Osmium(VII) fluorine compounds

Article abstract of DOI:10.1021/ic0608290

Of the four published osmium fluorine compounds in the oxidation state +7, OsO₃F, OsO₂F₃, OsOF₅, and OsF ₇, only one (OsOF₅) is a real Os(VII) compound. OsO ₃F has obviously been OsO₄. OsO₂F₃ in its two modifications is a mixed-valence Os(VI)/Os(VIII) compound, whereas a new compound Os₂O₃F₇ is a mixed-valence OsV/Os(VIII) compound. The molecular structures of OsO₃F, OsO ₂F₃, and OsO₃F₂ are calculated. OsO₃F₂ seems to exist in two forms, with D_3h and C_s symmetry. The original preparation of OsF₇ could not be reproduced, only OsF₆ has been obtained.

Full text of DOI:10.1021/ic0608290

Inorg. Chem. 2006, 45, 7929−7934
Osmium(VII) Fluorine Compounds
Hashem Shorafa and Konrad Seppelt*
Institut fu¨r Chemie der Freien UniVersita¨t, D-14195 Berlin, Germany
Received May 15, 2006
Of the four published osmium fluorine compounds in the oxidation state
+7, OsO3F, OsO2F3, OsOF5, and OsF7,
only one (OsOF5) is a real Os(VII) compound. OsO3F has obviously been OsO4. OsO2F3 in its two modifications
is a mixed-valence Os(VI)/Os(VIII) compound, whereas a new compound Os2O3F7 is a mixed-valence OsV/Os(VIII)
compound. The molecular structures of OsO3F, OsO2F3, and OsO3F2 are calculated. OsO3F2 seems to exist in two
forms, with D3h and Cs symmetry. The original preparation of OsF7 could not be reproduced, only OsF6 has been
obtained.
Introduction
is due to a charge-transfer excitation to Os(VI)/Os(VIII).
After all, isostructural Li5ReO6 and Na5ReO6 are only
yellow.6
The oxidation state +8 is easily accessible for osmium
because of the existence of stable OsO4. This can be
converted into a fair number of other osmium (VIII)
compounds.1 The number of Os(VII) compounds is much
smaller. In contrast to this, the related RuO4 is easily reduced
to stable RuO4 , which is even commercially available.
The reduction of OsO4 into OsO4 has finally been
achieved by means of (C6H5)4As+I-2. The resulting (C6H5)4-
As+OsO4- shows, according to vibrational data, a distorted
OsO4 tetrahedron. It would be worthwhile to reinvestigate
this synthesis and add a single-crystal determination to find
out whether the distortion of the anion is a Jahn-Teller effect
at work. Under typical high-temperature solid-state synthesis
2+
The cations (t-BuN)4Os2 and (t-ButN)6Os+ apparently
contain an Os(VII)-Os(VII) or Os(VII)-Os(VI) nucleus,
respectively.7,8 They are formed by the reduction of
(t-BuN)4Os.
We find four osmium (VII) fluorine compounds in the
literature: (1) OsO3F,9 (2) OsO2F3,10,11 (3) OsOF5,10,12-15 and
(4) OsF7.16
-
-
Here, these compounds are reinspected to see whether they
exist and whether they are true Os(VII) compounds.
Experimental Section
Caution! Handling anhydrous HF or compounds that produce
HF upon hydrolysis requires eye and skin protection.
5-
3-5
conditions, the oxo anion OsO6 is formed.
complete description is on Li5OsO6 and Na5OsO6.5
The most
Materials and Apparatus. Sample handling was performed
using Teflon-PFA ((poly)perfluoroether-tetrafluoroethylene) tubes
that are sealed at one end and equipped at the other end with a
metal valve; they are connectable to a stainless steel vacuum line.
These reaction vessels are pumped on the metal vacuum line and
These are prepared at high temperatures from Os, O2, and
Li2O or Na2O. The black materials are identified by single-
crystal structure determinations, elemental analysis, and
magnetic measurements and clearly seem to be Os(VII)
compounds. In light of the mixed-valence structure of
OsO2F3, one might speculate whether the intense black color
(7) Danopoulos, A. A.; Wilkinson, G.; Hussein-Bates, B.; Hursthouse,
M. B. J. Chem. Soc., Dlton Trans. 1991, 269-275.
(8) Danopoulos, A. A.; Wilkinson, G.; Hussein-Bates, B.; Hursthouse,
M. B. J. Chem. Soc., Dalton Trans. 1991, 1855-1860.
(9) Burbank, R. D. J. Appl. Crystallogr. 1974, 7, 4-44.
