436 BERNTKREBS,ACHIMMULLER,AND HANSH.BEYER TABLE V POSSIBLE HYDROGEN BONDDISTANCES IN a-ZrP (ANGSTROM UNITS)
3.06 (6) 2.78 (3) 07‘ 2.82 (6) 010”-010 3.07(6) 010’ 3.07 (6)’ Next shortest distance is Ol0-07 which is 3.24 (3) A. 012-07
010’’
a
bonds in projection is shown in Figure 5 . All three of these hydrogen bonds are intralayer bonds. The remaining hydrogen bonds form very weak interlayer bonds of the type P3-010-H---010” and Pa-010”H---Olo’. These hydrogen bonds form a zigzag array running parallel to the b axis. All of the interlayer hydrogen bonds involve Pa-type phosphate groups whereas the P2 type are all involved in intralayer hydrogen bonding. An alternative hydrogen-bonding scheme involves the assumption that the shortest hydrogen bonds originate from the two P-OH groups with water as the acceptor atom for both these bonds. Thus, the bonding
Inorganic Chemistry
scheme would be P2-07-H---012 and P~-OIO”-H---O~~. One hydrogen atom of the water forms a very long bond to 0, and the other hydrogen forms no hydrogen bond in this scheme. There are no interlayer hydrogen bonds so that the layers are held together by van der Waals forces. Both hydrogen-bonding schemes are speculative a t present and the correct choice must await a neutron diffraction study. However, in either case the forces between layers are quite weak and one might expect cleavage parallel to the layers (001). This is in fact observed. The zeolitic nature of the crystals and the weak forces between layers provide a basis for explaining their sieving behavior and the ready expansion of the lattice during e ~ c h a n g e . ~These aspects of the exchange mechanism will be discussed in a forthcoming paper. Acknowledgment.-The authors wish to express their sincere thanks to Dr. Crispin C. Calvo of McMaster University, Hamilton, Ont., Canada, for many helpful discussions.
CONTRIBUTION FROM
THE
ANORGANISCH-CHEMISCHES INSTITUT DER UNIVERSETAT, GOTTINGEN, GERMANY
The Crystal Structure of Rhenium(VI1) Oxide BY
BERNT KREBS, ACHIM M ~ ~ L L E RAND , HANS H. BEYER
Received September 23, 1968 The crystal structure of rhenium(VI1) oxide, Re207, has been determined from three-dimensional single-crystal X-ray data collected by film techniques. The least-squares refinement converges to a conventional R factor of 0.053 for 1126 nonzero reflections. The compound crystallizes in the orthorhombic system, space group P212121, with a = 12.508, b = 15.196, c = 5.448 A, and V = 1035.5 A3. The unit cell contains 8 formula units. Measured and calculated densities are 6.14 and 6.214 g/cm3, respectively. The structure consists of strongly distorted Re06 octahedra and fairly regular Re04 tetrahedra which are connected through corners to form polymeric double layers in the ac plane. The double layers have only van der Waals contacts to neighboring ones. Parallel zigzag chains of octahedra, every other chain being part of the upper or lower half of the double layer, are linked through corners of the tetrahedra in such a way that rings (0-t-o-t) (0,octahedron, t, tetrahedron) within the layers are formed. The octahedra contain three short (1.65-1.75 A) and three long (2.06-2.16 A) Re-0 bonds, while in the tetrahedra the bond distances range from 1.68 to 1.80 A. The structure is one of the relatively few known examples where metal atoms of the same oxidation number occur with coordination numbers 4 and 6 in the same structure. From the crystal structure the mechanisms of evaporation (to form OaReOReO3 molecules) and of hydrolysis (where Re207(OHz)zis formed) can be derived. In its structural and bond properties, rhenium(VI1) oxide represents an intermediate between the polymeric oxides Moo8 and WOs and the more covalently bonded 0~01, which forms a molecular structure.
Introduction The very simple crystal structure of rhenium(V1) oxidel has been known for a long time and has become the model for a basic structure type. Structure and bonding of solid rhenium(VI1) oxide, on the other hand, have been completely unknown up to now and have been the object of speculation and controversy. (1) K. Meisel, Z. Anorg. Allgem. Chem., 207, 121 (1932)
A mass spectroscopic investigation2 and vapor density measurements3 have shown that the oxide in the gas phase consists of Re207 molecules which have to be formulated as OaRe-0-Reo3 with tetrahedrally coordinated rhenium. From the physical and spectroscopic properties of the compound there were (2) B. Krebs and A. Mtiller, 2. Nafuvforsch., Bas, 415 (1968). (3) J. Noddack and W.Noddack, Z . Anorg. Allgem. Chem., 181, 1 (1929).
CRYSTAL STRUCTURE OF RHENIUM(VII) OXIDE 437
Vol. 8, No. 3, March 1969 indications that the solid oxide does not have a simple molecular structure,2 which could be compared with the structure of solid osmium(VII1) oxide.* The same properties, however, showed characteristic differences from the somewhat more heteropolar bonded do oxides of molybdenum (Moos5) and tungsten (WOa6). I n the course of our investigations on structural, coordination, and bond properties of transition metaloxygen compounds we therefore studied the crystal structure of rhenium(VI1) oxide in detail by X-ray diffraction (see also ref 7). The only previously known data on the structure were the cell dimensions and the probable space group which we reported some time ago,8 essentially in accordance with approximate data given by Wilhelmig Measurement and Reduction of the X-Ray Data Suitable transparent crystals of dirhenium heptoxide were prepared by oxidation of rhenium powder in a stream of oxygen a t 500°3210and subsequent slow sublimation a t normal pressure. All operations had to be done with great care to exclude moisture because of the high sensitivity of the oxide to hydrolysis. The most frequently observed crystal forms are very thin plates (plane of the plate: (010)) and more compact prisms with the dominating planes (loo), ( O l O ) , and (001). Rez07 forms an orthorhombic unit cell with the refined cell dimensions a = 12.508 f 0.008, b = 15.196 f 0.010, t = 5.448 rt 0.005A, and V = 1035.5 A3,with Z = 8 formula units, D, = 6.214 g/cm3, and D, (pycnometrically) = 6.14 f 0.02 g/cms. All values refer to room temperature. The cell dimensions were obtained from a least-squares refinement of 16 Weissenberg reflections (NaCl calibrated, X(Cu Kal) 1.54050 A). Errors indicated are 2a. Systematic absences were observed for reflections hOO with h odd, 0k0 with k odd, and 001 with 1 odd. This leads unambiguously to the noncentrosymmetric space group P212121,in which all atoms occupy the general positions 4(a): x , y, z; '/z - x, 7, '/z z; '/z x, '/z - y, Z; 2, '/z y, '/z - z. The result of the structure determination proved this space group to be correct. Owing to the extreme sensitivity of Re207 to moisture, the piezoelectric effect could not yet be measured. The reflection data for determining the structure were recorded on integrated equiinclination Weissenberg photographs, using the multiple-film technique (Ni-filtered Cu radiation). The intensities were
+
+
+
TABLE I FINALPOSITIONAL AND ISOTROPIC THERMAL PARAMETERS FOR X
Y
~e~0,5 5
B , .&a
0.36970(9) 0.5656 (3) 1.26 (2)b Re(1) 0.47488(9) 0.36367 (10) 0.0629 (3) 1.18(2)b Re(2) 0.26741 (9) 0.3255(3) 1.46 (3)b 0.52621 (10) 0.14267 (9) Re(3) 0.21967 (10) 0,13646 (10) 0.8080 (3) 1.52 (3)b Re(4) 0.3372 (17) 0.255 (5) 2 . 5 (4) 0(1) 0.3753 (17) 0.3687(17) 0.283 (4) 2 . 1 (4) O(2) 0.5837(17) 0.2349 (14) 0.5006 (19) 0.513 (4) 2 . 0 (4) O(3) 0.3822 (18) 0.766 (5) 3 . 2 (5) O(4) 0.5809 (21) 0.4482 (21) 0.4747 (17) 0.519 (5) 2 . 9 (5) O(5) 0.0496 (22) 0.463 (7) 4 . 8 (7) O(6) 0.4853 (24) 0.1579 (22) 0.079 (7) 4 . 7 (7) O(7) 0.4480 (24) 0.3375 (17) 0.762 (5) 2 . 4 (4) O(8) 0.3739 (17) 0.1378(18) 0.231 (4) 2 . 4 (4) 0.6614(18) 0(9) 0.2275 (17) 0.011 (5) 2 . 7 (5) O(10) 0.2478 (21) 0.3641 (18) 0.250 (5) 2 . 4 (4) O(11) 0.1606 (18) 0.4727 (16) 0.017 (4) 2 . 4 (4) O(l2) 0.2843 (18) 0.0437 (22) 0.967 (7) 4 . 9 (7) O(13) 0.2628 (25) 0.1500 (17) 0.542 (5) 2 . 8 (5) O(14) 0.2904 (18) a Numbers in parentheses here and in succeeding tables are standard deviations in the least significant digits. Values for the rhenium atoms before the start of refinement with anisotropic temperature factors.
