1 − 2 can be attributed to the contribution of the SpBs. This difference is 8.8, 9.2 and 11.2 kcal/mol using, − − − Int.Int. J. Mol.J. Mol. Sci. Sci.2020 2020, 21, 21, 7091, 7091 77 of of 14 14 Int. J. Mol. Sci. 2020, 21, 7091 7 of 14

difference ΔE1 − ΔE2 can be attributed to the contribution of the SpBs. This difference is −8.8, −9.2 and respectively,difference−11.2 kcal/mol ΔE the1 − Δ using, PBE0-D3,E2 can respectively, be B3LYP-D3 attributed the andto PBE0the MP2 contribution‐D3, methods. B3LYP ‐ofD3 Therefore, the and SpBs. MP2 This each methods. difference SpB stabilizes Therefore, is −8.8, the −each9.2 complex and SpB with−11.2 about kcal/mol5 kcal using,/mol, respectively, which is in thethe rangePBE0‐ ofD3, binding B3LYP energies‐D3 and MP2 recently methods. estimated Therefore, for intermolecular each SpB stabilizes −the complex with about −5 kcal/mol, which is in the range of binding energies recently SpBsstabilizesestimated [7]. the for complex intermolecular with about SpBs − [7].5 kcal/mol, which is in the range of binding energies recently estimated for intermolecular SpBs [7].

FigureFigure 6. 6.Reactions Reactions used used to to evaluate evaluate thethe SpBSpB energy in compound 22.. Figure 6. Reactions used to evaluate the SpB energy in compound 2. Finally,Finally, two two theoretical theoretical models models have have been been also also usedused toto evaluateevaluate the SpBs in in compound compound 11. .First, First, Finally, two theoretical models have been also used to evaluate the SpBs in compound 1. First, wewe have have evaluated evaluated thethe formationformation of of 11 fromfrom two two molecules molecules of ofthe the metalloligand metalloligand denoted denoted as LC as (see LC we have evaluated the formation of 1 from two molecules of the metalloligand denoted as LC (see (seeFigure Figure 7).7 Secondly,) . Secondly, a hypothetical a hypothetical complex complex was was calculated calculated where where the methoxy the methoxy group group is located is located para Figure 7). Secondly, a hypothetical complex was calculated where the methoxy group is located para parainsteadinstead of oforthoortho withwith respect respect to to the the phenolate phenolate O O-atom‐atom (denoted (denoted as as 1b1b).). In In this this complex complex the the spodium spodium instead of ortho with respect to the phenolate O‐atom (denoted as 1b). In this complex the spodium bondsbonds cannot cannot be be formed, formed, while while the the basicity basicity ofof thethe phenolatephenolate atomatom isis preserved.preserved. To To estimate estimate the the bonds cannot be formed, while the basicity of the phenolate atom is preserved. To estimate the intramolecularintramolecular SpB SpB energy, energy, the the formation formation energies energies of of complexes complexes1 1+ +AcO AcO− and 1b + AcO− fromfrom the the intramolecular SpB energy, the formation energies of complexes 1 + AcO−− and 1b + AcO− −from the reactionreaction of of two two molecules molecules of of metalloligands metalloligands Lc Landc and L dL(seed (see Figure Figure7) 7) have have been been estimated, estimated, which which are are in reaction of two molecules of metalloligands Lc and Ld (see Figure 7) have been estimated, which are in this case endothermic. Secondly, the difference ΔE1 − ΔE2 can be attributed to the contribution of this case endothermic. Secondly, the difference ∆E1 ∆E2 can be attributed to the contribution of the in this case endothermic. Secondly, the difference Δ−E1 − ΔE2 can be attributed to the contribution of SpBs,the SpBs, which which is 11.1 is −11.1 kcal /kcal/mol.mol. Therefore, Therefore, each each SpB SpB stabilizes stabilizes the the complex complex in in5.55 −5.55 kcal kcal/mol,/mol, which which is the SpBs, which− is −11.1 kcal/mol. Therefore, each SpB stabilizes the complex in− −5.55 kcal/mol, which is similar to the SpB bonding energy2 in 2. similaris similar to the to the SpB SpB bonding bonding energy energy in in. 2.

Figure 7. Reactions used to evaluate the spodium bond (SpB) energy in compound 1. FigureFigure 7. 7. ReactionsReactions used to evaluate the the spodium spodium bond bond (SpB) (SpB) energy energy in in compound compound 1.1 .