(10) Sunder, W. A.; Stevie, R. A. J. Fluorine Chem. 1975, 6, 449-463.
(11) Falconer, W. E.; DiSalvo, F. J.; Griffiths, J. E.; Sunder, W. A.; Vasile,
M. G. J. Fluorine Chem. 1975, 6, 499-520.
(12) Bartlett, N.; Jha, N. K. J. Chem. Soc. A 1967, 536-543.
(13) Bartlett, N.; Jha, N. K.; Trotter, J. Proc. Chem. Soc. 1962, 277.
(14) Bartlett, N.; Trotter, J. J. Chem. Soc. A 1967, 543-7.
(15) Holloway, J. H.; Selig, H.; Classen, H. H. J. Chem. Phys. 1971, 54,
4305-4311.
(16) Glemser, O.; Roesky, H. W.; Hellberg, K.-H.; Werther, H.-H. Chem.
Ber. 1966, 99, 2652-2662.
* To whom correspondence should be addressed. E-mail: seppelt@
chemie.fu-berlin.de
(1) Gerkan, M.; Schrobilgen, G. J. Inorganic Chemistry in Focus II; Meyer,
G., Naumann, B., Wesemann, L., Eds.; Wiley-VCH: Weinheim,
Germany, 2005; pp 243-246.
(2) Bilger, E.; Pebler, J.; Weber, R.; Dehnicke, K. Z. Naturforsch. 1984,
39b, 259-261.
(3) Sleight, A. W.; Longo, J.; Ward, R. Inorg. Chem. 1962, 1, 245-
249.
(4) Scholder, R.; Schatz, G. Angew. Chem., Int. Ed. 1963, 2, 164.
(5) Betz, T.; Hoppe, R. Z. Anorg. Allg. Chem. 1985, 524, 17-25.
(6) Betz, T.; Hoppe, R. Z. Anorg. Allg. Chem. 1984, 512, 19-33.
10.1021/ic0608290 CCC: $33.50
Published on Web 08/16/2006
© 2006 American Chemical Society
Inorganic Chemistry, Vol. 45, No. 19, 2006 7929

Shorafa and Seppelt
are passivated with elemental fluorine for several hours, filled with
dry argon, and then transferred to a drybox for use. The drybox of
the M. Braun Co., Germany, has water and O2 contents of less
than 1 ppm.
then attached to a metal vacuum line to remove traces of volatiles.
The autoclave was transferred to the drybox, opened there, and a
green material was scraped off the lid and stored in PFA containers.
Os2O3F7: 600 mg (2 mmol) of OsO4 and 900 mg (2.5 mmol)
and OsF6 are reacted as described above, except at 170 °C for 24
h. Green, crystalline material, mp 121.3°. IR (Nujol, cm-1): 997.9
m, 954.3 s, 668.0 s, 638.1 s, 527 s. Raman spectrum (cryst., cm-1):
997(15), 966(100), 711(30), 694(10), 652(10), 625(5), 606(3), 488-
(5,br), 374(60), 311(2), 276(5), 266(5), 230(2), 215(2), 119(20).
OsOF5: 377 mg (1.5 mol) of OsO4 mg and 1039 mg (4.2 mmol)
of OsF6 are reacted as described above at 230 °C for 24 h. All
volatile materials were removed between -20 and -30 °C within
10-15 min. Pure OsOF5 is obtained by sublimation at room
temperature into a PFA trap at -196 °C.
Attempts to prepare OsF7: 1000 mg (5.26 mmol) of osmium
metal powder was heated at 620 °C with 10 mL of fluorine in an
80 mL monel autoclave. The product was pumped directly at 600
°C through a 4 mm PFA tube that was cooled to -196 °C while
excess F2 was pumped away immediately. The tube was sealed off
at both ends, and the Raman spectrum was taken from the yellow
sublimate (OsF6). The pressure of F2 in the autoclave was calculated
assuming ideal gas behavior to be about 400 bar. To check this
value by an experiment, the autoclave has been equipped with a
pressure gauge and filled with the same molar amount (10 mL) of
Ar. The physical properties of Ar and F2 are very similar (bp, critical
data, mol wt), and indeed the measured pressure of 420 bar Ar is
very close to the calculated pressure of F2.