measured with a microdensitometer. A total of 1218 nonequivalent reflections, 92 of them with zero intensity, was obtained from two prism-shaped crystals, sealed into Lindemann capillaries, with the dimensions 0.031 X 0.022 X 0.104 mm (rotation axis c, data collected for hk0-hk4) and 0.080 X 0.023 X 0.030 mm (rotation axis a, Okl-9kl). The collected data represent 98% of the sphere with (sin O)/X < 0.62 A-l. Because of the very high linear absorption coefficient for copper K a radiation ( p = 872 cm-'; p R = 2-3), the absorption correction was applied with special care, using accurately measured dimensions for the six crystal faces. The program written by Burnham" was employed for this correction. No correction for extinction was applied. After scaling the layers following the method of Hamilton, Rollett, and Sparks,12 the data were reduced to structure factors in the usual way and brought to an approximate absolute scale according to Wilson's method.ls Determination and Refinement of the Structure Approximate parameters for the rhenium atoms could be derived from the direct interpretation of the three-dimensional Patterson synthesis, although some difficulties arose from the relatively large unit cell and the pseudosymmetries present in the structure. (11) C. W. Burnham, "General Absorption Program," modified for the
(4) T . Ueki, A. Zalkin, and D. H. Templeton, Acta Cvyst., 19, 157 (1965). (5) G. Andersson and A. Magnbli, Acta Chem. Scand., 4, 793 (1850); L. Kihlborg, Arkiv K e m i , 81, 357 (1963). (6) G. Andersson, Acta Chem. Scand., 7, 154 (1953); W. L. Kehl, R . G. Hay, and D.Wahl, J . A p p l . P h y s . , 88, 212 (1952). (7) B. Krebs, A. Muller, and H. Beyer, Chem. Commun., 263 (1968), (8) A. Muller, B. Krebs, and 0. Glemser, Natuvwissenschaften, 68, 55 (1965). (9) K . A. Wilhelmi, Acta Chem. ScQnd., 8 , 693 (1954). (10) J. Noddack and W. Noddack, Natuvwissenschaften, 17, 93 (1929); Z . Anorg. Allgem. Chem., 816, 129 (1933).
X-RAY 63 System (University of Washington and University of Maryland) by H. Takeda and J. Stewart. The X-RAY 63 system is operating on the IBM 7090 machine a t Darmstadt, Germany. (12) W. C. Hamilton, J. S. Rollett, and R. A. Sparks, Acta Cryst., 18, 129 (1965). (13) The calculations for this structure determination were done on an IBM 7040 (G6ttingen) and on an IBM 7090 computer (Darmstadt). Besides our own programs, modified versions of programs from the Brookhaven National Laboratory system were used, among them the B N L versions of the FORDAPER Fourier program written by A. Zalkin, the Busing-MartinLevy full-matrix least-squares program ORPLS, and the Busing-Martin-Levy ORFPE program.
438 BERNTKREBS,ACHIMMULLER,AND HANSH. BEYER
Inorganic Chemistry
TABLE I1 ANISOTROPICTHERMAL PARAMETERS,= ROOT-MEAN-SQUARE AMPLITUDESOF VIBRATION,AND DIRECTIONCOSINESOF AXES RELATIVE TO THE CRYSTALLOGRAPHIC AXES ReU) Re@) Re(3) W4)
811
822
Baa
0.00209 (7) 0,00188 ( 7 ) 0.00218 (8) 0,00243 (8)
0.00150 (5) 0,00133 (5) 0.00151 (5) 0.00153 (5)
0,0090 (4) 0.0092 (4) 0.0134 (5) 0.0142 (5)
Pia
812
0.00017 (6) -0.00001 (6) 0,00006 (6) 0.00001 (6)
A
Axis
1 2 3 Re@) 1 2 3 R43) 1 2 3 W4) 1 2 3 = The anisotropic temperature factors listed
After some cycles of least-squares refinement with the R e atoms alone, using individual isotropic temperature factors, first excluding data with (sin @/A > 0.40 A-l and afterward including all data, the conventional R factor R1 = I: F,] - ~ F c ~ Fol ~ / Isank : ~ to 0.11. A difference Fourier synthesis, subsequently calculated, showed a number of maxima, eight of which could be assigned unambiguously to oxygen atoms. Approximate parameters for the remaining six oxygen atoms could be taken from a second Fo F, synthesis which included the information from the first after two further cycles of refinement. The subsequent least-squares refinement including all atoms and with individual weight factors converged after several more cycles to an R1 value of 0.057 (without zero-intensity reflections: 0.053) and to a weighted R factor Rz = (I:w(lFo/ - [ F c ~ ) z / 2 w F o 2 ) of ” z 0.059. At first isotropic temperature factors for all atoms and six scale factors were refined ; later the anisotropic vibration of the Re atoms was included (constant scale factors). In this anisotropic part of the refinement the R factor dropped from 7.6y0 (R1) and 7.5% (R2)to the final values. The shifts of all 92 positional and thermal parameters were below 0 . 0 5 ~in the last cycle. The expression minimized was 2w(IFoI The atomic scattering factors for 0 were taken as given from Ibers in ref 14; for Re the values calculated by Cromer and Waberlsa were used. Anomalous dispersion corrections were applied in the case of the Re atoms (see below). The values of Af’ = -5.58 and Af” = 5.37 e for Cu K a radiation have been taken from C r ~ m e r . ’ The ~ ~ weighting scheme, which proved to be reasonable from an analysis of the relative
11
(14) J. A. Ibers in “International Tables for X-Ray Crystallography,” Vol. 111, T h e Kynoch Press, Birmingham, England, 1962, pp 202, 210. (15) (a) D. T. Cromer and J. T. Waber, Acta Cryst., 18, 104 (1965); (b) D. T. Cromer, i b i d . , 18, 17 (1965).
+
+ PaJ2 +
0.807 -0.538 -0.244 0.031 -0.969 0.245 0.849 -0.270 0.454
-0.055 0.988 -0.143 2pl~hk 2&hl
+
-0.00019 (17) -0.00032 (17) -0.00014(18) - 0.00008 (19)
-
Direction cosines----
7
0.137 (3) 0.577 0.124(3) 0.806 0.115 (3) 0.134 0.128 (3) -0.565 0.122 (3) -0.219 0.114 (3) -0.796 0.151 (3) -0.207 0.134 (2) -0.961 0.120(3) -0.185 0.153 (3) -0.572 0.134 (2) -0.148 0.131 (3) -0.807 are defined as exp[-((p1,k2 PZ2k2
PRINCIPAL
Pia
0.00001 (19) 0.00014 (17) 0.00109 (18) -0.00081 (19)
Rms amplitude,
ReU)
THE
+
-0,124 0.249 -0.960 -0,825 0.114 0.554 -0.486 -0.063 0.871 0.818 -0.038 -0,573 2PZ3kl)].
