Interstitial Bismuth Atoms in Icosahedral Rhodium Cages: Syntheses

May 18, 2017 - (3, 4) Moreover, the capability of rhodium compounds to interstitially host light p elements, such as C and N, gave rise to the fruitfu...
0 downloads 11 Views 3MB Size
Download PDF
Article pubs.acs.org/IC

Interstitial Bismuth Atoms in Icosahedral Rhodium Cages: Syntheses, Characterizations, and Molecular Structures of the [Bi@Rh12(CO)27]3−, [(Bi@Rh12(CO)26)2Bi]5−, [Bi@Rh14(CO)27Bi2]3−, and [Bi@Rh17(CO)33Bi2]4− Carbonyl Clusters Cristina Femoni,*,† Guido Bussoli,† Iacopo Ciabatti,† Marco Ermini,† Mohammad Hayatifar,‡ Maria C. Iapalucci,† Silvia Ruggieri,† and Stefano Zacchini† Department of Industrial Chemistry “Toso Montanari”, University of Bologna, Viale del Risorgimento, 4, 40136 Bologna, Italy S Supporting Information *

ABSTRACT: The reaction of [Rh7(CO)16]3− with BiCl3 under N2 and at room temperature results in the formation of the new heterometallic [Bi@Rh12(CO)27]3− cluster in high yields. Further controlled addition of BiCl3 leads first to the formation of the dimeric [(Bi@Rh12(CO)26)2Bi]5− and the closo-[Bi@Rh14(CO)27Bi2]3− species in low yields, and finally, to the [Bi@Rh17(CO)33Bi2]4− cluster. All clusters were spectroscopically characterized by IR and electrospray ionization mass spectrometry, and their molecular structures were fully determined by Xray diffraction studies. Notably, they represent the first examples of Bi atoms interstitially lodged in metallic cages that, in this specific case, are all based on an icosahedral geometry. Moreover, [Bi@Rh14(CO)27Bi2]3− forms an exceptional network of infinite zigzag chains in the solid state, thanks to intermolecular Bi−Bi distances.

■

INTRODUCTION Metal clusters are fascinating compounds demonstrating the versatility of metals in forming the most various architectures but still remaining at a molecular level. To quote F. A. Cotton, “there are no simple structural prototypes to which the majority of clusters could be a priori assigned”.1 However, once their compositions and structures are established, they can provide some help in understanding that wide area of chemistry that lies between molecular species and expanded solid-state structures, passing through metal nanoparticles and colloids.2 The chemistry of homometallic carbonyl clusters of rhodium has been widely investigated over the last four decades.3,4 Moreover, the capability of rhodium compounds to interstitially host light p elements, such as C and N, gave rise to the fruitful chemistry of carbide and nitride Rh clusters.5 However, less has been achieved when moving to the third-row p elements, and only a few examples are reported, namely, [Rh9P(CO)21]2−,6 [Rh10P(CO)22]3−,7 [Rh10S(CO)22]2−,8 and [Rh17S2(CO)32]3−.9 Similarly, with heavier and bulkier post-transition elements the species known so far are [Rh9(μ8-As)(CO)21]2−,10 [Rh10(μ8As)(CO)22]3−,10 [Sb@Rh12(CO)27]3− 11 (all by Vidal and coworkers), and [Sn@Rh12(CO)27]4−.12 The former two, prepared under high pressure of CO (250 atm) and at high temperature (140−160 °C), adopt As-centered square-antiprismatic frameworks that are mono- and bicapped, respectively. The latter two species, of which only the second one was synthesized in mild conditions, show identical E-centered (E = Sb, Sn) icosahedral frameworks, stabilized by the same number © 2017 American Chemical Society

of carbonyl groups, and possess 170 cluster valence electrons (CVEs) in line with the Wade−Mingos rule.13 Notably, from the [Sn@Rh12(CO)27]4− derivative, the electronically unsaturated [Sn@Rh12(CO)26]4− and [Sn@Rh12(CO)25]4− can be obtained.14 Not only they represent the first examples of icosahedral clusters with less than 170 CVEs, likely thanks to the extra stability imparted by the interstitial heteroatom, but they could be relevant in the field of CO-releasing molecules (CORMs).15 Finally, most recently, two additional Rh−Sb compounds derived from the parent [Sb@Rh12(CO)27]3− were synthesized: the unsaturated [Sb@Rh12(CO)24]4− and the dimeric [{Sb@Rh12Sb(CO)25}2Rh(CO)2PPh3]7− species.16 To widen the chemistry of rhodium compounds containing post-transition elements, we investigated the chemistry of heterometallic Rh−Bi carbonyl clusters. Notably, Bi is an active heterogeneous catalyst in oxidation reaction,17 and bismuth nanoparticles mixed with other active metals can be prepared and employed in such reactions by using Bi-containing metal compounds as precursors.18 In the field of heterometallic carbonyl clusters, bismuth has received growing attention over the past 30 years, and several scattered examples of transition-metal carbonyl compounds containing bismuth are known in the literature, albeit none with rhodium. Among them it is worth mentioning BiIr3(CO)9,19 [Bi 4 Fe 4 (CO) 1 3 ] 2 − , 2 0 [ Co 1 4 B i 8 (CO) 2 0 ] 2 − , 2 1 [ Ni x @ Received: February 17, 2017 Published: May 18, 2017 6343

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry Bi6Ni6(CO)8]4−,22 Ir5Bi3(CO)10,18 [Bi12Ni7(CO)4]4−,23 and Bi3Ir6(CO)13.24 Their structures vary from (capped) tetrahedral, to (capped) octahedral, to centered cubic, and to icosahedral geometries. In all those species Bi contributes to the polyhedron forming the cluster framework, alongside the metal atoms, and it is never lodged inside the cavities. Their syntheses largely differ, as they were prepared in various pressure (from atmospheric to 400 bar) and temperature (from room to 160 °C) conditions and starting from completely different reactants. Here we report the syntheses, characterizations, and crystal structures of the first examples of rhodium carbonyl clusters where bismuth atoms are present not only on the surface but also inside icosahedral metal cavities.

