Crystal structure and absolute configuration of (+) 546-trans-dinitro (1

Muster aus H-Brücken: ihre Funktionalität und ihre graphentheoretische Analyse in Kristallen. Joel Bernstein , Liat Shimoni , Raymond E. Davis , Nin...
0 downloads 0 Views 810KB Size
1376 Inorganic Chemistry, Vol. 11, No. 6, 1972

NICHOLAS C. PAYNE

COKTRIBUTION FROM THE

DEPARTMENT O F CHEMISTRY, UNIVERSITY OF WESTERN OSTARIO, LONDON 72, OSTARIO,CANADA

The Crystal Structure and Absolute Configuration of (+)546-trans-Dinitro(l,l0-diamino-4,7-diazadecane)cobalt(III) Bromide BY h-ICHOLAS C. PAYNE Received June 24, 1.571

+

The crystal structure and absolute configuration of ( )aa~-truns-dinitro(l,l0-diamino-4,7-diazadecane)cobalt(III) bromide, ( + ) [ C o { N H ~ ( C H ~ ) ~ N H ( C H ~ ) Z N H ( C( N HO~P) )~ZJ]H Bhave ~ ,] been determined from three-dimensional X-ray data collected by counter methods The crystals are monoclinic, with a = 7.848 (5), b = 14 756 (12), c = 6.397 ( 5 ) 8, p = 99 5 (l)', and two formula units in space group P21. The structure was solved by application of Patterson and Fourier techniques and refined on F by full-matrix least-squares methods to a final residual RI = 0.038. The absolute configuration of the cation was determined by the Bijvoet method. The inner coordination sphere around the cobalt atom is approximately octahedral, with the tetradentate linear tetramine occupying the equatorial plane and two nitro ligands the apices. The conformation of the five-membered chelate ring is gauche, with absolute configuration 6. The two six-membered chelate rings assume the chair conformation. The configurations of the asymmetric secondary nitrogen atoms are both R.

Introduction absolute configuration determination of the (+) isomer ] + ion, which was underThe linear tetramine I,l0-diamino-4,7-diazadecane1 of the [(N02)2Co(3,2,3-tet) taken t o determine the absolute configurations of the NH2(CH2)3NH(CH2)2NH(CH2)3NH2 (hereafter 3,2,3asymmetric secondary nitrogen atoms and the contet), can act as a tetradentate ligand and form comformations of the chelate rings. The optical properties plexes with Co(III).l There are three possible geoof the enantiomer studied are presented in Figure 1.' metrical forms of the [Co(3,2,3-tet)X2In+ion, two cis isomers, designated CY and 8, and one trans isomer. Experimental Section Coordination of 3,2,3-tet in complexes where X is a A crystalline sample was supplied by Dr, J . N . MacB . Harrowunidentate ligand appears to occur with a preference field and Dr. B . Bosnich. trans-Dinitro(l,lO-diamino-4,Tfor the trans topology. A source of further asymmetry diazadecane)cobalt(III) bromide crystallizes as transparent orange plates, elongated along [ O O l ] . A preliminary photoarises in the trans structure from the two secondary graphic X-ray examination showed monoclinic symmetry, Laue amine groups. There are three possible isomers, group 2 / m . Systematic absences restricted to OkO when k = 2n depending upon the configurations at the nitrogen 1 suggested space groups P ~ ~ / W Zor- P21-C22.8 C ~ ~ ~ Since the maatoms, the antimeric RR and SSforms and the internally terial is optically active, space group P21 (no. 4) was chosen. Unit cell parameters were obtained from a least-squares refinecompensated RS (meso) form.' h study of scale ment of diffractometer setting gngles,Qusing Mo Ka1 radiation models suggests that for the RS (meso) configuration, (X 0.70926 8); u = 7.848 (5), b = 14.756 (la), c = 6.397 (5) A, the ethylene fragment in 3,2,3-tet will assume an and p = 99.5 (1)'. The density of t h e crystals was determined eclipsed form, while both the RR and SS configurations by flotation in a mixture of chloroform and 1,2-dibromoethane, result in a gauche conformation. I n many cases, in and the value observed, 1.87 (1) g is in agreement with the value 1.84 g calculated for two formula units per cell. the absence of strong hydrogen-bonding effects, the There are no crystallographic symmetry conditions imposed upon gauche conformation of a chelated ethylenediamine the ions. ligand has been demonstrated to be energetically A crystal was selected for data collection of approximate difavored over the eclipsed form.3 Tf the gauche conmensions 0.15 mm X 0.05 mm X 0.05 mm and mounted with formation occurs, then resolution of the complex t h e long dimension [ O O l ] approximately parallel t o the diffractometer 4 axis. An optical examination identified the crystal faces should be possible since the RR and SS configurations as (no), (?TO), (120), ( n O ) , a n d (S00)andforms { O l O ) and { O O l ) . are antimeric. The two six-membered chelate rings The crystal was measured with a micrometer eyepiece in prepformed upon coordination can assume skew, boat, or aration for an absorption correction, for which p(Mo K a ) = 41 chair conformations. A consideration of nonbonding cm --I. A series of w scans through intense reflections showed an average intramolecular repulsions suggests t h a t the chair and width a t half-height of 0.1'. This value was considered satisskew forms will be more stable thermodynamically than factory.10 the boat form.4 The chair conformation has been obIntensity data were recorded on a Picker four-circle automatic served in the tris(l,3-diaminopropane)cobalt(III) ion diffractometer. Molybdenum radiation was used for data collecfor the six-membered chelate rings.5 I n a recent tion, and t h e diffracted beam was filtered through 0.11 mm of letter, the optical properties of trans-RR-[co(3,2,3- Zr foil. At the takeoff angle of 1.1' the intensity of t h e beam was SO^, of the maximum attainable. T h e counter was placed tet)Cla]+ and trans-RR-[Co(3,2,3-tet)Brz]+ were re32 cm from t h e crystal, with an aperture 4 mm X 4 mm. A ported.6 symmetric scan range of 1' was sufficient, employing the 8-28 We present here the result of a crystal structure and scan technique, a t a scan rate of S"/min. Stationary-crystal,

