Jahn–Teller Effect in the B12F12 Radical Anion ... - ACS Publications

Nov 25, 2015 - Freie Universität Berlin, Institut für Chemie und Biochemie, Fabeckstrasse 34-36, 14195 Berlin, Germany. ‡. Department of .... pubs...
0 downloads 0 Views 882KB Size
Article pubs.acs.org/IC

Jahn−Teller Effect in the B12F12 Radical Anion and Energetic Preference of an Octahedral B6(BF2)6 Cluster Structure over an Icosahedral Structure for the Elusive Neutral B12F12 Moritz Malischewski,*,† Eric V. Bukovsky,‡,§ Steven H. Strauss,‡,§ and Konrad Seppelt† †

Freie Universität Berlin, Institut für Chemie und Biochemie, Fabeckstrasse 34-36, 14195 Berlin, Germany Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523 United States



S Supporting Information *

ABSTRACT: The B12F12− radical anion was generated by oxidation of [CoCp2+]2B12F122− with AsF5 in SO2. In the crystal structure of [CoCp2+]B12F12−, the anion displays a lowered symmetry (D2h) instead of an Ih-symmetric dianion as a result of Jahn−Teller distortion. Moreover, shortening of the B−F bonds and subtle changes of the B−B bonds are observed. DFT calculations show that, for the unknown neutral B12F12, unprecedented structural isomers [e.g., octahedral B6(BF2)6] are energetically favored instead of an icosahedral structure. The structures and energetics are compared with those of the analogous chlorine compounds.



performed in 2003.9 In 2014, Jenne and co-workers reported the electrochemical potential for the B12F12−/2− couple versus FeCp2+/0 (Cp = cyclopentadienyl) as 1.68 V for (N(nBu)4+)2(B12F122−) in SO2 at −59 °C.21 Additionally, they reported that the oxidations of K2B12F12 with AsF5 in SO2 or anhydrous HF were hampered by the limited solubility of K2B12F12 in these solvents21 [although the solubility in HF is sufficient for the crystallization and determination of the X-ray structure of K2(HF)3B12F12, which was found to contain HF− K+ bonds15b]. To overcome the solubility problem, Jenne and co-workers oxidized (N(n-Bu)4+)2(B12F122−)9 with AsF5 in SO2 and isolated N(n-Bu)4+B12F12−,21 but they were unable to obtain crystals of the radical monoanion salt suitable for X-ray diffraction. By replacing N(n-Bu)4+ by CoCp2+, we succeeded in isolating and characterize the B12F12− radical monoanion as crystalline CoCp2+B12F12−. Additionally, CoCp2+ is ideally suited as an inert counterion under extremely oxidative conditions, because the oxidation potential of CoCp2+/2+ (3.15 V vs SCE in SO222 → 2.69 V vs FeCp20/+) is higher than the oxidation potential of B12F122−/−. CoCp2+ already has been used successfully as a robust counterion for the oxidation of B9Hal92−.23 Addionally, we report DFT calculations on different isomers of B12X12m− clusters (m = 0, 1; X = F, Cl) using the B3LYP functional with the 6-311+G(d,p) basis set. The success of isolating the Jahn−Teller-distorted24 cation radicals C6X6+ (X = F, Cl, Br, I)25,26 and their structural determination encouraged us to approach the problem of the B12F12− radical anion.

INTRODUCTION Since the seminal publications on the synthesis of the deltahedral cluster B12H122−,1 its derivatives, B12X122−,2 have been studied for both fundamental and applied science (see leading references for X = D,3 CH3,4 OH,5 OR,6 OCOR,7 F,8,9 Cl,10−12 Br,10,11 and I13). Although it is widely accepted that “isolated” B12X122− dianions with X = H, D, F, Cl, Br, and I have Ih symmetry,2,3,14 crystallographic C5 axes are not possible. Nevertheless, the anions in K2B12H121b and various compounds containing B12F122− 9,15 have either equal B−B bond lengths at the 3σ level of confidence or minor deviations from it, which can be explained by interionic packing forces. In the past several years, one-electron oxidations of several substituted closo-dodecaborates have been investigated because they can be understood as three-dimensional aromatics.16 B12(CH3)122− undergoes one-electron oxidation by (NH4)2Ce(NO3)6 to the blue radical anion.17 Using stronger electrondonating alkoxy substituents, B12(OR)122− can be oxidized even by the mild oxidant Fe3+, yielding a series of stable and isolable purple B12(OR)12− radicals, as well as neutral B12(OR)12 (e.g., R = CH2Ph).6,18 One-electron oxidation of B12(OH)122− to the yellow radical anion is accomplished by 30% H2O2.19 Recently, it was shown that, with electron-withdrawing Cl and Br substituents, the B12Cl122− and B12Br122− dianions require the very strong oxidizing medium AsF5 in anhydrous SO2 to prepare the isolable and stable blue radicals B12Cl12− and B12Br12−, respectively.20,21 This medium was sufficiently strong to oxidize the radical monoanions to the neutral clusters B12Cl2 and B12Br12. Interestingly, the structure of Na(SO2)6(B12Br12)2 contains a B12Br12− monoanion and a neutral B12Br12 cluster in the same lattice.21 The first indication of a possible B12F122− radical anion was provided by cyclic voltammetry of the B12F122− anion, first © XXXX American Chemical Society