■

and [Rh12Bi(CO)25]2− ions. ESI-MS analysis of 3 presents main signals at 1566, 1341, 1072, 922, and 851 m/z, due to the {[Rh14Bi3(CO)26][NEt4]2[NMe4]}2−, [Rh14Bi3(CO)22]2−, {[Rh14Bi3(CO)27][NEt4]3}3−, [Rh14Bi3(CO)25]3−, and [Rh14Bi2(CO)25]3− ions, respectively. Synthesis of [Bi@Rh17(CO)33Bi2]4− (4). An acetonitrile suspension of BiCl3 (8 mL) was added to a solution of 1[NEt4]3 in the same solvent (15 mL), under either CO or N2 atmosphere, reaching a 1.5:1 molar ratio, respectively. After ∼5 h the brown mother solution was dried under vacuum, and the solid was washed with water (10 mL), methanol (10 mL), and THF. Cluster 4 was extracted in acetone, and black crystals of 4[NEt4]4 were obtained by layering hexane onto the acetone solution (yield ≈ 30% based on Rh). Cluster 4 is well-soluble in acetone, acetonitrile, and DMF and stable in water. Its IR spectrum presents νCO absorptions in CH3CN at 2002(sh), 1996(s), 1990(sh), 1872(ms), and 1870 (ms) cm−1. ESI-MS analysis results of 4 show characteristic signals at 1673, 1566, 1134, and 1091 m/z, assigned to the {[Rh17Bi3(CO)30][NEt4]}2−, [Rh17Bi3(CO)27]2−, {[Rh17Bi3(CO)32][NEt4]}3−, and [Rh17Bi3(CO)32]3− ions, respectively. Molecular Structure Determination. X-ray single-crystal diffraction experiments were performed at 100 K on a Bruker Apex II diffractometer, equipped with a CCD Detector, by using Mo Kα radiation. Data were corrected for Lorentz polarization and absorption effects (empirical absorption correction SADABS).26 Structures were solved by direct methods and refined by full-matrix least-squares based on all data using F2.27 Hydrogen atoms were fixed at calculated positions and refined by a riding model. All non-hydrogen atoms were refined with anisotropic displacement parameters, including disordered atoms. In refining the crystal structures of 2, 3, and 4 some distance and anisotropic parameter restraints were applied. In case of 2 restraints were mostly applied to [NEt4]+ and acetone molecules, as they presented a high degree of positional disorder. Crystals of 3 had a very low diffraction power, which also accounts for a high Rint. Attempts made with other crystalline samples did not give better results. For this reason, many restraints were applied to carbonyl distances, besides the cation and acetonitrile molecules, to obtain a better structural model. Cluster 4 presents four Rh atoms disordered over three positions, with a consequent heavy repercussion on the carbonyl ligands, many of which are also disordered. For this reason, a high number of distance and anisotropic parameter restraints was required. Structure drawings were made with SCHAKAL99.28

EXPERIMENTAL SECTION

General Procedures. All reactions and compounds were handled using standard Schlenk technique and under nitrogen (or carbon monoxide, when stated) atmosphere. Solvents were dried and degassed before use; tetrahydrofuran (THF) was dehydrated with Na-benzophenone and distilled under nitrogen. Ammonium salts and BiCl3 reagents were commercial products. The [Rh7(CO)16]3− cluster precursor was prepared according to literature.25 IR spectra were recorded on a PerkinElmer Spectrum One interferometer in CaF2 cells. Positive/negative-ion mass spectra were recorded in CH3CN solutions on a Waters Micromass ZQ 4000 by using electrospray (ES) ionization. Experimental conditions: 2.56 kV ES-probe voltage, 10 V cone potential, 250 L h−1 flow of N2 spray-gas, incoming-solution flow 20 μL min−1. Synthesis of [Bi@Rh12(CO)27]3− (1). An acetonitrile suspension of BiCl3 (0.39 mmol) was slowly added to a solution of [Rh7(CO)16][NEt4]3 (0.51 mmol) in the same solvent, under N2 atmosphere, and in a 3:4 molar ratio, respectively. After ∼6 h the resulting brown solution was dried under vacuum, and the solid was washed with water (10 mL), ethanol (10 mL), and THF (20 mL). The residue was extracted with ∼20 mL of acetonitrile, and di-isopropyl ether was layered on top, to obtain black crystals of 1[NEt4]3 (yield ≈ 70% based on Rh). Cluster 1, with both [NEt4]+ and [NMe4]+ counterions, is soluble in acetone and acetonitrile and stable, but not soluble, in water. Its IR spectrum registered in CH3CN shows νCO absorptions at 1992(s), 1857(m), 1829(m), and 1774(sh) cm−1. ESI-MS analysis shows characteristic signals at 705, 855, 1095, 1146, and 1240 m/z (accompanied by satellite signals due to CO losses), attributable to the [Rh12Bi(CO)24]3−, {[Rh12Bi(CO)26][NEt4]3}3−, {[Rh12Bi(CO)22][NEt4]}2−, {[Rh12Bi(CO)21][NEt 4] 2}2−, and {[Rh12Bi(CO)23][NEt4]3}2− ions, respectively. Synthesis of [(Bi@Rh12(CO)26)2Bi]5− (2) and [Bi@Rh14(CO)27Bi2]3− (3). To an acetonitrile solution of 1[NEt4]3 under CO (or N2) atmosphere (15 mL), an acetonitrile suspension of BiCl3 was added in the same solvent (8 mL), reaching a 1:1 molar ratio. After ∼4 h the brown mother solution was dried under vacuum and dissolved in 10 mL of N,N-dimethylformamide (DMF). A water solution of [NMe4] Br was added to it (30 mL) to perform the cation exchange. After the metathesis the suspension was filtered and washed with water (10 mL), ethanol (10 mL), and THF. The dimeric cluster 2 was extracted in acetone, and, by layering hexane on the solution, black crystals of 2[NEt4][NMe4]4·4(CH3)2CO were obtained (yield ≈ 20% based on Rh). The residue was dried in vacuum and extracted in acetonitrile, and black crystals of 3[NMe4]3·3CH3CN were obtained, by layering di-isopropyl ether onto the acetonitrile solution (yield ≈ 15% based on Rh). Clusters 2 and 3 are well-soluble in acetonitrile and DMF (2 is also soluble in acetone), and are both stable in water. Their IR spectra registered in acetonitrile show νCO absorptions at 1991(s), 1865(ms), and 1828(m) cm−1 for compound 2; 1989(s), 1896(s) and 1863(ms) cm−1 for compound 3. ESI-MS analysis of 2 shows signals at 1190, 1095, and 1072 m/z (accompanied by satellite ones due to CO losses) attributable to the {[Rh12Bi(CO)26]Bi}2−, {[Rh12Bi(CO)22]NEt4}2−,

■

RESULTS AND DISCUSSION Our approach for building Rh−Bi heterometallic clusters exploited the redox condensation method, first reported by Hieber and Schubert,29 but effectively coined and developed by Chini.30 More specifically, we prepared an acetonitrile solution of a Rh cluster precursor, that is, [Rh7(CO)16][NEt4]3, and added a second solution of BiCl3 in the same solvent in an ∼1:0.75 ratio. This way the Rh-cluster precursor is still subject to oxidation but also provides the necessary CO ligands for the reaction to occur, so no carbon monoxide atmosphere was required. In fact, the above reaction was performed at room temperature and under N2 pressure of ∼1.5 bar. The choice of the [NEt4]+ counterion was dictated by the different solubility that the cation imparts to the various final products, making them separable by extraction in solvents of increasing polarity. The BiCl3 solution was prepared in the desired quantity and completely added to the cluster solution, as BiCl3 is only sparingly soluble in CH3CN and actually forms a suspension. After ∼6 h the final IR spectrum registered in solution showed the characteristic carbonyl absorptions of the [Rh(CO)2Cl2]− complex at 2067 (s), 1991 (s) cm−1, which represents an expected byproduct, plus other peaks of an unknown species with a main absorption at 1992 cm−1. The workup of the mother solution consisted in drying it under vacuum, washing 6344

