5366
tion around the aluminum atoms may be explained by considering the electrostatic repulsion between pairs of bonding electrons. Since oxygen is more electronegative than carbon, the AI-C bonding electron pairs are closer to the aluminum atom than the more strongly polarized AI-0 bonding pair. Thus the C-A1-C angles become larger (1 16-1 17 ") and the C-AI-0 angles become smaller ( 100") than the tetrahedral value. Figures 2 and 3 show the molecular structure of [AI( CH&I2.C4H8O2and packing in the crystals, respectively. Three-coordinate oxygen is usually found in a pyramidal configuration; a n exception is reported" for tetramethyl-0,O '-bistrimethylsilylcyclodialuinoxane, where the three oxygen valencies are coplanar. I n the present study, the AI-0-C(1) angle (122") and the (17) P. J. Whzatley, J . Chem. Soc., 2562 (1963)
C(1)-0-C(1) angle (108") are not what would be expected on the basis of sp3 hybridization for the oxygen atoms, nor d o they indicate that the oxygen is in a planar configuration. The nmr spectra using tetramethylsilane as an internal standard indicate a chemical shift of T 10.93 for the protons of the (CHJaAl- group. This suggests a higher electron density at these partly negative carbon atoms than is found at the terminal carbon atoms in the trimethylaluminum dimerIa ( T 10.67). The protons on the dioxane ring carbon atoms in the complex are in resonance at lower fields with a chemical shift of T 6.19 compared with the protons in p-dioxane (7 6.30). Acknowledgment. The support of the National Science Foundation is gratefully acknowledged. (18) N. Muller and D. E. Pritchard, J . An?. Chem. Soc., 82, 248 ( 1960).
A Compound Containing a Tetrahedral Cluster of Nickel Atoms' M. J. Bennett, F. A. Cotton, and B. H. C. Winquist Contribution from the Department of Chemistrj., Massachusetts Institute of Technology, Cambridge, Massachusetts 02139. Receioed May 6, 1967
Abstract: The crystal structure-and simultaneously the true empirical and molecular formulas-for a compound, Ni,(CO),[P(C,H,CN),],, first reported but incorrectly formulated in 1959, has been determined. The compound crystallizes in the space group R3 with hexagonal unit cell dimensions a = 18.78 0.02 A and c = 26.12 =E 0.02 A , and six formula units per cell. The molecules have nearly perfect tetrahedral symmetry aside from small deformations attributed to packing effects, although, crystallographically, only CI symmetry is required. In the tetrahedron of nickel atoms the mean Ni-Ni distance is 2.508 r 0.004 A. There is a symmetrical bridging carbonyl group across each edge of the tetrahedron and a P-Ni bond collinear with each threefold axis of the tetrahedron. A qualitative discussion of the bonding would suggest that the molecule should be inherently diamagnetic. Observed paramagnetism is thus attributed to contamination by nickel metal and other magnetic impurities in bulk samples of this rather unstable substance. _t
s
ome years ago Meriwether, Colthup, Fiene, and Cotton2 reported that the reaction of nickel carbonyl with tris(fl-cyanoethyl)phosphine, r(CH2CH2CN)3,led not only to the expected product of the Ni(C0)?(PR3)' type, which may be obtained with phosphines generally, but also to a red-orange product (mp 160" dec) containing less C O and less phosphine per metal atom. This compound, which was then formulated as [Ni(CO)P(C2H4CN)a],, was found to have n o infrared absorption attributable to terminal CO groups but it did exhibit one strong band at 1815 cm-'. Moreover, only a single C=N stretching band was observed at 2245 cm-', suggesting that none of the cyano groups was coordinated. It was also reported that the substance was paramagnetic, but with less than one unpaired electron per metal atom. It was suggested on the basis of this evidence that the substance was likely to be a polymer of some sort, either linear or
cyclic, with the polymerization number, 71, equal to at least 2. In the years since this preliminary study was reported, the widespread occurrence of metal atom cluster compounds in genera13a'4and those of the polynuclear metal carbonyl typezb in particular has been recognized. This prompted a reconsideration of the nature of this substance, Since it did form crystals, albeit rather unstable ones, a reinvestigation of it was undertaken using, as the principal tool, single-crystal X-ray diffraction. As will be seen, the compound has now been shown to have the formula Ni,(C0),[P(C2H,CN),], and to contain a tetrahedral cluster of nickel atoms.
( I ) This \vork was supported, i n part, by Contract No. AT(30-111965 with thc U. S. Atomic Energy Commission. (2) L. S. Meriwetlier, E. C. Colthup, M. L. Fiene, and F. A. Cotton, J . l i i o r g . .Virci. Chem., 11, 181 (1959).
(3) Cf. F. A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry," 2nd ed, Interscience Publishers, John Wiley and Sons, Itlc., New York, N.Y . , 1966: (a) p 656; (b) p 719. (4) F. A. Cotton, Quorl. Rec. (London), 20, 389 (1966).