Single crystals were grown from dry HF by slow cooling from
0 to -78 °C over a period of 2-3 days. Crystals have also been
obtained by sublimation. Crystals were handled under nitrogen
cooling to approximately -140 °C in a special device,17 and with
this mounted on a Bruker SMART CCD 1000 TU diffractometer,
using Mo KR irradiation, a graphite monochromator, a scan width
of 0.3° in ω, and a measuring time of 10 or 20 s per frame. Each
compound was measured up to 2θ ) 61° by 1800 frames (OsOF5
up to 2θ ) 86° and 3600 frames), covering a full sphere.
Semiempirical absorption corrections (SADABS) were used by
equalizing symmetry-equivalent reflections. Because the refractive
power of the compound was high, very small crystals (0.02 mm)
were chosen, to minimize absorption effects. The structures were
solved and refined with the SHELX programs.18
Calculations. The program package GAUSSIAN 03 has been
used.14 Basis sets: Os, Re: Electron core potentials for 60 core
electrons from Institut fu¨r Theoretische Chemie, Universita¨t Stut-
tgart, and scalar relativistically corrected basis sets 8s7p6d for
valence electrons. F, O: 6-31+G(d,p) and aug-cc-pVTZ basis sets,
as implemented in the GAUSSIAN program. B3LYP-DFT cal-
culations according to Becke20 in the version of Lee, Yang, and
Parr,21 and coupled cluster calculations, CCSD(T), all as imple-
mented in the GAUSSIAN Program.
Raman spectra were recorded on a Bruker RFS 100 FT-Raman
spectrometer.
Results and Discussion
OsO3F. The osmium(VII) oxide fluoride with the highest
oxygen content would be OsO3F. There is only one report
about its possible existence.9 A colorless crystalline material
has been obtained from OsF6 by reaction with glass and has
considerable volatility. The single-crystal structure determi-
nations could not distinguish between OsO3F and OsO2F2.
It has existed in two crystallographic modifications, mono-
clinic, C2/c, and cubic, P4h3n. In the meantime, the crystal
structures of OsO4 and RuO4 have been (re)investigated,22-24
and the data for OsO4 are identical to those of the claimed
OsO3F/OsO2F2, see Table 1, including all positional param-
eters. Also, the colorless nature and volatility of the
compound in question point to OsO4. We have often
observed that OsF6 and the oxide fluorides convert eventually
to OsO4 in the case of slow hydrolysis. According to our
calculations, molecular OsO3F is a minimum on the potential
hypersurface. Its structure is quite distorted from ideal C3V
symmetry; it has only Cs symmetry, see Figure 1. This is
further evidence that the material obtained in 1974 was not
OsO3F.
OsO4 has been purchased from Chempur, Karlsruhe, Germany,
and used without further purification. OsF6 is prepared from osmium
powder and excess elemental fluorine in an 80 mL monel autoclave
at 250 °C. HF was redistilled in a metal vacuum line and stored
over BiF5.
OsO2F3: OsO4 (140 mg, 0.59 mmol) was loaded into an 80 mL
stainless steel autoclave in the drybox. OsF6 (169.5 mg, 0.56 mmol)
is condensed onto the OsO4 at -196 °C with the help of a metal
vacuum line. The autoclave was allowed to warm to room
temperature and then heated for 16-20 h at 150 °C. At the end of
the reaction, the autoclave was cooled to room temperature and
(17) Kottge, T.; Stalke, G. J. Appl. Crystallogr. 1996, 29, 465-468.
(18) Sheldrick, G. H. SHELXS-93, Program for Crystal Structure Solution;
Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1986 and 1993.
(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.
A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P.
Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J.
B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision
B.04; Gaussian, Inc.: Pittsburgh, PA, 2003.