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry

Second of all we investigated its stability at high temperature by refluxing its acetone solution under N2 atmosphere. By monitoring the reaction path via IR, we registered a progressive lowering of the νCO absorptions of the terminal carbonyls from 1992 to ∼1980 cm−1, coherent with a possible loss of some CO ligands and the formation of an unsaturated species. The fact that under CO atmosphere the new cluster returned to the parent compound sustained our conclusion. This behavior is reminiscent of that of the [Sn@Rh12(CO)27]4− cluster, from which the isostructural, unsaturated [Sn@Rh12(CO)26]4− and [Sn@Rh12(CO)25]4− 14 can be thermally prepared, as previously mentioned. Unfortunately, any attempt to obtain suitable crystals for X-ray diffraction studies failed, and this species is still under investigation. Finally, we added further BiCl3 salt to 1[NEt4]3, basically treating the cluster as a starting material for a new reaction. The first stoichiometric ratio we investigated was 1:1, and the reaction conditions paralleled the ones used for the synthesis of the starting material, such as room temperature, CH3CN as solvent, but under CO atmosphere (∼1.5 bar). The reaction path was monitored via IR spectroscopy, and the targeted stoichiometric ratio corresponded to the complete disappearance of the characteristic νCO absorptions of the starting cluster, replaced by others not related to any known species. The final mother solution was dried under vacuum and dissolved in DMF to perform a metathesis from [NEt4]+ to [NMe4]+. Subsequently, the mixture was washed with water to eliminate organic salts and dried under vacuum, and the residue was treated with THF, which extracted large quantities of the [Rh(CO)2Cl2]− complex, together with traces of an unknown species with νCO absorptions at 1991(s), 1865(ms), and 1829(m)m cm−1. This new compound was then completely extracted in acetone. The last solid residue was dried under vacuum and finally extracted with acetonitrile, and its IR spectrum revealed a yet unknown product with νCO absorptions at 1989(s) and 1864(ms) cm−1. Crystallizations were performed on both acetone and acetonitrile solutions, by layering hexane on the former and di-isopropyl ether on the latter. In both cases we obtained crystals suitable for X-ray analyses, and the two compounds were identified as [(Bi@ Rh 12 (CO) 26 ) 2 Bi][NEt 4 ][NMe 4 ] 4 ·4(CH 3 ) 2 CO and [Bi@ Rh14(CO)27Bi2][NMe4]3·3CH3CN. The residual presence of some [NEt4]+ cation in 2 showed that the metathesis was not complete. We repeated the above reaction in the same conditions but under N2 atmosphere and obtained the same results. Unfortunately, in both cases the product yields turned out to be extremely low and prevented us from performing any reactivity tests. Clusters 2 and 3 were also characterized by ESI-MS. In the case of 2 the molecular-ion signal is not present, indicating some instability in the experimental conditions; however, signals due to its partial fragmentation are well-observed. In the case of 3, the positive-ion region showed the presence of some residual [NEt4]+ in solution. Furthermore, its more compact structure makes it more stable in the experimental conditions (except for some CO loss), and the signal at 1566 m/z is attributable to the {[Rh14Bi3(CO)26][NEt4]2[NMe4]}2− molecular ion. Characteristic signals of clusters 2 and 3 are specified in the Experimental Section, while their ESI-MS spectra are reported in Figures 2S and 3S in the Supporting Information. Encouraged by these positive results, we performed a second reaction of 1[NEt4]3 with BiCl3, in the same conditions described above, but with a 1:1.5 stoichiometric ratio. By

the solid residue with water, ethanol, and THF to eliminate the above-mentioned byproduct, and in extracting the unknown species with acetonitrile. Some residual black precipitate, not soluble in any solvent, remained at the end of the workup. The crystallization of the acetonitrile solution was achieved by layering on di-isopropyl ether. After a few days beautifully shaped black crystals of [Bi@Rh12(CO)27][NEt4]3 were obtained, as determined by X-ray diffraction. The cluster was later crystallized also in acetone with [NMe4]+ as counterion, resulting in [Bi@Rh12(CO)27][NMe4]3·2(CH3)2CO. Considering the applied stoichiometry ratio and the isolated products, we can confidently assume that the following reaction occurred: 4[Rh 7(CO)16 ]3 − + 3BiCl3 → 2[Bi@Rh12(CO)27 ]3 − + 3Cl− + +3[Rh(CO)2 Cl 2]− + Bi + Rh + 4CO

The formation of metallic species (Rh and Bi in this case) is expected and quite common in these kinds of reactions, that is, where Rh is oxidized to various oxidation states by Bi, which, in turn, is reduced to zero. More specifically, the starting Rh atoms, all formally in a −3/7 oxidation state, were oxidized to +1 (in [Rh(CO)2Cl2]−), zero (in metallic Rh), and −3/12 (in the cluster product). On the other side Bi was reduced from +3 to zero. This redox behavior is consistent with what can be expected by relating the standard reduction potentials of the [Rh7(CO)16]3−/[Rh7(CO)16]2− and Bi3+/Bi couples (−0.04 and 0.308 V, respectively). More in details, it was experimentally established in the past 31 that whenever [Rh7(CO)16]3− is treated with aquo-complexes featuring E°(Mn+/M) < ca. −0.20 V, the Rh cluster is oxidized to an homometallic species because of the intrinsic Brønsted acidity of the metal salt, with the concurred reduction of H+ to H2. Conversely, when the standard potential of the couple Mn+/M is higher than −0.20 V, irrespective of the pKa of the salt, it is Mn+ that oxides the starting [Rh7(CO)16]3− and that, in turn, is formally reduced to M(0) and becomes part of the heterometallic cluster product. The synthesis of [Bi@ Rh12(CO)27]3− is in line with the above findings. Cluster 1 in its [NEt4]+ salt was also characterized by ESIMS, showing minor fragmentations in the experimental conditions mainly due to CO loss. Actually, the loss of ligands is essential to correctly assign the negative charge to the cluster within the peaks, which is hardly ever the original one. As a matter of fact, in the ESI-MS analysis 1 was detected as bi- and trianion, holding up to three counterion molecules. The main peaks at 705, 855, 1095, 1146, and 1240 m/z can be assigned to {[Rh12Bi(CO)24]}3−, {[Rh12Bi(CO)26][NEt4]3}3−, {[Rh12Bi(CO)22][NEt4]}2−, {[Rh12Bi(CO)21][NEt4]2}2−, and {[Rh12Bi(CO)23][NEt4]3}2− ions. The other peaks are associated with their derived species but with different number of CO ligands (see Figure 1S in the Supporting Information section). The smooth and high-yield synthesis of 1 (∼70% based on Rh) allowed us to have fair amounts of sample at our disposal to extend the study of its reactivity in different conditions and various agents. First of all we tested its stability under CO atmosphere (∼1.5 bar). The cluster proved to be completely stable, even for a prolonged period of time (up to one week), as inferred by IR spectroscopy. 6345

