Structure and chemical bond of carbodiylide complexes [W(CO)5{C(ECp*)2}] (E = B to Tl): DFT calculations - Nguyen Thi Ai Nhung

Vietnam Journal of Chemistry, International Edition, 55(5): 561-568, 2017 DOI: 10.15625/2525-2321.2017-00508 561 Structure and chemical bond of carbodiylide complexes [W(CO)5{C(ECp*)2}] (E = B to Tl): DFT calculations Nguyen Thi Ai Nhung Department of Chemistry, Hue University of Sciences, Hue University, Hue, Vietnam Received 2 March 2017; Accepted for publication 20 October 2017 Abstract The bonding of the carbodiylide complexes [(CO)5W-{C(ECp*)2}] (W5-C1E) was calculated at the BP86 level with the basis sets def2-SVP, def2-TZVPP, and TZ2P+. The nature of the (CO)5W-{C(ECp*)2} bonds was analyzed by energy decomposition method. The calculated structures of complexes show that all ligands C(ECp*)2 (C1E) are bonded in a tilted orientation relative to the fragment W(CO)5 in W5-C1E and the tilting angle become much more acute when E becomes heavier. Analysis of the bonding reveals that [(CO)5W–{C(E’Cp*)2}] donation in W5-C1B come from the -lone-pair orbital of C(BCp * )2, while [(CO)5W–{C(E’Cp*)2}] donation in the strongly tilted bonded complexes when E’ = Al to Tl comes from the -lone-pair orbital of the carbodiylides C(E’Cp*)2. The W-C bonds have not only (CO)5W C(ECp*)2 strong -donation but also a significant contribution π-donation and the trend of the W-C bond strength in W5-C1E complexes. EDA-NOCV calculations reveal that C(ECp*)2 ligands in W5-C1E complexes are strong -donors and weak -donors which make them good spectator ligands that are well-suited for synthesizing robust catalysts for a variety of applications. Keywords. Carbodiylides, energy decomposition analysis, bond dissociations energy, bonding analysis. 1. INTRODUCTION The recent experimental studies of main-group elements pentamethylcyclopentadienyl (Cp*) suggested that the steric requirements of the - or - bound Cp* group enable the kinetic stabilization of highly reactive species and represents a very important substituent [1]. Moreover, the chemistry of Cp* with transition metal complexes has advanced significantly in the fields of organometallic catalysis [2-4]. The chemistry of group-13 diyl ligand ECp* (E = B to Tl) is a topic of interest to both synthetic and theoretical chemists [5- 8]. Transition metal (TM) complexes with ECp* ligands have been the subject of extensive experimental and theoretical investigations of the first stable complex [(CO)4Fe-{AlCp*}] which was isolated and characterized by X-ray analysis in 1997 by Fischer et al. [5]. Further work was reported with group-13 homologues [(CO)4Fe-{ECp*}], where E = B and Ga [6, 7]. Very recently, the first homoleptic complex with an ECp* substituent [Mo(GaCp*)6], was synthesized [9]. Numerous other group-13 complexes with ligands ER, where R is either a strong donor or a very bulky substituent, have since been reported [1, 5]. Theoretical studies clearly showed that diyl ligands ER were strong donor and weaker -acceptors than CO [10]. It has been known that there is another class of stable carbones CL2, where L is a group-13 diyl ligand ECp* (E = B to Tl), that has been studied in the recent past [10, 11]. Theoretical studies clearly showed that the diyl ligand ECp* is a stronger - donor and weaker -acceptor than CO [11, 12], and ECp* was considered as a ligand for stabilizing a divalent carbon(0) atom in carbodiylides C(ECp * )2. The coordination chemistry of monovalent group-13 elements has received significant attention and is presently a topic of intensive experimental research [7-13]. The diyl ligand ER where the coordinated atom E has the formal oxidation state +1 is analogous to CO [11-14]. Numerous other group-13 complexes with ER such as model ligands ECp, EN(SiH3)2, and ECH3 [10, 11] where R is either a strong donor or a very bulky substituent, and their electronic