Int.Int.Int.Int. J. J.J.J. Mol. Mol.Mol.Mol. Sci. Sci.Sci.Sci.2020 20202020, 21,,, 212121, 7091,,, 709170917091 88 of8 ofof 14 1414

2.4.2.4. CSDCSD SearchSearch 2.4. CSD Search TheThe CSDCSD versionversion 5.415.41 (including(including twotwo updatesupdates untiluntil MayMay 2020)2020) waswas inspectedinspected usingusing ConQuestConQuest versionversionThe CSD2.0.42.0.4 version(build(build 270014).270014). 5.41 (including AllAll searchessearches two updateswerewere limitedlimited until to Mayto singlesingle 2020) crystalcrystal was inspected X-rayX-ray diffractiondiffraction using ConQuest structuresstructures versionwherewhere 2.0.43D-coordinates3D-coordinates (build 270014). werewere All determined.determined. searches were InIn limited totatotal,l,l, tothethethe single CSDCSDCSD crystal containscontainscontains X-ray 22,05522,05522,055 diffraction CrystallographicCrystallographicCrystallographic structures whereInformationInformation 3D-coordinates FilesFiles (CIFs)(CIFs) were withwith determined. atat leastleast oneone In total,SpXSpX444 structurestructure thestructure CSD contains (entry(entry(entry 1a1a1a 22,055 ininin TableTableTable Crystallographic 3),3),3), wherewherewhere SpSpSp == Information= Zn,Zn,Zn, Cd,Cd,Cd, Hg,Hg,Hg, FilesXX cancan (CIFs) bebe anyany with atom,atom, at least thethe oneSp–XSp–X SpX bondsbonds4 structure werewere setset (entry atat anyany 1a type intype Table ofof bondbond3), where andand Spthethe =numbernumberZn, Cd, ofof Hg, bondedbonded X can atomsatoms be any toto atom,SpSp werewere the Sp–Xsetset toto bonds fourfour (denoted(denoted were set atbyby any thethe type“T4”“T4” of superscrsuperscr bond andiptiptipt the ininin thethethe number table).table).table). of AAA bonded similarsimilarsimilar atoms searchsearchsearch to returnedreturnedreturned Sp were set22,44922,44922,449 to fourCIFsCIFs (denoted forfor SpXSpX555 by structuresstructuresstructures the “T4” (entry(entry(entry superscript 2a)2a)2a) andandand in 13,50513,50513,505 the table). CIFsCIFsCIFs forfor Afor similar SpXSpXSpX666 StructuresStructuresStructures search returned (entry(entry(entry 3a).3a).3a). 22,449 WithinWithinWithin CIFs thesethesethese for SpX threethreethree5 structuressearches,searches, (entrythethe Zn–OZn–O 2a) bondandbond 13,505 distancesdistances CIFs werewere for SpX measuredmeasured6 Structures reresultingsulting (entry inin 3a). thethe data Withindata shownshown these inin three entriesentries searches, 1b,1b, 2b,2b, the andand Zn–O3b3b forfor bond XX333Sp–O,Sp–O, distances XX444Sp–O,Sp–O, were andand measured XX555Sp–OSp–O resultingstructuresstructures in respectively.respectively. the data shown TheThe in reresulting entriessulting 1b,distancedistance 2b, and distributionsdistributions 3b for X3Sp–O, havehave Xbeen4beenSp–O, plottedplotted and X as5asSp–O relativerelative structures frequenciesfrequencies respectively. asas showshow Thenn inin resulting FigureFigure 88 distance(see(see greygrey distributions highlightedhighlighted havearea).area). been plotted as relative frequencies as shown in Figure8 (see grey highlighted area). TableTable 3.3. NumericalNumerical overviewoverview ofof CSDCSD data.data. SpSp == Zn,Zn, CdCd oror Hg,Hg, XX == anyany atom,atom, allall bondsbonds couldcould bebe anyany Table 3.typetypetypeNumerical ofofof bond.bond.bond. overview of CSD data. Sp = Zn, Cd or Hg, X = any atom, all bonds could be any type of bond.