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry monitoring the reaction path it was possible to see the νCO absorptions of the above-described new compounds disappearing, in favor of another species with unknown signals. The workup of the mother solution was performed as previously described but skipping the metathesis of [NMe4]+ with [NEt4]+, and, apart from large quantities of [Rh(CO)2Cl2]−, a new compound was isolated in acetone. Nothing was left to be extracted in CH3CN. By layering hexane onto the acetone solution we were able to obtain single crystals suitable for X-ray diffraction analysis, identifying the new compound as [Bi@ Rh17(CO)33Bi2][NEt4]4. We repeated the reaction between 1[NEt4]3 and BiCl3 in a 1:1.5 stoichiometric ratio but under N2 atmosphere, by keeping all the other parameters unchanged, and got the same results. The ESI-MS spectrum, registered on dissolved crystals of 4 and reported in Figure 4S in the Supporting Information, presents intense characteristic signals due to the molecular ion that withstood loss of some CO ligands and retains up to one counterion molecule (see details in the Experimental Section).

not significantly different from the ones with [NEt4]+ in the lattice. The comparison of the Rh−E bond distances (E = Sn, Sb, Bi) of 1 with other icosahedral heterometallic Rh clusters of similar geometry12,16 is coherent with the subsequent swelling of the metal cage on going from the smaller Sb atom to Sn and finally to the larger Bi atom. As a matter of fact, the average Rh−E bond lengths for [Sb@Rh 12 (CO) 27 ] 3− , [Sn@ Rh12(CO)27]4−, and [Bi@Rh12(CO)27]3− are 2.8227, 2.8298, and 2.8593 Å, respectively. As far as the Rh−Rh distances are concerned, both the average value (2.9832 Å) and the longest contact (3.470(6) Å) are also higher in the case of cluster 1 with respect to the analogous with Sb and Sn, which present average Rh−Rh distances of 2.9472 and 2.9405 Å and extra longer distances of 3.368(1) and 3.308(4) Å, respectively. The icosahedral arrangement is not new for clusters of such nuclearity32 and well beyond, and it is observed not only with Rh but also in Ni, Pd, Au, and Ag clusters. Among them it is worth mentioning [Ni 12M(CO)22 ]2− (M = Ge, Sn), 33 Pd145(CO)x(PEt3)30,34 and the most recent Pd55L12(CO)20,35 [M(AuPMe3)11(AuCl)]3+ (M = Pt, Pd, Ni),36 [Au25(SCH2CH2Ph)18]−,37 Au38(SC2H4Ph)24,38 [Au60Se2(Ph3P)10(SeR)15]+,39 and [Ag44(p-MBA)30]4− (pMBA = p-mercaptobenzoic acid).40 Notably, the latter is the only one having an empty icosahedral core. The molecular structure of 2 is illustrated in Figure 2. It consists of two icosahedral {Bi@Rh12(CO)26} units connected

■

CRYSTAL STRUCTURES The molecular structure of 1 is illustrated in Figure 1. As 1[NEt4]3 crystallized in the centric monoclinic C2/c space

Figure 1. Molecular structure of [Bi@Rh12(CO)27]3−, 1. Rh is depicted in blue; Bi is in pink, C is in gray, and O is in red.

group, the asymmetric unit contains half a cluster, with the Bi atom laying on a twofold axis, and one and a half molecule of cation. The clusters and the cations are arranged in an ionic fashion in the solid state, with no voids left for solvent molecules, unlike in 2, 3, and 4. The molecular structure of 1 consists of an icosahedral metal skeleton of Rh atoms centered by the unique Bi. The metal polyhedron is to some extent distorted, with an average Rh−Rh distance of 2.9832 Å, but having the shortest bonding contact at 2.8119(6) Å and the longest at 3.0792(6) Å. Furthermore, there are three extra, longer Rh−Rh distances at 3.388(6) and 3.470(6) Å. The Rh− Bi bond lengths span between 2.7358(4) and 2.9781(4) Å, with an average of 2.8593 Å. The metal frame of 1 is stabilized by 27 carbonyl ligands, of which 12 are terminally bonded (one for each vertex) and 15 are edge-bridged. This ligand distribution is shared with the analogous [Sb@Rh12(CO)27]3− derivative and one isomer of [Sn@Rh12(CO)27]4−, while the second isomer presents 13 terminal and 14 edge-bridging COs. Notably, in all the above clusters there are no edge-bridging carbonyls connecting Rh atoms separated by bond distances greater than 3 Å. As stated earlier, cluster 1 was also structurally characterized with [NMe4]+ counterion, but its geometrical parameters do

Figure 2. (a) Molecular structure of [(Bi@Rh12(CO)26)2Bi]5−, 2; (b) metal skeleton of 2. Rh is depicted in blue; Bi is in pink, C is in gray, and O is in red.

via a Bi atom, which shows a tetra-coordination, rotated by ∼90° one with respect to the other. The 26 carbonyl ligands of each unit are divided into two groups: 12 are terminally bonded, and 14 are edge-bridged. The overall, delocalized charge of the cluster is −5, counterbalanced by four [NMe4]+ and one [NEt4]+ cations. In the unit cell there are also positionally disordered acetone molecules, four per cluster. As 2[NEt4][NMe4]4·4(CH3)2CO crystallizes in the tetragonal 6346

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry P4̅n2 space group, the asymmetric unit contains half a {Rh12Bi(CO)26} unit plus the bridging Bi atom (with both Bi atoms lying on a fourfold axis), one [NMe4]+ cation and a quarter of the [NEt4]+ one, and finally one acetone molecule. The molecular structure of the single {Rh12Bi(CO)26} unit coincides with the one of 1, both in terms of geometry and metal contacts. As a matter of fact, the Rh−Rh bond lengths vary from 2.8050(17) to 3.188(3) Å, with an average of 2.9703 Å, only slightly shorter than in 1. There is also one longer Rh−Rh contact per unit of 3.368(0) Å, indicating the deviation from the ideal Ih symmetry. The Rh−Bi bond distances within the icosahedra vary from 2.7258(10) to 2.9697(11) Å, with an average of 2.8521 Å (similarly to 1), while those involving the bridging Bi atom are all equivalent and much shorter, 2.6678(11) Å. With respect to 1, the missing CO ligand in the {Bi@Rh12(CO)26} monomeric unit is replaced by the coordination with the bridging Bi atom. The molecular structure of 3 is shown in Figure 3. Again, it is based on an icosahedral geometry with the addition of two Bi−

The [Bi@Rh14(CO)27Bi2]3− anion differs from all the others, because it shows a very interesting packing in the solid state. As a matter of fact, due to exceptionally short intermolecular Bi− Bi interactions of 3.591(2) Å and 3.774(2) Å involving the two Bi capping atoms in each cluster, the network results in infinite zigzag chains spaced by cations and solvent molecules, making [Bi@Rh14(CO)27Bi2]3− a molecular wire in the solid state. The overall architecture of 3 is illustrated in Figure 4.