errors, was: w = ( 2 0 0 / F 0 ) 2for Fo > 200, w = 1 for 75 I Fo 5 200, and w = (F0/75) for F, < 75. The final standard deviation of an observation of unit weight is 4.2 e. A final F, - F, difference Fourier synthesis did not show any significant peaks higher than 1.5-2.0 e A-3, that is, 15-20% of the electron density of one oxygen atom in this structure. The final parameters after the refinement, including the standard deviations obtained from the inverse matrix, are summarized in Table I. Table I1 gives the anisotropic temperature factor coefficients of the Re atoms together with the dimensions and orientations of the vibration ellipsoids. It can be assumed that the vibration parameters of the Re atoms have some physical significance in spite of the refinement being only partially anisotropic, considering the fact that scattering by these atoms dominates by far and that correction for the large absorption effect has been made. The observed structure factors are compared with the final calculated ones in Table 111. In order to apply the anomalous dispersion correction properly, the absolute configuration of the structure was determined according to the Bijvoet methodx6in the course of the final refinement. After an approximate correction f = [(fo Af’)z (Af”)2]l’a had been applied initially during the refinement, a complete set of structure factors was calculated for both possible configurations, using the exact correction f = fo 4Af’ iAf“.l7 Although for most of the structure factors the Bijvoet differences were negligible, a number of them were predicted to show easily measurable
+
+
+
(16) J. M. Bijvoet, Koninkl. Ned. Akad. Welenschafi. Proc., 62, 313 (1949). (17) The modification to t h e ORFLS program written by J. A. Ibers was used for these calculations. See J. A . Ibers and W. C. Hamilton, Acta Cryst., 17, 781 (1961).
CRYSTAL STRUCTURE OF RHENIUM(VII) OXIDE 439
VoZ. 8, No. 3, March 1969
....,. ....
-*
a 151 I C 1 0
1 . 0
1
0
b o w o m L O I 0
b4
Ibb
IO 0
I1
II 11. Y 1b
0 0
16
I
4
I I I I
3 b
I
1 1
I1
I11 11
I
I*# 51
0 114
I
I
1I
\ I I 101 I1 I I5 I 1 I I15 Ib I 11 I5 I In 0 1 151 I 1 I07 1 1 1 1 b
a
0 117
n
I I01 1 0 I 11b 5* 1 I 101 0 I O 1 11 I I l l 0 I 1 1 I1 I 1 1 ) 0 l b l I5 I IM 1 1 b* I 3 I1 1 3 LIS b J bb 4 I 105 b 1 Y 1 1 1*1 I 1 45 1 BII IO 1 b* I1 3 I l l I 1 1 0 I 3 1 2*5 Ib 1 17 I5 1 11 5 ( I b
*
0 1 1 1 4 5 b
I
I IO
11
I1 I1
4 4 4 4 b 4 b b b
.
b 4 b
l b b I5 b
I
19
*a
111 .I W 0
5
*I
* 5 1 0 5
IIb
i s I 5
I1
5
11
5
I1 5 I4 5 0 b 1 b I b 5 7
b b b b
L
11 71
10 15
LIS
a 0
0 .I 3) IW 1. b11
115 21 bI 101 131 I1 ll*
m
I
*
9
6
17b
LO b 11 b
25
I1 I1 I* 1
. 1 1
47
*
1.Y 31 *b lb IIb
I U
OBSERVED AND -*0 IbK I *CI I C 1 7 111 I D S 7
b b
11.
I
lb 111 (11 b0
I I I
: I
O
b
0
. 11 4I 15
*1 ** *
1
*I
55
60
0 0
I5
9
DT
*I I lob
1-7
10 10
IIb SI 1lb
IO
TI
I1
IO IO IO
10
I1
1*1
M
190 5b
111 5
11 57 111 5)
Y
I1
IO
bl 11 I7
Ill IO1
.. 9s 5
I77 3) 11.
15 411
I21 11 b9
210 I27 13
211
bi
107 Ibb 11 I41
IO
I1
10
115
IO
10 I1
I1 II
11 11
I1 I1 II
I1 I1 I1 I1
I1 I1 I1
13 13 13 I3 13 I1
I5 11
I1 I1
12
7
101 50
I5 1*
I1 111 b1
*7
la
I57
lbb
a
1*1 171 bl I1 11 I1 Y b1 191 17b
I1
49
11 10
1% 0
0 0 1. 15 0 0
0
40 17 11
0
1.
14
114 I* 45
16 16 14
Ib I4 14 Ib Ib Ib
111 14-
I*
61 (2 I9 16
15 1%
41
I5 15
I51
I5
1.
I7 11
17 I1 17
3,
I3 b7 n1 1b1 IY 0
I Ib
I I* 1 I1
IO0
110
113 4.
.* 111 13
.... . .... I1
4
. I
I
0
o
I O 0
5
0
7
0
. a
*I 00 L O O I1
1. 1) W
0
. I
1 3
0 171 114
IO U *I 0 164
5%
171
I
lb4 119 10 bl
*a
Ib 159 510 lkb
IO.
.L
111 104 lbl
I1
I I 1
111 I10 111 170 H 51 IH 104 bI lb
I
1
I 1
111
loo
1 .
70
I1
110 0
llb b 110
.*
a IO+
I 2 I 1 b 1 5 I b 1 7 I I 1 1 10 1 11 I 1 1 1 I3 1 I1 I
314
I
10
I5
SI1
I17 III 110
41
*I7
w1
I* 110 27b 318 1.
111 ID1 11 11 I l l I15 0 1 I* #I I51 I.*
I* 1W 1 1 1bl 1 1 I11
0 1 1 1
*I
b
1
119
54
5
1.1
161
11
b
2Ia
Tb 119
1 0 111
V
1 3 1 1 1 3 3
I15
lab
I?+
1 b
I1 IO.
10
b
I2 I1 I.
b1 SI
*o
15 11 I4 In 1b 10
*
3.5 126
w
55 105 141
1. LO I1 100 15
50
I,
Ill
n 9.
.I
1.1 1.3
I1
I1
4
4
1
5 b 7
I
* IO
+
1
. 4 4
b 4
'1 0
10) I1 50
0
I1
15n
151 1.1
111
*I
15
117
ra
1b1 ll*
LbI
0
*I
Sb 111 I. 117 111 117 Ib
lb
11. 11 55
I9 0
I
4
+
5
5
2 I 1 5 1 5
. 5 6
I
7
5 5 5 5
V
10 I1
¶
5
. ..
I2
2
15 21
111 171 01 I5
(0
5
,I
1 1 1
I1
116
I$
0 11 6 . 91 11
I*
+
I2
9
31 21
15% 1.1
11 4 I4 4
I1
101 11. $1 111
lbl
1 10.
..
I
0
32 21
1
3
15 0
11
5 5 5
12.
I1
15-
I1
0 IW
171
1.1
171
I10
Ill
61
lb7 217 111 5s 125 3b
1w
51
b5 Vb 31
5
66
I.. I2b 11. .I I14 41
111 5. bb 91
*6
lb1
2. Ibb
631 142
64b I++
6 1
6
0 1%
I2 134
111
2,.
1
6
2I'I 0
ID.
I
115
I20 b5
I
I
I
b
L
60
9
b b
23
LO
b
41
bo
#Y5 73
I51
I47
1II
IIb
I1 I
I I
I
1 I 7 I 7 1 1 b 7 5 I b I I I
I1
*b 0
b1
1.1
Ib. 111
I3 I4 15
*I
I51
I
I2
I1
IbO
I I
I I
111
15
1% I11
5 b
IO
1W
19.
ll* I** lo ab
0 0 1 1
I I
io
b 6 b b
11 101
0
I5
I**
11 I1 I3 1b 0
loo
I2 0 I1 0 16
I1 lab
1
L
LOO
113 Ib 0 1Y
BO1
lb 101 111 lbl 141
1.
:: E s
I7
I
n
I*) I.*
I.
15
5. I04 50
Y
IO1
1b I12
17.
im
I1 I3
11
111 b.
66
11 0
I11
16 16
I5
I5
I1 I1 I1 I1 I1
I)# 39
3..
bi
11 11 11 1bB Ibl I 1 177 l a 1
lb
7
In It5 174
I12 51
I
11 BI b** 51. 11. I l b 10 11 0 I1 111 315
101
112
7
15
IO
26.
75 115
I?
IO
170 I 104 15 11 LIS
11)
l
7 111
IO
bb
b*
0 10
I11
1 7
*I
b4
*
t
16
IO I1
5.