+

(1) M. D. Alexander a n d H . G. Hamilton, Jr., Inovg. Chem., 8, 2131 (1969). (2) K . N. Raymond, P. W. R . Corfield, a n d J. A. Ibers, ibid., 7 , 842 (1968). ( 3 ) E. J. Corey and J. C. Bailar, Jr., J . Amer. Chem. SOL.,81,2620 (1959). (4) B. Bosnich, W. R . Kneen, and A. T. Phillip, Inovg. Chem., 8, 2567 (1969). ( 5 ) Y . Saito, T. Nomura, and F. Marumo, Bull. Chem. Soc. J U p . , 41, 530 (1968). (6) G. R . Brubaker and D . P. Schaefer, Inovg. Nucl. Chem. L e l f . , 6, 237 (1970).

stationary-counter background measurements of 10 sec were unique data set ( h k l ) was first made a t each end of a scan. collected, for 20 less than 50". At higher 20 values, there was (7) J. N. MacB. Harrowfield a n d B. Bosnich, private communication. (8) "International Tables for X - R a y Crystallography," Kynoch Press, Birmingham, England, 1962. (9) P. W. R . Corfield, R . J. Doedens, and J. A. Ibers, I n o v g . Chem., 6, 197 (1967). (10) T. C. Furnas, "Single Crystal Orienter Instruction Manual," General Electric Co., Milwaukee, Wis., 1957.

Inorganic Chemistry, Vol. 11, No. 6, 1972 1377 x(nm1 4SO

400

I

550

500

I

I

I

J

1

25

24

23

22

5

21

20

19

(cm-ll

18

Figure 1.-Absorption and CD spectra of (+)54s[Co(3,2,3tet) (N02)2]Br : , electronic spectrum in water; - - - - -, CD spectrum in water; -.-.-, CD spectrum of a 1% KBr disk, sensitivity 5 mdeg cm-l. ~

little or no intensity observed. The pulse height analyzer was set to accept a window of approximately 90% when centered on the Mo Kcul peak. Coincidence losses for strong reflections were minimized by insertion of copper foil attenuators with attenuator factors of approximately 2.0. During the collection of the data, three intense, axial reflections (loo), (040), and (002) were monitored after every 100 reflections recorded. The intensities observed showed only a statistical variation throughout data collection. The Friedel pair data set ( A b ) was then collected under the same conditions. In all, 1438 ( h k l ) reflections were collected. The 1151 unique reflections for which F2 > 2u(F2) were used in the solution and refinement of the structure. The data were then corrected for background and Lorentzpolarization effects. Standard deviations were assigned as described previously,g using a value of p = 0.04.11 Absorption correction trials using p = 41 .O cm-l for Mo Kcu radiation showed transmission factors varying from 0.67 to 0.79, so an absorption correction was applied.12