structures have been analyzed. This paper provides the detailed calculations on quantum-chemical investigations of the model complexes [(CO)5W-{C(ECp*)2}] (W5-C1E) where E = B to Tl (Scheme 1). The aim of the study presented in this study was to investigate the nature of bonding and extent of and interactions between ligands C(ECp*)2 and the TM fragment VJC, 55(5), 2017 Nguyen Thi Ai Nhung 562 W(CO)5 (scheme 1). The structures of the complexes and the bond dissociation energies are predicted with DFT. The electronic structures determined by charge- and energy decomposition analysis of the systems are also presented. Scheme 1: Complexes investigated in this study: [(CO)5W-{C(ECp*)2}] W5-C1E (E = B to Tl) 2. COMPUTATIONAL METHODS Geometry optimizations of the molecules have been carried out without symmetry constraints using the Gaussian03 [15] optimizer together with Turbomole 6.0.1 [16] energies and gradients at the BP86 [17, 18] /def2-SVP [19] level of theory (denoted BP86/SVP). For the heavier group-13 atoms In, Tl, and for W, small-core quasi-relativistic effective core potentials (ECPs) were used [20]. The stationary points on the potential energy surface (PES) obtained at this level of theory was denoted as BP86/def2-SVP. All structures presented in this study turned out to be minima on the PES. Single point calculations with the same functional but the larger def2-TZVPP [21] basis set and the small core ECPs for In, Tl and W atoms were carried out with Gaussian03 on the structures derived on BP86/def2- SVP level of theory. The bond dissociation energies and molecular orbitals were calculated and plotted at the BP86/TZVPP//BP86/SVP level of theory using the NBO 3.1 program [22, 23]. The complexes were re-optimized for the EDA-NOCV with the program package ADF 2009.01 [24] with BP86 in conjunction with a triple-z-quality basis set using uncontracted Slater-type orbitals (STOs) augmented by two sets of polarization function, with a frozen- core approximation for the core electrons [25]. An auxiliary set of s, p, d, f, and g STOs was used to fit the molecular densities and to represent the Coulomb and exchange potentials accurately in each SCF cycle [26]. Scalar relativistic effects were incorporated by applying the zeroth-order regular approximation (ZORA) [27]. The calculations have been carried out at the BP86/TZ2P+ level of theory on the BP86/def2-SVP optimized geometries which were used for the bonding analysis in term of the EDA [28]-NOCV [29] method of C1 symmetric geometries. 3. RESULTS AND DISCUSSION The optimized geometries of W5-C1B to W5-C1Tl complexes and free C1B to C1Tl ligands are shown in Figure 1. The theoretically predicted W-C bond length decreases from W5-C1B (2.434 Å) to W5- C1Tl (1.989 Å). The equilibrium geometries of W5- C1B to W5-C1Tl in Figure 1 shows that only the ligand C1B has the B1 atom 1-bonded to the central C atom which the longest and shortest B1-C bonds of Cp* ring are 3.420 and 1.577 Å, while the B2 atom is 3-bonded to the central C atom with longest and shortest bond of 2.308 and 1.649 Å, respectively. In contrast, the Al atoms in W5-C1Al are both 5-bonded to the central C atom of the respective Cp* rings, which have values between 2.222 and 2.237 Å, while the heavier homologues C(ECp*)2 where E = Ga to Tl, suggest that there is a trend toward bonding between 3 and 1 for E-Cp* when E becomes heavier. This remains the five E-C bonds to the carbon atoms of the Cp* ligand which exhibit between 2.272 and 2.364 Å for W5-C1Ga, 2.495 and 2.802 Å for W5-C1In, and 2.748 and 2.845 Å for W5-C1Tl. The bending angle, , is 156.9° in W5-C1B and becomes much more acute in the heavier homologues which the value decreases from = 138.3° in W5-C1Al to = 132.8° in W5- C1Ga and then increases a bit for W5-C1In 135.8°, and is 138.7° for W5-C1Tl. This means all ligands are bonded in a tilted orientation to W(CO)5 