Journrrl oftlie American Chemical Society
1
89.21
Experimental Section Crystals of Nir(CO)b[P(CH?CH2CN)3]4 were prepared by a modification of the method of Xleriwether, ei 0 1 . ~ Dry methanol (75
October 11, 1967
5367 ml) was distilled into a flask containing 2.9 g (0.015 mole) of tris(2-cyanoethy1)phosphine. The mixture was heated to reflux temperature, and 2.75 ml (0.022 mole) of Ni(CO)* dissolved in 50 ml of methanol was added slowly. Soon after addition of the Ni(C0)4 was initiated white, crystalline Ni(CO)?[P(CH2CH2CN)& precipitated with evolution of CO. The reaction mixture was then allowed to stand for 24 hr at 70” after which time small orange crystals of Ni4(CO)G[P(CHrCH2CN)3]4 had formed. Many of the crystals grew o n the surface of the magnetic stirring bar in the reaction flask. The stirring bar was carefully removed from the flask and stored in a vial under liquid nitrogen to reduce the rate of decomposition of the crystals. Suitable crystals for X-ray study were then selected from the stirring bar. When quantities of the compound other than single crystals are desired, it is necessary to remove the N~(CO)?[P(CHZCH.CN)~]* which is also formed in the reaction. The most satisfactory method found was to dissolve the contaminant in hot methanol, in which the cluster compound is insoluble, However, even after repeated purification, very weak bands of the impurity were still present in the infrared spectrum. Solubility. The compound is insoluble in most organic solvents and only slightly soluble in acetonitrile and acetone. These solutions undergo slow decomposition even in the absence of air Infrared Spectrum. A mull spectrum of Ni4(CO)s[P(CH2CH2CN)Jr in Nujol has absorption bands at 2239 (medium) and 1802 cm-1 (strong). They correspond to the C-N stretching frequency of the tris(2-cyanoethy1)phosphine and the C-0 stretching frequency of the bridging carbonyl? respectively. Very weak bands from the Ni(CO)2[P(CH2CHrCN)&were also present in the spectrum. Magnetic Properties. Magnetic measurements were made by the Gouy method. The magnetic susceptibility was measured at temperatures from 80 to 305°K and at field strengths from 2000 to 6700 oersteds, A considerable field dependence was observed over the range of field strengths, suggesting the presence of a ferromagnetic impurity. Nickel metal has been observed in the reaction flask during preparation of the compound, very likely resulting from the decomposition of Ni(C0)4 or the disubstituted carbonyl. The insolubility of the cornpound has prevented the removal of the ferromagnetic contamination. The nonlinearity of the CurieWeiss plot over a temperature range of 80-305X prevents any meaningful calculation of the magnetic moment. A mean value of xS = 1.64 x 10-6 was found at 19” which is close to that (1.8 X 10V) found previously. * An attempt was made to determine the magnetic behavior in acetone solution by observing the proton magnetic resonances of the cyanoethyl groups. but a saturated solution was too dilute to produce an observable resonance. Collection and Reduction of X-Ray Data. Single crystals were examined by precession photography and were found to belong to the rhombohedral system. The reflections were indexed with reference to hexagonal axes in the obverse rhombohedral setting. The space group extinctions are - A k I = 3n. The diffraction symmetry indicated the space group t o be R3 (No. 146) or RS (No. 148) with unit cell dimensions a = 18.78 i 0.02 A and c = 26.12 + 0.02 A. (The rhombohedral cell dimensions are u 13.89 A and a = 84.98”.) The density measured (by flotation) in aqueous barium iodide solution is 1.45 glee. uhich agrees well with the calculated density of 1.47 gicc for six molecules per hexagonal unit cell. The compound was found to decompose rather rapidly on exposure to the X-ray beam, and it was thereforeconsidered that threedimensional data could not be satisfactorily collected by photographic methods. Initially, an attempt was made to determine heavy atom (Ni, P) positions using two-dimensional precession data from the hkO and Okl zanes. To decrease the error introduced by decomposition of the compound, different crystals were used for collecting data from each zone. For each, a series of timed photographs was taken. However, when these data were read, it became apparent that considerable decomposition had occurred, and reduction of the data was not deemed worthwhile. The rate of decomposition of the crystals suggested, however, that a useful set of intensities could be collected by counter methods. Intensity data were thus collected at room temperature on a G.E. XRD-5 diffractometer using nickel-filtered Cu K a radiation. The intensities were monitored with a scintillation counter with the pulseheight discriminator set fer 95 % with the window centered on cuK~ radiation. Three single crystals, each in the fGrm of a Parallelepiped, were mounted with [W11 parallel to the 9 axis ofthe diffractometer. Reflection intensities were recorded from each crystal until the intensities of two standard reflections 0, 1, 2 and 1, 10, 0
+ +
7
decreased by approximately 10% of their original value, at which time a new crystal was mounted and data collection continued. The moving crystal-moving counter method of Furnass was used, but so modified that background counts were taken at the maximum and minimum values of 28 for each scan. The diffraction characteristics of the crystals indicated that a 40-sec scan of 2.66” would adequately record the integrated intensities. The intensity of each reflection is given by the total number of counts during the scan minus the sum of the two 20-sec background counts. All independent reflections (1860) within the range Ocu 5 50.4” were measured. The diffractometer settings and Lorentz-polarization corrections were calculated with Shoemaker’s MIXG-2program.6 The intensities were reduced to iFi,al’ and ‘F,i,lz by Elder’s program, RAWRE-2.‘ Solution of the Structure. For the purpose of calculating the Patterson function, all reflections from the three crystals were put on a single scale based upon the intensities of ten reflections which had been measured from each crystal. The possible positions of the general nickel and phosphorus atoms were found from the Harker plane ( x , y , 0). Vectors from each of these nickel atoms to the nickel atom on the threefold axis indicated the presence of tetrahedral arrangements of the metal atoms and, further, that adjacent tetrahedra on a threefold axis are related by a center of sqmmetry. It was not apparent from the Patterson map that the phosphorus atoms were also related by a center of symmetry, since vectors between center-related phosphorus atoms were either much weaker than other phosphorus-phosphorus vectors or did not appear. However, after a series of trial-and-error Fourier electron-density maps phased on the nickel atoms and an incomplete set of phosphorus atoms had been studied, it was seen that each phosphorus atom was actually related to another by a center of symmetry. Two successive difference Fourier maps then revealed the positions of all the remaining atoms and indicated that the proper space group was R3. The subsequent successful refinement of he structure has substantiated this choice. A structure factor calculation based on the atom positions found from the difference Fourier map had the residuals