EntryEntryEntry SearchSearchSearch CIFs CIFs CIFs HitsHitsHits 1a SpT4T4X4 22,055 1a1a1a SpSpSp XXX444 4 22,05522,05522,055 1b  X3ZnT4T4–O 9557 37,255 1b1b1b  XXXX3333ZnZnZn –O–O–O 955795579557 37,25537,25537,255 a a aa  3 T4T4 1c1c1c  XXX333ZnZn –X–O–X–O (2(2(2 bonds) bonds)bonds) 62 6262 134134134 a,b a,ba,b T4T4 1d1d1d a,b  XXX333ZnZnT4–X–X–O–X–X–O (3 (3(3 bonds) bonds)bonds) 788078807880 31,12031,12031,120 a T4 1e a aa X 3Zn T4T4–X–X–X–O (4 bonds) 724 2123 1e1e  XXX3333ZnZn –X–X–X–O–X–X–X–O (4(4 bonds)bonds) 724724 21232123 T5T5 2a2a2a SpSpSpT5XXX555 5 22,44922,44922,44922,449 T5T5 2b2b2b  XXXX4444ZnZnZnT5–O–O–O 864986498649 36,51936,51936,519 a aa T5 2c2c2c a  XXX444ZnZn T5–X–O–X–O (2(2(2 bonds) bonds)bonds) 240 240 240 818 818 818 a,c T5 2d a,ca,c 4 T5 6377 21,227 2d2d a,c  XXX444ZnZnT5–X–X–O–X–X–O (3 (3(3 bonds) bonds)bonds) 6377 6377 21,227 21,227 a T5 2e aa X44Zn T5–X–X–X–O (4 bonds) 1136 2463 2e2e a  XXX444ZnZnT5–X–X–X–O–X–X–X–O (4(4 bonds)bonds) 1136 1136 2463 2463 T6T6 3a3a3a SpSpSpT6XXX666 6 13,50513,50513,50513,505 3b T6T6 10,347 44,770 3b3b  XXXX5555ZnZnZnT6–O–O–O 10,34710,347 44,77044,770 a a aaThea TheTheThe intramolecular intramolecularintramolecularintramolecular Zn Zn···OZn···OZn···OO distance distandistandistan wascece was limitedwas limitedlimited to 4.41 totoA ˚ 4.414.41 (i.e., ÅÅ The (i.e.,(i.e., sum TheThe of sumsum the Bondi ofof thethe van BondiBondi der van Waalsvan derder radii WaalsWaals of Zn (1.39 A)˚ and O (1.52 A)˚··· plus a tolerance of 1.5 A);˚ b 5549 Crystallographic Information Files (CIFs) involve a radii of Zn (1.39 Å) and O (1.52 Å) plus a tolerance of 1.5 Å); bbb 5549 Crystallographic Information Files carboxylateradiiradii ofof ZnZn (OC(R)O) (1.39(1.39 Å)Å) fragment; andand OO (1.52(1.52c 4778 Å)Å) CIFs plusplus involve aa tolerancetolerance a carboxylate ofof 1.51.5 Å);Å); (OC(R)O) 55495549 Crystallographic fragment.Crystallographic InformationInformation FilesFiles c (CIFs)(CIFs)(CIFs) involveinvolveinvolve aaa carboxylatecarboxylatecarboxylate (OC(R)O)(OC(R)O)(OC(R)O) fragment;fragment;fragment; ccc 477847784778 CIFsCIFsCIFs involveinvolveinvolve aaa carboxylatecarboxylatecarboxylate (OC(R)O)(OC(R)O)(OC(R)O) fragment.fragment.fragment. These distributions reveal clear peak-shapes that are centered around about 1.95 A˚ in tetracoordinated Zn (n = 3), 2.00 A˚ for pentacoordinated Zn (n = 4) and 2.10 A˚ for hexacoordinated Zn TheseThese distributionsdistributions revealreveal clearclear peak-shapespeak-shapes thatthat areare centeredcentered aroundaround aboutabout 1.951.95 ÅÅ inin (n = 5). These distances are well below the sum of the van der Waals radii of O (1.52 A)˚ and Zn (1.39 A˚ tetracoordinatedtetracoordinatedtetracoordinated ZnZnZn (((nn === 3),3),3), 2.002.002.00 ÅÅÅ forforfor pentacoordinatedpentacoordinatedpentacoordinated ZnZnZn (((nn === 4)4)4) andandand 2.102.102.10 ÅÅÅ forforfor hexacoordinatedhexacoordinatedhexacoordinated according to Bondi [18,19] or 2.01 A˚ according to Hu and Robertson [20]; see also vertical dashed lines). ZnZn ((nn === 5).5).5). TheseTheseThese distancesdistancesdistances areareare wellwellwell belowbelowbelow thethethe sumsumsum ofofof thethethe vanvanvan derderder WaalsWaalsWaals radiiradiiradii ofofof OOO (1.52(1.52(1.52 Å)Å)Å) andandand ZnZnZn The increased bond distances are likely a result of increased steric crowding of the Zn complexes with (1.39(1.39 ÅÅ accordingaccording toto BondiBondi [18,19][18,19] oror 2.012.01 ÅÅ accordingaccording toto HuHu andand RobertsonRobertson [20];[20]; seesee alsoalso verticalvertical increasing coordination number. In all cases, virtually no data is found with a Zn–O bond distance dasheddashed lines).lines). TheThe increasedincreased bondbond distancesdistances areare likelylikely aaa resultresultresult ofofof increasedincreasedincreased stericstericsteric crowdingcrowdingcrowding ofofof thethethe ZnZnZn above 2.5 A.˚ complexescomplexes withwith increasingincreasing coordinationcoordination number.number. InIn allall cases,cases, virtuallyvirtually nono datadata isis foundfound withwith aa Zn–OZn–O To inspect the distance distributions of formally non-bonded Zn O intramolecular distances, bondbond distancedistance aboveabove 2.52.5 Å.Å. ··· three distinct searched were performed for SpX4 (entry 1c–e in Table3) and SpX 5 structures (entry 2c–e in Table3) where the Zn and O atoms were separate by two (entries c), three (entries d) or four (entries e) bonds. The cutoff for the intramolecular Zn O distances was set at the sum of the van ··· der Waals radii of Zn (1.39 A)˚ and O (1.52 A)˚ according to Bondi [18,19], plus a tolerance of 1.5 A.˚ The resulting distance distributions are also plotted in Figure8 (top = ZnX4, bottom = ZnX5). Hardly any data was found with a Zn O distance of 2.5 A˚ or below, implying that none of the cases found are ··· genuine coordination bonds (found at ~2.0 A,˚ see grey highlight). The data involving two bonds of separation between Zn and O are shown as blue spheres, and appear to display a grouping of data near 2.9 A˚ for ZnX4 and around 3.1 A˚ for ZnX5 structures. Both are well below the sum of the van der Waals radii according to Hu and Robertson (3.53 A)˚ and near the van der Waals benchmark according to Bondi (2.91 A).˚ It must be noted however, that an O atom removed two bonds away from Zn in a linear fashion (i.e., Zn–X–O = 180◦) is found at about 2.9 A,˚