in the complexes. This implies that there is not only a possible interaction with the -lone-pair of C1E, but also with the -lone-pair [10, 11, 32]. Figure 1 also shows the optimized geometries of free C(ECp*)2 molecules. There is a significant difference between boron compound C1B and the heavier homologues. The former has a nearly linear B1-C-B2 moiety (178.5°), whereas the latter, the heavier species, are strongly bent. The bending angle, E-C-E, of the heavier homologues varies between 101.1° for C1Ga and 104.5° for C1Tl. The calculated bending angles are clearly smaller than those in C(NHCMe)2 (131.8°) and in C(PPh3)2 (136.9°) [30]. The geometry of C(BCp*)2 suggests that the compound can be considered as a substituted homologue of HB=C=BH, which has been synthesized by laser- ablated of boron atoms with methane in a low- temperature matrix by Andrews [31]. The boron atoms are 1-bonded to one carbon atom of the Cp* ligand. The calculated B1-C and B2-C bonds in C(BCp*)2 are 1.380 Å. The interatomic B-C VJC, 55(5), 2017 Structure and chemical bond of carbodiylide 563 distances to the other carbon atoms of the ring are much longer [32], and should not be considered as genuine boron-carbon bonds. The C-C bonds in the Cp* groups, which are rotated with respect to each other by about 90° around the C-B-C axis, show the characteristic pattern of alternating distances in a 1,3-butadiene moiety that are bonded to the carbon atoms of Cp* rings, which exhibit between 1.553 and 3.177 Å [32]. This situation is strikingly different to the C-C bonds in the Cp* rings of C(AlCp*)2, which have nearly identical values between 1.442 and 1.445 Å. The same trend holds for each of the five Al-C bonds to the carbon atoms of the Cp* ligand, which lie between 2.259 and 2.273 Å. The calculated equilibrium structure of C(AlCp*)2 clearly shows that the Cp* ligands are 5-bonded to aluminum with calculated Al1-C and Al2-C bonds of 1.844 and 1.843 Å. The optimized geometries of the remaining homologues C(ECp*)2, where E = Ga, In, and Tl, suggests that there is a trend toward 3 or 1 bonding for E-C (Cp* rings) when E becomes heavier. This becomes obvious by an increasing distortion of the cyclic ligands toward bond alternation of the C-C distances in the ring and particularly by the differences among the E-C bonds to the Cp* ligand. The ligand in C(GaCp*)2 have one short (2.066 Å) Ga1-C bond, two rather long Ga-C bonds (2.449 and 2.472 Å), and two very long Ga-C distances (3.002 and 3.023 Å). The GaCp* bonding can be interpreted as intermediate between 3 and 1. Note that the similar situation is found for the indium and thallium ligands C(InCp*)2 and C(TlCp*)2. A comparison between the geometry of W5-C1B to W5-C1Tl and the free ligands C1B to C1Tl shows that the E-C bonds in all ligands CE are clearly longer in complexes W5-C1B to W5-C1Tl (0.4 to 0.7 Å) than those in the free ligands. Note that free C(BCp*)2 ligand has a nearly linear B1-C- B2 moiety (178.5°), whereas the ligand in the complex has a bent B1-C-B2 moiety (150.1°). The calculated B1-C and B2-C bonds in C1B are 1.380 Å, which is slightly longer than the calculated values of 1.374 Å for the linear equilibrium structure of HB=C=BH at BP86/SVP [33]. The calculated B1-C and B2-C bonds in W5-C1B are 1.403 and 1.424 Å. The optimized geometries of the free ligands in figure 1, together with the calculated values for the most important bond lengths and angles, are similar to the calculated values of carbodiylides C(ECp*)2 investigated recently [32]. Figure 1: Optimized geometries of the complexes W5-C1E and the free ligands C1E at the BP86/def2-SVP level. Bond lengths are given in Å; angles in degrees. Calculated metal-ligand BDEs, De (kcal/mol), at the BP86/def2-TZVPP//BP86/def2-SVP level for the (CO)5W-C(ECp*)2 bonds (E = B to Tl) Figure 1 also gives the theoretically predicted bond dissociation