For the refinement, all reflections were rejected for which ~-~ v‘(P $- B)/(P - B ) > 0.5, where P is the number of counts measured during the scan, and B is the total background count. Of a total of 1860 reflections measured, 843 were rejected using the above criterion. The atom coordinates, three scale factors, and isotropic atomic thermal parameters were refined by a full-matrix, least-squares procedure, minimizing Z w [ F,,; - ~ F C ~ l Atomic-scattering 2. factors for neutral Ni: P, C, N, and 0 used are those given by Ibers.8 The real and imaginary terms for Ni and P9 arising from anomalous dispersion were included in calculating the structure factors.10 No correction was made for absorption of X-rays by the crystals, since prmaxwas less than 0.15 ( p = 18.7 cm-1 for Cu K a radiation), After four cycles of isotropic least-squares refinement, the residuals R1and R2 decreased to 0.114 and 0.1 33, respectively, Since a unit weighting scheme has little justification in practice, an attempt was made to introduce a scheme more closely related to the experiment in which the weight, w , of each reflection is approximated by the expression [0.95 - 0.195iF0’~ 0.038~F,’2]-1. Applying this expression, another cycle of refinement resulted in values of R, = 0.107 and RP = 0.127. The weighting scheme was then checked by plotting [ ~ F o i ‘FC,l2 as a function of IFo!. It was found that the weighting scheme would be more satisfactorily __-~
+
(5) T. G. Furnas, Jr., “Single Crystal Oricnter Instruction Manual,” X-ray Department, General Electric, Schenectady, N. Y., 1957. . (6) D. P. Shoemaker, “MIXG-2, .M.I.T. X-ray Goniometer Package,”
__ 19(674).R. c. Elder, Ph.D. Thesis, Massachusetts Institute of Technology. 1y6L,
(8) J. A. Ibers in “IIlternational Tables for X-ray Crystallography,” VOI. 111, I 12.0, w = 12/(3.3 1.151F0’!). In the expressions for the weight of a reflection, F,’ is related to the structure amplitude on an absolute scale, IF,l, by IF,’l = 1/31F01 X (scale factor). A further cycle of refinement produced R1 = 0.107 and R2 = 0.112. In order to determine the effect of crystal decomposition on the data, the reflections were groupedaccording to the crystal fromwhich each was obtained. Each group was then used separately as data for a cycle of refinement in which only the scale factor and an overall temperature factor Bo were allowed to vary. The initial value of Bo was set at 2.0 A2, which also was subtracted from each atomic thermal parameter. The initial and final values of the parameters are shown in Table I. The small divergence in the three over-all temperature factors indicates that the effect of decomposition was practically equivalent for the three crystals, and it is clear from the behavior of the scale factors that no compensation was necessary to bring the three sets of data to a common scale other than individual scale factors.
+
Figure 2. A perspective view of the Ni4(CO)6[P(CH2CHzCN)3]4 molecule. Numbering of atoms corresponds to that in Tables I1 and IV-VI.
Description of the Structure A crystal of Ni4(CO)G[P(CH2CH2CN)3]4 is composed of distinct molecules stacked along the threefold axis of the space group with adjacent molecules in the stack related by a center of symmetry. Table 11. Final Atom Coordinates and Thermal Parameters“
Ni(1) Ni(2) P(1)
Table I. Refinement of Over-all Temperature Factor and Scale Factor for Data Sets from Three Crystals Crystal
-Bo, Initial
A2Final
-Scale Initial
factor-Final
1 2 3
2.00 2.00 2.00
1.98 2.04 1.97
0.232 0.281 0.307
0.232 0,283 0.306
The final cycle of refinement using all the data produced no change in the residuals: RI = 0.107, RZ = 0.112. The largest parameter shift in the last cycle was 1/4thof its estimated standard deviation. The largest feature on a difference Fourier map calculated from the final atom positions was a peak of 1.2 e/A3 near atom C(8). This together with the high temperature factors and poor bond lengths associated with atoms C(7) and C(8) suggest some disorder in this cyanoethyl chain. The final atomic positional and thermal parameters and their estimated errors are listed in Table 11. The estimated standard deviations were calculated from the usual least-squares formula, u 2 ( j ) = a,,(ZwAZ)/(m - n), where a j j is the appropriate element of the matrix which is inverse to the normal equation matrix. The observed and calculated structure amplitudes times 12 are listed in Table 111. Fourier and least-squares programs used in the calculations were M I F R - 2a ~modification , of Sly, Shoemaker, and Van den Hende’s Fourier program, and Prewitt’s full-matrix least-squares crystallographic program SFLSQ-3. l2 (1 1) W. G. Sly, D. P. Shoemaker, and J. H. Van den Hende, “A Twoand Three-Dimensional Fourier Program for the IBM 709,7090, ERFR-2,” 1962. (12) C. T. Prewitt, “A Full-Matrix Crystallographic Least Squares Program for the IBM 709/7090,” 1962.
Journal of the American Chemical Society
1 89:21 J October
0.0 0.08495 (16) 0.0 0.20235 (30) -0.0066 (11) 0.0513 (11) 0.0451 (14) 0.2978 (15) 0.2913 (18) 0.3764 (15) 0.2249 (15) 0.1582 (19) 0.2025 (31) 0.2305 (11) 0.2321 (12) 0.1501 (14) 0.1136 (11) 0.0234 (13) 0.0425 (13) 0.4375 (20) 0.2084 (1 8) 0.0878 (13) 0.1721 (8) 0,0363 (7)
0.0 0.06592 (16) 0.0 0.15951 (31) 0.0841 (11) 0.1675 (11) 0.2313 (14) 0.1704 (16) 0.0895 (18) 0.1067 (21) 0.1571 (15) 0.1457 (19) 0.1668 (30) 0.2688 (11) 0.2977 (12) 0.2821 (13) 0.0825 (11) 0.1101 (13) 0.2811 (14) 0.1114 (18) 0.1622 (18) 0.2701 (11) 0.1204 (8) 0.1686 (7)
0.21778 (18) 0.29606 (11) 0.13493 (31) 0.32556 (20) 0.1012 (6) 0.1254 (7) 0.0995 (9) 0.2982 (9) 0.2929 (11) 0.2741 (12) 0.3916 (9) 0,4235 (13) 0,4820 (22) 0.3195 (7) 0,2651 (8) 0.2487 (8) 0,2260 (7) 0.3247 (8) 0.0750 (8) 0.2635 ( 1 1 ) 0.5218 (13) 0.2360 (7) 0.1991 (5) 0.3504 (5)
2.9 (1) 3 . 0 (1) 3 . 2 (2) 3 . 6 (1) 4 . 2 (4) 4 . 2 (4) 6 . 2 (6) 7 . 6 (6) 9 . 5 (7) 11.1 (10) 7 . 2 (6) 10.0 (8) 17.6 (17) 4 . 2 (4) 4 . 8 (4) 5 . 4 (5) 4 . 2 (4) 5 . 6 (5) 8 . 8 (6) 12.6 (8) 1 2 . 2 (8) 6 . 7 (5) 5 . 3 (3) 4 . 2 (3)
Numbers in parentheses are estimated standard deviations and occur in the last significant figure for each parameter.