Received: October 1, 2015

A

DOI: 10.1021/acs.inorgchem.5b02256 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 1. Synthesis of (CoCp2)2B12F12 and (CoCp2)B12F12



RESULTS The reaction of K2B12F12 with O2+Sb2F11− in anhydrous HF (aHF) led to the destruction of the cluster and the formation of BF4−. No reaction occurred when K2B12F12 was mixed with the milder oxidizing agent NO2+SbF6− in propionitrile. However, the cobaltocenium (1+) (CoCp2+) salt of B12F12− could be prepared successfully as shown in Scheme 1 [see the Supporting Information (SI) for more details]27−29 by reaction of [CoCp2+]2[B12F122−] with excess AsF5 in SO2 and was found to have good solubility in SO2. The desired product, CoCp2+B12F12−, was crystallized from the reaction mixture and was characterized spectroscopically and by single-crystal Xray crystallography.30 A frozen SO2 solution of CoCp2+B12F12− at −196 °C exhibited an electron paramagnetic resonance (EPR) spectrum with a sharp, apparently isotropic signal at g = 2.008 and a peakto-peak separation of 14 G (see SI for more details). For comparison, solid N(n-Bu)4+B12F12− exhibited an axial EPR signal at −203 °C with g|| and g⊥ values of 2.00 and 2.02, respectively, and line widths of ca. 20 G.21 The X-ray crystal data for CoCp2+B12F12− afforded the first X-ray structural characterization of the B12F12− monoanion.30 The cation/anion packing in CoCp2+B12F12− (shown in the SI) is a distorted CsCl-type lattice, with distances of 7.42, 7.42, and 9.07 Å and angles of 90°, 90°, and 96.5° between the centers of the cations and anions. The asymmetric unit contains one-half of a CoCp2+ cation and one-quarter of each of two crystallographically independent B12F12− clusters, each defined by four B−F moieties and each having crystallographic 2/m (C2h) symmetry. This results in eight unique B−B bond distances per cluster, which range from 1.805(6) to 1.841(6) Å for the B1−B4 cluster and from 1.805(4) to 1.832(5) Å for the B5−B8 cluster. In comparison to those in the B12F122− dianion,15e the B−B bond lengths in B12F12− are longer, and the B−F bond lengths are shorter. No abnormabilities in Co−C bond lengths were observed for the CoCp2+ cation [1.999(3)− 2.017(3) Å]. Based on the crystal structures of the CoCp2+ and B12F12− ions (Figure 1) as well as the electrochemical data, we conclude that the product is CoCp2+B12F12− and not CoCp22+B12F122−. Both crystallographically different B12F12− clusters have two B−B bond lengths that are obviously longer than all others, although in cluster II (B5−B8), this difference is on the edge of the 3σ range. To interpret these differences, DFT calculations were performed on the three lowest-energy isomers of B12F12− (see the calculated relative energies and B−B and B−F bond lengths in the SI). The Th isomer was found to be marginally the most stable of the three, with the D2h and D3d being higher in energy by 3.5 and 3.9 kJ mol−1, respectively.

Figure 1. Crystal structures of the CoCp2+ salts of (left) B12F122− and (right) B12F12−.