Structure Solution and Refinement Initial coordinates for the cobalt and bromine atoms were determined from a Patterson synthesis. The y coordinate of the cobalt atom was chosen as l / 4 to fix the origin of the unit cell. Of the two possible Br positions determined from the Patterson synthesis, related by the mirror plane y = l/4, one was arbitrarily chosen a t this point. One cycle of full-matrix, least-squares refinement varying positional and isotropic thermal parameters of the two atoms gave values of RI = Z[lFul - ~ F o ~ ~ /=Z 0.43 ~ F oand ~ RZ = ( Z w ( F, F, ) 2 / Z w F o a ) 1 /= z 0.53. Thefunction minimized was Zw( F, F ) 2 , and the weight w was calculated as 4FO2/u2(Fo2).The atomic scattering factors for Br, Co, 0, N, and C were taken

IJ

!I

(11) W. R. Busing and H . A. Levy, J . Chem. Phys., 26, 563 (1957). (12) T h e programs used in this work were local modifications of a program library for t h e C D C 6000 series kindly supplied by Dr. J. A. Ibers. T h e y include Hamilton's GONOSabsorption program, Ibers' NUCLS least-squares program, Zalkin's FORDAPfourier program, Johnson's ORTBP thermal ellipsoid plotting program, and t h e Busing-Levy ORBBE error function program.

from Cromer and \Vaber;la that of H was from Stewart, et d . 1 4 The real and imaginary anomalous dispersion corrections for Co and Br of Cromer were used.'s Aseries of least-squares refinements and difference Fourier syntheses located the remaining nonhydrogen atoms. Two cycles of least-squares refinement of the 20 nonhydrogen atoms, with isotropic thermal parameters, reduced R1 and 122 t o 0.097 and 0.128, respectively. Two possible enantiomeric structures exist-that chosen up to this point, referred to as structure R, and its mirror image referred to as structure L. These are related in space group P21 by a mirror plane perpendicular to b. To distinguish between the two, the refinement on structure R was continued and, with isotropic thermal parameters, coverged a t R1 = 0.080 and R2 = 0,109. The atomic positional parameters were then reflected across y = to give structure L and refined using isotropic thermal parameters and the same data. The refinement of structure L converged a t R1 = 0.102 and R1 = 0.136. The R-factor shows that ratio test applied to the weighted agreement factors, Rz, the L structure may be conclusively rejected a t the 0.005 level.16 It has been shown that the choice of the wrong enantiomer for the value of Af" applied to the scattering factor will lead to polar dispersion errors in the coordinates of atoms for which there are anomalous scattering effects.17 An examination of the geometry of the refined cations showed the Co-N(l) and Co-K(3) vectors to be within 7' of the y axis. I n space group P21 the error will be in the y coordinate.l8 An incorrect choice of enantiomer should therefore result in a discrepancy between the Co-K(1) and Co-N(3) bond lengths. For the R structure values of Co-N(l) and c o - x ( 3 ) of 2.01 and 1.98 A were observed, and for the L structure the values were 1.81 and 2.21 A, respectively. The bond lengths observed for the L structure differ by 0.4 A, and this model may therefore be rejected. On these two criteria the R structure was chosen as the correct enantiomer. A difference Fourier synthesis a t this stage revealed clear evidence of the 22 hydrogen atoms, thus justifying their inclusion in the model. Idealized coordinates were used, assuming tetrahedral coordination about both nitrogen and carbon atoms and C-H and N-H bond lengths of 1.0 b. The contributions from the hydrogen atoms were thereafter included in calculations of F,. One cycle of least-squares refinement, allowing the 22 nonhydrogen atoms to vibrate anisotropically, reduced Rl and R2 to 0.064 and 0.072. On the basis of this model the hydrogen atom coordinates were recalculated, and the refinement on F , employing 1151 reflections with Fo2 > 2 u ( F O 2 )converged , a t R1 = 0.038 and R2: = 0.039. For all 180 variables in the full-matrix leastsquares refinement the final shift was less than 0.1 esd. -4 statistical analysis of Ra over various ranges of 1 F,/ and X -l sin 0 confirmed that the weighting scheme was satisfactory and showed no unusual trends. The error in an observation of unit weight is 1.11 electrons. A comparison of F, and F, values showed that secondary extinction was not significant. A difference Fourier synthesis computed from structure factors based on the final model showed no features of chemical significance. a t fracThe highest peak was of electron density 0.42 (9) e k 3 tional coordinates (0.192, 0.467, 0.311). The final positional and thermal parameters are given in Table I. The derived hydrogen atom positional parameters, which were not varied in the refinement, are listed in Table 11. Structure amplitudes 'are presented in Table 111, as 10IFoI and 10lF,I in electrons . l a