I.
I1
LO
33 1.3
n
a*
10
lb 1.3
I7 I10
I 7 7
1: I31 11.
*
I.
I b 7
45
. . ..
61
4) 1..
I11
1b1 4be
I
115
141
:1
lb
I I
7 1
Lb
49
51 101 51
1*1 11 100 $1
I
I5 I* I* lb 1b
6 6 6
41
I* H O
0 I 0 18 I1 I1 I14 l b * IO7 * b
b
I
Ib4 I* IO
IIO Ill 9I1 555
5
6
11. 11 I5 10
111 bI 11 *os 1008 b1 51 0
5 1 5 5 5
1 3
41
ia III
I I 1 I
10
11 111
TABLE 111 CALCULATED STRUCTURE FACTOR AMPLITUDES(IN ELECTRONS) FOR DIRHENIUM HEPTOXIDE *b 1CI *Cb .* I *C8 *c1 .* I I C 1 *Cb 1C1 I C 1 .I5 42
I¶
9
I
7 7
IO
7 7
*
13 I4 0
7 1 6
I
I
5
I a b
.
I I 1 I
b
1
I
n
10
b b
I1 I I1 I 11 I
o *
. I
V
1
*9
I 1 b
I *IOI **
I I1 I I1
* I1
IO I2 11 I1
*I 0 1.2 lb1 III
b
I15 117 IO I2 I10 I01 41 3n
10 11 I1 I1
b
0
I5
0
*I
11. 112 11
15
I*
I77
1bO 77
9
IO
I3
13
I1 I3 0 II
I I.
144 111
117
LII
115
131 I* 117 111 b* 3
0
5
0 0
b
0
0 1
1% 451
Ib
*1
10
10
I75
* IO
I1 14.
IO*
n
0 15.
51
IIb 111 1.1 51
111
I01
bl
b0
lbb
17)
lI0 51 +b
41
111
15. 112 18 Y
I1 17 M
111 ab
I23
2 I*
I11
3 1,
lb?
IbO 111
1
4
I
o
I O
11 I1
0 0 0 0
I1 0 11 0 0 I 1 I
2
3
b
5 b
I
I
I
1 I I
I I
*
101 160
1%
I1
I
107
14 0
I
105
5I *I IU
I 5. I I lob I 1 35 1 2 140
w
5
I
Ibl
6
I 2 I
47 14. 11 bo 0 I1 18
I
b V
1 0 I1 11 11 I4 0
I
I
2
I
1 I I 3
5.
103 I7
a**
11 lb7 4b 141 I* 53 b 15 37
bo
37
55
32 415
45*
I**
.
a
1
IY 157
5
3
111
1
110
I I
3 3
I17 IbO I 2 1 117
9
149 211
IT#
121
I21
17I
1
51
55
3
61 5,
0
110 IIb
11. 111
I* I45 b7
40 14) 60
IbO
151 bl
1% Ibb 101 01
.b
61
112 b3 116
I10 7b 11
10
137
I
* IO I1 12
I3 1.
0 1 1
14 16 13
106
3
15
b9 52
51 b9
I 5 b 7
I10
I
1Ob 3b .I 12 31
LI
*
*
7 I I 7 7 1
IO 7
I I
71
** ** **
* 0 10
1 10 I 10 3 10 4
IO
1 10 b 10
I10
5 I1
611
I I1 I I1 I1 10 11 I1 It 0 12 I 11
4
:1: in 110
n
I5
101 b1 IO. 0 Ib9
197 45 a19
.b
M
15. 41
IOI
lo 4b 21 71
.
be1 25) I21 15210
41
I11 11. b~
IM
17 10. I7 b5 lb
b9
.#I
llb 50
111 11,
37
39 14 31 I7 111
15
11. 5+
I+*
I5 I*
IO
101
H
181 b*
I1
47
111 M Ib IM
55
141
I1 III bb
I14 I7 1-1 b3 11 5b
I11 58
be
I
1 3 6
5 b
I I
IO I1 I1 I3 0
I
b
lb5 LI1
5
lbb bI
* IO
1u
21 158 l Y 1
15.
I. I51
111 I13 I l l I* 11 41 65 1 4 1 11. Y 41 11. 111 Y sa 191 145 5. 5* 110 IO7 0 4b 1.0 IBL
IW
151 0
*I 1w
4)
bl 41 3 H
I01 lb 1.7 303 157 1
I5 .I
>I I.
31
11 74
n.
11 14
0 0 0 0 0
0 0 0
1
I
1 I 1 I
152 51
*I
o
* a
11. 1.5
0 0 0
b
I I
I
I
1 1
I I I
I1 I I1 I 11 I
I1
bb 155 ¶*
6
157
11
I5
1)
151 5.
15.
** .4*1
I
b0 119 67
I15
5
5
)I* I+*
212 117
¶
154
I55
5 5 5 1 5 5 5
161 1.0 191
1.1
6
31b 166 5.
1,.
*I
I11 187 324 115 58 1.1
*3
I7
2 I1
u
1 I2 L I2 ¶ 11 b I1 ? I1 I I1 + 11 IO I 1 I t 12
I01
0 11 1 13
bI W
W
I10
I12
91
I* 31 111 1H
1 I1 3 13
3 b 7 I
11 13
I1 I1
L3 9 11
bb I4 7.
I15 I1 ¶ .
.I
ita lob 12b I.
I27
IO 11 I V 5 0 I+
I I4
2 I. 3 14
I4
differences. For the 16 reflections with the largest calculated differences the observed Fh,, - f i r 1 values were larger than the estimated errors of the observations, With one exception they all indicated the same configuration to be present. As a check, their hkl and Lkl intensities were carefully remeasured on the Weissenberg films. The results which are summarized in Table TV proved (for both crystals investigated) that the configuration described here is the correct one, and the final refinement was based on this. The values in Table I V refer to the final parameters. In assessing the reliability of this determination of the absolute configuration, it should be noted that the crystals investigated have a morphological center of symmetry (dominating planes (loo), (OlO), (001) and centric equivalents). Therefore, possible inprecise
5I I01
I1
IO.
17 b5 02 71
I10
51 41
32
I20
18 126
I56
55 17 Y
45 71
I.
I5
51
52
1I
IO1
IO
111
2%
175
luI W
**
I I b 111 I 1 I5 111 I10
1w
157
0 I1 I? 0 41 41 111 111 W *b 0
I51
IW
I1
1.1
I15
a+z an
191 IM 11. I+* I ? 11 111 I.*
llb
37 5.
*a
III
41
*a
*I
111 15. b1 bl b5 0 2 b1 I 1 2*1 lb5 I 1 155 15. 1 2 151 1*0 * I w a* 5 1 113 m b I le 1Y 7 1 Y 15
a
I
I IO I I1 2 I1 2 I1 2 0 3 I 3 2 3 1 1 3 1 3 b 1 I 3 I 1 9 1 10 I II 1 11 3 ?
.
11 I 0 4 I b I 4 3 4 6
*
5
1 b
6
b 1
IO b 11 12 13 0
I
2
I
5 b 7
b 4 5 5 5
5 5 5
5 5 I 5
b1 ?I W
101
11 I5 0 111
a
101 IH
1. 15 11. la1 I* I.
19
Ira
7b 0
1.1
58 bb
51 101 I1 I5
*
llb
11
AI
I12 1.b I6 I31 241
I
1*0
12*
I11
21 101 211 52
lob 2 0
in
10
12b b¶ 171
I)*
17 45 Y* IH -4
I1 Ill 211
18
45
61 W 1b
W Ib
I1 I1
5
IO.
H bl
in
H
11 b* .I
217 10
I2
*I 111 b7
a)
55 7.
I b7
I I 1 I
5b 11 I1 1w I*
I
1
I1
17
1.1 lb 24
1bO
0
b
I
2
*
I
10
I
I1 I I1 I 0
1
I1
11 11 I1 I1
I1 I1
1st 17
5I
14*
I7
M
bb
151
110
I11 1.5
1I 11 45 I 1 101 I1 11 I1 I2 11 0
17 *1 214 b5 14
12
195 7b I01
I2 I2
I1 I1 I2 13 13 I1 I3 I1
I3 I1 I1 3
I O
1b 101 1.