Table 1. Crystallographic Data of Claimed OsO3F or OsO2F2,3 in
17,19
18
Comparison to OsO4
and RuO4
modification
OsO3F/OsO2F2
OsO4
RuO4
a (pm)
b (pm)
c (pm)
â (deg)
a (pm)
C2/c
942
449.1
860
117.5
859.5
937.4(4)
451.5(2)
863.0(3)
116.58(4)
856.8(1)
930.2(4)
439.6(1)
845.4(4)
116.82(3)
850.9(1)
(20) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
P4h3n
(21) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(22) Krebs, B.; Hasse, K. Acta Crystallogr., Sect. B 1976, 32, 1334-1337.
(23) Play, M.; Wickleder, M. S. J. Solid State Chem. 2005, 178, 3206-
3209.
(24) Play, M., Wickleder, M. S.; Shorafa, H.; Seppelt, K. Unpublished
results.
OsO2F3. OsO2F3 is a green sublimable material, first
prepared in 1975. No conclusive structure has been presented,
even in the most detailed description,11 although elemental
7930 Inorganic Chemistry, Vol. 45, No. 19, 2006

Osmium(VII) Fluorine Compounds
Figure 1. Calculated structures of the OsO3F, OsO2F3, ReO2F3, and OsO3F2 molecules, method B, as indicated in Table 4; italic numbers, CCSD(T)
calculation, numerical values in picometers and degrees.
Table 2. Experimental Data of the Single-Crystal Structure Determinations
[OsO3F2‚OsOF4]∞
[OsO3F2‚OsOF4]2
[OsO3F2‚OsF5]2
OsOF5
fw
color
558.4
green
558.4
green
561.4
red-brown
542.3(2)
996.0(3)
1220.3(4)
99.21(5)
650.7
P21/c
4
39.1
110
8063
301.2
green
923.4(2)
492.9(1)
848.3(1)
90
386.1
Pna21
4
a (pm)
b (pm)
c (pm)
â (deg)
V (1 × 106 pm2)
space group
Z
540.57(5)
501.31(5)
1213.73(12)
99.896(3)
324.02
Pc
2
39.3
111
10 796
3322
541.8(1)
993.3(2)
1229.6(2)
99.776(4)
652.13
P21/c
4
39.0
110
10 957
1989
µ (mm-1
)
33.0
66
no. of variables
no. of reflns measured
no. of unique reflns
GOF
16 159
2767
1.044
0.029
0.095
1975
1.072
0.037
0.087
1.071
0.029
0.063
1.079
0.031
0.080
R (I > 4σ(I))
wR2 (all data)
Table 3. Important Bond Lengths (pm) and Angles (deg) of [OsO3F2·OsOF4]∞, [OsO3F2·OsOF4]2, [OsO3F2·OsF5], and OsOF5
[OsO3F2‚OsOF4]∞
[OsO3F2‚OsOF4]2
[OsO3F2‚OsF5]2
OsOF5
Os(VIII)dO
Os(VIII)-F
170.6(9)
171.1(8)
171.1(11)
183.4(9)
214.0(8)
214.2(7)
170.7(10)
182.0(9)
182.8(8)
185.5(9)
202.9(7)
204.4(8)
136.8(4)
142.2(4)
168.o(5)
169.5(6)
173.0(5)
183.2(4)
217.8(4)
220.1(4)
176.7(5)
182.5(5)
182.7(5)
184.7(5)
198.8(5)
199.8(5)
134.7(2)
140.6(2)
167.9(6)
168.5(6)
171.7(6)
183.8(5)
222.3(5)
225.2(5)
183.4(5)
185.3(5)
185.3(5)
185.6(5)
195.2(5)
195.5(5)
135.3(2)
141.3(3)
OsdO
Os-Feq
170.1(12)
182.1(10)
183.6(4)
183.9(7)
184.4(4)
187.2(7)
94.2(5)-94.8(5)
179.1(6)
a
a
Os-Fax
OdOs-Feq
OdOs-Fax
Os(VI,V)dO
Os(VI,V)-F
Os-FdOs
a Feq ) equatorial fluorine atom, Fax ) axial fluorine atom.
analyses, mass spectra, magnetic measurements, vibrational
spectra, and X-ray powder data have been obtained. It has
been remarked that the lattice constants of OsO2F3 are very
close to those of OsO3F2. This is indeed correct for one of
the two modifications of OsO2F3. But because the structure
of OsO3F2 became known only later,25 no conclusion could
be drawn from this fact.