The weighted average B−B and B−F distances in the calculated Th, D2h, and D3d B12F12− structures were found to be the same, 1.813 and 1.363 Å. These values are larger and smaller, by 1.06% and 1.07%, respectively, than the 1.794-Å B− B distance and the 1.386-Å B−F distance in the DFT-optimized structure of Ih B12F122−.21 The expansion of the B12 cage upon oxidation of B12F122− to B12F12− can be attributed to the removal of an electron from the 4-fold-degenerate B−B bonding highest occupied molecular orbital (HOMO)14 of Ih B12F122−. The resultant asymmetrically occupied degenerate HOMO gives rise to Jahn−Teller distortion24 of B12F12−. When the geometry was constrained to Ih symmetry, the relative energy increased by about 18 kJ mol−1. The longer weighted-average B−B distance and larger volume of B12F12− relative to B12F122− notwithstanding, the ranges of DFT-predicted B−B distances for all three B12F12− isomers vary strongly: for Th symmetry, six at 1.748 Å and 24 at 1.829 Å; for D2h symmetry, eight at 1.777 Å, two at 1.778 Å, eight at 1.802 Å, eight at 1.838 Å, two at 1.839 Å, and two at 1.906 Å; for D3d symmetry, six at 1.773 Å, 12 at 1.798 Å, six at 1.817 Å, and six at 1.877 Å. In Figure 2, drawings of the B12 clusters (three DFT structures and X-ray structure) are shown. The resemblance of the experimental structure to the DFT D2h structure is striking. The experimental B−B bond length differences are smaller than the DFT-calculated values. This could result from partial disorder of the B 12 F 12 − clusters in the crystal. The interpretation of the crystal structure must be tentative until a more precise structure of CoCp2+B12F12− can be acquired and the differences in short and long B−B distances are unambiguous at the ±3σ level. Calculated Structures of Neutral B12F12 and B12Cl12. In all of our oxidation attempts, we found no indication of the formation of neutral B12F12. This species seems to be hardly producible even by electrochemical methods.21 Previous calculations of neutral B12F12 showed deviations from an icosahedral structure larger than in the monoanion.14b,c Surely, singlet-state B12F12 with Th symmetry is a local minimum, whereas D2h-symmetric (triplet) and D3d-symmetric (singlet) B12F12 are about 23 and 27 kJ/mol, respectively, higher in B

DOI: 10.1021/acs.inorgchem.5b02256 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Because fluorine is a better π donor than chlorine, the B−BX2 fragment is more favorable for X = F. The calculated structure of B6(BF2)6 displays very short B−F bonds (1.32−1.33 Å). External B−B bonds have lengths of 1.68−1.69 Å, whereas the cluster-internal bond lengths are less consistent. Most of them are in the range of 1.68−1.72 Å but there is one very short bond (1.59 Å) and one very long bond (1.76 Å). In comparison, the B−B bond lengths in Th B12F12 are significantly longer (24 bonds at 1.88 Å and 6 bonds at 1.71 Å). The B−F bonds have a length of 1.34 Å. Although the aforementioned findings are no explanation for why B12F12 cannot be isolated, they show that nearly icosahedral B12F12 is not the global energetic minimum. Also, it is questionable whether B6(BF2)6 can withstand powerful oxidants without the formation of BF3. Nevertheless, the unexpected variety of structural isomers of B12F12 seems to be unprecedented for any B12 cluster.



SUMMARY In harmony with these and previously reported14,20,21 DFT calculations, the X-ray structure of CoCp2+B12F12− demonstrates that B12F12− has longer B−B bonds and shorter B−F bonds, at the ±3σ level of confidence, than B12F122−. The crystal structure of CoCp2+B12F12− is consistent with a static Jahn−Teller-distorted D2h structure for B12F12−, in contrast to the Jahn−Teller-distorted Th structures observed and/or predicted for the B12 cages in all other B12X12− radical monoanions.6,17−21 The unknown neutral B12F12 is predicted not to have an icosahedral structure; possibly, it can be formulated as B6(BF2)6.

Figure 2. Illustrations of different symmetries in B12F12−.

energy. Surprisingly, we found that constitutional isomers of B12F12 can have significantly lower energies than the local minimum of Th B12F12. In particular, 1,2-fluorine shifts (formation of B−BF2 fragments) seem to be energetically favored (see Figure 3). Migration of one (B11F10−BF2) or two



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b02256. Synthetic details, crystallographic data, EPR spectrum, mass spectra, and data from DFT calculations (PDF) Crystallographic data for CoCp2+B12F12− (C10H10B12CoF12) (CIF) Crystallographic data for (CoCp2+)2B12F122−·1/2C3H6O (CIF)

Figure 3. Possible isomerizations of neutral B12F12 (from left to right): B12F12, B11F10−BF2, B10F8(BF2)2, B6(BF2)6. See Supporting Information (SI) for details.