Determination of the Absolute Configuration The absolute configuration of the cation has been determined by the Bijvoet absorption-edge technique.20 The values of the (13) D. T. Cromer a n d J. T. Waber, Acta Cvystallogr., IS, 104 (1965). (14) R. F. Stewart, E. 11. Davidson, and W. T. Simpson, J. Chem. P h y s . , 42, 3175 (1965). (15) D. T. Cromer, Acto CvystaElogv., 18, 17 (1965). (16) W. C. Hamilton, ibid., 18, 502 (1965). (17) T. Ueki, A. Zalkin, a n d D. H. Templeton, ibid., 20, 836 (1966). (18) D. W. J. Cruickahank and W. S. McDonald, ibid.,23, U (1867). (19) Table 111,a listing of structure factor amplitudes, will appear following these pages in t h e microfilm edition of this volume of t h e journal. Single copies may he obtained from the Business Operations Office, Books and Journals Division, American Chemical Society, 1156 Sixteenth S t . , N.W., Washington, D. C. 20036, by referring t o code number INORG-72.1376, Remit check or money order for $3.00 for photocopy or 12.00 for microfiche. (20) J. M. Bijvoet, Nature (Londoiz), 173, 888 (1954); A. F. Pcerdeman, A. J. Van Bommel, and J. M . Bijvoet, Proc. A c a d . S c i . Amsteidam, S e d . 13, 84, 16 (1951).

1378 Inorganic Chemistry,Vol. 11, No.6, 1972 FINAL ATOMIC Y

X

NICHOLAS C. PAYNE TABLE I THERMAL PARAMETERSa Ullb uzz U3a 255 (5) 411 (6) 241 (5) 625 (7) 579 (6) 590 (7)