1% I.
I1
15
IO1 160
51 11
I22 125 .b 0
I15
u
1.1 0 b4
I5
110 I.
I4 2 111
*I I71 I1 *2 3* I 2. II
M
.a I7 111
I4 I4
I# 104 lb 122 $0
1. 15
I. SI 5%
121 3b 7. b5 5.
101
lob
.I
*I
I* It 14 1.
I.
I¶
10
%I
bb 112
111
115 .1
116 .1
a
b 4
5 .
.
4
is
I1
I*
14 I4 I1 1b I5 15
W
I1 I1
6 b
lb 0
10 b7 14) 50
0
$
I
5
1
5
1 4
5 5 5
5 b 5 1 5
I ID
*
5 5
11 O 1
5 b
.. b
I
b 1 b 5
L
b
I
b 6
b
* b
10
I1
O
I
2 3
b f
I I 7
bI I21 48
231
bI 117
.*
115 0 10)
27 2.
51 21 III 13I
I11 Y
b2 11b 11 0 18
3M 29 0 77 27
I1
b
.I 55 25 17
kt
*
110 114
12b $0 b7 120 77
11 I1 111 21 I1 bI 35
I*
7 I 1 1 I
27
I4 *I
52 +I .L
7
31
I7
IO 11 7
IP b 0
b31 b I
. 4 5 b
7
*
O I
I. 19
20 70
b I b
12. I¶
I 210 1
1 b I I
155 IO 111 1.1
112
I I
1 I 1
I
1 I
I
I1 111 177
I. ?1 I It 1.
.
I 197 Ib
1.1
5I 111 W IO 11 1)
1-
lb*
Ib 150 72
.n
67 I5
I11 bI 151
I?. I01
15 115 120 Y
0
I15
11
101 102 I.
25 130 37 5b 210
bb 67
'b
171
6
.
I*.
4 6
101 24
¶
1
17 I11 1.1 \I* I11
5 b 0
w
I*
I2 *I
lb1
151
0
1.5
**
26 $1
11. 50 101
,e,
I*
26
I21
115
b2 0
9
I1 0
9
bl
IO IO
IIb
IO IO IO 10 IO 11 I1 I1 11
151
21
LOI
0
71
Ill 11 lb lb bb I 111
Ill I14
101
*I 52 I3
.1 16
n
IO. b4
115
a
bI
I01
11
51
bY
117 (1
114 I T
w i in
I5
I1
0
n
I* 11 1
L
0
41
0 0
11 148
*I
Y0
o
n
I
. *n
b4
b1 100 51 *I 100
IO 17, 110
19 11 Y I *B Y
b5
75
131 .I 1.0 bl
I31 101 I51 75
0 116 17
1.3
I I
.*
11 35
l bH 2
I I
51
I10
11
1 I I
116
I.
0
L. 170
I I
w
1 I 2 2
I45
1 1 1 1
111 *I 116
.I
*I II 1 I71
I
*63
. 4 4 b 4 5
5 5 I
*L
5 b
20
6
T.
4
I.
IYo $1 *I 10
Ill
110
5I
n.
I**
150
18
I07 3. 51 21
1.
3. bb 0 59 2.
*I
6.
.I
11 7.
11
21
*?
.I
b# 7I .I I1 o
b1 bo
I11
b
I)# b4
203
L
IO*
21 43
7 1
II*
loo LOO IVI
17
. ..
1b7 *J II 10. 111
Y
1
5
15.
0
IO
I**
151 30 I10 I52 117
102
..
I 1 1.
I I b I
.*
I1 101
kl
I40
1
1%
I11
*I .I Y 2b
3
3
? b
lb
.I
159
1 1
1
.*
I* I51 I11 .o 5.
LOT
60
3
121
1b1
1% 112
I
7 I 1
+I
IIb
I1 I1
1.
1
11b 23 I1 IbI
55 -115
11
1.0 111 115
2 2 2
bI
16 I1
I71
2 2
Y 11 I1
I 7 I I I
.... .... ......
I I
51
15
I1 b.
b 12I
I1 II 111 II* 12. 111 12b 12b 1. *I
37
Y IW 11. 130
*
1.1
0 101
7. H
.
loo
11 I1
2 2 2 2
b b b
b.
2
b9
b 5
6
0 0 0
72 1bI
10 I.
5141
5
11
I05
*.b
10
,.
I5
1.
b
22 110 bI
52 0
31
I4 7
* IO
I,%
51 21
16
11 *I
b
bI
7
0
11 I1 I3 I1
13
0 0
bl 111
*b
10
aaa 11
119 11 150 I1 .I I*
n
10
10
I15 101
I1 0 3 11 141 12. I1 IbO l * O I1 11 H 0 *I3
11
aw
lb
11 0 1 0 110 I53 11
0 I15 10
L
.I
20
12
0 0 0 0
110 59
24
42
I1
11
111 I1 117 IO
11) 71
1)
I...
b
0 0 0 *I
n
b
I5 2,
14 .37
I1 I7 1.1
11
11 I1
I1 11 I1
111 7.
Ill
2.0
5b
IO*
U
111
I1
21 bI
5
101
I1
55 1
IC. *I
101 2b 5.
IC)
L
LO 11 IO
I M
b1
5 5
6.
10
I1
5
71 110 1. .I
I1
44
1 1 1 1 I54 10 I53 II I > I11 10. b 1 11 15 5 1 101 114 Y 4) b l 7 s 71 n I 3 70 bb t I 15 lb IO 1 b1 b l 11 1 I ? 3. I1 1 w 11 0 4 15 II
111 II I. bl
10
IO LO IO
b+
aI
I71
IO
IO IO
BO
I
IO
10
11.
In.
¶
1
I I 4
I5 IS
5 5
40
11
51
31 3#
10. bb
Ill 72
9
111
161
111 la*
11
I* bI
107 bb
I5 I5 I$
10
I1
Ill Il*
14
10
IO
I11
I05
I4
IO b i ii IO 1U I11 IO 1)) 15. IO 111 118 10 IW U l bI D1
I1
n
30
101 IW 1Y 10)
*
30 111 bI
ibm
I01 11
9
IO
11
II7
*
I2 I2
III I10 I31
I,
* * **
10
*
10
. 5 5 b b
M l** w7
9
I5
b4 5 11 15
17 57 0 1% bz
5
I1 I S
0 19
0 15
29 101 70
I
b
11
11
IO1
1 1
I I
0
I I I
Y
IO1
l l b 21b 1 0 I1 I 117 IbI I1 111 55 115 .I
I2
nr
170 I5
101 Y
IO I1
It4
0
110
17b Ib 11.
V
1 0
9
1
I
5 0o b L O 1 0 1 0 t 0 1 0 0 I1 0 I2 0 0 I I 1
b I I
....,. ....
w1
n
o
I 4 9
.I Ill 17b
70 1
**
I
14b
SI *2
bI
5b 111 71 lb7
i
37
IO
71 1b
0 0 1 0
111
Ib
7 10 I IO LO Y 10 10 LI I 1 19 I1 10 41 0 I1 1.1 1 I 1 117 21 2 I1 1 11 l b 1 4 I1
....,. ....
1:;
. . .. . .. . . . .
1 2 1 4 b 5 4 6 4 7 I
+
I
I1
101
w7
I
0
1
I2 I3 I4
5 b
IO I1 I1
2 3
4
6
*B
I55
171 IW
1
I I I I 1 7
*
1 1 1 Ill
. a
b b b b b b b b b b 1
14) 1.1 111
9 10
11 I 11 I
6
11 7 11 I 11 I 0 I 51 96 1 I in in 2 I H 1I 1 n 1Ib I** I b1 bb 5 I b I 101 101 58 51 I I 17) I17 I I b1 I 611 .a7 IO I I71 l b l I1 I 11 I 111 111 I l l I80 13 1 119 110 1 I # 111 I 9 11. 111 I 9 1.1 1.1 3 107 *I b5 111 I15
o w
11. I1 Ib
25
77
0
b1 II
IO.
121
I5
(1 25
3 3
7 lb
1,.