Int. J. Mol. Sci. 2020, 21, 7091 9 of 14

while this distance is 2.7 A˚ for a Zn–X–O angle of 90◦ (for X = N or O). This means that a relatively short distance (below the Hu and Robertson van der Waals benchmark) is inevitable, as is reflected by the grouping of data. The longer Zn O distance observed for ZnX structures (bottom) is likely a ··· 5 resultInt. J. Mol. of the Sci. longer2020, 21, Zn–X7091 distance on pentacoordinated structures (as found for Zn–O bonds). 9 of 14

FigureFigure 8. 8. PlotsPlots ofof thethe relativerelative frequenciesfrequencies asas aa function function of of the the Intramolecular Intramolecular Zn Zn∙∙∙OO distances distances for for ··· tetracoordinatedtetracoordinated (top) (top) and and pentacoordinated pentacoordinated (bottom) (bottom) Zn Zn complexes complexes where where Zn Zn and and O O are are separated separated byby two two (blue (blue circles), circles), three three (red (red diamonds) diamonds) or or four four (orange (orange hexagons) hexagons) bonds. bonds. For For reference reference purposes, purposes,

thethe distribution distribution of of the the Zn–O Zn–O bond bond lengths lengths found found in in complexes complexes ZnX ZnX4 4(white (white circles) circles) ZnX ZnX55(grey (grey circles) circles) oror ZnX ZnX66(black (black circles) circles) are are also also plotted plotted (highlighted (highlighted in in grey, grey, X X can can be be any any atom). atom). The The dashed dashed vertical vertical lineslines indicate indicate the the sum sum of of the the van van der der Waals Waals radii radii of Oof (1.52 O (1.52A)˚ andÅ) and Zn accordingZn according to Bondi to Bondi (1.39 (1.39A)˚ or Å) Hu or andHu Robertsonand Robertson (2.01 (2.01A).˚ Å).

TheTo inspect data for the Zn distanceO distances distributions separated of formally by three non bonds‐bonded are shown Zn∙∙∙O as intramolecular red diamonds distances, and are ··· nearlythree distinct evenly searched distributed were in betweenperformed Zn forO SpX= 2.5–4.54 (entryA ˚1c–e with in aTable small 3) hill-like and SpX feature5 structures around (entry 3.3 2c–A.˚ ··· Furthere in Table inspection 3) where of the these Zn data and revealed O atoms that were most separate consist by of two carboxylates (entries c), (OC(R)O; three (entries 5549/ 7880d) or CIFs four for(entries ZnX4 e)and bonds. 4778 The/6377 cutoff CIFs for for the ZnX intramolecular5). Other ligands Zn∙∙∙O involve distances nitro was groups, set at the amides, sum of perchlorates, the van der phosphatesWaals radii andof Zn polyoxometallates. (1.39 Å) and O (1.52 In Å) these according structures, to Bondi a short [18,19], Zn plusO distance a tolerance is not of inevitable. 1.5 Å. The 1 ··· Forresulting example, distance a simple distributions molecular are mechanics also plotted model in Figure of a η -coordinated8 (top = ZnX4 carboxylate, bottom = ZnX ligand5). Hardly indicates any thandata the was uncoordinated found with a Zn∙∙∙OO distance of (which 2.5 Å are or below, three bonds implying apart) that can none vary of from the cases about found 4.1 A˚ are to ··· 3.1genuineA.˚ The coordination data below bonds about (found 3.5 A˚ (the at ~2.0 Hu Å, and see Robertson grey highlight). van der Waals benchmark) could thus be seenThe as data cases involving of intramolecular two bonds interactions. of separation Manual between inspection Zn and O of are these shown data as revealed blue spheres, that these and featuresappear to are display an artifact a grouping arising of largely data near from 2.9 uncoordinated Å for ZnX4 and carboxylate around 3.1 O-atoms Å for ZnX in5 five-memberedstructures. Both chelateare well Zn-structures. below the sum of the van der Waals radii according to Hu and Robertson (3.53 Å) and near the vanData der involving Waals benchmark four bonds inaccording between to Zn Bondi and O (2.91 are shownÅ). It must as orange be noted hexagons however, and that for both an O ZnX atom4 (top)removed and ZnX two5 bonds(bottom) away structures from Zn there in a is linear a feature fashion at about (i.e., 4.1Zn–X–OA.˚ This = 180°) feature is cannotfound at be about ascribed 2.9 toÅ, anwhile interaction this distance as this is is 2.7 above Å for the a Zn–X–O van der Waalsangle of benchmarks 90° (for X = (according N or O). This to both means Bondi that and a relatively Hu and short distance (below the Hu and Robertson van der Waals benchmark) is inevitable, as is reflected by the grouping of data. The longer Zn∙∙∙O distance observed for ZnX5 structures (bottom) is likely a result of the longer Zn–X distance on pentacoordinated structures (as found for Zn–O bonds). The data for Zn∙∙∙O distances separated by three bonds are shown as red diamonds and are nearly evenly distributed in between Zn∙∙∙O = 2.5–4.5 Å with a small hill‐like feature around 3.3 Å. Further inspection of these data revealed that most consist of carboxylates (OC(R)O; 5549/7880 CIFs for ZnX4 and 4778/6377 CIFs for ZnX5). Other ligands involve nitro groups, amides, perchlorates,