The closest intermolecular contacts are between cyanoethyl chains of neighboring molecules. The shortest of these, 3.21 A, is between the nitrogen atom of one chain and a methylene carbon atom of another. All contacts up to 4.5 A have been c a l c ~ l a t e d ,and ~~ pertinent values are listed in Table IV. A projection of the complete molecule down the c axis is shown in Figure 1. A perspective drawing, in (13) D. P. Shoemaker, “DtSTAN-CryStallOgraphiC Bond Distance, Bond Angle, and Dihedral Angle Program,” 1963.
11, 1967
5369 Table 111. Observed and Calculated Structure Amplitudes Times 12 H -3 b 12
1
-2 4 -5 7 10 -1
K fOBS fCAL 6 486 753
H
6 1257 6 887 7 1221 7 1612 7 1160 7 589 1 987 7 806
1161 793
3
362
6 9 12
4083
8 716 8 1279
664 1120
-4
-; 8:2 %191 8
3 -1
-6 1
-5 -8
1877
9
9 682 9 788 1 0 15*1 10 .I8
l o 799
-1
I591 PO6
11 11 11 11 -3 12 -6 12 -9 12 -2 13 i 13 -5 13 -11 13 -1 1 * -7 Ir -10 1 * -3 15 -6 15 -5 16 2 -4 -7
637
hi3 914 51L 1905 19LO
0
1051 1634 1372 233
15 1
805 217
4
7
IO -1
:
la:;
0
662 1667 679 783 1636 90
3 -3 6
I2
1
870
-2
3179 1147 563 79 1241
7 10
+
-1 2 -I
bo
1688 1989
5 8
974
583
1Sl5 $LO
1709
11 0
,799
3
11122
1153
951
-3 6
987
517
-6
Pi* 1859
1045
1
907
2052 $97
-4
,I
-7
17 1171 18 1536
-2
1182
-5 -1 2
2OOk
995 ,301 1525
.... ...... -K
L
1 4
7
-C
5
1.
i
0 1001 0 1019 0 1827
1113
-7
856
0
177' '*0
fOBS
fCAL
6452
6471 18 314b 1195
1253 770 401 975 142 664 1028 5682 6014 1551 1664 1076 1088 408 426 1737 1569 2211 2 3 7 6 1163 iiai 3185 3237 13i9 1415 1181 1066 987 894 911 13b 952 952 1509 1b76 1239 1075 830 6L3 711 *9* 2403 2235 925 785 1610 1*91 927 736 l O L b 1138 2577 2716 1061 980 2 3 8 9 2378 1533 1370 1110 1195 1157 1101 1561 l51L 2000 2254 227* 3160 z 3 n v 2219 2131 2189 8OI 651 I903 lBli 812 812
3 -3 6
,*is
1518
515 983
-6
23&* 1607
686 761 233a 1611
d
I 1307 -4 8 1053 -1 8 1698 I 8 976 0 9 666 3 9 749 -3 9 921 -b 9 660 -2 10 2622 -8 10 616 -4 11 2 7 0 1 -7 11 * 2 1 -10 11 5 5 1 -9 12 I593 -2 13 762 -13 1* 898 15 896 -3 -6 15 1 8 2 -I 16 1 1 1 5 -L 17 1543 I I
10 13
2 5 11 0 3
6 9 1
-2
+ 7
-1 2 D
-3 9
12 1 -2
13
0
2
1
5
1
56
1
8 0
1 2238 2 1551
2112 129%
-2
3
2 ,137 2 3500
?I35 3 ~ 8 0
-8 2
i:ii 1027
-7
9'6
-10
2353 1095 13*D
2312
lD*l 1130
-2
939
539
1131
-5 -1
6 9
2
ii 1 -2 7 10
z
13 -1
1
2
1.
5 3 -1
6 5
L
12 1
-2 4 -5
-1 -i
11 1
-3 -6 9
1
-2 i
-5 -8
-1
-* 5
-7 D
-3 -9 -5 -8
-L
-7 -9 -12 1 -4 -6
-11
1.1.L
2 5
e
11
14
8119 280
e71
,e)
3 1129
3 3 3
+
-*
**a
954
657
1365 1007
1197
1127
693 1511 907
1626
382 806 825 988 3288 2300 20*0 il'6
828 3086
5 2IIL
2318
-*
1166 1060 2213 695 1070 1191
5 1218 5 1'50
1267
-13
977
L
313 589
221Q lBl* li*S
'7'1 L ,1*P9 1 * B 1 502
143
d 2920
2962
L 1220 7 1260 7 ,689 7 2318
1359 ,598
d
B 2761 B 2077
8 1066 B
1 -3
1C5L
-6 -9
-12 -5 -8
-1,
-6
-12
1"l ,:s7 25LB
... -8
I
0 3
6 9
2094
:2 i
911
6
7
988 672
623 33s
10
9 2981 9 881 9 2037
3212
-1
9
P ,155 10 1317 i o 161s LO 1 5 7 2 10
395
11 1602 11 2 1 8 7 11 9'7 12 6 7 9 12 151' 1, 1c29 1) lil.3
BIB
2
,
-4
-7 3
1 3 1'93 I * Pi9
3475
-9
1410 1637 iU6
.12 1 -8 -8
19 9 4 1 l * 1366 15 IC72 1 5 1028
-8
156 2**
-11 -1
D
3
1396
6 1
2226 475
2511
0
1 3511
631
3
2121 1995
2191 1663 93
6 9
1 1290 1 3682
2503
5
8
11 14
QBii
2155 L71
1171 ,669 719
726
IO
813
96,
975 225
809
171
7
3105
2175 115 2077 2261
-1 2
9'0 926
-4
553
5
3091 3016 1336 1713 1193 599
= 5...,*. 0 2231 2199 0 2211 2 3 4 5
1652 1072 617
0
3 -3
6 1
-2
632
-5
1137 2065 986
c
10
1313 1458 995 258 801 1133 1901 935
1
* 10 13
5 8
I1 0
3 -3 6
1 1512 189
2 3097 2 1565 2 1113 3 191, 3 2536 3 3295 3 1664 3 686 & 911 4 2973 4 1181
4 2005
P ..