41
0
1 1
a I.
I¶ I¶
0
11
b1 I21
+
I
IO
I.
.b 16 IY 5.
*
2
1) 115 7 llb
b 11 1 lb b 14
10 0 1 2 3
....,. ....
I.*
101 0 I 1 111 I3 I 9 1 I1 b l 1, SI 1 1 131 11 I72 11 106
5 b 7 b I1
I b
145 1.0
. 2 1
5
I..
013
I
a
I15
*
6 12
5 5 16 5 O b I b 1 b
#I 100 113 171 1bI
a1 1 I O 151 148 I ID 1Y 170 3 IO 1151 I1 4 IO 10 5 10 1 0 I51 b IO I Y 111 I IO 1 0 I*¶ I 1 IO IM IS* 1s 41 I1 10 11 I. 0 II 15 I1 1 I1 Y 51 2 I1 J 11 111 111 -7 H b 11 u I5 5 I1 0 I1 b 10 IO I Y I14 bb 41 10 IO b 11 15 51 1 11 IU l b b b 11 117 I I I 129 1 11 11. I 0 IO I1 11 11 a io3 55 I 2 I1 51 0 1 1 0 13 1b I I2 I I2 151 l b 7 1 I t 11n 117 11 I1 I1 bI b1 1 I1
.
I12
5
IY
bb
IO
155
I1 la
ID
I11 I l *
I I 9 I81 I1 I b* 11 0
IO 5 a b 1
.I
*.I5 b 61 111 110 111
111 37 ,,I
absorption corrections cannot influence the result. Description of the Structure and Discussion The crystal structure of dirhenium heptoxide consists of an arrangement of polymeric double layers perpendicular to the crystallographic b axis, related to each other by the twofold screw axes. Between the layers there are only oxygen-oxygen van der Waals contacts. This layer structure is expressed in the macroscopic shape and the cleavability parallel to the (010) face of the crystal slabs and plates. The structure is built of equal numbers of Re00 octahedra and Re04 tetrahedra in the following way. Zigzag chains of octahedra linked through corners run along the crystallographic c axis, every other chain being part of the upper and lower half of the double
440 BERNTKREBS, ACHIMMULLER,
AND
HANSH. BEYER
Inorganic Chemistry TABLE IV DETERMINATION OF THE ABSOLUTE CONFIGURATION^ h
Figure 1.-Crystal structure of rhenium(VI1) oxide: one of the two double layers in the unit cell, projected along the b axis. The layers are parallel to the ac plane.
a
b
1
2
3
L
i
X
Figure 2.-Projection of the structure along the short c axis. The zigzag chains of Re06 octahedra and the layers are perpendicular to the plane of paper.
layer, respectively. These chains are connected by the Re04 tetrahedra to form the double layers, every tetrahedron sharihg one oxygen with each of two octahedra, one in the upper chain and one in the lower. Every octahedron is thus linked to two octahedra and two tetrahedra through corners. The double layers can also be described as a system of “rings” of four polyhedra (t-o-t-o) connected through corners. The rings themselves, which are perpendicular to the plane of the layers, are linked through corners of the octahedra resulting in a polymeric arrangement. I n one unit cell there are two identical double layers with different orientations due to the twofold screw axis along b. Figures 1 and 2 show projections of the structure parallel and perpendicular to the layers. The structural ring unit is pictured in Figure 3. All bond distances, the distances between neighboring oxygen atoms within the polyhedra, and the shortest
5 9 1 1 1 9 9 12 13 1 12 3 4 11 8 3
k
1
Fc(kk1)
F, cS;S)
1 1 7
1 1 1 1 1 2 2 2 2 2 2 2 2 2 3
197.4 165.6 69.4 92.6 26.5 142.4 47.0 41.0 58.7 83.4 57.3 119.5 45.2 58.5 62.0 89.3
204 5 170.3 73.2 88.4 28.6 135.9 53.5 37.7 55.0 78.1 54.0 112.7 53.9 62.6 58.4 86.3
8 11 1 2 2 3 4 8 9 10 10 2 11
3
I
Io(!@ I~ (hki)
1 >1 >1 1
“ T h e data were collected on the quadrants hkl and hkl. The numbers in Table I11 refer to the quadrant hkl.
Re-Re separations with their standard deviations are listed in Table V. Bond angles in and between the polyhedra are given in Table VI. The values show that the tetrahedra are reasonably regular, whereas the octahedral environment is strongly distorted. The central rhenium atom is displaced from the center parallel to one of the threefold axes of the octahedron resulting in three short Re-0 bond distances (1.651.75 A) and three long ones (2.05-2.16 A) in every Re06 octahedron and giving a “long” and a “short” Re-0 bond distance in every Re(o)-0-Re(o) bridge. The terminal Re-0 bond lengths in the tetrahedra (average value 1.70 A) are shorter than those associated with bridging oxygens (average 1.78 A), as should be expected owing to the higher formal orders of the terminal bonds. The bond distances are very similar to those in the perrhenate ion of KRe04 (1.77 or in gaseous Re03CI (1.702 A)23(see also ref 24). Just as for these compounds, one has to assume a considerable amount of r bonding in the terminal as well as in the bridging bonds. The same is true for the three short bonds in the octahedra. Here again the (18) The standard deviations obtained do not allow quantitative conclusions to be drawn about slight variations in bond strengths from differences in experimental bond lengths. These differences must he greater than 0.07-0.09 A to he significant a t the 1 % level. However, the over-all quality of the determination seems to be rather better than that of similar heavy-atom structures solved from film data (usual c values 0.04-0.06 A). Given a complex very heavy atom-light atom structure like this only very carefully collected (and multiply observed), counter data can give substantially higher accuracy. Ibers and coworkers’Q’20have also pointed this out reporting from highly accurate counter data Re-N bond distances with c 0.011 A. Similarly accurate Re-0 bond lengths have been reported recently.21 (19) P. W. R. Corfield, R. J. Doedens, and J. A. Ibers, Inoug. Ckem., 6 , 197 (1967). (20) R. J. Doedens and J. A. Ibers, ibid., 6, 204 (1967). (21) M. J. Bennett, W. K. Bratton, F. A. Cotton, and W. R. Robinson, ibid., 1 , 1570 (1968). (22) J. C. Morrow, A d o Crysl., 13, 443 (1960). (23) E. Amble, S. L. Miller, A. L. Schawlow, and C. H. Townes, J . Chem. P k y s . , 20, 192 (1952); R. Javan and A. Engelbrecht, P h y s . Reu., 96, 649 (1954). (24) A. Muller, B. Krebs, and W. Hdltje, Spectrochim. Acto, 2 8 8 , 2753 (1967).