Int. J. Mol. Sci. 2020, 21, 7091 10 of 14

Robertson). For the data involving tetracoordinated Zn (top) there is some data in between 2.5–3.7 A˚ that is consistent with an interaction geometry. Data with such short distances are less prevalent for structures involving pentacoordinated Zn (bottom). Given the analyses above it is clear that there is no strong directional trend in the intramolecular SpB for Zn O interactions. Moreover, the Zn O distances found in 1 (2.664 and 2.692 A)˚ and 2 (2.667 ··· ··· and 2.688 A)˚ are on the short side of the distributions shown in Figure8 and can thus be considered as rare. This is particularly relevant for the distributions with four bonds of separation between Zn and O (orange spheres), which is also the case in 1 and 2. It must be noted however, that the 1,2-relationship of the two O-donor atoms in the ligands deployed preorganizes the alkoxy groups involved in the SpB interaction.

3. Materials and Methods

3.1. Materials All chemicals were of reagent grade and used as purchased from Sigma-Aldrich without further purification.

3.2. Preparation of Reduced Schiff Base Ligands

3.2.1. Preparation of 2-(((3-(dimethylamino)propyl)amino)methyl)-6-methoxyphenol (HL1) For the synthesis of HL1 a solution of N,N-dimethyl-1,3-diaminopropane (2 mmol, 0.25 mL) with 3-methoxysalicylaldehyde (2 mmol, 0.310 g) in methanol (20 mL) was refluxed for ca. 2 h. The resulting yellow colored solution (20 mL) was then cooled to 0 ◦C in an ice bath, and solid sodium borohydride (4 mmol, 0.150 g) was added slowly with constant stirring until the yellow color of the solution disappeared. The resulting reaction mixture was acidified with glacial acetic acid (3 mL) and then evaporated to dryness under reduced pressure. The resulting slurry was suspended in water (15 mL) and extracted with dichloromethane. Finally, the solvent (i.e., dichloromethane) was evaporated under reduced pressure in a rotary evaporator to obtain the reduced Schiff base ligand, HL1.

2 3.2.2. Preparation of 2,20-[(1-Methyl-1,2-ethanediyl)bis(iminomethylene)]bis[6-ethoxyphenol] (H2L ) A methanol solution containing 1,2-diaminopropane (2 mmol, 0.17 mL) and 3-ethoxysalicylaldehyde (2 mmol, 0.330 g) was refluxed for ca. 2 h resulting in a yellow solution. NaBH4 (4 mmol, 0.150 g) was added and the reaction mixture was stirred the solution was colorless. The resulting reaction mixture was acidified with glacial acetic acid (3 mL) and the volatiles were evaporation under reduced pressure. The remaining mass was suspended in water (15 mL) and extracted with dichloromethane. Finally, the solvent (i.e., dichloromethane) was evaporated under reduced pressure in a rotary evaporator to 2 obtain H2L .

3.2.3. Preparation of the Complex [Zn2(µ1,3-OAc)(L1)2]I MeOH (1) · A methanol solution of zinc acetate dihydrate (2 mmol, 0.440 g) was added to a stirred methanol solution of the ligand HL1. Stirring was continued for an additional 2 h. An aqueous methanol (1:1) solution of potassium iodide (1 mmol, 0.170 g) was then added and the mixture was stirred for ca. 2 h. The solution was then kept in open air at room temperature for a few days, resulting in the formation of white crystals that were isolated by filtration. X-ray quality single crystals were collected from this crystalline product. Yield: 0.550 g, 68% (based on zinc). Anal. Calc. for C29H49N4O7Zn2I (823.40): C, 42.30; H, 6.00; N, 1 6.80%. Found: C, 42.1; H, 5.9; N, 6.9%. FT-IR (KBr, cm− ): 3158 (νN H); 2997-2878(νC H); 1473 {νsym 1 1 − −3 (COO )}; 1557 {νasym (COO )}. λmax (nm) [εmax(lit mol cm )] (DMF): 291 (6.25 10 ). − − − − × Int. J. Mol. Sci. 2020, 21, 7091 11 of 14