Figure 3. (a) Molecular structure of [Bi@Rh14(CO)27Bi2]3−, 3; (b) metal skeleton of 3. Rh is depicted in blue; Bi is in pink, C is in gray, and O is in red.

Figure 4. Molecular packing of 3 in the solid state along the c axis, showing infinite zigzag chains. Rh is depicted in blue; Bi is in pink. Carbonyl groups, cations, and solvent molecules were omitted for sake of clarity.

Rh capping fragments on two different sides of the polyhedron. The metal frame is stabilized by 27 carbonyl ligands, of which 12 are terminally bonded, and 15 are edge-bridged. The 3[NMe4]3·3CH3CN salt crystallizes in the orthorhombic Pna21 space group, and the asymmetric unit contains six cations, six acetonitrile solvent molecules, and two clusters, which only marginally differ in terms of bond lengths. The main icosahedral core is slightly less compact with respect to 1 and 2, being the average Rh−Bi distances of 2.872 and 2.869 Å for the two clusters in the asymmetric unit. Significantly shorter are the bond lengths involving the two Bi capping atoms: the Rh−Bi average are 2.784 and 2.724 Å, for the first cluster, and 2.783 and 2.731 Å for the second one. The Rh−Rh distances within the icosahedral unit vary from 2.754(4) to 3.117(4) Å, for the first cluster, and from 2.761(4) to 3.170(5) for the second one (average values of 2.968 and 2.962 Å, respectively, slightly shorter than in 1 and 2). The Rh−Rh distance of the capping rhodium atoms are shorter, being their average 2.749 and 2.743 Å for the two independent clusters. Four longer nonbonding Rh−Rh distances are observed where the capping Bi−Rh fragments lay, up to 3.404 Å in the first cluster and 3.442 Å in the second one.

Similar long-range intermolecular Bi−Bi interactions are quite rare and, to our knowledge, were described in [Bi4Fe4(CO)13]2−20 (3.978−3.981 Å) and [Bi4Fe5(CO)13Cp2]41 (3.725 Å), making the distance of 3.591(2) Å among the shortest reported in the scientific literature. Furthermore, the steric hindrance due to carbonyl ligands is likely to prevent a further neighboring of the Bi atoms. Note that Bi−Bi bond lengths in crystalline bismuth are in the 3.071−3.516 Å range.42 The molecular structure of 4 and its shell anatomy are shown in Figure 5. The cluster crystallized in the orthorhombic Cmca space group, and its asymmetric unit contains one-quarter of the cluster and one cation molecule. The metal skeleton is again based on a Bi-centered icosahedron of Rh atoms (Figure 5a), symmetrically decorated by two Rh−Bi−Rhx fragments (x = 3/2), presenting a Rh atom in the asymmetric unit with a partial occupation factor of 75%. As a result, there are three Rh atoms disordered over four positions (indicated in yellow in Figure 5b) within the cluster anion, breaking its idealized D3d symmetry. Reasonably, this partial occupation factor has a repercussion on the entire cluster, so five of the eight independent carbonyl 6347

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry

this has five additional atomic orbitals (the d orbitals), taking the whole to 5N + N + N + 1 = 7N + 1 cluster valence molecular orbitals (CVMOs). According to this rule, an icosahedral carbonyl metal cluster (N = 12) should exhibit 85 CVMOs and 170 cluster valence electrons (CVEs). Note that, according to the inclusion principle formulated by D. M. P. Mingos,43 interstitial heteroatoms can be considered as an internal source of electrons; therefore, their valence electrons should be included in the CVEs counting. The [Bi@ Rh12(CO)27]3− cluster (1) perfectly conforms to the above electron counting rule and presents 170 CVEs, given by the 9 × 12 rhodium atoms (108), the 2 × 27 carbonyl ligands (54), the interstitial Bi atom (5), and the negative charge (3). As for cluster 2, because it consists of two icosahedra that do not share any face or vertex, we expect it to have 340 CVEs in total.44 Each [Bi@Rh12(CO)26] unit possesses 165 CVEs, and then 5 + 5 electrons must be added: 5 from the bridging Bi atom and 5 from the negative charge. The dimeric species again conforms to the cluster counting rule. In the case of cluster 3, the rule predicts 194 CVEs, being an icosahedron (170 CVEs) capped by two metal atoms (2 × 12 CVEs). The same electron counting is obtained when considering cluster 3 as a condensed polyhedral molecule formed by the fusion of one icosahedron (170 CVEs) with two surface tetrahedral fragments (60 × 2 CVEs), sharing one triangular face each (−48 × 2 CVEs). Indeed, the total electron counting is the following: 9 × 14 (Rh atoms) + 2 × 27 (CO ligands) + 5 (interstitial Bi atom) + 3 × 2 (surface Bi atoms) + 3 (negative charge), giving a total of 194 CVEs. The only species that does not follow the electron counting rule is cluster 4. In fact, as it can be seen as an icosahedron capped by five metal atoms, 230 CVEs are expected. The same number of valence electrons is obtained when using the condensed polyhedral formalism. However, the actual whole electron counting gives 234 CVEs: 9 × 17 (Rh atoms) + 2 × 33 (CO ligands) + 5 (interstitial Bi atom) + 3 × 2 (surface Bi atoms) + 4 (negative charge). This deviation from the rule is not unprecedented for clusters of similar nuclearity.45 For instance, bimetallic Rh−Ni carbonyl clusters sharing the same nuclearity and metal architecture but with different numbers of valence electrons are known.46 Notably, the PSEPT approach is not suitable for gold clusters, even those closer to [B12H12]2− with 12 ligands like [Pd@Au12(PR3)8Cl4]47 and [M@(AuPMe3)11(AuCl)]3+ (M = Pt, Pd, Ni), for which alternative models can be applied.48,49

Figure 5. [Bi@Rh17(CO)33Bi2]4−, 4: (a) icosahedral kernel; (b) full metal skeleton; (c) molecular structure. Rh is depicted in blue; Bi is in pink, and Rh atoms with 75% occupation factor are in yellow.

groups, as well as the [NMe4]+ counterion molecule, were split in two positions, and the overall structural model presents numerous distance and anisotropic displacement parameter (ADP) restraints. The icosahedron kernel shows Bi−Rh and Rh−Rh bond lengths similar to 1, 2, and 3, being their average 2.9025 and 2.9215 Å, respectively. As shown in Figure 5a, there are also six longer Rh−Rh distances of 3.488 and 3.490 Å. Conversely, the Rh−Rh and Rh−Bi bond distances on the icosahedral surface are significantly shorter (average values 2.7153 and 2.7944 Å) than those in the core. Notably, the Bi− Rh bond lengths involving the disordered Rh atoms are sensibly longer, having an average value of 3.0113 Å. The metal frame of 4 is finally stabilized by 33 carbonyl ligands, of which, notably, only five are terminally bonded, while the remaining 28 are edge-bridged. Crystallographic data for clusters 1 (both as [NEt4]+ and [NMe4]+ salt), 2, 3, and 4 are reported in Table 1, while their most relevant bond lengths are reported in the Supporting Information.

