Supporting Information

Transcription

1 Supporting Information An Aerobic Synthetic Approach Toward Bis-Alkynyl Cobalt(III) Compounds Sean N. Natoli, Matthias Zeller and Tong Ren Contents 1. Characterization S2 2. ESI-MS of reaction leading to 3c S27 3. Reactivity of Ag salts S28 4. Crystal Data S29 5. Electrochemistry S32 6. Computational Details S34 S1

2 1. Characterization 1.1 IR Figure S1. IR absorption spectrum of 1a, ATR. S2

3 Figure S2. IR absorption spectrum of 1b, ATR. S3

4 Figure S3. IR absorption spectrum of 2a, ATR. S4

5 Figure S4. IR absorption spectrum of 2b, ATR. S5

6 Figure S5. IR absorption spectrum of 3a, ATR. S6

7 Figure S6. IR absorption spectrum of 3b, ATR. S7

8 Figure S7. IR absorption spectrum of 3c, ATR. S8

9 Figure S8. IR absorption spectrum of 3d. Inset shows close up view of C C stretch, ATR. S9

10 Figure S9. UV-vis spectra of compounds [1-3]b, 3c, and 3d in MeCN. S10

11 1.2 ESI-MS Figure S10. ESI-MS of compound 1a. S11

12 Figure S11. ESI-MS of compound 1b. S12

13 Figure S12. ESI-MS of compound 2a. S13

14 Figure S13. ESI-MS of compound 2b. S14

15 Figure S14. ESI-MS of compound 3a. S15

16 Figure S15. ESI-MS of compound 3b. S16

17 Figure S16. ESI-MS of compound 3c. S17

18 Figure S17. ESI-MS of compound 3d. S18

19 1.3 1 H NMR Figure S18. 1 H NMR of compound 1a in CD 3 CN. S19

20 Figure S19. 1 H NMR of compound 1b in CD 3 CN. S20

21 Figure S20. 1 H NMR of compound 2a in CD 3 CN. S21

22 Figure S21. 1 H NMR of compound 2b in CD 3 CN. S22

23 Figure S22. 1 H NMR of compound 3a in CD 3 CN. S23

24 Figure S23. 1 H NMR of compound 3b in CD 3 CN. S24

25 Figure S24. 1 H NMR of compound 3c in CD 3 CN. S25

26 Figure S25. 1 H NMR of compound 3d in CD 3 CN. S26

27 2. ESI-MS of reaction leading to 3c Figure S26. MS data for the reaction mixture of 3c after 24h shows no observable scrambling of acetylides as is the case of lithium reagents. No peaks corresponding to 3a (m/z = 641) or 3b (m/z = 547) are present. S27

28 3. Reactivity of Ag Salts Cl HN NH Co F F F (i) X HN NH Co F F F (ii) HN NH Co F F F HN NH F F HN NH F F HN NH F F 1 2 3d Conditions: (i) 3 equiv AgX, MeCN, reflux 48h; (ii) 4 equiv HC 2 Ph, Et 3 N, MeCN, reflux 24h. Ag(X) salts 2 %yield a (3d) %yield a AgOTf [Co(cyclam)(C 2 C6F 5 )(NCMe)](OTf) AgNO 3 [Co(cyclam)(C 2 C6F 5 )(NCMe)](NO 3 ) AgCF 3 CO 2 [Co(cyclam)(C 2 C6F 5 )(CF 3 CO 2 )](CF 3 CO 2 ) 2 84 * 25 b AgBF 4 [Co(cyclam)(C 2 C6F 5 )(F)](BF 4 ) 86 No RXN AgPF 6 [Co(cyclam)(C 2 C6F 5 )(F)](PF 6 ) 88 No RXN a Based on isolated yields confirmed by ESI-MS. b Yield based on 72h of reflux. * Sparingly soluble in MeCN and slowly degrades. Table S1. Evaluation of silver salts to facilitate the formation of bis-alkynyl Co(III) compounds. Figure S27. ESI-MS of the crude reaction of AgPF 6 and [Co(cyclam)(C 2 C 6 F 5 )Cl]Cl + revealing a m/z that correlates to the complex [Co(cyclam)(C 2 C 6 F 5 )F] +. Thus, providing insight into the lack of reactivity in yielding the bis-alkynyl complex 3d. S28