3.2.4. Preparation of the Complex [Zn2(µ1,3-OAc)(L2)(NCS)] (2) A methanol solution of zinc acetate dihydrate (2 mmol, 0.440 g) was added to a stirred solution 2 of ligand H2L in methanol. Stirring was continued for an additional 2 h. An aqueous methanol (1:1) solution of sodium thiocyanate (1 mmol, 0.080 g) was added and stirring was continued ca. 2 h. The resulting solution was kept in open air at room temperature for few days, resulting in the formation of a white crystalline product that could be isolated by filtration. X-ray quality single crystals were collected from this crystalline product. Yield: 0.430 g, 69% (based on zinc). Anal. Calc. for C24H31N3O6SZn2 (620.35): C, 46.47; H, 5.04; N, 1 6.77%. Found: C, 46.3; H, 4.9; N, 6.9%. FT-IR (KBr, cm− ): 3223 (νN H); 2979-2878 (νC H); 2102 (νNCS); −1 1 − 3 1472 {νsym (COO )}; 1566 {νasym (COO )}. λmax (nm) [εmax(lit mol cm )] (DMF): 286 (6.93 10 ). − − − − × 3.3. Details of Instrumentation Elemental analysis (, hydrogen and ) was performed using a Perkin-Elmer 240C 1 elemental analyzer. IR spectrum in KBr (4500–500 cm− ) was recorded with a Perkin-Elmer Spectrum Two spectrophotometer. Electronic spectra in DMF were recorded on a JASCO V-630 spectrophotometer. The powder XRD data were collected on a Bruker D8 Advance X-ray diffractometer using Cu Kα radiation (λ = 1.548 A)˚ generated at 40 kV and 40 mA. The PXRD spectrum was recorded in a 2θ range of 5–50◦ using 1-D Lynxeye detector at ambient conditions.

3.4. Crystal Data Collection and Refinement Suitable single crystals of both complexes were used for data collection as described earlier [21]. Direct methods were employed for determination of molecular structures and refinements were done by full-matrix least square methods using the SHELX-18 package [22]. Multi-scan empirical absorption corrections were accomplished using the SADABS program [23]. The details of crystallographic data and structure refinement parameters are provided in Table4.

Table 4. Crystal data and refinement details of complexes 1 and 2.

Complex 1 2

Formula C29H49N4O7Zn2IC24H31N3O6SZn2 Formula Weight 823.40 620.37 Temperature (K) 273(2) 273(2) Crystal System Orthorhombic Triclinic Space group P212121 P1 a (A)˚ 12.744(6) 10.5900(8) b (A)˚ 16.260(7) 11.7781(9) c (A)˚ 17.008(8) 12.0470(9) β (◦) 90 90.588(2) β (◦) 90 101.556(2) γ (◦) 90 109.376(2) Z 4 2 3 dcal (g cm− ) 1.552 1.489 1 µ(mm− ) 2.284 1.850 F(000) 1680.0 640 Total reflection 29800 49447 Unique Reflections 7136 6157 Observe data[I>2σ(I)] 6500 5165 R(int) 0.052 0.032 R1, wR2 (all data) 0.0392, 0.0675 0.0408, 0.1052 R1, wR2 [I>2σ(I)] 0.0338, 0.0655 0.0308, 0.0922 Int. J. Mol. Sci. 2020, 21, 7091 12 of 14

3.5. Theoretical Methods The energies and geometries of the complexes included in this study were computed at the PBE0 [24]-D3 [25]/def2-TZVP [26] level of theory by means of the TURBOMOLE 7.0 software [27]. The Bader’s “Atoms in molecules” theory and NCI method [28] have been used to study the interactions discussed herein by means of the AIMall calculation package [29]. Calculations related to the wavefunction properties were carried out using the Gaussian 16 calculation package [30] at the same level of theory. In particular, the density and potential energy cubes used to generate the MEP surfaces and the wavefunction used as input to perform the QTAIM/NCI analyses were obtained using Gaussian-16. The PBE0-D3 level of theory has been recently used by us to analyze σ/π-hole interactions in the solid state [31–38]

4. Concluding Remarks Two new dinuclear Zn(II) complexes have been synthesized and characterized. They exhibit intramolecular spodium bonds that have been described herein for the first time. Moreover, the intramolecular SpBs have been characterized and differentiated from coordination bonds using a combination of QTAIM and NCI method analyses. The existence of σ-hole in both tetracoordinated and pentacoordinated Zn(II) atoms has been evidenced using MEP surfaces. Due to the intramolecular nature of the interaction, the strength of the spodium bond in 2 has been estimated using two theoretical models that evidence a moderately strong interaction (about 5 kcal/mol). Finally, the CSD analysis of − intramolecular spodium bonds revealed that the short Zn O interactions observed in compounds 1 ··· and 2 are rare and likely largely affected by the pre-organization of the ligand.