297*
0 1121 0 1595 0 1532 I 1169 P 1926 1 *36
3'76
806
2 2184 2 3167
-2
-51
-1
-4 2 0
1313 1304
140' 1784
212 867 29G3
3391
foes FCAL 136 1163 996 279 850 159 793 I65 413 1816 465
535
1228
-7
629
159
~ L P S 1580 1801 1131
889
78i
-2 1
969 661 l28i 1231 ,069
306
3
-1, -1 -1 -7 - 1-9 2
-5
-le -4 ...I),
I
112,
1164 1618 964
I 2
0
- 33
728
118
1515 2011
-2
7 2019 7 1131 I 736
2109 1069
,262
-5
7
2,s'
2097
16*8
IO
1511
$73
-1
i 1113 8 202,
1207
2 -4 5
B 1982 8 2356 8 831
2060
-7 B
8
IlCL
-52 1'20 2317 1199
9 lS03
117s
2522
1111 ,089
815 la*?
118
255
2369
6
6
"
98,
8 1275 9 1212
121 777
3 -3
9
671
-6
560 1430
1 -2
9 10 10 10 '10 11 11 11 12 12 12
1367 730 81' 1317 424 901
318
3461 1172 3532
1515 89 2919 2762 1503 783
i
-5 -1 -6 -7
-3 -6
-9
-5 .*..L 1 4 7
10 13 2 5
0 3 L 9
1811
12
1578
1
1334
-2
1401 5
973 3094 1011 2113
which six of the cyanoethyl chains are omitted for clarity, is presented in Figure 2. The intramolecular bond lengths and bond angles14 and their estimated standard deviations are listed in Tables V and VI. The molecule contains a tetrahedral cluster of nickel atoms with phosphorus-bonded tris(P-cyanoethy1)phosphine ligands at the apices and bridging carbonyl groups at the edges of the tetrahedron. The space group requires the molecule to have threefold symmetry, but it is only slightly distorted from tetrahedral symmetry with the exception of the arrangement of the cyanoethyl chains of the phosphine ligands. Each nickel atom interacts with seven neighboring atoms forming bonds with three nickel atoms, three carbon atoms, and one phosphorus atom. The average nickel-nickel bond length is 2.508 f 0.004 A, and the values of the two crystallographically independent types differ by an amount (0.0056 A) of the order of only one estimated standard deviation. Consequently, (14) J. S. Wood, "MGEOM, A Molecular Geometry Program for the IBM 7094," Massachusetts Institute of Technology, 1964.
7 -1
932 il$L 2193
2692
2958
l"6 276,
27LI
811 2558
1159 PO67 1638 170; 137
.
1'07 1 3 1198
...I:
0 iil9
211 2675 1013 3382
,611 ldll
1288 1ZR6 1311
5iB 0 1399
"42 747 1>87
,591
210
1380
,086 2866
0 0
o
1 3113 1 1789 2 3165
12
- 21
480
253
1?98 263
1 1 1878
-* -7
8
1st.
3 0
7 -10 1
52 11 0 3
-1
ero
2
978 1692
-&
817
-1
1**8 1789 835
- 30
815
-12 -5
916 2183 1960
-11 -10
-8
10 10 10 11 I1 12 12 12 12 13
19 0.89 5 1358 842 977 1362
1086
7
-2 -5
2 7
21 ,473 880
,710 690
1201 1712 l5E5
8 5 0 3 6
I 2II5 2 1293
233' 1355
593
i
3 b18 3 911 3 2*15 (i 1117 5 1191 6 660 6 838 6 971 7 1757 8 180 8 1155
396 1096 2770 167T
1528 36*
13 1525 14 1657 I* 1283 1 * 1522 15 1*17
1114
652
-1
I646
0
6
733
*I*
767
1
I
1141
-2 4 -5 -6
1321 1121
I
4 1510
5
756
5
:rii
1663 2590 318 ,969
3 -3 -6
-2 -5
26!3 912 i191 i795
-e
!lii 1 1 2 5
-7
:Pbl
2 5
-6
576 952
1351
-8
0 -9
1667
1571
23'9
2361
-7
1970
iCS7
J?CC
:i*i
-3 -11 -5
1665
ldll
i51'
15'1 Oil I>.". 1801
-539
.
3161
1594 852
8'3 19es 3298
1757 l2b2 l5Il 1575
-i
"..tL
2 5 8
11 0 I
6 9 1
962
4
Et1
LCS
3119
3142
3578
3981
6l6 982 1808
776
7 10 -1 2 5
e87 1487
519 1714 529 882
1684
o
3 -3 4
1737
8 2756 e 629
9 lil* 9 931
P
7*6
10 118C
1368
-i 1:
idll 1785 121
lil0
e
10
bo
11 11
898 677 963
853
-3 88, i i i ~ -L -6 525
iiir )Oh1 515 1525
7 1110 7 BPS 7 265, a 723 8 676
821
71: 1215 12'8 ,008 5'9 1218 106
1 1 122'1 1 2 1566
1151
13 1 3 2 6 13 9 6 6
1+;2
1525
3
876
1
I
ll0L
16'0 1216
1 2285
ic99
1 26*3
2Q29
1956
15'9 2 2136 2302 2 2721 3211 2 i?' 626 1 1334
1:5 8:I 1118 ,092 115' LBi
3 571 3 1278
4 7
-1
1'B*
896
,931 1386
1L:7
' I 0b 0l 5L L
' , 6'350 2 9
5
594
i ,191 6
919
-2 -5
6 :,PI
-1
7 2 x 6
-6
7
3
8
61
-*- 2
607
-7
,110 1119
-6 I . 8 . L
6
i:bb 055 969
156'
lo98 175i 67,
212 9''
99 ,655 96'
zoca e47
I ?L91 B9 9,s e61
557 221? 5nc ,I!