energies (BDEs) for the W-C bond of W5-C1B to W5-C1Tl and exhibit an interesting non-steric trend. The calculated bond energies suggest that the tungsten-cabodiylides bond strength increases from W5-C1B to W5-C1Al, decreases for E = Ga, and then increases again for W5-C1In and W5-C1Tl. The data thus suggest that the heavier complexes have stronger bonds than the lighter homologues. Continuously, the EDA-NOCV calculations give a more insight into the nature of metal-ligand bonding in W5-C1B to W5-C1Tl. Bending angle ( ) is the angle W-C-X where X is the midpoint between the E-E distance W5-C1B α = 156.9°; De = 25.5 W5-C1Al α = 138.3°; De = 46.1 W5-C1Ga α = 132.8°; De = 43.5 W5-C1In α = 135.8°; De = 46.9 W5-C1Tl α = 138.7°; De = 52.9 C1B C1Al C1Ga C1In C1Tl VJC, 55(5), 2017 Nguyen Thi Ai Nhung 564 Table 1 shows the numerical results of EDA-NOCV calculations for the (CO)5W-carbodiylide bonds. The EDA-NOCV data demonstrates that the increase in metal-ligand bonding comes from the intrinsic interactions Eint, which clearly increases from W5- C1B to W5-C1Tl. The intrinsic interaction in W5- C1Ga is even smaller than that in W5-C1Al and increases for the heavier homologues. The increase of Eint from W5-C1B to W5-C1Al is not as steep as the BDE, which strongly increases from W5-C1B to W5-C1Al. This is because the aluminum complex has a significantly smaller preparation energy of Eprep=8.7 kcal/mol, while for the boron complex it is Eprep = 23.1 kcal/mol. From this, it follows that linear Cp*B=C=BCp* has to be bent in complex W5-C1B. The small decrease of the BDEs (De) from W5-C1Al (45.0 kcal/mol) to W5-C1Ga (42.8 kcal/mol) is due to the small increase in the preparation energy Eprep and the small decrease in Eint for the complexes. In contrast, the increase of De in W5-C1E (E = Ga to Tl) comes from the larger intrinsic interactions Eint (-52.2 kcal/mol) for W5- C1Ga to -66.5 kcal/mol for W5-C1Tl), and are nearly canceled out by the preparation energies Eint in the complexes. Table 1: EDA-NOCV results at the BP86/TZ2P+ level for complexes W5-C1B to W5-C1Tl using the moieties [W(CO)5] and [C(ECp*)2] as interacting fragments. The complexes were analyzed with C1 symmetry. Energy values in kcal/mol Compound W5-C1B W5-C1Al W5-C1Ga W5-C1In W5-C1Tl Fragments W(CO)5 C(BCp * )2 W(CO)5 C(AlCp * )2 W(CO)5 C(GaCp * )2 W(CO)5 C(InCp * )2 W(CO)5 C(TlCp * )2 Eint -48.7 -53.7 -52.2 -64.6 -66.5 EPauli 100.3 101.6 102.4 148.0 156.1 Eelstat [a] -92.4 (62.0 %) -95.2 (61.3 %) -92.3 (59.7 %) -128.5 (57.7 %) -133.6 (62.8 %) Eorb [a] -56.6 (38.0 %) -60.1 (38.7 %) -62.3 (40.3 %) -79.0 (42.4 %) -94.1 (37.2 %) E [b] -35.9 (63.4 %) -42.3 (70.3 %) -45.0 (72.2 %) -61.1 (78.0 %) -76.7 (81.5 %) E [b] -18.0 (31.8 %) -15.0 (24.9 %) -13.8 (22.2 %) -14.5 (18.3 %) -14.3 (15.2 %) Erest [b] -2.7 (4.8 %) -2.8 (4.8 %) -3.5 (5.6 %) - 2.9 (3.7 %) -3.1 (3.3 %) Eprep 23.1 8.7 9.4 17.5 18.2 E(= -De) -25.6 (-25.5) [c] -45.0 (-46.1) [c] -42.8 (-43.5) [c] -47.1 (-46.9) [c] -48.3 (-52.4) [c] [a] The values in parentheses are the percentage contributions to the total attractive interactions Eelstat + Eorb; [b] The values in parentheses are the percentage contributions to the total orbital interactions Eorb ; [c] The values in parentheses give the dissociation energy at the BP86/def2-TZVPP//BP86/def2-SVP level. The three main terms EPauli, Eelstat, and Eorb are considered to inspect their contribution to the interaction energy Eint of the molecules. Inspection of the three main terms indicated that the Pauli repulsion EPauli was similar for the lighter species where E = B, Al, and Ga and became larger for the heavier atoms when E = In and Tl. This can be explained that the increase in the bond strength for the heavier carbodiylides comes from the stronger attraction rather than weaker repulsion [33]. The attractive interactions Eelstat increase from W5- C1B to W5-C1Tl except for the small