■

CONCLUSIONS In this paper we presented the synthesis and full characterization of four new Rh−Bi carbonyl clusters, featuring unprecedented fully interstitial bismuth atoms in a transitionmetal rhodium cage. As illustrated in Scheme 1, the mother compound 1 originated by the condensation reaction between [Rh7(CO)16]3− and BiCl3, performed in very mild conditions of temperature and pressure. Its high yield and straightforward synthesis made it possible to use it as a starting material for the other carbonyl species 2, 3, and 4, by addition of further Bi3+ salt in similar reaction conditions. All compounds share the same structural motif consisting of 12 rhodium atoms, distributed on an icosahedral cage, hosting a fully interstitial bismuth. This base architecture, which represents the actual metal structure of 1, is then differently decorated by Rh and Bi atoms in 3 and 4, while compound 2 is a dimeric species bridged by a Bi atom. The full encapsulation of Bi in a metal cluster is unprecedented for such a large atom,

■

ELECTRON COUNTING Electron counting in deltahedral clusters derives from the borane chemistry analogy, as suggested by Wade and Mingos within the Polyhedral Skeleton Electron Pair Theory (PSEPT).13 For instance, a closo borane features N + (N + 1) filled molecular orbitals (where N = number of boron atoms), of which N molecular orbitals (MOs) for the localized B−H bonds and N + 1 for the delocalized MOs to sustain the boron cage. When the counting is moved to a transition metal, 6348

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry Table 1. Crystallographic Data for Clusters 1, 2, 3, and 4 1[NEt4]3

1[NMe4]3 ·2(CH3)2CO

2[NEt4][NMe4]4 · 4(CH3)2CO

3[NMe4]3·3CH3CN

4[NEt4]4

C51H60BiN3O27Rh12 2590.92 monoclinic C2/c 20.0343(4) 22.6293(5) 15.8131(3) 90 102.6850(10) 90 6994.1(2) 4 2.461 5.333 4904 1.377−24.999 −23 ≤ h ≤ 23, −26 ≤ k ≤ 26, −18 ≤ l ≤ 18 46 309

C45H48BiN3O29Rh12 2538.76 monoclinic P2(1)/c 12.5792(7) 22.5805(12) 24.8173(13) 90 104.4950(10) 90 6824.8(6) 4 2.471 5.464 4776 1.24−25.00 −14 ≤ h ≤ 14, −26 ≤ k ≤ 26, −29 ≤ l ≤ 29 65 094

C88H92Bi3N5O56Rh24 5212.45 tetragonal P4̅n2 17.4735(13) 17.4735(13) 22.3715(17) 90 90 90 6830.5(9) 2 2.534 6.739 4864 1.48−25.000 −20 ≤ h ≤ 20, −20 ≤ k ≤ 20, −26 ≤ l ≤ 26 62 807

C45H45Bi3N6O27Rh14 3169.55 orthorhombic Pna21 38.119(5) 16.966(2) 22.150(3) 90 90 90 8 8 2.939 10.544 11 616 1.31−24.000 −43 ≤ h ≤ 43, −19 ≤ k ≤ 19, −25 ≤ l ≤ 25 122 815

C65H80Bi3N4O33Rh17 3821.74 orthorhombic Cmca 24.6007(14) 23.7403(13) 15.3839(9) 90 90 90 8984.6(9) 4 2.825 8.957 7104 1.656−25.000 −29 ≤ h ≤ 29, −28 ≤ k ≤ 28, −18 ≤ l ≤ 18 41 787

6141 [R(int) = 0.0509]

12008 [R(int) = 0.1754]

6014 [R(int) = 0.0626]

22470 [R(int) = 0.2462]

4077 [R(int) = 0.0587]

100.0%

100.0%

99.9%

100.0%

100.0%

6141/0/432

12008/120/811

6014/528/457

22470/1686/1711

4077/661/487

1.073 0.0271 0.0698 1.625 and −0.882

0.979 0.0535 0.0957 1.128 and −0.870

1.067 0.0517 0.1263 2.142 and −0.857

0.993 0.0733 0.1621 1.792 and −1.681

1.042 0.0536 0.1522 1.827 and −1.909

compound formula Fw crystal system space group a (Å) b (Å) c (Å) α (deg) β (deg) γ (deg) cell volume (Å3) Z D (g/cm3) μ (mm−1) F(000) θ limits (deg) index ranges reflections collected independent reflections completeness to θ max data/restraints/ parameters goodness-of-fit R1 (I > 2σ(I)) wR2 (all data) largest diff. peak and hole, e Å−3

to clusters 1, 2, and 3. However, the fact that cluster 4 does not follow the above rules is another hint of the complexity of binding in cluster frameworks. Finally, cluster 1 can be added to the small library of related compounds of the type [E@ Rh12(CO)27]n− (E = Sb or Bi, n = 3; E = Sn, n = 4), which are isoelectronic and share very similar synthetic strategy and structural features. However, the chemical reactivity of the Bi and Sn derivatives diverges. As a matter of fact, in case of 1 further addition of oxidant salt leads to significant cluster aggregation, while excess of SnCl2 to [Sn@Rh12(CO)27]4− mainly results in the disruption of the metal cage and partial ligand substitution. As for [Sb@Rh12(CO)27]3−, preliminary tests seem to point toward a complex cluster growth, but further investigation is needed and is currently ongoing.

Scheme 1

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00409.

making the icosahedral geometry one of the most versatile for homo- and heterometallic molecular metal packing. Another remarkable feature is represented by the solid structure of 3, which shows significantly short intermolecular Bi−Bi bond distances that, to our knowledge, are among the shortest reported in the scientific literature. Those close contacts create a polymer-like arrangement in the solid state, with infinite zigzag chains of clusters spaced by counterions and solvent molecules. Electron-counting rules according to the PSEPT seem to apply to nearly all title compounds, more specifically,

ESI-MS spectra and most relevant bond distances for compounds 1−4 (PDF) Accession Codes

CCDC 1530614−1530618 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The 6349

