Angewandte Communications Chemie International Edition:DOI:10.1002/anie.201706323 Metalloradicals Very Important Paper German Edition:DOI:10.1002/ange.201706323 Ta(CNDipp)6 :AnIsocyanide AnalogueofHexacarbonyltantalum(0)** Khetpakorn Chakarawet, Zachary W. Davis-Gilbert, Stephanie R. Harstad, Victor G. Young,Jr. , Jeffrey R. Long, and John E. Ellis* In memory of Alan Davison (1936–2015) Abstract: Hexakis(2,6-diisopropylphenylisocyanide)tantalum Xyl = 2,6-dimethylphenyl, which was far more thermally [1i] is the first isocyanide analogue of the highly unstable Ta(CO)6 stable (m.p.159–1608Cdecomp.) than V(CO)6 (decomp. [8] and represents the only well-defined zerovalent tantalum 60–708C), suggested that Ta(CNXyl)6 and the niobium complex to be prepared by conventional laboratory methods. analog might be isolable.However,despite success in the 0 + Twoprior examples of homoleptic Ta complexes are known, synthesis and isolation of [Ta(CNXyl)7] ,[Ta(CNXyl)6]À,and [11, 12] Ta(benzene)2 and Ta(dmpe)3,dmpe = 1,2-bis(dimethylphos- related complexes, attempts to isolate the corresponding phano)ethane,but these have only been accessed via ligand co- elusive neutral homoleptic tantalum and niobium complexes condensation with tantalum vapor in asophisticated metal- uniformly failed.[13] Many years ago Mann, Gray,and atom reactor.Consistent with its 17-electron nature,Ta- Hammond reported on the zerovalent complex W(CNDipp)6, [14] (CNDipp)6 undergoes facile one-electron oxidation, reduction, Dipp = 2,6-diisopropylphenyl, and it was of interest to or disproportionation reactions.Inthis sense,itqualitatively determine whether this more sterically encumbered isocya- resembles V(CO)6,the only paramagnetic homoleptic metal nide could stabilize the corresponding neutral tantalum carbonyl isolable under ambient conditions. complex. Initial results were promising and provided astruc- tural characterization of the unique ditantalum salt [Ta- [1] [15] Incontrast to vanadium and niobium, isolable homoleptic (CNDipp)7][Ta(CNDipp)6], atype of compound previously complexes of zerovalent tantalum are quite rare,with only unknown for homoleptic metal carbonyls and isocyanides,[16] [2] two examples presently known:Ta(benzene)2 and Ta- and of significance as aformal disproportionation product of [3] (dmpe)3, dmpe = 1,2-bis(dimethylphosphano)ethane.Also, the desired Ta(CNDipp)6 complex. Unfortunately,this salt these species are very poorly accessible because they are only could not be obtained as apure bulk solid and the neutral available via tantalum vapor synthesis methods,which require species was not isolated, so this project was temporarily [4] specialized metal-atom reactors. Although Ta(bipy)3,bipy = abandoned. Now more than adecade later,independent [5] a,a’-bipyridine, and Ta(iPr2-dad)3, iPr2-dad = 1,4-diiso- syntheses of both the cation and anion in the ditantalum salt, [6] [17] propyl-1,4-diazabuta-1,3-diene, were originally suggested as [Ta(CNDipp)7][BF4](1), and K[Ta(CNDipp)6](3), to be zerovalent tantalum complexes,later experimental and respectively,were developed, as well as the synthesis and computational studies indicate that both compounds contain isolation of neutral Ta(CNDipp)6.Ta(CNDipp)6 is the first V [7] high-valent Ta . fully substituted derivative of Ta(CO)6 containing monoden- Perhaps the most interesting comparisons of homoleptic tate ligands,[18a] and is the only one to be obtained by zerovalent complexes of Group 5metals are for the 17- aconventional synthesis.[18b] Theniobium analogue has also [19] electron hexacarbonyls of these species:V(CO)6 is well- been isolated and structurally characterized. known as the only paramagnetic unsubstituted metal carbonyl Thedeep red-brown and very air sensitive precursor [8] [9] isolable under ambient conditions, while Ta(CO)6 and complex TaI(CNDipp)6 (2), was obtained in 90%yield by the [10] [20a] Nb(CO)6 have only been obtained at low temperatures as reaction of I2 with [Et4N][Ta(CO)6] in the presence of six matrix isolated species and may be unstable above 50 K. equivalents of CNDipp in tetrahydrofuran, THF.Asimilar Barybinssynthesis of the 17-electron complex V(CNXyl)6, procedure was previously used for the synthesis of TaI- [11] (CNXyl)6. However,the reaction mixture leading to 2 also [*] Z. W. Davis-Gilbert, S. R. Harstad, Dr.V.G.Young Jr., Prof. J. E. Ellis had to be heated at reflux for 20 htoavoid contamination by Department of Chemistry mixed carbonyl-isocyanides.[20b] IR and NMR spectra, as well University of Minnesota as the molecular structure of 2,[21] are very similar to those Minneapolis, MN 55455 (USA) previously reported for the CNXyl analogue.