29 4. Crystallographic Details General Procedures. X-ray diffraction data for 1-3 were collected using a Nonius Kappa CCD diffractometer with Mo-Kα radiation (λ = Å), or a Rigaku Rapid II curved image plate diffractometer with Cu-Kα radiation (λ = Å). The Nonius Kappa CCD instrument features a fine focus sealed tube X-ray source with graphite monochromator. The Rigaku Rapid II diffractometer is equipped with an X-ray microsource with a laterally graded multilayer (Goebel) mirror for monochromatization. Single crystals were mounted on Mitegen micromesh or loop mounts using a trace of mineral oil and cooled in-situ to 100(2) K for data collection. Data on the Kappa CCD diffractometer were collected using the Nonius Collect software 4. Data obtained from the Rigaku Rapid II instrument were collected using the dtrek option of CrystalClear. 5 Both kinds of data sets were processed using HKL and data were corrected for absorption and scaled using Scalepack 6. The space groups were assigned and the structures were solved by direct methods using XPREP within the SHELXTL suite of programs 7 and refined by full matrix least squares against F 2 with all reflections using Shelxl2013 or using the graphical interface Shelxle 9. If not specified otherwise H atoms attached to carbon and nitrogen atoms and hydroxyl hydrogens were positioned geometrically and constrained to ride on their parent atoms, with carbon hydrogen bond distances of 0.95 Å for and aromatic C-H, 1.00 and 0.99 Å for aliphatic CH 2 and CH 3 moieties, respectively. Methyl H atoms were allowed to rotate but not to tip to best fit the experimental electron density. U iso (H) values were set to a multiple of U eq (C) with 1.5 for CH 3, and 1.2 for C-H and CH 2 units, respectively. Non meroheric twinning was handled using the programs Rotax 10 and WinGX 11. In compound 2b one of the two triflate anions is disordered over one major and two minor orientations. The three moieties were restrained to have similar geometries, and U ij components of ADPs were restrained to be similar for atoms closer than 1.7 Å. Subject to these conditions the occupancy rates refined to 0.817(2), 0.058(2) and (18). In compound 3b the crystal under investigation was found to be slightly twinned by nonmerohedry. Programs compatible with the Nonius KappaCCD diffractometer and HKL3000 lack the ability to simultaneously integrate more than one twin domain. With no data set obtainable through simultaneous integration of both twin domains, the data were instead handled as if not S29

30 twinned, with only the major domain integrated, and converted into an hklf 5 type format hkl file after integration using the "Make HKLF5 File" routine as implemented in WinGX. The twin law and matrix were obtained using the program ROTAX as implemented in WinGX. The twin operation was identified as a 180 rotation around in the reciprocal lattice direction, the twin matrix as , 0-1 0, The overlap R1 and R2 values in the "Make HKLF5 File" routine used were 0.18, i.e. reflections with a discriminator function less or equal to overlap radius of 0.18 were counted as overlapped, all others as single. The discriminator function used was the "delta function on index non-integrality". No reflections were omitted. The structure was solved using direct methods with all reflections of component 1. The structure was refined using the hklf 5 routine with all reflections of component 1 (including the overlapping ones) as obtained from WinGX, resulting in a BASF value of (9). No R int value is obtainable for the hklf 5 type file using the WinGX routine. Disorder is observed in the structure, affecting the triflate anions, the solvate ether and acetonitrile molecules, as well as one of the dimethyl amine substituents of one of the four crystallographically independent cations. The two triflate cations are disordered by rotation. The molecule of S1A was refined as disordered over two orientations. That of S1B as disordered over three orientations. All five triflate moieties were restrained to have similar geometries (SAME command in Skelxl). U ij components of ADPs of atoms at each site were restrained to be similar if closer than 1.7 Å. Subject to these conditions the occupancy rates refined to 0.667(9) and 0.333(9) for the moieties of S1A and S1C, and to 0.513(4), 0.364(3) and 0.122(2) for the moieties of S1B, S1D and S1E. One of the dimethyl amine substituents was refined as disordered by inversion at the nitrogen. The two moieties were restrained to have similar geometries, and U ij components of ADPs were restrained to be similar if closer than 1.7 Å. Subject to these conditions the occupancy rates refined to 0.55(3) and 0.45(3). Solvate acetonitrile and ether molecules are extensively disordered, with one site occupied by mostly acetonitrile, and a neighboring site by mostly ether. The latter site is in close proximity to an inversion center, and part of the disordered ether molecules are incompatible with their own symmetry equivalent counterparts. Due to the extensive disorder and overlap of neighboring disordered molecules no attempts were made to ensure exact full occupancy for each site. All ether molecules and all acetonitrile molecules were each restrained to have similar S30