decrease from W5-C1Al to W5-C1Ga. The increase in the attractive interactions Eelstat and Eorb of the heavier carbodiylide ligands can be traced back to the -lone-pair orbital, which leads to stronger - orbital interactions E and to stronger electrostatic attraction Eelstat. Inspection of the trend of the electrostatic term Eelstat, and the orbital term Eorb shows that the stronger bonds are mainly caused by the latter term. The -orbital contribution E is much stronger for the heavier carbodiylides which means they increase from W5-C1B to W5-C1Tl. In contrast, the -orbital contributions E are much weaker than those of E and decrease for the heavier group-13 diyl ligands. The Eorb term of the EDA-NOCV results was examined further to obtain more detailed information on the bonding in W5- C1B to W5-C1Tl. Figures 2 shows plots of those pairs of orbitals k/ -k that yield the NOCVs with the largest contributions to the - and -orbital terms E and E in W5-C1B and W5-C1Tl. The associated deformation densities, , and stabilization energies are also given. The adducts W5-C1Al, W5-C1Ga, and W5-C1Tl exhibit similar shapes to those of the complex W5-C1Tl. Therefore, the NOCV pairs of W5-C1Al, W5-C1Ga and W5- C1In are not shown in Figures 2. Note that the green/red colors for k/ -k indicate the sign of the orbitals, while the yellow/blue colors in the VJC, 55(5), 2017 Structure and chemical bond of carbodiylide 565 deformation densities, indicate the charge flow. The yellow areas of designate present charge depletion while the blue areas indicate charge accumulation. The charge flow takes places in the direction yellow blue. Figure 2a gives the NOCV pairs 1/ -1 and the deformation densities 1 of the most important pair of s orbitals for E of W5-C1B. The associated stabilization energies of 1 are approximately 90% of the total energies E (Table 2). Thus, the orbital pairs 1/ -1 can be considered as dominant sources of s bonding for the C1B ligands in the two complexes. The shape of the orbital pairs clearly indicates that -orbital interactions take place between the donor orbitals of C1B ligands, and the acceptor orbital of W(CO)5. Note that the charge flow (CO)5W C(ECp*)2 involves not only donor C and acceptor W atoms. In particular, there is charge flow into the W-CO bonding and C-O anti-bonding regions, particularly for the trans-CO bond, which agrees with the change in the bond lengths between W(CO)6 and W5-C1E. Figure 2-a clearly shows that the -type interaction has clearly the direction (CO)5W C(ECp*)2. The deformation density reveals that the charge flow comes from the C(ECp*)2 ligands (E = B) toward the W(CO)5 fragment; this is in good agreement with the calculated partial charges which were shown in Table 1. The NOCV pairs were analyzed for the W(CO)5-carbodiylides because the ligands C(ECp*)2 are double donors, and there should be no significant contribution from (CO)5W C(ECp*)2 -back- donation. Figure 2-b and 2-c show that two NOCV pairs k/ -k (k = 2, 3) dominate the total stabilization E in W5-C1B. The shape of the NOCV pairs 2/ -2 and the deformation densities, which reveal the charge flow 2, are shown in Figure 2-b and indicate that the stabilization of -8.7 kcal/mol can be assigned to the (CO)5W C(BCp*)2 -backdonation where the C-B vacant anti-bonding orbital serves as acceptor. This contributes to the weakening of the C- B bonds, which become longer in W5-C1B than that of the free ligand. In contrast, figure 2-c shows the shape of the charge flow 3, which indicates that the stabilization of -3.7 kcal/mol comes mainly from relaxation of the W(CO)5 fragment. The EDA- NOCV results for W5-C1Tl, which are shown from Figures 2-d to 2-f, are interesting because they give detailed insight into the bonding situation of the tilted bonded thallium complex, which exhibits the shortest W-C bond of the lighter species. Figures 2- d, and 2-e show