[11] Reduction of E-mail:[email protected] [22] 2 by aminimum of 3equivalents of KC8 in toluene over K. Chakarawet, Prof. J. R. Long aperiod of 5days at 208Cwas necessary for high conversion Department of Chemistry,UniversityofCalifornia, Berkeley to satisfactorily pure dark violet microcrystalline K[Ta- Berkeley,CA94720 (USA) (CNDipp) ], 3,isolated in 70–80%yields.[23] [**] Dipp= 2,6-diisopropylphenyl. Highly Reduced Organometallics, 6 Treatment of cold, 658C, deep red solutions of 3 in part 67. Part 66:W.W.Brennessel, J. E. Ellis,Inorg. Chem. 2012, 51, À 9076. toluene with one equivalent of solid Ph3PAuCl afforded Supporting information and the ORCID identification number(s) for within 2habright green solution of Ph3PAuTa(CNDipp)6 (4) the author(s) of this article can be found under: which was isolated in 79%yield as violet-brown micro- https://doi.org/10.1002/anie.201706323. crystals.Although gold derivatives of metal carbonyls have Angew.Chem. Int.Ed. 2017, 56,10577 –10581 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim 10577 Angewandte Communications Chemie Scheme 1. Synthesis of Ph3PAuTaL6 and TaL 6,L= CNDipp, solvent = to- luene. reaction was carried out to determine whether the previously [15] isolated ditantalum salt [Ta(CNDipp)7][Ta(CNDipp)6] could be obtained by arational route,but the only products observed were 5 and asmall amount of free CNDipp.For this and other reasons,webelieve the formal disproportionation product of 5 is stable only in the solid state.Figueroasgroup reported very similar comproportionation reactions of [Co- + (CNAr’)4(THF)n] , n = 0, 1, with [Co(CNAr’)4]À to quantita- tively produce the second known 17-electron homoleptic Figure 1. Molecular structure of 4.Thermal ellipsoids are set at 50% isocyanide metal(0) complex, Co(CNAr’)4,where Ar’ is an probability,with hydrogen atoms omitted for clarity.Selected bond extremely bulky m-terphenyl group,2,6-bis(2,4,6-trimethyl- lengths []and angles [8]: Ta1–Au1 2.7207(2), Au1–P1 2.2737(8), Ta1– [30] phenyl)phenyl. Thesecond reaction with [Et3NH][BPh4] C66 2.111(3), Ta1–C40 2.117(3), Ta1–C53 2.113(3), Ta1–C27 2.137(3), was an attempt to protonate 3 to give the unknown hydride, Ta1–C1 2.148(3), Ta–C14, 2.150(3);avTa-C-N 174(2), av C-N-C 165(8), [20a] Ta-Au-P 176.20(3). HTa(CNDipp)6.But unlike HTa(PF3)6 which is fairly robust, no evidence for the isocyanidehydride was obtained. Isocyanide metal hydrides are rare species and Figueroa and [24] [31a] been known since 1964, 4 appears to be the first gold adduct co-workers recently reported [HFe(CNAr’)4]À and [HCo- [31b] of ahomoleptic metal isocyanide.X-ray structural character- (CNAr’)4]. ization of 4,Figure 1, confirmed the presence of seven Solution IR spectra of 5 are nearly superimposable in coordinate tantalum bound to six discrete CNDipp ligands unreactive solvents,including alkanes,arenes,and THF.In 1 and the gold atom of the Ph3PAugroup.The geometry about THF,one intense broad n(CN) peak is present at 1934 cmÀ , 1 tantalum is best described as amonocapped octahedron, similar to that of V(CNXyl)6,1939 cmÀ ,and those reported 1 where Ph3PAuisthe capping group and the M–L distances fall for tungsten analogs,W(CNAr)6,Ar= Xyl, 1934 cmÀ ;Dipp, 1 [14] 1 13 into the expected 1:3:3pattern. Structural features of 4 are 1944 cmÀ (KBr disk). Hand CNMR spectra of 5 in similar to those of Ph3PAuV(CO)6,where the vanadium also [D8]THF show paramagnetically shifted resonances in posi- [25] [1i] has aslightly distorted monocapped octahedral geometry. tions similar to those reported for V(CNXyl)6. Attempts to TheTa–Audistance,2.7207(2) ,the first to be reported, is obtain reliable solution magnetic data for 5 by the Evans longer than the V–Aubond, 2.690(3) ,inPh3PAuV(CO)6, NMR method were unsuccessful due to its poor solubility in but is quite similar to the Ti–Audistance,2.719(1) ,in hydrocarbons,ethers and other unreactive solvents.However, [26] [Et3PAuTi(CO)6]À , owing to the nearly identical atomic as measured by SQUID magnetometry,the effective mag- [27] radii of Ti,1.45 ,and Ta,1.43 . netic moment of solid 5 at 300 Kis2.03 mB,and decreases with Discovery of areagent that would cleanly oxidize K[Ta- decreasing temperature to 1.76 mB at 2K,consistent with an 1 [31c] (CNDipp)6](3)toTa(CNDipp)6 (5)was challenging.With S = =2 spin centered on the Ta atom. Thecyclic voltammo- two exceptions,tobedescribed below,soluble oxidants gram of 5 in 1,2-difluorobenzene displays areversible [Ta- 0/1 À 1 invariably caused overoxidation of 3.For this reason, possible (CNDipp)6] redox couple centered at E =2 = 2.03 Vversus 0/1+ À poorly soluble or insoluble oxidants were examined. Gratify- Cp2Fe .Interestingly,this value is close to the reversible 0/1 ingly,asuspension of oligomeric MoO3 in toluene worked couple observed for [V(CNXy)6] À, 1.97 V, but is about À 0/1 well to selectively oxidize 3 to satisfactorily pure 5 (see 1.5 Vmore negative than that of [V(CO) ] À, 0.53 V, both 6 À Supporting Information).