31 geometries. For the acetonitrile molecules the N-C triple bond length was restrained to approximately 1.15(2) Å, the 1,3 N...C distance to at least 2.50(2) Å. U ij components of ADPs were restrained to be similar if closer than 1.7 Å. Subject to these conditions the occupancy rates refined to 0.396(7), 0.220(6) and 0.076(6) for the three ether molecule moieties, and to 0.400(14), 0.299(14), 0.150(8) and 0.184(10) for the four acetonitrile moieties. The forth of these moieties does overlap mostly with the ether molecules, the other three take up the other solvate occupied site. Computer programs: Nonius Collect (Nonius, 1998), CrystalClear-SM Expert 2.1 b32 (Rigaku, 2014), HKL-3000 (Otwinowski & Minor, 1997), SHELXS97 (Sheldrick, 2008), SHELXL2014/7 (Sheldrick, 2014), SHELXLE Rev714 (Hübschle et al., 2011). S31

32 5. Electrochemistry Cyclic Voltammogram Figure S28. Cyclic (CV; black) and differential pulse voltammograms (DPV; grey) recorded for compounds 1a (top), 2a (top middle), 3d (bottom middle), and 3c (bottom) in a 0.10 M MeCN solution of Bu 4 NPF 6 at a scan rate of 0.10 V s -1 for CV and a pulse width of 0.05 s for DPV. S32

33 Figure S29. CV of complex 3d in in a 0.10 M MeCN solution of Bu 4 NPF 6 at a variable scan rates. Figure S30. CV of complex 1a and 2a in a 0.10 M MeCN Solution of n-bu 4 N[B(C 6 F 5 ) 4 ] at a scan rate of 0.10 V s -1. Differential Pulse Voltammogram (DPV) S33

34 Figure S31. Differential pulse voltammagram showing the 1 st oxidation (N/N + ) of [3b](OTf) in DCM with 0.1M NBu 4 PF 6 under Taube-Richardson conditions (pulse amplitude = 10 mv). S34

35 6. Computational Details The geometries of [3a'] +, [3b'] +, and [3c'] + in the ground state were fully optimized from the crystal structures reported in this work, using the density functional method B3LYP 1 (Beck s 3 parameter hybrid functional using the Lee-Yang-Parr 2 correlation functional) and employing the LanL2DZ and 6-31G(d,p), 3,4 basis sets as described in the experimental section. The calculation was accomplished by using the Gaussian03 program package. 5 Figure S32. Fully optimized structure of [3a'] + using DFT methods. Table S2. Relevant bond length (Ǻ) and angle (deg) computed for [3a'] +. Co1-N Co1-N Co1-N Co1-N Co1-C C1-C C2-C C3-C C3-C C4-C C5-C C6-C C7-C C1-Co-C Co1-C1-C C1-C2-C S35

36 Figure S33. Fully optimized structure of [3b'] + using DFT methods. Table S3. Relevant bond length (Ǻ) and angle (deg) computed for [3b'] +. Co1-N Co1-N Co1-N Co1-N Co1-C C1-C C2-C C3-C C3-C C4-C C5-C C6-C C7-C C6-N C9-N C10-N C1-Co-C Co1-C1-C C2-C3-C C6-N5-C S36

37 Figure S34. Fully optimized structure of [3c'] + using DFT methods. Table S4. Relevant bond length (Ǻ) and angle (deg) computed for [3c'] +. Co1-N Co1-N Co1-N Co1-N Co1-C C1-C C2-C C3-C C3-C C4-C C5-C C6-C C7-C Co1-C C9-C C1-Co-C Co1-C1-C C1-C2-C C2-C3-C Co1-C9-C C9-C10-C C10-C11-C C14-N5-C S37

38 References (1) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields. J. Phys. Chem. 1994, 98, (2) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 1988, 37, (3) Hariharan, P. C.; Pople, J. A. The influence of polarization functions on molecular orbital hydrogenation energies. Theor. Chim. Acta 1973, 28, (4) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self-consistent molecular orbital methods. XXIII. A polarization-type basis set for second-row elements. J. Chem. Phys. 1982, 77, (5) Frisch, M. J., et al. Gaussian 03. Gaussian , Gaussian, INC., Wallingford, CT,. S38