that the -type interaction have surprising pairs in either the direction of (CO)5W C(TlCp*)2 -donation or (CO)5W C(TlCp*)2 backdonation in the W5-C1Tl complex. The shapes of the -1 and -2 donor fragment of the NOCV pairs of W5-C1Tl suggest that the -donation comes from the C1Tl ligand toward the W(CO)5 fragment. The acceptor fragment 1 of W5-C1Tl looks very similar to the - acceptor fragment of W5-C1B (Figure 2-a) together with the shapes of the -1 donor fragment. However, the shape of the 2/ -2 acceptor fragment of the NOCV pair of W5-C1Tl suggests that -donation comes from the HOMO of C(TlCp*)2, which has - symmetry with respect to the free ligand. The deformation densities 1 and 2, which indicate stabilization of -49.5 and -21.2 kcal/mol, and not only exhibit a significant area of charge donation (yellow area) from the C1Tl fragment toward the W(CO)5 moiety, but also exhibit an area of charge backdonation (blue area) from (CO)5W to C(TlCp*)2. Figure 2-f shows very weak -type orbital interactions in W5-C1Tl, which indicate that the stabilization of 3 = -11.4 kcal/mol comes mainly from typical -back-donation (CO)5W C(TlCp*)2. The bonding analysis in continuously examined by considering the molecular orbitals with the energy levels of the energetically highest lying and orbitals of C1E ligands. Figure 3 shows the shape of the energetically highest-lying occupied orbitals HOMO and HOMO- 1. These MOs of C(BCp*)2 exhibit the shape of a nearly degenerate pair of orbitals that have approximate -symmetry. HOMO and HOMO-1 are strongly delocalized over the whole molecule, and thus, do not resemble lone-pair orbitals. In contrast to this, the highest-lying occupied MOs of C(AlCp*)2 are easily identified as -lone pair (HOMO) and -lone-pair (HOMO-1) orbitals at the carbon atom. This weakly attractive E-C-E interaction leads to the rather acute bonding angle. A similar situation of C(AlCp*)2 is found for the HOMO and HOMO-1 of the heavier homologues C(ECp*)2 (E = Ga to Tl). The main difference is that the HOMO-1 in the heavier species has increased contributions from the -orbitals of the Cp* moieties. Figure 3 also shows the energy levels of the two highest-lying occupied MOs which have - or -symmetry of the ligands C(ECp*)2. The energies of the -orbitals get lower in energy from Al to Ga, and then they increase slightly from Ga to Tl. The -orbitals are lower in energy than the - orbitals and become lower in energy when E changes from Al to Ga, and then nearly do not change in energy. The lower energy of the -lone- pairs is one reason for the change to tilted bonding VJC, 55(5), 2017 Nguyen Thi Ai Nhung 566 of the C(ECp*)2 ligands. We realize that -donation [(CO)5W {C(ECp*)2}] in the latter complexes takes place from the -lone-pair orbitals of the ligands C(ECp*)2, which have pure -character. There is some s/p hybridization at the carbon donor atom in the complexes that becomes smaller when the bending angle, , becomes more acute. The carbodiylides C(ECp*)2 have two lone-pair orbitals, but they can use their -lone-pair electrons for donor-acceptor interactions in the side-on complexes. The increase in the donation (CO)5W C(ECp*)2, which is manifested in the calculated values for E and electrostatic attraction Eelstat provides a rationale for the stronger bonding of the ligands C(ECp*)2 when E becomes heavier. Figure 2: Most important NOCV pairs of orbitals -k, k with their eigenvalues – k, k, which is given in parentheses, and the associated deformation densities, k, and orbital stabilization energies, E, for the complex W5-C1B and W5-C1Tl. The charge flow in the deformation densities is from the yellow blue region. (a) -NOCV of W5-C1B; (b) and (c) -NOCVs of W5-C1B. (d) and (e) -NOCVs of W5-C1Tl; (f) -NOCV of W5-C1Tl. Energy values in kcal/mol Figure 3: Plot of the energy levels of the energetically highest lying and orbitals of free ligands