[(B10F8(BF2)2] fluorine atoms results in structures of very similar energy (both being about 19 kJ/mol more stable than Th B12F12. The energetic gain of rearranging Th B12F12 to a distorted octahedral B6 cluster B6(BF2)6 with six terminal BF2 groups is about 134 kJ/mol! For the known neutral B12Cl12, this pathway is highly endothermic. Migrations of one, two, and six chlorine atoms are 72, 132, and 330 kJ/mol endothermic, respectively, compared to Th B12Cl12. This finding is in accordance with the structures of B4Cl431 and B9Cl9,23 which consist of B−Cl fragments, whereas the neutral boron subfluorides B8F12 and B10F12 contain all fluorine atoms in terminal B−BF2 groups.32 We attribute the different stabilities of B−Cl and B−BF2 fragments to the higher π-bonding ability of fluorine atoms compared to chlorine atoms. Assuming sp hybridization of boron in icosahedral B12X12, the p orbitals of boron are involved in cluster bonding, whereas one sp orbital radiates into the center of the cluster and the other sp orbital points toward the X substituent.33 The B6 core in B6(BX2)6 is formed analogously. The boron atoms in the BX2 groups are sp2-hybridized. This leads to an unfilled p orbital perpendicular to the B−BX2 plane that can interact with a p orbital of X.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions §

E.V.B. and S.H.S. contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS M.M acknowledges funding of Fonds der Chemischen Industrie (FCI) and Deutsche Forschungsgemeinschaft (GRK 1582), helpful discussions with Prof. Dr. Dieter Lentz (FU Berlin), and computing time made available by High-Performance Computing at ZEDAT/FU Berlin.



REFERENCES

(1) (a) Pitochelli, A. R.; Hawthorne, M. F. J. Am. Chem. Soc. 1960, 82, 3228−3229. (b) Wunderlich, J. A.; Lipscomb, W. N. J. Am. Chem. Soc. 1960, 82, 4427−4428.

C

DOI: 10.1021/acs.inorgchem.5b02256 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (2) Sivaev, I. B.; Bregadze, V. I.; Sjoberg, S. Collect. Czech. Chem. Commun. 2002, 67, 679−727. (3) Muetterties, E. L.; Merrifield, R. E.; Miller, H. C.; Knoth, W. H., Jr.; Downing, J. R. J. Am. Chem. Soc. 1962, 84, 2506−2508. (4) Peymann, T.; Knobler, C. B.; Hawthorne, M. F. J. Am. Chem. Soc. 1999, 121, 5601−5602. (5) Peymann, T.; Herzog, A.; Knobler, C. B.; Hawthorne, M. F. Angew. Chem., Int. Ed. 1999, 38, 1061−1064. (6) Peymann, T.; Knobler, C. B.; Khan, S. I.; Hawthorne, M. F. Angew. Chem., Int. Ed. 2001, 40, 1664−1667. (7) (a) Thomas, J.; Hawthorne, M. F. Chem. Commun. 2001, 1884− 1885. (b) Maderna, A.; Knobler, C. B.; Hawthorne, M. F. Angew. Chem., Int. Ed. 2001, 40, 1661−1664. (8) Solntsev, K. A.; Mebel, A. M.; Votinova, N. A.; Kuznetsov, N. T.; Charkin, O. P. Koord. Khim. 1992, 18, 340−364. (9) Ivanov, S. V.; Miller, S. M.; Anderson, O. P.; Solntsev, K. A.; Strauss, S. H. J. Am. Chem. Soc. 2003, 125, 4694−4695. (10) Avelar, A.; Tham, F. S.; Reed, C. A. Angew. Chem., Int. Ed. 2009, 48, 3491−3493. (11) Geis, V.; Guttsche, K.; Knapp, C.; Scherer, H.; Uzun, R. Dalton Trans. 2009, 2687−2694. (12) Gu, W.; Ozerov, O. V. Inorg. Chem. 2011, 50, 2726−2728. (13) Knoth, W. H.; Miller, H. C.; Sauer, J. C.; Balthis, J. H.; Chia, Y. T.; Muetterties, E. L. Inorg. Chem. 1964, 3, 159−167. (14) (a) Fujimori, M.; Kimura, K. J. Solid State Chem. 1997, 133, 178−181. (b) McKee, M. L. Inorg. Chem. 2002, 41, 1299−1305. (c) Lee, T. B.; McKee, M. L. Inorg. Chem. 2012, 51, 4205−4214. (15) (a) Peryshkov, D. V.; Popov, A. A.; Strauss, S. H. J. Am. Chem. Soc. 2010, 132, 13902−13913. (b) Peryshkov, D. V.; Friedemann, R.; Goreshnik, E.; Mazej, Z.; Seppelt, K.; Strauss, S. H. J. Fluorine Chem. 2013, 145, 118−127. (c) Peryshkov, D. V.; Goreshnik, E.; Mazej, Z.; Strauss, S. H. J. Fluorine Chem. 2010, 131, 1225−1228. (d) Peryshkov, D. V.; Strauss, S. H. J. Fluorine Chem. 2010, 131, 1252−1256. (e) Bukovsky, E. V.; Fiedler, S. R.; Peryshkov, D. V.; Popov, A. A.; Strauss, S. H. Eur. J. Inorg. Chem. 2012, 2012, 208−212. (16) King, R. B. Chem. Rev. 2001, 101, 1119−1152. (17) Peymann, T.; Knobler, C. B.; Hawthorne, M. F. Chem. Commun. 1999, 2039−2040. (18) Farha, O. K.; Julius, R. L.; Lee, M. W.; Huertas, R. E.; Knobler, C. B.; Hawthorne, M. F. J. Am. Chem. Soc. 2005, 127, 18243−18251. (19) Van, N.; Tiritiris, R. F.; Winter, R. F.; Sarkar, B.; Singh, P.; Duboc, C.; Munoz-Castro, A.; Arratia-Perez, R.; Kaim, W.; Schleid, T. Chem. - Eur. J. 2010, 16, 11242−11245. (20) Boeré, R. T.; Kacprzak, S.; Keßler, M.; Knapp, C.; Riebau, R.; Riedel, S.; Roemmele, T. L.; Rühle, M.; Scherer, H.; Weber, S. Angew. Chem., Int. Ed. 2011, 50, 549−552. (21) Boeré, R. T.; Derendorf, J.; Jenne, C.; Kacprzak, S.; Keßler, M.; Riebau, R.; Riedel, S.; Roemmele, T. L.; Rühle, M.; Scherer, H.; Warneke, J.; Vent-Schmidt, T.; Weber, S. Chem. - Eur. J. 2014, 20, 4447−4459. (22) Bard, A. J.; Garcia, E.; Kukharenko, S.; Strelets, V. V. Inorg. Chem. 1993, 32, 3528−3531. (23) Binder, H.; Kellner, R.; Vaas, K.; Hein, M.; Baumann, F.; Wanner, M.; Winter, R.; Kaim, W.; Hönle, W.; Grin, Y.; Wedig, U.; Schultheiss, M.; Kremer, R. K.; von Schnering, H. G.; Groeger, O.; Engelhardt, G. Z. Anorg. Allg. Chem. 1999, 625, 1059−1072. (24) (a) Jahn, H.; Teller, E. Proc. R. Soc. London, Ser. A 1937, 161, 220. (b) Bersuker, I. B. Chem. Rev. 2001, 101, 1067−1114. (25) Shorafa, H.; Mollenhauer, D.; Paulus, B.; Seppelt, K. Angew. Chem., Int. Ed. 2009, 48, 5845−5847. (26) Molski, M. J.; Mollenhauer, D.; Gohr, S.; Paulus, B.; Khanfar, M. A.; Shorafa, H.; Strauss, S. H.; Seppelt, K. Chem. - Eur. J. 2012, 18, 6644−6654. (27) Peryshkov, D. V.; Popov, A. A.; Strauss, S. H. J. Am. Chem. Soc. 2009, 131, 18393−18403. (28) Peryshkov, D. V.; Bukovsky, E. V.; Strauss, S. H. In Efficient Preparation of Fluorine Compounds; Roesky, H., Ed.; Wiley: New York, 2012; pp 16−18.