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry

(13) Mingos, D. M. P. Polyhedral skeletal electron pair approach. Acc. Chem. Res. 1984, 17, 311−319. (14) Femoni, C.; Iapalucci, M. C.; Longoni, G.; Tiozzo, C.; Zacchini, S.; Heaton, B. T.; Iggo, J. A.; Zanello, P.; Fedi, S.; Garland, M. V.; Li, C. The loss of CO from [Rh12(μ12-Sn) (CO)27]4−: Synthesis, spectroscopic and structural characterization of the electron-deficient, icosahedral [Rh12(μ12-Sn) (CO)25]4− and [Rh12(μ12-Sn) (CO)26]4− tetra-anions. Dalton Trans. 2009, 2217−2223. (15) García-Gallego, S.; Bernardes, G. J. L. Bernardes CarbonMonoxide-Releasing Molecules for the Delivery of Therapeutic CO In Vivo. Angew. Chem., Int. Ed. 2014, 53, 9712−9721. (16) Femoni, C.; Ciabatti, I.; Iapalucci, M. C.; Ruggieri, S.; Zacchini, S. Alternative synthetic route for the heterometallic CO-releasing [Sb@Rh12(CO)27]3− icosahedral carbonyl cluster and synthesis of its new unsaturated [Sb@Rh12(CO)24]4− and dimeric [{Sb@Rh12Sb(CO)25}2Rh(CO)2PPh3]7− derivatives. Prog. Nat. Sci. 2016, 26, 461− 466. (17) Grasselli, R. K. Site isolation and phase cooperation: Two important concepts in selective oxidation catalysis: A retrospective. Catal. Today 2014, 238, 10−27. (18) Adams, R. D.; Chen, M.; Elpitiya, G.; Potter, M. E.; Raja, R. Iridium- Bismuth Cluster Complexes Yield Bimetallic Nano- Catalysts for the Direct Oxidation of 3-Picoline to Niacin. ACS Catal. 2013, 3, 3106−3110. (19) Kruppa, W.; Bläser, D.; Boese, R.; Schmid, G. Heteronucleare Clustersysteme, XX [1] μ3-Bismutio-cyclo-tris(tricarbonyliridium) (3 Ir-Ir), BiIr3(CO)9 Darstellung und Strukturuntersuehung eines neuartigen Iridiumclusters. Z. Naturforsch., B: J. Chem. Sci. 1982, 37, 209−213. (20) Whitmire, K. H.; Churchill, M. R.; Fettinger, J. C. Synthesis and Crystal Structure of [Et4N]2[Bi4Fe4(CO)13]. Discovery of a Hybrid Zintl-Metal Carbonyl Cluster. J. Am. Chem. Soc. 1985, 107, 1056− 1057. (21) Whitmire, K. H.; Eveland, J. R. Structural characterization of two large bismuth−cobalt carbonyl clusters: (PPN)2[Bi4Co9(CO)8(μCO)8]·2THF and (PPN)2[Bi8Co14(CO)12(μ-CO)8]·1.08THF. J. Chem. Soc., Chem. Commun. 1994, 1335−1336. (22) Goicoechea, J. M.; Hull, M. W.; Sevov, S. C. Heteroatomic Deltahedral Clusters: Synthesis and Structures of closo[Bi 3 Ni 4 (CO) 6 ] 3− , closo-[Bi 4 Ni 4 (CO) 6 ] 2− , the Open Cluster [Bi3Ni6(CO)9]3−, and the Intermetalloid closo[Nix@{Bi6Ni6(CO)8}]4−. J. Am. Chem. Soc. 2007, 129, 7885−7893. (23) Perla, L. G.; Sevov, S. C. [Bi12Ni7(CO)4]4−: Aggregation of Intermetalloid Clusters by Their Thermal Deligation and Oxidation. Inorg. Chem. 2015, 54, 8401−8405. (24) Adams, R. D.; Elpitiya, G. Iridium-bismuth carbonyl cluster complexes. J. Organomet. Chem. 2016, 812, 115−122. (25) Martinengo, S.; Chini, P. Synthesis and characterization of the [Rh6(CO)15]2− and [Rh7(CO)16]3− anions. Gazz. Chim. It. 1972, 102, 344−354. (26) Sheldrick, G. M. SADABS, Program for empirical absorption correction; University of Göttingen: Germany, 1996. (27) Sheldrick, G. M. SHELX97, Program for crystal structure determination; University of Göttingen: Germany, 1997. (28) Keller, E. SCHAKAL99; University of Freiburg: Germany, 1999. (29) Hieber, W. O.; Schubert, E. H. Absorptionsmessungen an Carbonylferrat-Lösungen im Sichtbaren und UV-Gebiet. Z. Anorg. Allg. Chem. 1965, 338, 32−36. (30) Chini, P. Large Metal Carbonyl Clusters (LMCC). J. Organomet. Chem. 1980, 200, 37−61. (31) Collini, D.; De Biani, F. F.; Fedi, S.; Femoni, C.; Kaswalder, F.; Iapalucci, M. C.; Longoni, G.; Tiozzo, C.; Zacchini, S.; Zanello, P. Synthesis and Electrochemistry of New Rh-Centered and Conjuncto Rhodium Carbonyl Clusters. X-ray Structure of [NEt4]3[Rh15(CO)27], [NEt4]3[Rh15(CO)25(MeCN)2].2MeCN, and [NEt4]3[Rh17(CO)37]. Inorg. Chem. 2007, 46, 7971−7981. (32) Zhang, H.; Teo, B. K. Stereochemical and electronic evidence of icosahedricity and polyicosahedricity. Inorg. Chim. Acta 1997, 265, 213−224.

Cambridge Crystallographic Data Centre, 12, Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Cristina Femoni: 0000-0003-4317-6543 Stefano Zacchini: 0000-0003-0739-0518 Present Address ‡

Italiëlei 235, 2000 Antwerpen−Belgium.

Author Contributions †

These authors contributed equally. The manuscript was written through contributions of all authors, who have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank the Univ. of Bologna for the funding (RFO 2016). REFERENCES