-
CRYSTAL STRUCTURE OF RHENIUM(VII)OXIDE 441
Vol. 8, No. 3, March 1969
TABLE V PRINCIPAL INTERATOMIC DISTANCES (IN A) WITHIN
THE
Re00 AND Re04 POLYHEDRA'
Bond distances
Re( 1)-0( 1) Re( 1)-O(2) Re( 1)-O(3) Re( 1)-0(4) Re( 1)-O( 5) Re( 1)-O(8) Re(3)-O( 3) Re( 3)-O( 9) Re(3)-O( 6) Re( 3)-O( 7)
2.160 (26) 2.054 (23) 2.093 (22) 1.727 (28) 1.650 (26) 1.725 (24) 1.764 (22) 1.769 (22) 1.680 (33) 1.677 (33)
Ab B B
C C A B B D D
Mean
A B B C C A B B D D
Re( 2)-O( 8) Re( 2)-O( 9) Re( 2)-O( 10) Re( 2)-O( 11) Re(2)-0( 12) Re( 2)-O( 1) Re(4)-O( 2) Re(4)-O( 10) Re( 4)-O( 13) Re(4)-O( 14)
0(1)-0(8) 0(2)-0(3) o(2)-o(4) O(2)-O(5) O(3)-0(4) O(3)-0(8) 0(4)-0(8) 0(4)-0( 5) O(5)-0( 8) Mean
2.66 (3) 2.62(3) 2.70(4) 2.76(4) 2.60(3) 2.64(4) 2.67(3) 2.81 (4) 2.61 (3) 2.68 (3) 2.56 (4) 2.64(4) 2.66
1.742 (34)\ (25) 1.709
0(8)-0(9) O(8)-O( 10) O(8)-O( 12) O(1)-O(8) O(9)-O( 10) O(9)-O( 11) O(9)-O( 12) O(10)-O( 11) O(1)-O( 10) O(1)-O(11) O(11)-O(12) O(1)-O( 12) Mean
Re(3) tetrahedron
0(3)-0(6) 0(3)-0( 7) O(3)-0(9) 0(6)-0(7) O(6)-0(9) 0(7)-0(9) Mean
1.806 (25) *773(21)\
2.16 2.08 1.69 1.74 1.78 1.70
Re(2) octahedron
Re(1) octahedron
O(1 )-0(2) O(1)-OW O(1)-0(5)
2.151 (25) 2.080 (24) 2.103 (25) 1.682 (25) 1.690 (24) 1.753 (24)
2.84(4) 2.72(4) 2.93(3) 2.71 (5) 2.88(4) 2.81(4) 2.82
2.69(3) 2.67(3) 2.72 (3) 2.69 (4) 2.67(4) 2.62 (3) 2.65(3) 2.68 (4) 2.66(4) 2.72 (4) 2.60(3) 2.69(3) 2.67
Re(4) tetrahedron
O(2)-O( 13) O(2)-O( 14) O(2)-O( 10)
O(13)-O( 14) O(10)-O( 13) O(10)-O( 14) Mean
2.94 (4) 2.77 (3) 2.99(3) 2.84(5) 2.81 (4) 2.86(4) 2.87
Rhenium-rhenium separations between linked polyhedra
Re( 1)-Re( 2) 3.753 (3) Re( 1)-Re( 4) 3.678 (3) Re( 1)-Re( 2) 3.774 (3) Re( 2)-Re( 3) 3.686(3) Re( 1)-Re( 3) 3.745 (3) Re( 2)-Re( 4) 3.769 (3) 0 All distances in the lists for the two crystallographically independent equal polyhedra are given in the same order to facilitate easy comparison. b Functions of the oxygen atom: A, bridging atom, octahedron-octahedron; B, bridging atom, octahedron-tetrahedron; C, terminal atom within octahedron; D, terminal atom within tetrahedron.
Figure 3.-Asymmetrical unit of the Rea07 structure with experimentally determined bond distances in &ngstrGms. The direction of the view is approximately along c.
TABLE VI BONDANGLES(IN DEG) IN AND BETWEEN THE Reon POLYHEDRA O(1)-Re( 1)-O(2) 78.1 (9) O(@-Re( 2)-O( 9) 78.8 ( 11) 0(1)-Re(l)-0(3) 76.0(9) 0(8)-Re(2)-0(10) 77.8(10) O(1)-Re( 1)-O(4) 164.3 (11) 0(8)-Re(2)-0( 11) 163.O (11) O(1)-Re( 1)-O(5) 89.0 (12) 0(8)-Re(2)-0( 12) 89.5 (11) O(1)-Re( 1)-O(8) 90.0 ( 12) O(1)-Re( 2)-O( 8) 86.3 ( 12) O(2)-Re( 1)-O(3) 77.8 ( 10) O(9)-Re( 2)-O( 10) 79.2 ( 11) 88.0 (11) 0(9)-Re(2)-0( 11) 87.7 (11) O(2)-Re( 1)-O( 4) 0(2)-Re(l)-0(5) 91.5(12) 0(9)-Re(2)-0(12) 88.6(12) 0(2)-Re(l)-0(8) 161.2(11) 0(1)-Re(2)-0(9) 161.3(11) 0(3)-Re(l)-0(4) 94.4(11) 0(10)-Re(2)-0(11) 89.5(12) 0(3)-Re(l)-0(5) 162.9(11) 0(10)-Re(2)-0(12) 163.9(11) 0(3)-Re(l)-0(8) 85.4(11) 0(1)-Re(2)-0(10) 86.7(11) 0(4)-Re(l)-O(S) 98.4(13) O(ll)-Re(2)-0(12) 100.7(11) 0(4)-Re(l)-0(8) 101.7 (12) O(1)-Re(2)-O(l1) 104.5(11) O(5)-Re( 1)-O(8) 102.8 (12) O(1)-Re( 2)-O( 12) 102.5 (12) 0(3)-Re(3)-0(6) 110.8(14) 0(10)-Re(4)-0(13) 104.8(14) 0(3)-Re(3)-0(7) 104.3(13) 0(10)-Re(4)-0(14) 109.1 (12) 0(3)-Re(3)-0(9) 112.1 (12) 0(2)-Re(4)-0(10) 113.1 (12) 0(6)-Re(3)-0(7) 107.2 (16) 0(13)-Re(4)-0(14) 110.9 (15) 0(6)-Re(3)-0(9) 112.7 (14) 0(2)-Re(4)-0(13) 113.6(13) 0(7)-Re(3)-0(9) (14) 0(2)-Re(4)-0(14) 105*4(11) Re(1)-O(1)-Re(2) 149.2(15) Re(l)-0(2)-Re(4) 147.7 (15) Re(l)-O(8)-Re(2) 150.9(15) Re(2)-0(9)-Re(3) 146.6(15) Re(l)-0(3)-Re(3) 152.4(15) Re(2)-0(10)-Re(4) 149.0(16)
442 BERNTKREBS,ACHIMMULLER,AND HANSH. BEYER bridging bond (mean 1.74 A) seems to be somewhat longer than the two terminal bonds (mean 1.69 A). The strong distortion of the coordination octahedra is a common feature in complexes and oxides expecially of group IV-VII transition metals. Reasons and examples of these distortions have been discussed.25 The kind of distortion observed in the present case has been found by Magn&lizB and KihlborgZ7in a similar form in the two modifications of M04O11. The Re207 structure has features in common with the M o ~ 0 7 ~ ion in Na2Moz07 which consists of polymeric chains with (like Rez07) equal proportions of MOBoctahedra and M01 tetrahedra connected through corners. 28 However the connection scheme and the distortion of the octahedra in the dimolybdate ion (two long, two intermediate, two short bond distances) are different. If the partly ionic character of the long Re-0 bonds >2.0 A is stressed, the structure of rhenium(VI1) oxide can be described as being close to an arrangement of ReOd- anions (the tetrahedra) and Reo3 + cations. It is interesting to note that there is a strong pseudosymmetry in the structure. Each of the two double layers in the unit cell has additional pseudoinversion centers a t approximately 3/8, 3/4, l / 4 , etc., and 5/8, l / 4 , l/4, etc., which, taken as real, apply only to half of the structure and do not lead to a higher symmetry of the structure as a whole. The interpretation of the observed bond distances in terms of quantitative .ir-bond contributions is difficult. The very complicated infrared s p e ~ t r u mcan ~ ~only ~~ lead to approximate force constants. An estimation of the bond orders can be done on the basis of the ideas given by Cotton and Lippard30 and Doedens and IbersZ0 Starting from data for some octahedral Re(V) halo complexes, a theoretical Re-0 single bond length of 2.04 A and a double bond length of 1.86 A were obtained. This extrapolates according to Pauling’s empirical rule to a theoretical triple bond distance of 1.75 A. As was pointed out, however,20 these values, derived from systems in which nonbonded interactions of the ligands determine the mean bond distances, predict very high bond orders in less “stressed” polyhedra with lower coordination numbers. In the case‘of Re207 the observed bond lengths would correspond to an unreasonably high sum of Re-0 bond orders in the octahedra (12.5) as well as in the tetrahedra (12.3). I t seems that a Re-0 bond orderbond length curve not far from the one derived for Mo-0 bonds31 is closest to reality, giving approximate
(25) A. Magnhli, J . I E O Y g . Nucl. Chem., a, 330 (1956); L. E. Orgel, Discussions Faraday Soc., 26, 138 (1958); H.D. Megaw, Acta Cryst., B24, 149 (1968); J. D . Dunitz and L. E. Orgel, Aduan. Inorg. Chem. Radiochem., a, 45 (1960). (26) A. Magn&li, Acta Chem. Scand., 2 , 861 (1948). (27) L. Kihlborg, Avkiv K e m i , a i , 365 (1963); S. Asbrink and L. Kihlborg, Acta Chem. Scand., 18, 1571 (1964); see also L.Kihlborg, Avkiv K e m i , ai, 471 (1963). (28) I. Lindquist, Acta. Chem. Scand., 4, 1066 (1950); 14, 960 (1960); M . Seleborg, ibid., 21, 499 (1967). (29) K. Ulbricht and H. Kriegsmann, 2.Chem., I , 224 (1967). (30) F. A. Cotton and S. J. Lippard, Inorg. Chem., 6, 416 (1966). (31) F. A. Cotton and R. IM. Wing, ibid., 4, 867 (1965).