The original preparation is easily repeated, only the
outcome is strongly dependent on the ratio of the starting
materials OsO4 and OsF6 and the reaction temperature.
modifications, both emerald green. One is a fluorine-bridged
polymer, the other a fluorine-bridged tetramer, see Tables 2
and 3 and Figures 2 and 3. Often the polymer is obtained
by sublimation, the tetramer by recrystallization from dry
HF. At -100 °C, the polymer has a slightly smaller cell
volume, so one could assume it is the more stable modifica-
tion. In both structures, we find two different osmium centers.
According to their oxygen and fluorine environments, one
is an Os(VIII) atom and the other an Os(VI) atom, because
oxygen and fluorine can readily be distinguished by the
distances toward Os, as well as by the displacement and
R-factors if positions are exchanged. The entire compound
can be described as [OsO3F2‚OsOF4]. The polymer modifica-
tion is reminiscent of OsO3F2,25 TcO2F3,26 ReO2F3,27 and
some other oxide fluorides. In Figure 2, the structural
similarity is shown between the polymer structures of
OsO3F2, OsO2F3, and OsOF4.28 It is interesting to know that
(25) Bougon, R.; Buu, B.; Seppelt, K. Chem. Ber. 1993, 126, 1331-1336.
(26) Mercier, H. P. A.; Schrobilgen, G. J. Inorg. Chem. 1993, 32, 145-
151.
Crystals of OsO2F3 are readily obtained by sublimation
or by recrystallization form dry HF. It comes in two
Inorganic Chemistry, Vol. 45, No. 19, 2006 7931

Shorafa and Seppelt
Figure 3. ORTEP representations of the crystal structures of [OsO3F2‚
OsOF4]2 (top) and [OsO3F2‚OsF5]2 (bottom) at the 50% probability level.
Because OsO2F3 can be sublimed, it might exist as a
monomeric Os(VII) molecule in the gas phase. IR spectra
of mixtures containing matrix-isolated OsO2F3 have been
tentatively interpreted in terms of a D3h structure for this
molecule (trigonal-bipyramidal structure with the two oxygen
atoms in axial positions).31 We consider this structure as
unlikely, and have therefore made DFT calculations on
OsO2F3; for comparison, we have also made calculations on
ReO2F3 and OsO3F2, see Table 4 and Figure 1. The predicted
structures for OsO2F3 and ReO2F3 have C2V symmetry, with
two equatorially positioned oxygen atoms in a distorted
trigonal-bipyramidal overall structure. These calculations
agree well with previous ones on ReO2F3.32 A D3h structure
for OsO2F3 is no minimum on the potential hypersurface and
is about 62.7 kcal mol-1 higher in energy. A few remarkable
differences can be found by comparing OsO2F3 and
ReO2F3: The angle OdMdO is predicted to be 10° larger
in OsO2F3. In ReO2F3, counterintuitively, the axial Re-F
bond is slightly shorter than the equatorial Re-F bonds, in
contrast to the situation in OsO2F3. These differences have
to be the “sterical” influence of the single d-electron in
OsO2F3.
Figure 2. Crystal structure of (OsO3F2‚OsOF4)∞ (above), in comparison
to OsO3F2 (middle25), and OsOF4 (below28). ORTEP representations, 50%
probability plots.
by a qualitative interpretation of the Raman spectra one of
the several possible structures that have been proposed is
indeed the same as in the tetrameric modification.11 The
formulation of these oligomeric species as, for example,
[OsO3F2‚OsOF4]2 is arbitrary. Because the fluorine bridges
are asymmetric, an alternative formulation could be [OsO3F+]-
-
[OsOF5 ]. The relation to other derivatives of eight-valence
-
OsO3F2, such as [OsO3F+][SbF6 ], is then more evident.29
In the reaction between OsF6 and OsO4, a new compound
is also formed that, according to its crystal structure, has
the composition of Os2O3F7. It has a structure similar to that
of the cyclic OsO2F3, but it consists of alternating Os(VIII)
and Os(V) units and could be described as [OsO3F2‚OsF5],
see Tables 2 and 3 and Figure 3. The archetype of such
tetrameric fluorine-bridged structures is that of RhF5.30
(30) Morell, B. K.; Zalkin, A.; Tressaud, A.; Bartlett, N. Inorg. Chem. 1973,
12, 2640-2644.