C1E 4. CONCLUSION The calculated structures of the W5-C1E complexes show that ligands C1E are bonded in an arrangement that is tilted with respect to the metal fragment W(CO)5. The theoretical calculation of BDEs suggests that the bond strength of complexes increases from the boron complex W5-C1B to the strongest bonded thallium adduct W5-C1Tl. Analysis of the bonding situation reveals that the (CO)5W C(BCp*)2 donation in W5-C1B comes from the -lone-pair orbital of C(BCp*)2, while the donation (CO)5W C(ECp*)2 in the strongly tilted bonding complexes where E = Al to Tl comes from the -lone-pair orbital of the carbodiylides C(ECp*)2. The EDA-NOCV results suggest that the trend of the W-C bond strength W5-C1B < W5- C1Al < W5-C1Ga < W5-C1In < W5-C1Tl comes from the increase in (CO)5W C(ECp*)2 donation VJC, 55(5), 2017 Structure and chemical bond of carbodiylide 567 and from lower preparation energies than that of W5-C1B and the carbodiylides ligands C(ECp*)2 in the complexes W5-C1E are strong -donors and weak -donors. Acknowledgements. Nguyen Thi Ai Nhung especially thanks Prof. Dr. Gernot Frenking and Dr. Ralf Tonner for their support and helpful discussion. The program of the studies was run via the Erwin cluster operated by Reuti at the Philipps-Universität Marburg-Germany. REFERENCES 1. D. Weiss et al. [(dcpe)Pt(ECp*)2] (E = Al, Ga): Synthesis, Structure, and Bonding Situation of the First Aluminum(I) and Gallium(I) Complexes of Phosphine-Substituted Transition Metal Centers, Organometallics. 19(22), 4583-4588 (2000). 2. Y. Bunno, N. Murakami, Y. Suzuki, M. Kanai, T. Yoshino, S. Matsunaga, Cp*Co III -Catalyzed Dehydrative C−H Allylation of 6-Arylpurines and Aromatic Amides Using Allyl Alcohols in Fluorinated Alcohols, Org. Lett., 18 (9), 2216-2219 (2016). 3. K. Fujimura, M. Ouchi, J. Tsujita, M. Sawamoto. Cationic Cp*–Ruthenium Catalysts for Metal- Catalyzed Living Radical Polymerization: Cocatalyst- Independent Catalysis Tuned by Counteranion, Macromolecules, 49(8), 2962-2970 (2016). 4. S. Figueiredo et al. Bis(pyrazolyl) methanetetracarbonyl-molybdenum(0) as precursor to a molybdenum(VI) catalyst for olefin epoxidation, J. Organomet. Chem. 723, 56-64 (2013). 5. J.Weiss, D. Stetzkamp, B. Nuber, R. A. Fischer, C. Boehme, G. Frenking, [(η5-C5Me5)Al-Fe(CO)4] Synthesis, Structure, and Bonding, Angew. Chem. Int. Ed. Engl., 36(12), 70-72 (1997). 6. J. Su, X.-W. Li, R. C. Crittendon, C. F. Campana, G. H. Robinson, Experimental Confirmation of an Iron−Gallium Multiple Bond: Synthesis, Structure, and Bonding of a Ferrogallyne, Organometallics, 16(21), 4511-4513 (1997). 7. A. H. Cowley, V. Lomeli, A. Voigt, Synthesis and Characterization of a Terminal Borylene (Boranediyl) Complex, J. Am. Chem. Soc., 120(25), 6401-6402 (1998). 8. T. A. N. Nguyen et al. Structures and Bonding Situation of Iron Complexes of Group-13 Half- Sandwich ECp* (E = B to Tl) Based on DFT Calculations, Z. Anorg. Allg. Chem., 642(8), 609-617 (2016). 9. T. Cadenbach, T. Bollermann, C. Gemel, I. Fernndez, M. von Hopffgarten, G. Frenking, R. Fischer. Twelve One-Electron Ligands Coordinating One Metal Center: Structure and Bonding of [Mo(ZnCH3)9(ZnCp*)3], Angew. Chem. Int. Ed., 47(47), 9150-9154 (2008). 10. C. Boehme, J. Uddin, G. Frenking. Chemical bonding in mononuclear transition metal complexes with Group 13 diyl ligands ER (E=B-Tl): Part X: Theoretical studies of inorganic compounds, Coord. Chem. Rev., 197(1), 249-276 (2000). 11. J. Uddin, G. Frenking. Energy Analysis of Metal- Ligand Bonding in Transition Metal Complexes with Terminal Group-13 Diyl Ligands (CO)4Fe-ER, Fe(EMe)5 and Ni(EMe)4(E = B−Tl; R = Cp, N(SiH3)2, Ph, Me) Reveals Significant π Bonding in Homoleptical Molecules, J. Am. Chem. Soc., 123(8), 1683-1693 (2001) . 