(29) Peryshkov, D. V. Ph.D. Dissertation, Colorado State University, Fort Collins, CO, 2011. (30) Crystal data for CoCp2+B12F12−: C10H10B12CoF12, Mr = 546.83 g·mol−1, monoclinic, C2/m, a = 14.8447(18) Å, b = 14.8356(18) Å, c = 11.0527(14) Å, β = 125.359(6)°, V = 1985.1(4) Å3, Z = 4, T = 143(2) K, 12157 reflections, 3014 independent reflections, 171 parameters, R1 [I > 2σ(I)] = 0.0473, wR2 = 0.1114, Bruker Smart 1000 TU diffractometer (Mo Kα, λ = 0.71073 Å; graphite monochromator). A semiempirical absorption correction was applied using SADABS. The structure was refined using SHELXL-97. (31) Atoji, M.; Lipscomb, W. N. J. Chem. Phys. 1953, 21, 172. (32) Pardoe, J. A. J.; Norman, N. C.; Timms, P. L.; Parsons, S.; Mackie, I.; Pulham, C. R.; Rankin, D. W. H. Angew. Chem., Int. Ed. 2003, 42, 571−573. (33) Jemmis, E. D.; Balakrishnarajan, M. M.; Pancharatna, P. D. J. Am. Chem. Soc. 2001, 123, 4313−4323.

D

DOI: 10.1021/acs.inorgchem.5b02256 Inorg. Chem. XXXX, XXX, XXX−XXX