Inorganic Chemistry
bond orders of -1 for the long octahedral bonds above 2 A and of 1.6-3.0 for the short ones below 1.80 A, This predicts approximate total Re-0 bond orders of 10.2 in the octahedra and of 9.0 in the tetrahedra, these values being reasonable and close to the numbers to be expected from molecular orbital considerations, The packing of the space-determining oxygens in the Re207 structure is relatively open. The 0-0contact distances in the octahedra (mean values 2.66 and 2.67 A) are not far from the possible minimum (thus causing the “rattling” effect and distortion in the concept of OrgelZ5). All other 0-0 contacts in the tetrahedra (mean values 2.82 and 2.87 A) and extrapolyhedral 0-0 distances (>2.75 A) are significantly longer. Every 0 atom in the structure is surrounded by 10-12 others within 2.56-3.5 The quotient V/no (VI volume of the unit cell; no, number of oxygens in the cell), which gives a measure for the packing density, is 18.9 A3 for Red&. Values for other do oxides are (A3): MOO3, 16.9; WO3 (monoclinic), 17.6; WO8 (tetragonal), 18.2; OsOd, 20.7. Like other structural features these numbers place rhenium (VII) oxide between the polymeric oxides Moo3 and Won and the more covalently bonded 0 ~ 0 4 ,confirming its character as being a compromise between an ionic and a molecular structure. On the basis of the crystal structure determined, it is possible to derive a plausible evaporation mechanism and to explain the relatively high volatility of rhenium(VII) oxide. Thermal cleavage of two of the long, weak Re-0 bonds per octahedron, together with a slight change in hybridization a t these Re atoms, is sufficient to yield the Re207 molecule with tetrahedral coordination around all Re atoms as is assumed in the gas phase (see below). The configuration of the Re207 molecule is already “pre-formed” in the structure of the polymeric solid. In this context it may be of significance that in the process of preparing the oxide one obtains, besides the yellow crystalline Re2O7, a small quantity of a poorly defined white amorphous solid, which had been erroneously described to be rhenium(VII1) oxide by its first investigator^.^ I t is very probable that this is a second “molecular” modification of solid RezO7, formed by direct condensation (quenching) of the gas-phase molecules, without rearrangement of the bonding system. I t is interesting to compare the features of the structure with those of solid perrhenic acid which is the primary hydrolysis product of Re207 and which is also formed on evaporation of aqueous solutions. Recently we succeeded in characterizing this compound as dirhenium diaquoheptoxide Rez07(OHz)z. The crystal structure shows isolated molecules which consist of one Re04 tetrahedron and one Re04(OHz)z octahedron connected through one oxygen corner.
A.
(32) H.Beyer, 0.Glemser, and B. Krebs, Angew. Chem., 80, 286 (1968); Angew. Chem. Intevn. E d , Engi., 7 , 295 (1968); see also B. Krebs, Angew. Chem., 80, 291 (1968); Angew. Chem. Intevn. E d . End., I , 308 (1968).
Vol. 8,No. 3, March 1969
Figure 4.-Configuration and bond distances in %ngstromsof the primary hydrolysis product of RezO7: dirhenium diaquoheptoxide, RezO~(0Hz)z(“solid perrhenic acid”).a2
If the dimensions of the molecule, as given in Figure 4, are compared with the bond distances in RezO7, a close resemblance is evident. Of special theoretical interest is the fact that the kind of distortion of the octahedra is analogous in both compounds. The mechanism of the partial hydrolysis of rhenium(VII) oxide to Rez07(OH~)~ can be assumed to be similar to the evaporation mechanism. The Re-0 bonds which are cleaved hydrolytically are the same which are broken in the process of evaporation : Re(1)0(1),Re(2)-O(8) plus either Re(1)-0(2), Re(2)-O(9)
SYNTHESIS OF TRIFLUOROPHOSPHAZO COMPOUNDS 443 or Re(1)-0(3), Re(2)-O(10). If Figures 3 and 4 are compared, the structural relationships among Rez07(s),Re~07(g),and Rez07(OHz)z can be visualized. Thermal Vibrations.-The temperature factor coefficients are significantly larger for the tetrahedrally coordinated Re atoms than for those in the octahedra (see Tables I and 11). The shortest axes of the vibration ellipsoids for the octahedral Re atoms were found to be along the directions in which the metal atoms are displaced from the centers of the octahedra. These features are physically reasonable, and they are observed almost identically in two crystallographically independent polyhedra. In view of the esd’s the parameter differences of Table I1 are to be regarded as possibly significant. The oxygen atoms 0 ( 6 ) , 0(7), and 0(13), representing terminal atoms in the tetrahedra, show substantially (and possibly significantly) higher temperature factors than the rest. Acknowledgments.-We wish to thank Professor Dr. 0. Glemser for his generous support of our work, and we thank Dr. W. C. Hamilton and Dr. D. F. Koenig for supplying programs from the Brookhaven National Laboratory library. The assistance of Dip1.-Chem. G. Wagner in preparing the oxide is greatly appreciated.
CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY, MEMPHISSTATE UNIVERSITY, MEMPHIS,TENNESSEE 3811 1
The Synthesis of Trifluorophosphazo Compounds BY MAX LUSTIG Received August 19,1968 Trifluorophosphazosulfuryl fluoride, PFs=NSOzF, trifluorophosphazophosphoryl fluoride, PFa=NP( O)Ft, and trifluorophosphazothiophosphoryl fluoride, PFa=NP( S)Fz, are prepared by the reaction between phosphorus dichloride trifluoride, PFaClz, and sulfuramidic fluoride, FSOZNHZ,phosphoramidic difluoride, F2P( O)NHz, and thiophosphoramidic difluoride, FzP(S)“2, respectively. Some properties of these new compounds, including FzP(S)NHz, have been studied.
Trichlorophosphazophosphoryl difluoride, PCla=NP(O)Fz, and several of its derivatives‘ and trichlorophosphazosulfuryl fluoride, PCla=NS02F, have recently been reported. However, attempts to exchange the chlorine atoms bound to the phosphorus with fluorine have been unsuccessful.2 A method has now been found for the preparation of perfluorinated phosphazosulfuryl and -phosphoryl compounds PFaC1z
FSOzNHz PFa=NS02F + FzP + + 2HC1 (E)NHz PFa=NP (E)Fz
(1) 0.Glemser, H. W.Roesky, and P.R. Heinz, Inorg. NucE. Chem. Letters, 4, 179 (1968). (2) J. K.Ruff,Inorg. Chem., 6,2108 (1967).
where E = 0 or S. Thiophosphoramidic difluoride, a new compound and the precursor to trifluorophosphazothiophosphoryl difluoride, is prepared by ammonolysis of M-oxo-bis(thiophosphory1difluoride), FzP(S)OP(S)Fz, in a manner similar to the synthesis of phosphoramidic difluoride from pyrophosphoryl fluoride. Experimental Section Reagents.-The amidic compounds FSOzNHz2 and POF2NHz3 were prepared by the literature methods. Established methods were used to synthesize pyrophosphoryl fluoride4 and p-oxo-bis(thiophosphoryl difluoride).6 Lecture bottles of PFa and Clz were (3) S. Kongpricha and C. Preusse, ibid., 6, 1915 (1967). (4) E. A. Robinson, C a r . J . Chem., 40, 1725 (1962). (5) W.E. Hill, t o be submitted for publication.