(1) Metal Clusters in Chemistry; Braunstein, P., Oro, L. A., Raithby, P. R., Eds.; Wiley-VCH: Weinheim, Germany, 1999. (2) Goicoechea, J. M.; Sevov, S. C. [Zn9Bi11]5−: A Ligand-Free Intermetalloid Cluster. Angew. Chem., Int. Ed. 2006, 45, 5147−5150. (3) Dolzhnikov, D. S.; Iapalucci, M. C.; Longoni, G.; Tiozzo, C.; Zacchini, S.; Femoni, C. New High-Nuclearity Carbonyl and Carbonyl-Substituted Rhodium Clusters and Their Relationships with Polyicosahedral Carbonyl-Substituted Palladium and GoldThiolates. Inorg. Chem. 2012, 51, 11214−11216. (4) Martinengo, S.; Ciani, G.; Sironi, A. Synthesis and X-ray structural characterization of the [Rh22(μ3-CO)7(μ-CO)18(CO)12]4− anion, containing a large close-packed cluster with an ABAC sequence of compact layers. J. Am. Chem. Soc. 1980, 102, 7564−7565. (5) Hughes, A. K.; Wade, K. Metal−metal and metal−ligand bond strengths in metal carbonyl clusters. Coord. Chem. Rev. 2000, 197, 191−229. (6) Vidal, J. L.; Walker, W. E.; Pruett, R. L.; Schoening, R. C. [Rh9P(CO)21]2−. Example of encapsulation of phosphorus by transition-metal-carbonyl clusters. Inorg. Chem. 1979, 18, 129−136. (7) Vidal, J. L.; Walker, W. E.; Schoening, R. C. [Rh10P(CO)22]3−. A transition-metal carbonyl cluster with a metal polyhedron based on the bicapped square antiprism as illustrated by the structural study of the benzyltriethylammonium salt. Inorg. Chem. 1981, 20, 238−242. (8) Ciani, G.; Garlaschelli, L.; Sironi, A.; Martinengo, S. Synthesis and X-ray characterization of the novel [Rh10S(CO)10(μ-CO)12]2− anion; a bicapped square-antiprismatic cluster containing an interstitial sulphur atom. J. Chem. Soc., Chem. Commun. 1981, 0, 563−565. (9) Vidal, J. L.; Fiato, R. A.; Cosby, L. A.; Pruett, R. L. [Rh17(S)2(CO)32]3−. 1. An example of encapsulation of chalcogen atoms by transition-metal-carbonyl clusters. Inorg. Chem. 1978, 17, 2574−2582. (10) Vidal, J. L. [Rh10As(CO)22]3−. Example of encapsulation of arsenic by transition-metal carbonyl clusters as illustrated by the structural study of the benzyltriethylammonium salt. Inorg. Chem. 1981, 20, 243−249. (11) Vidal, J. L.; Troup, J. M. [Rh12Sb(CO)27]3−. An example of encapsulation of antimony by a transition metal carbonyl cluster. J. Organomet. Chem. 1981, 213, 351−363. (12) Femoni, C.; Iapalucci, M. C.; Longoni, G.; Tiozzo, C.; Zacchini, S.; Heaton, B. T.; Iggo, J. A. Sn-centred icosahedral Rh carbonyl clusters: synthesis and structural characterization and 13C−{103Rh} HMQC NMR studies. Dalton Trans. 2007, 35, 3914−3923. 6350

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351

Article

Inorganic Chemistry (33) Ceriotti, A.; Demartin, F.; Heaton, B. T.; Ingallina, P.; Longoni, G.; Manassero, M.; Marchionna, M.; Masciocchi, N. Nickel carbonyl clusters containing interstitial carbon-congener atoms: synthesis and structural characterisation of the [Ni12(μ12-E) (CO)22]2−(E = Ge, Sn) and [Ni10(μ10-Ge) (CO)20]2− dianions. J. Chem. Soc., Chem. Commun. 1989, 786−787. (34) Tran, N. T.; Powell, D. R.; Dahl, L. F. Nanosized Pd145(CO)x(PEt3)30 Containing a Capped Three-Shell 145-Atom Metal-Core Geometry of Pseudo Icosahedral Symmetry. Angew. Chem., Int. Ed. 2000, 39, 4121−4125. (35) Erickson, J. D.; Mednikov, E. G.; Ivanov, S. A.; Dahl, L. F. Isolation and Structural Characterization of a Mackay 55-Metal-Atom Two-Shell Icosahedron of Pseudo-Ih Symmetry, Pd55L12(μ3-CO)20 (L = PR3, R = Isopropyl): Comparative Analysis with Interior Two-Shell Icosahedral Geometries in Capped Three-Shell Pd145, Pt-Centered Four-Shell Pd−Pt M165, and Four-Shell Au133 Nanoclusters. J. Am. Chem. Soc. 2016, 138, 1502−1505. (36) Puls, A.; Jerabek, P.; Kurashige, W.; Förster, M.; Molon, M.; Bollermann, T.; Winter, M.; Gemel, C.; Negishi, Y.; Frenking, G.; Fischer, R. A. A Novel Concept for the Synthesis of Multiply Doped Gold Clusters [(M@AunM′m)Lk]q+. Angew. Chem., Int. Ed. 2014, 53, 4327−4331. (37) Zhu, M.; Lanni, E.; Garg, N.; Bier, M. E.; Jin, R. Kinetically Controlled, High-Yield Synthesis of Au25 Clusters. J. Am. Chem. Soc. 2008, 130, 1138−1139. (38) Qian, H.; Eckenhoff, W. T.; Zhu, Y.; Pintauer, T.; Jin, R. Total Structure Determination of Thiolate-Protected Au38 Nanoparticles. J. Am. Chem. Soc. 2010, 132, 8280−8281. (39) Song, Y.; Fu, F.; Zhang, J.; Chai, J.; Kang, X.; Li, P.; Li, S.; Zhou, H.; Zhu, M. The Magic Au60 Nanocluster: A New Cluster-Assembled Material with Five Au13 Building Blocks. Angew. Chem., Int. Ed. 2015, 54, 8430−8434. (40) Desireddy, A.; Conn, B. E.; Guo, J.; Yoon, B.; Barnett, R. N.; Monahan, B. M.; Kirschbaum, K.; Griffith, W. P.; Whetten, R. L.; Landman, U.; Bigioni, T. P. Ultrastable silver nanoparticles. Nature 2013, 501, 399−402. (41) Gröer, T.; Scheer, M. Transition-Metal-Substituted Dichlorobismuthanes as Starting Materials for Novel Bismuth−TransitionMetal Clusters. Organometallics 2000, 19, 3683−3691. (42) Cucka, P.; Barrett, C. S. The crystal structure of Bi and of solid solutions of Pb, Sn, Sb and Te in Bi. Acta Crystallogr. 1962, 15, 865− 872. (43) Halet, J. F.; Evans, D. G.; Mingos, D. M. P. Effect of cavity size on the charge distribution in carbido-metal carbonyl clusters and its possible catalytic implications. J. Am. Chem. Soc. 1988, 110, 87−90. (44) Mingos, D. M. P. Theoretical aspects of metal cluster chemistry. Pure Appl. Chem. 1991, 63, 807−812. (45) Femoni, C.; Kaswalder, F.; Iapalucci, M. C.; Longoni, G.; Zacchini, S. The possible role of metal carbonyl clusters in nanoscience and nanotechnologies. Coord. Chem. Rev. 2006, 250, 1580−1604. (46) Collini, D.; Femoni, C.; Iapalucci, M. C.; Longoni, G.; Svensson, P. H.; Zanello, P. Tuning Electronic Behavior of Carbonyl Metal Clusters by Substitution of Interstitial and Capping Atoms. Angew. Chem., Int. Ed. 2002, 41, 3685−3688. (47) Laupp, M.; Sträle, J. Synthese und Struktur des Au12Pd-Clusters [(PPh3PAu)8(AuCl)4Pd]. Z. Naturforsch. 1995, 50b, 1369−1372. (48) Pyykkö , P. Understanding the eighteen-electron rule. J. Organomet. Chem. 2006, 691, 4336−4340. (49) Häkkinen, H. Atomic and electronic structure of gold clusters: understanding flakes, cages and superatoms from simple concepts. Chem. Soc. Rev. 2008, 37, 1847−1859.

6351

DOI: 10.1021/acs.inorgchem.7b00409 Inorg. Chem. 2017, 56, 6343−6351