(27) Supel, J.; Seppelt, K. Z. Anorg. Allg. Chem. 2005, 631, 2979-2986.
(28) Shorafa, H.; Seppelt, K. To be published.
(29) Gerkan, M.; Dixon, D. A.; Schrobilgen, G. J. Inorg. Chem. 2002, 41,
259-277.
(31) Hope, E. G.; Levason, W.; Ogden, J. S. J. Chem. Soc., Dalton Trans.
1988, 997-999.
(32) Casteel, W. J., Jr.; Dixon, D. A.; Mercier, H. P. A.; Schrobilgen, G.
Inorg. Chem. 1996, 35, 4310-4322.
7932 Inorganic Chemistry, Vol. 45, No. 19, 2006

Osmium(VII) Fluorine Compounds
a,b
Table 4. DFT Calculations on OsO3F, OsO2F3, ReO2F3, and OsO3F2
energy (a.u.)
frequencies (cm-1
)
OsO3F, Cs
A
B
-416.2838737
-416.3551804
A′: 1012.9, 914.7, 647.2, 314.6, 271.5, 204.2
A′′: 985.9, 304.8, 140.1
OsO2F3, C2V
A
B
-540.8012203
-540.8864519
A1: 1028.4, 674.0, 614.1, 317.9, 239.3
A2: 330.5
B1: 649.7, 323.8, 269.5
B2: 908.9, 159.0, 131.7
ReO2F3, C2V
A
B
-528.5297848
-528.6200437
A1: 1060.1, 688.5, 610.1, 355.0, 252.2
A2: 330.8
B2: 289.0, 1029.1, 271.7, 93.2
B1: 672.0, 369.3, 289.0
OsO3F2, D3h
A
B
-516.1361704
-516.2251073
A1′: 997.6, 604.2
A2′′: 642.5, 270.0
E′: 984.6, 326.9, 198.7
E′′: 362.9
C
-514.6527956
OsO3F2, Cs
A
B
-516.1327411
-516.2231704
A′: 1019.1, 981.9, 635.4, 568.1, 421.3, 360.4, 345.3, 263.8
A′′: 999.2, 403.5, 290.6, 62.0
C
-514.6467762
a For geometrical parameters, see Figure 1. b A: B3LYP method. Os, Re: electron core potential for 60 core electrons, 8s7p6d basis set for valency
electrons; F, O: 6-311+G(d,p) basis set. B: B3LYP method. Os, Re: as above; F, O: aug-cc-pVTZ basis set C:CCSD(T) method, basis sets and electron
core potentials as in A.
Surprisingly OsO3F2 with eight-valence osmium represents
a special case. Two minima with D3h and Cs structures are
observed that differ by only between 2.1 and 3.8 kcal mol-1
in energy. The lowest-energy transition state that converts
the D3h into the Cs structure lies 34.3 kcal mol-1 above the
assumption of a 2-fold disorder.14 We have calculated our
data in space group Pna21 as a twin, resulting in better
crystallographic characteristics, including quite precise OsdO
and Os-F distances, see Table 3. These crystallographic
results are very similar to those of TcOF5.35
D
3h state. On the other hand, the Cs structure has a very soft
OsF7. There is only one report on the preparation of this
compound. It has been obtained from a high-temperature/
high-pressure fluorination of Os and is assumed to decom-
pose into OsF6 and fluorine slowly at -100 °C and rapidly
at room temperature.16 Attempts to generate it by other
means, e.g., by the reaction of OsF6 or OsOF5 and KrF2 or
KrF+, have failed.36 It would be only the third existing neutral
heptafluoride, after IF7 and ReF7.