POSITIONAL AND

2

u 1 2

Uza

u13

0.21342 (15) -0.08396 (13) '/4 14 (6) 74 (4) 11 (6) 0.30487 (13) 0.47063 (11) 0.44233 (15) -105 (6) 21 (5) 44 (6) 0.3828 (5) 0.1800 (12) 495 (50) 424 (49) 440 (45) -0.0924 (11) -16 (38) 49 (38) 64 (38) 335 (38) 629 (55) 291 (35) 0.2330 (5) 0.0064 (10) 53 (36) 27 (28) -64 (34) K(2) -0.2971 (9) 0.1161 (5) 0.2495 (12) 383 (46) 525 (48) 507 (52) 57 (36) 224 (40) -43 (38) N(3) -0.0882 (10) 0.2623 (6) 0.4146 (10) 373 (36) 504 (47) 388 (37) 0.1347 (9) -6 (41) 25 (30) 13 (39) N(4) 313 (36) 0.2643 (5) 0.4545 (10) 422 (44) 391 (37) 127 (38) 118 (30) 157 (41) K(5) -0.2033 (8) -0.0176 (11) 750 (55) 300 (35) 0.0524 (9) 0.2453 (7) 350 (42) 105 (48) 27 (30) 93 (50) K(6) 0.3278 (5) 0.5792 (10) 709 (46) 603 (44) 393 (35) 41 (40) 193 (32) -193 (36) 0 ( 1 ) -0.1695 (9) 0.2115 (5) 0.4919 (11) 624 (48) 685 (49) 607 (45) -40 (37) 359 (39) -16 (35) O ( 2 ) -0.3144 (9) 0.3088 (7) -0.0460 (11) 501 (44) 1332 (72) 557 (44) -225 (52) 206 (35) 119 (49) O(3) 0.1434 (9) 0.1813 (7) -0.1387 (12) 1254 (77) 897 (65) 552 (48) 316 (59) 546 (50) 57 (49) O(4) 0.0418 (12) C ( l ) -0.2648 (14) 0.4314 (7) 0.1451 (17) 694 (71) 465 (55) 579 (63) 225 ( 5 5 ) 198 (54) 123 (50) 555 (67) 735 (80) 464 (61) 380 (59) 0.3943 (8) -0.0387 (16) 63 (52) -6 (53) C(2) -0.3848 (14) 0.2987 (9) 331 (53) 970 (89) 472 (55) 0.0004 (15) 73 (56) 19 (46) -53 (59) C(3) -0.4380 (12) 0.0267 (17) 547 (68) 29 ( 5 2 ) -282 (57) 0.1387 (8) 689 (74) 584 (65) -207 (57) C(4) -0.3622 (14) 627 (69) 24 (54) -88 (53) 0.0761 (7) 0.0723 (18) 510 (64) 611 (70) -105 (57) C(5) -0,2119 (14) 613 (66) 0,0630 (7) 0.2809 (16) 484 (59) 654 (64) 146 (54) 253 (52) 52 (54) 0,0756 (13) C(6) 0.4530 (18) 714 (78) 694 (80) 574 (65) 422 (66) 166 (57) 177 (59) 0.2080 (16) 0.0996 (9) C(7) 0.1928 (9) 0.4173 (16) 286 (52) 903 (85) 551 (62) 192 (55) 65 (46) 39 (61) C(8) 0.2699 (11) a Estimated standard deviations in this and other tables are given in parentheses and correspond to the least significant digits. Uij = p:j/2r2ai*aj* (A). The values have been multiplied by IO4. The thermal ellipsoid is given by exp[ - (pllhz f h k 2 &Z2 2p1zhk 2Piahl 2Pzakl)l. CO Br N(1)

+

+

+

TABLE I1 DERIVEDHYDROGEN ATOMPOSITIOXAL PARAMETERS X

Y

B

X

Y

B

-0.035 -0.023 -0,244 -0.319 -0.326 -0.490 -0.534 -0,480 -0,247 -0.437 -0.431

0.397 0.409 0.497 0.425 0.395 0.433 0.281 0.207 0.238 0.120 0.136

0.056 0.311 0,119 0.275 -0,166

-0.156 -0.251 -0.140 0.124 0.050 0.310 0.157 0.312 0.368 0.187 0.104

0 070 0.015 0 114 0.064 -0 001 0 058 0 100 0.194

-0,056

-0.066 -0.115 0.140 -0.127 -0.109 0.145

0 207 0 322 0 264

0.114 0.382 0.146 0.317 0.471 0.587 0,279 0.534 0.383 0.560

aHydrogen atoms are numbered in order around the tetradentate ligand. Thus H ( l ) and H(2) are bonded to N ( l ) and H(3), H(4) is bonded to C(1), etc. For each atom B = 5 Az.

+

As discussed earlier, the wrong application of the anomalous dispersion correction to the atomic scattering factors leads to polar dispersion errors. I n structure R , final values of the Co-?; bond lengths, Co-N(l) of 1.971 (8) h and co-n'(3) of 1.990 (8) A, were determined. For structure L, values of Co-N(l) of 1.929 (10) dk and Co-N(3) of 2.031 (9) A were found. The difference for structure L is 0.102 (13) A, or approximately 8 u . Accordingly, we can also reject structure L on the basis of the cation geometry. TABLE IV DETERMIXATIOX OF ABSOLUTE CONFIGURATION

Indices 1 1 3 1 37 1 5 0 115 1 2 22

F,(hkl) 22.64 58.31 43.59 16.52 46.41

Obsd relationship

<


> >

Possible A-H.

*

Figure 2.-Perspective view of the [Co(3,2,3-tet)(N02)2] cation, showing the atom-numbering scheme. H ( 9 ) is bonded t o N(2); H(14) is bonded t o K(3). Atoms are plotted as 40% probability ellipsoids of thermal motion.

Indices 242 2 60 3 2 i 3 50 39 1

Obsd relationship

F,(hkl) 44.15 31.92 6.11 20.50 14.22

F,(&j)