Supplementary Materials: The supplementary materials are available online at http://www.mdpi.com/1422-0067/ 21/19/7091/s1. Author Contributions: Conceptualization, A.F., A.B., T.J.M. and S.C.; methodology, M.K., A.F., A.B., and T.J.M.; software, A.F., A.B., and T.J.M.; formal analysis, all authors; investigation, all authors; writing—original draft preparation, A.F., S.C., and T.J.M.; writing—review and editing, all authors; project administration, A.F. and T.J.M.; funding acquisition, A.F. and T.J.M. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by MICIU/AEI, grant number CTQ2017-85821-R, FEDER funds. Acknowledgments: M.K. thanks the UGC, India, for awarding Junior Research Fellowship (228/(OBC)(CSIR-UGC NET DEC. 2016)]. S.C. gratefully acknowledges UGC-CAS II program, Department of Chemistry, Jadavpur University, for financial support under the head (Chemicals/Consumables/Glassware). T.J.M. thanks NWO (VIDI project 723.015.006) for funding. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Muller-Dethlefs, K.; Hobza, P. Non-Covalent Interactions: Theory and Experiment. In Theoretical and Computational Chemistry Series, 1st ed.; Hirst, J., Jordan, K.D., Thiel, W., Lim, C., Eds.; RSC Publishing: Cambridge, UK, 2010; Volume 2. 2. Mahmudov, K.T.; Kopylovich, M.N.; Pombeiro, A.J.L. (Eds.) Noncovalent Interactions in Catalysis, 1st ed.; RSC Publishing: Cambridge, UK, 2019. 3. Maharramov, A.M.; Mahmudov, K.T.; Kopylovich, M.N.; Guedes da Silva, M.F.C.; Pombeiro, A.J.L. (Eds.) Non-covalent Interactions in the Synthesis and Design of New Compounds, 1st ed.; John Wiley & Sons Inc.: Hoboken, NJ, USA, 2016. 4. Murray, J.S.; Lane, P.; Politzer, P. Expansion of the sigma-hole concept. J. Mol. Model. 2009, 15, 723–729. [CrossRef][PubMed] 5. Bauzá, A.; Mooibroek, T.J.; Frontera, A. The bright future of unconventional σ and π-hole interactions. Chem. Phys. Chem. 2015, 16, 2496–2517. [CrossRef][PubMed] 6. Alkorta, I.; Elguero, J.; Frontera, A. Not Only Hydrogen Bonds: Other Noncovalent Interactions. Crystals 2020, 10, 180. [CrossRef] Int. J. Mol. Sci. 2020, 21, 7091 13 of 14

7. Joy, J.; Jemmis, E.D. Contrasting Behavior of the Z Bonds in X–Z Y Weak Interactions: Z = Main Group ··· Elements Versus the Transition Metals. Inorg. Chem. 2017, 56, 1132–1143. [CrossRef][PubMed] 8. Joy, J.; Jemmis, E.D. Designing M-bond (X-M Y, M = transition metal): σ-hole and radial density distribution. ··· J. Chem. Sci. 2019, 131, 117. [CrossRef] 9. Bauzá, A.; Alkorta, I.; Elguero, J.; Mooibroek, T.J.; Frontera, A. Spodium Bonds: Noncovalent Interactions Involving Group 12 Elements. Angew. Chem. Int. Ed. 2020.[CrossRef] 10. Bader, R.F.W. Atoms in Molecules: A Quantum Theory; Clarendon Press: Oxford, UK, 1990. 11. Johnson, E.R.; Keinan, S.; Mori-Sanchez, P.; Contreras-Garcia, J.; Cohen, A.J.; Yang, W. Revealing Noncovalent Interactions. J. Am. Chem. Soc. 2010, 132, 6498–6506. [CrossRef] 12. Groom, C.R.; Bruno, I.J.; Lightfoot, M.P.; Ward, S.C. The Cambridge Structural Database. Acta Cryst. 2016, B72, 171–179. [CrossRef] 13. Hazari, A.; Das, L.K.; Kadam, R.M.; Bauzá, A.; Frontera, A.; Ghosh, A. Unprecedented structural variations in trinuclear mixed valence Co(II/III) complexes: Theoretical studies, pnicogen bonding interactions and catecholase-like activities. Dalton Trans. 2015, 44, 3862–3876. [CrossRef] 14. Banerjee, A.; Frontera, A.; Chattopadhyay, S. Methylene spacer regulated variation in molecular and crystalline architectures of (iii) complexes with reduced Schiff base ligands: A combined experimental and theoretical study. Dalton Trans. 2019, 48, 11433–11447. [CrossRef] 15. Mondal, I.; Basak, T.; Banerjee, S.; Chattopadhyay, S. A theoretical insight on the rigid hydrogen-bonded network in the solid state structure of two zinc(ii) complexes and their strong fluorescence behaviors. Cryst. Eng. Comm. 2020, 22, 3005–3019. [CrossRef] 16. Orpen, A.G.; Brammer, L.; Allen, F.H.; Kennard, O.; Watson, D.G.; Taylor, R. Supplement. Tables of bond lengths determined by X-ray and neutron diffraction. Part 2. Organometallic compounds and co-ordination complexes of the d- and f-block metals. J. Chem. Soc. Dalton. Trans. 1989, S1–S83. [CrossRef] 17. Roeleveld, J.J.; Lekanne Deprez, S.J.; Verhoofstad, A.; Frontera, A.; van der Vlugt, J.I.; Mooibroek, T.J. Engineering Crystals Using sp3-C Centred Tetrel Bonding Interactions. Chem. Eur. J. 2020, 26, 10126–10132. [CrossRef][PubMed] 18. Bondi, A. Van der Waals volumes and radii. J. Phys. Chem. 1964, 68, 441–451. [CrossRef] 19. Bondi, A. Van der Waals volumes and radii of metals in covalent compounds. J. Phys. Chem. 1966, 70, 3006–3007. [CrossRef] 20. Hu, S.-Z.; Zhou, Z.-H.; Robertson, B.E. Consistent approaches to van der Waals radii for the metallic elements. Z. Kristallogr. 2009, 224, 375–383. [CrossRef] 21. Karmakar, M.; Frontera, A.; Chattopadhyay, S. Methylene spacer regulated variation in supramolecular assembly of zinc(II) dicyanamide complexes with reduced Schiff base ligands: Synthesis, structure and DFT study. Cryst. Eng. Comm. 2020.[CrossRef] 22. Sheldrick, G.M. SHELXT-Integrated space-group and crystal-structure determination. Acta Cryst. 2015, C71, 3–8. [CrossRef] 23. Sheldrick, G.M. SADABS, V2014/5, Software for Empirical Absorption Correction; University of Göttingen, Institute fur Anorganische Chemie der Universitat: Gottingen, Germany, 1999. 24. Adamo, C.; Barone, V. Toward reliable density functional methods without adjustable parameters: The PBE0 model. J. Chem. Phys. 1999, 110, 6158–6169. [CrossRef] 25. Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, 132, 154104. [CrossRef] 26. Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297–3305. [CrossRef] [PubMed] 27. Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Electronic structure calculations on workstation computers: The program system turbomole. Chem. Phys. Lett. 1989, 162, 165–169. [CrossRef] 28. Contreras-García, J.; Johnson, E.R.; Keinan, S.; Chaudret, R.; Piquemal, J.-P.; Beratan, D.N.; Yang, W. NCIPLOT: A program for plotting non-covalent interaction regions. J. Chem. Theory Comput. 2011, 7, 625–632. [CrossRef] [PubMed] 29. Todd, A.; Keith, T.K. (Eds.) AIMAll (Version 13.05.06); Gristmill Software: Overland Park, KS, USA, 2013. 30. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Petersson, G.A.; Nakatsuji, H.; et al. Revision B.01; Gaussian, Inc.: Wallingford, CT, USA, 2016. Int. J. Mol. Sci. 2020, 21, 7091 14 of 14