12. J. Uddin, C. Boehme, G. Frenking. Nature of the Chemical Bond between a Transition Metal and a Group-13 Element: Structure and Bonding of Transition Metal Complexes with Terminal Group-13 Diyl Ligands ER (E = B to Tl; R = Cp, N(SiH3)2, Ph, Me), Organometallics., 19(4), 571-582 (2000). 13. J.Weiss, D. Stetzkamp, B. Nuber, R. A. Fischer, C. Boehme, G. Frenking. [(η5-C5Me5)Al-Fe(CO)4]- Synthesis, Structure, and Bonding, Angew. Chem. Int. Ed. Engl., 36(12), 70-72 (1997). 14. R. Kinjo, B. Donnadieu, M. A. Celik, G. Frenking, G. Bertrand. Synthesis and Characterization of a Neutral Tricoordinate Organoboron Isoelectronic with Amines, Science, 333(6042), 610-613 (2011). 15. Gaussian 03, Revision D.01. M. J. Frisch, J. A. Pople, Gaussian Inc., Wallingford, CT, 2004. 16. R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kölmel. Electronic structure calculations on workstation computers: The program system turbomole, Chem. Phys. Lett., 162(3), 165-169 (1989). 17. A. D. Becke. Density-functional exchange-energy approximation with correct asymptotic behavior, Phys. Rev. A, 38(6), 3098-3100 (1988). 18. J. P. Perdew. Density-functional approximation for the correlation energy of the inhomogeneous electron gas, Phys. Rev. B, 33(12), 8822-8824 (1986). 19. A. Schäfer, H. Horn, R. Ahlrichs. Fully optimized contracted Gaussian basis sets for atoms Li to Kr, J. Chem. Phys., 97(4), 2571 (1992). 20. F. Weigend, R. Ahlrichs. 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., 7(18), 3297 (2005). 21. B. Metz, H. Stoll, M. Dolg. Small-core multiconfiguration-Dirac–Hartree–Fock-adjusted pseudopotentials for post-d main group elements: Application to PbH and PbO, J. Chem. Phys., 113(7), 2563 (2000). 22. A. E. Reed, L. A. Curtiss, F. Weinhold. Intermolecular interactions from a natural bond orbital, donor-acceptor viewpoint, Chem. Rev., 88(6), 899-926 (1988). 23. K. Wiberg. Application of the pople-santry-segal CNDO method to the cyclopropylcarbinyl and cyclobutyl cation and to bicyclobutane, Tetrahedron, 24(3), 1083-1096 (1968). 24. G. te Velde et al. Chemistry with ADF, J. Comput. Chem., 22(9), 931-967 (2001). VJC, 55(5), 2017 Nguyen Thi Ai Nhung 568 25. J. G. Snijders, E. J. Baerends, P. Vernoojs. Roothaan- Hartree-Fock-Slater atomic wave functions: Single- zeta, double-zeta, and extended Slater-type basis sets for87Fr-103Lr, At. Data. Nucl. Data Tables, 26(6), 483- 509 (1982). 26. J. Krijn, E. J. Baerends. Fit Functions in the HFS- Method, Internal Report, Vrije Universiteit Amsterdam, The Netherlands (1984). 27. E. van Lenthe, A. Ehlers, E. J. Baerends. Geometry optimizations in the zero order regular approximation for relativistic effects, J. Chem. Phys., 110(18), 8943 (1999). 28. T. Ziegler, A. Rauk. CO, CS, N2, PF3 and CNCH3 as donors and acceptors. A theoretical study by the Hartree-Fock-Slater transition-state method, Inorg. Chem., 18(7), 1755-1759 (1979). 29. A. Michalak, M. Mitoraij, T. Ziegler. A Combined Charge and Energy Decomposition Scheme for Bond Analysis, J. Chem. Theory. Comput., 5(4), 962-975 (2009). 30. R. Tonner, G. Frenking. Divalent Carbon(0) Chemistry, Part 1: Parent Compounds, Chem. Eur. J., 14(11), 3260-3272 (2008). 31. P. Hassanzadeh, L. Andrews. Reactions of pulsed laser evaporated boron atoms with methane. 1. Synthesis and characterization of a novel molecule with carbon-boron double bonds: HBCBH, J. Am. Chem. Soc., 114(23), 9239-9240 (1992). 32. S. Klein and G. Frenking. Carbodiylides C(ECp*)2 (E = B – Tl): Another Class of Theoretically Predicted Divalent Carbon(0) Compounds, Angew. Chem. Int. Ed., 49(39), 7106-7110 (2010). 33. W. A. Herrmann, M. Denk, J. Behm,W. Scherer, F.-R. Klingan, H. Bock, B. Solouki, M. Wagner. Stable Cyclic Germanediyls (“Cyclogermylenes”): Synthesis, Structure, Metal Complexes, and Thermolyses, Angew. Chem. Int. Ed. Engl., 31(11), 1485-1488 (1992). Corresponding author: Nguyen Thi Ai Nhung Hue University of Sciences, Hue University No. 77, Nguyen Hue, Hue City, Thua Thien Hue Province E-mail: nguyenainhung.hueuni@gmail.com.