9 1CR7
1095
1: >:>I I: eel 13 1 2 6 4
109s
=
957
1703 1i . . .1..
2701
n
1242
BSO
eio
2
L
i
0
982
4
1250
,036
0
1 1363
L 1237
1'06 1960
3
1
I
1'1
1 2192
1
-3
1152
3 1'51
6
8s.
5 19Lh
,060
1
3 3206 3 770 3 2716
2
1165 1221
ier9 693
2
1'180
360
BPf
5 0 -9
130' 0 1 '68 o iisi 1 1921
56,
2 2648 2
....
1 lD90
in19 0 ' 8 9Bi
2 2
-2
1 El5 3 1228 5 Ill8
61"
6 9
z~
-2 0 -2
2818
0 3
812
0
B98
1 2'30 I i z
980 1296
830
e
1200
2696
0 1286
1886 2060
5
li
.
lies
1
I4
806 736 15 i 2 i T ......il 0 2113
:a
.... I0
11'3 732
I36* 1190
IL 8 7 7 1 2 1269 14 , 2 6 8 1 5 ,187 L .=..'..bl
965
28Y
970
1530
-L
i 2280 7 862
722
906 .'.*91
8 lL83
10 !CIP
6
5
.
--i 2 10 9 ,I205 396 1 .=.....nj
6
io
8
P 11
1 965 7 1236
2005 800 1*03
-i
-7
-i
6 1916
-1
-7 -3
L..*.
-3
' -5
6 llli
-6
820
5&2 2l8
-1
-7
-2
323
1851
876 1153 1-36 1517
llli 2300
716
770
906 1065
928
139 2110
763
6
0
1628
L
-2
5 -3 6 -2 -1
,516 5 252, 5 66*
II
691
1566
880 ?I79
2
I*82 9Lb
' I956
1600
>SO2
2256
L
2262 $31
.... . ...... -6 -5
679 122' 118* 1305 1208 1210
3 2178 3 1789 3 lCdl L ill* 719
1596
1109 el5 2'03 23:e 1 0 3 1 1025 > i l l ,712 i o & & $91
22i8 ziie
9 1 -2
5 654 5 1066
2792
2097
6
1193 1991 636 2587
I 2591 2Lil I 1'150 1 1 3 & 3 1022 735
5 2L2L
9 i
1210 3053
I951
1991 702 2886
7 10 -1 2 5 8 0 3
1206 1291 ,192 1099 603 1871 11 118b
,604
851 ,481
0
6
827
-1 3 -3
15
1 22.2 1 1112 1 901 1 799 2 695 2 1511
1086 "1
61
881
1688
1008 1077 912
0 1199
LO 891
1061 1263 1*71 456 179 1593
802 9 8I 7 1217 8 1327 8 1142 9 966 15 168 10 1589 10 630 I 1 ,061 L 18 D1 1465 866
4 836 4 1325
ill4
1639
1153 2538 2025 2661
I
1009 175 1319 797
1167 1087 987
0 616 0 liZ2 0 2116
4
6 134 6 1630
-1 2
2702
1907
61
4 5
3
973 612 801 903 1313 lil0 617
7
899 ,286 686
i
0 3 6 12 1
1583 1975 1336
6
.... . .....
766
1150 1706 1543 16bP
1067 1001
3
-30 -6
1617 1'19
818
-2 4
.
785
1425
I05
16 16 14 15
,374 2811 2325
18:Y
-7
3 0
-3 -6 1
FCAL
2 1108
1620
1619
-9 -6
-10 --9 5
921 1339 1030 2208
1. F09S
3
5
7
Z'iB
--i e
-1
7
-*
4
-3
- 21
5
H -1 2 8 3 -2 -1
1 -2
191.
2498
50
0 9
-4
lr2
1599 1102
2
622 ioii
- 42
-I
id07
811 2116
7
P
1
1 -2
811 la19
10
-3 6
5
b -6
367
I II
239 962 7*5
3202
1
5
199L
1 . 1 .
1315
-2
1351 731 1792 1613 1094 1010 1186 181 944 1411
1930
-9
-12 -11
1509
P - 39
fcAL 765
9
-10 -13
2889 17G6 2617
101
14
5 961 6 I660 512 6 1603 7 1701 1 1018 1 101" 1 1367 9 763 9 1144 9 1385 9 1751
5 0 6
974
237
6600
6 1300 5 333
-I1
3756
1
4
1 3 20113
176
85
5 0
1119
1065 I863 996
-8
864
1015
1915 1871
2
1082 1173 1041
3968
9
806 559 1098 1074 2998 586 130
1250
762 * li*.**** 920 ,,3
12
2'70
1195
9 973 9 1396 1 10 1065 L 1" 2017 -5 1 0 111 -n i o 1058 -1 1 1 1>81 2 11 1193 5 1 1 153L -10 11 7 1 0 12 1 2 8 6 0 3 12 1613 -6 I 2 931 1 13 1091 -2 13 1061
K pons
1
1920 1602
,066 897 799
1816
2 1035 2 985 3 1320 3 853 3 2700 3 685 4 627
629
-3 6
:1 ::
H -1 2
1300
IS69
3098
2 157 2 2171
-1
-' 2
213
irC8
-5
*19 1111 e26 2289 169 1686 431 1216
111* 1479 1676 li3h 15 19b6 1I 1 3 6 1 L . ...*..,I 0 1312 0 1166 0 1888 0 BE8 1 1306 1 2417 2 2035 2 2539 2 9r5 3 1967 3 862 3 2198 3 ll0I
lBb8
6 2325
d"8
1727 1138
0
-1
6
267 1062 699 1281 551
-P
9
I
-3 4
5 41b 5 1143 5 789 6 2211 6 791 1 1628 1 707 8 1118 :1 :;
2 -4 5
13
3i50
5 8 3
K FOBS F c i L 5 414 36 5 860 I86
1163
6 3
1058
756 932 707 49* 11 1971
2 -k
-6
0
isi
I04
H
-I
-2 -8 -5
5
i i iii
831 1333 951
;;;: :::: -; :
-8
0 1671
1
3 6 1
1063
1366 55L
714 1600 a99
2 2939
984
1718 1659 101P
?Be2
330
2
1083
5 986 6 2928
711 " 1679
e
I5h*
11 0 3 -3 6
808
0 897 3 1SP5
llnl 1*71
1s
i =
2 8 11 '1
-2 LI
158
2lW
.... 2 ......