31. Zelenkov, L.E.; Ivanov, D.M.; Sadykov, E.K.; Bokach, N.A.; Galmés, B.; Frontera, A.; Kukushkin, V.Y. Semicoordination Bond Breaking and Bond Making Changes the Supramolecular Architecture of Metal-containing Aggregates. Cryst. Growth Des. 2020.[CrossRef] 32. Thakur, S.; Gil, D.M.; Frontera, A.; Chattopadhyay, S. Exploration of Br O halogen bonding interactions in ··· dinuclear (V) complexes with Schiff base ligands. Polyhedron 2020, 187, 114676. [CrossRef] 33. Rozhkov, A.V.; Ananyev, I.V.; Gomila, R.M.; Frontera, A.; Kukushkin, V.Y. π-Hole dz2 [PtII] Interactions ··· with Electron-Deficient Arenes Enhance the Phosphorescence of PtII-Based Luminophores. Inorg. Chem. 2020, 59, 9308–9314. 34. Navarro-García, E.; Galmés, B.; Velasco, M.D.; Frontera, A.; Caballero, A. Anion Recognition by Neutral Chalcogen Bonding Receptors: Experimental and Theoretical Investigations. Chem. Eur. J. 2020, 26, 4706–4713. [CrossRef] 35. Galmés, B.; Juan-Bals, A.; Frontera, A.; Resnati, G. Charge assisted chalcogen bonds: CSD and DFT analyses and biological implication in glucosidase inhibitors. Chem. Eur. J. 2020, 26, 4599–4606. [CrossRef] 36. Katlenok, E.A.; Haukka, M.; Levin, O.V.; Frontera, A.; Kukushkin, V.Y. Supramolecular Assembly of Metal Complexes by (Aryl) I dz2[PtII] Halogen Bond. Chem. Eur. J. 2020, 26, 7692–7701. [CrossRef] ··· 37. Bofill, L.; Prohens, R.; Barbas, R.; Frontera, A. DFT analysis of uncommon π H-bond array interaction in a ··· new pterostilbene/theophylline cocrystal. Cryst. Growth Des. 2020.[CrossRef] 38. Soldatova, N.S.; Postnikov, P.S.; Suslonov, V.V.; Kissler, T.Y.; Ivanov, D.M.; Yusubov, M.S.; Galmés., B.; Frontera, A.; Kukushkin, V.Y. Diaryliodonium as a double σ-hole donor: The dichotomy of thiocyanate halogen bonding provides divergent solid state arylation by diaryliodonium cations. Org. Chem. Front. 2020, 7, 2230–2242. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).