bending mode of 62 cm-1, namely, the OsF2 twist, which
has a barrier of only 0.33 kcal mol-1, see Figure 4. The
energetics of the D3h-Cs interconversion and the OsF2 twist
in the Cs modification are shown in Figure 4. Looking at
the solid-state structure of OsO3F2, the Cs structure of the
molecule looks to be the more likely one. However, the IR
and Raman spectra of OsO3F2 in N2 or Ar matrixes have
been assigned to the D3h structure.33 Previous calculations
on the OsO3F2 have only found the D3h molecule.34
IF7 and ReF7 can be obtained readily and purely if certain
precautions are taken. The most precise structural information
is from neutron powder diffraction data.37-39 IF7 is almost
regular pentagonal-bipyramidal, with only small deviations
from the ideal structure, mainly among the five equatorial
fluorine atoms (“ring puckering”). ReF7 is, however, quite
distorted. The description as a pentagonal bipyramidal
structure holds only as a first approximation. Already, in light
OsOF5. There are several reported preparations for this
compound.12-14 At first, it was obtained by simultaneous
oxidation and fluorination of Os metal.13 We made it from
OsO4 and OsF6 at fairly high temperatures between 220 and
230 °C, which always results in the formation of OsOF4 as
a byproduct. Purification is readily done by sublimation of
this highly volatile compound (bp 100.6 °C).13 It has almost
the same green color as OsO2F3, but previous experiments,
including a single-crystal structure determination, prove it
to be a monomeric species, so it is definitely an Os(VII)
compound. The previous single-crystal structure determina-
tion has been calculated in space group Pnma and under the
(35) Le Blond, N.; Mercier, H. P. A.; Dixon, D. A.; Schrobilgen, G. J.
Inorg. Chem. 2000, 39, 4494-4509.
(36) Bougon, R.; Cicha, W. V.; Isabey, J. J. Fluorine Chem. 1994, 67,
271-276.
(37) Vogt, T.; Fitch, A. N.; Cockcroft, J. K. J. Solid State Chem. 1993,
103, 275-279.
(38) Marx, R.; Mahjoub, A.-R.; Seppelt, K.; Ibberson, R. M. J. Chem. Phys.
1994, 101, 585-593.
(33) Beattie, I. R.; Blayden, H. E.; Crocombe, R. A.; Jones, P. J.; Ogden,
J. S. J. Raman Spectrosc. 1976, 4, 313-322.
(34) Veldkamp, A.; Frenking, G. Chem. Ber. 1993, 126, 1325-1330.
(39) Vogt, T.; Fitch, A. N.; Cockcroft, J. K. Science 1994, 263, 1265-
1267.
Inorganic Chemistry, Vol. 45, No. 19, 2006 7933

Shorafa and Seppelt
Figure 4. Energy diagram of the OsO3F2 molecule. R refers to the bond angle between the fluorine atoms, â to the twist (dihedral) angle of the two fluorine
atoms versus one OsdO bond.
of this structure, it seemed worthwhile to repeat the original
preparation of OsF7.
ments. However, the only compound we observed was very
pure OsF6. Its solid state Raman spectrum is identical to an
authentic sample and to a previously published spectrum:40
It consists of a very strong and sharp band at 731 cm-1 (A1g),
a weak and complex broad feature at around 620 cm-1 (Eg)
broadened by Jahn-Teller or crystal site splitting, and a
medium to weak and broad T2g deformation mode at 321
cm-1. Because it is safe to assume that OsF7 would have a
somewhat similar structure and Raman spectrum to ReF7,
we can state that within the sensitivity of the Raman
experiment, no other compounds have been present.
We used an all monel autoclave with 80 mL volume and
heated Os (99.99%) with HF-free elemental fluorine to 620
°C and 400 bar. The equipment is very similar to that
described in ref 16. As in ref 16, the only part where the
interior gas has contact to anything other than the monel are
a few square millimeters of the needle of the outlet valve,
which was heated to only ∼100 °C. Possible impurities of
the stainless steel fluorination could be volatile CrF5 and
MoF6, but these have not been detected at all. We have
expanded the mixture, while holding it at 620 °C and 400
bar, into a vacuum and immediately froze out the condens-
able material at -190 °C, obtaining a yellow powder. This
has been subjected to Raman spectroscopy immediately at
-140 °C without any warming in between. In case OsF7
was obtained, we were prepared for neutron powder experi-
Acknowledgment. We gratefully acknowledge support
from the Deutsche Forschungsgemeinschaft and the Fonds
der Chemischen Industrie.
Supporting Information Available: Crystallographic data in
CIF format. This material is available free of charge via the Internet
at http://pubs.acs.org.
(40) Shamir, J.; Malm, J. G. J. Inorg. Nucl. Chem., Suppl. 1976, H. H.
Hyman memorial issue, 107-111.
IC0608290
7934 Inorganic Chemistry, Vol. 45, No. 19, 2006