!LJL >*I
1160 1193 1181
553 1605 11 11.81 1057 18 895 627
0 1290
2
620
891
6918 2166 659 1330 1941 706
20'9
16 ,395
o lose
2099
'
...
"56
1069 1209
811 9 1331 9 78' 2792 13 1603 10 1978 10 i l P 5
-5
731 1002 610
2'20
9
121
928
1283
9
-6 -9
619 1053 1170 969
'8
1
1 1 1352 1 1 2640 1 1 920 11 PO2 12 Pi9 1 2 1901 12 161 1 2 PI0 13 8 7 2 13 979 13 1050
4 1
.
1315
720 1563
H
708
1077 13'
13 1 5 8 5 11 B L I I
-6 -5 L+ ..
1012 1013 1963 1601 1111
fCAL 699 933 2282 680 1091 1440 1295
$31
B 1563
i
13 13 1* 15 16 17
-9
-5 -4
13 -1 2
2,ai
7*0
14
1610
860
(111. .
12 150%
-3
963
20b9 1902 1895 665 126i 2021
7
-,o
9 1
a 2090
6 2091 6 248 6 1216
. 1.,-
b
1124
s
-1
856
101.7
5 2GO3 I 1191 5 776 5 1197
846 1051 5 2687' b 942 5 6 925 3 7 1318 -3 7 1153 6 7 836 -6 1 1631 -5 8 109 2 9 158 3 10 1315 -3 10 1179 6 10 1200 -6 10 665 1 I1 1301 -8 11 5 5 2
121 1901 869 801
926 1 1563 3 1109 L 622
K fO0l 4
4 2
0 3
2097
685
1455 1529
700
H 9 -1
6960 536
8 ,926 8 2161
-7
4113
613
8 2198
LO3
,895
537
0
Pi9
798
7011
**I 1197
-1 -6 1
1166
1630 550 3G10 lCh3
1 I
1299 lb71
2067
0
1131 839 308 2113
0 616, 0 1228' 0 836 0 936 0 962 1 6695 L 615 1 7*8 2 e75 2 1809 2 I55 2 600
727
7 1113 7 2119
3 -3
a
236
211 1630 883 821 190
6 2362
-6 9
.....l
955 1021 570 2932 116 2686
2
-*
2117
950
501
-5 -1
1368 1058 1091
16
o
6
ID58
13 -1 2
716 1117 15 1190
BO5
*
*L
1
0
7C9 3216
1290 931 1363 969
... ,......
10
567 Sib
I! FOBS fCAL
-1
601
$07 3096
137, 159
z 1507 isei
within experimental error the nickel atom cluster has undistorted tetrahedral symmetry. The average nickel-phosphorus bond length is 2.161 0.005 A, and the tetrahedral symmetry is slightly distorted by the phosphorus atoms. The interatomic distance between phosphorus atoms related by the threefold axis is 6.008 f 0.011 A, and the distance from one of these to the phosphorus atom on the threefold axis is 6.067 f 0.008 A ; thus there is a slight elongation of the tetrahedron composed of phosphorus atoms. Since there is no apparent chemical reason for it, this distortion is assumed t o be due t o intermolecular forces in the crystal. The tetrahedral symmetry of the molecule is further destroyed by the arrangement of the cyanoethyl chains. Those attached to the phosphorus atom in the special position (on the threefold axis) are related by the threefold axis. However, those on the three phosphorus atoms which occupy general positions have no internal symmetry. Thus the chain composed of atoms C(lO), C(11), and C(12) folds back toward the molecule, while the other two
*
Bennett, Cotton, Winquist
1
Crystal Structure o f Ni4(CO)6[P(C2H4CN)3]4
5310 Table IV. Intermolecular Contacts
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Atom in given molecule
Atom (2) in neighboring molecule
C(1) C(1) C(2) C(3) C(4) C(4) C(5) C(5) C(6) C(6) C(7) C(7) C(8) C(8) C(9) C(9) C(9) C(9) C(l0) C(10) C(10) C(10) C(11) C(11) C(11) C(12) N(1) N(2) N(3) N(3) N(3)
N(2) N(2) N(2) N(2) N(1) N(1) C(10) N(1) N(1) C(10) N(1) N(2) N(3) N(3) C(9) N(3) C(8) N(3) C(12) N(4) O(2) N(1) N(3) O(2) N(1) N(3) N(2) O(1) N(3) N(4) O(2)
Table VI. Bond Angles, Degrees Interatomic distance, A
Coordinates of atom (2) relative to those of same atom in given molecule (Y
-
('13
('/a ('15
('13 ('13 ('18
('13 ('13 ('13 ('13
('13
+Y -
-
Z)
-
Z)
+x - y, x - Y , + x - Y, '13 + - Y, + x y, + +x y,x - Y, + x - ,v, +
X)
'/3,
-
'13
X , '/3 .Y,
'/j
X,
'13
X,
'13
- Y, '/3 - Y , '13 - y , '13 -
+x - Y, + - Y, + s '13
'13
Z) Z)
i)
.Y, '/a
.Y,
'13,
'13
+
'13
'13
-
'/j
-
'13
'/3,
'/a
+ y , '13 + y - x ,
Z) Z)
Z)
' / 3 - Z) '/3
Z)
2/3 - Z )
(y,y -
X, 1 - Z) - JJ, X , 1 - Z ) (x - y , x , 1 - z ) ( X - y , .Y, 1 - Z ) (y,y - X, 1 - Z) ( y . y - .Y. 1 - Z ) ( ' 1 3 - X , '13 - ,v. 'I,{ - Z ) ( ' 1 3 - X , '13 - y , ' 1 3 - Z ) ( ' 1 3 - x , ' 1 3 - y , '13 (*/3 y - X, '/z X , '/A - Z ) ('13 y - X , ' 1 3 - X, z - I/,,) ( ' 1 3 - X , '13 - y , ' 1 3 - Z ) ( ' / j x - y , '/a X, '/:I- Z ) ('/j y - X. '/: - X, z - '/a) ('13 y - X, ?/.i- S, z - ' 1 3 ) ( ' 1 3 - X, ' 1 3 - y , ' 1 3 - Z ) ( y , JJ - X , 1 - Z ) ( ' 1 3 - Y , '13 .Y - y, ' / 3 Z) ( x , y - X, 1 - Z ) (X
+ + + + +
+
+
+
+
3.64 3.94 3.30 3.63 3.54 3.92 3.91 3.47 3.52 3.93 3.77 3.85 3.91 3.21 3.64 3.40 3.87 3.68 3.75 3.68 3.84 3.59 3.32 3.58 3.21 3.40 3.88 3.76 3.74 3.85 3.40
Atoms forming angle
Angle, deg 60.15 59.93 144.64 146.22 143.00 145.01 82.2 137.4 140.3 84.6 137.6 137.8 118.1 119.5 118 9 118.7 111.0 110.4 111.5 114.8 110.5 105.7 101.5 111.5 174.2 171.9 158.9 179.6 96.4 96 9 95.8 96.9 99.6 98.5 99.8 97.0
Est std dev 0 . I7 0.09 0.32 0 . I8 0.20 0.20 0.7 1.5 1.5 0.9 1.7 1.7 0.7 0.8 0.8 0.6 1.2 1.8 1.9 1.3 1.6 2.4 2.9 1.6 2.4 3.4 4.8 2.2 0.6 0.6 0.7 0.7 0.9 1.1 1 .o 1 .o
Table V. Bond Distancese Bond
Distance, A
Ni( 1)-Ni(2) Ni(2)-Ni(2) ' Ni( 1)-C( 13) Ni(2j C ( 1 3 ) Ni(2)-C(14) Ni(2)'-C(14) Ni( l)-P( 1) Ni( 2)-P( 2) C(13)-0(1) C( 1 4 t 0 ( 2 ) P( 1t C ( 1) ~(2)-c(4) ~(2)-c(7)
2,506 ( 5 ) 2.512 (6) 1.922(18) 1.889(19) 1.881 (22) 1.852 (22) 2.163 (IO) 2 . I61 (6) 1.194(23) 1,206 (25) 1.866(19) 1.844 (25) 1.783 (25)
Bond
Distance, A
P(2)-C(10) C(I)-C(2) C(4)-C(5) C(7)-C(8) C(lO)-C(11) C(2)-C(3) C(5)-C(6) C(X)-C(9) C(ll)-C(12) C(3)-N(1) C(6)-N(2) C(9tN(3) C( 12)-N(4)
1.852 (19) 1 . 5 3 (3) 1 47 (4) 1 , 4 3 (4) 1.52(3) 1.43 (3) I , 54 (5) 1 .69 (7) 1 .48 (3) 1.16 (3) 1 . 1 4 (5) I .05 (7) 1.12 (3)
The bond lengths found between other atoms of the cyanoethyl groups are close t o those expected with the exception of that group composed of atoms C(7), C(8), C(9), and N(3), which shows evidence of disorder. Thus the bond lengths and errors calculated for these atoms are not considered reliable. There is no evidence that any coordination of the nickel atoms by nitrogen occurs. The closest nickelnitrogen distance is 4.12 A between N(4) and Ni(2). The bridging carbonyl groups are symmetrical; each carbon atom is equidistant, within experimental uncertainty, from each of the two nickel atoms to which a Numbers in parentheses are the estimated standard deviations it is bonded. The mean of these Ni-C distances is and occur in the last significant figure of each bond length. 1.890 i 0.003 A. This figure is similar to those found in other cases, e.g., in dicobalt octacarbonyl'j (1.92 f 0.01 A) and [(CjH,)Fe(CO)s]2'6(1.85 =I= 0.03 A). chains extend out from the nickel cluster. The arThe average carbon-to-oxygen distance, 1.20 f 0.02 rangement of the carbon chains appears to be the result A, is quite comparable to those'5a16(1.20 f 0.02, 1.21 of interatomic repulsions, particularly those between i 0.04 A) found in similar cases. the C(lO), C(ll), C(12) chain of one molecule and that Ideally, that is, for perfect tetrahedral symmetry, composed of atoms C(1), C(2), C(3), and N ( l ) located the six carbon atoms and the six oxygen atoms would at ( 2 / 3 x - y, x, 1/3 z) in the neighboring each define a regular octahedron. There is, however, molecule. Atom N ( l ) approaches within 3.21 A minor distortion which takes the form of a slight bendof C( 11) in the folded chain. ing of some of the CO groups out of the plane defined Bond angles involving atoms Ni-P-C(a) and C(a)by the two nickel atoms to which they are bonded and P-C'(a) have been calculated and are shown in Table the midpoint of the opposite edge of the tetrahedron of VI. The average Ni-P-C(a) angle is 118.7 * 0.4", which is within the range, 114-120", commonly found (15) G.G.Sumner, H.P. ]
