Pentafluoronitrosulfane, SF5NO2

Article abstract of DOI:10.1021/ic0516212

The synthesis of pentafluoronitrosulfane, SF₅NO₂, is accomplished either by reacting N(SF₅)₃ with NO ₂ or by the photolysis of a SF₅Br/NO₂ mixture using diazo lamps. The product is purified by treatment with CsF and repeated trap-to-trap condensation. The solid compound melts at -78°C, and the extrapolated boiling point is 9°C. SF₅NO₂ is characterized by ¹⁹F, ¹⁵N NMR, IR, Raman, and UV spectroscopy as well as by mass spectrometry. The molecular structure of SF ₅NO₂ is determined by gas electron diffraction. The molecule possesses C_2v symmetry with the NO₂ group staggering the equatorial S-F bonds and an extremely long 1.903(7) A S-N bond. Calculated bond enthalpies depend strongly on the computational method: 159 (MP2/6-311G++(3df)) and 87 kJ mol^-1 (B3LYP/6-311++G-(3df)). The experimental geometry and vibrational spectrum are reproduced reasonably well by quantum chemical calculations.

Full text of DOI:10.1021/ic0516212

Inorg. Chem. 2006, 45, 1783−1788
Pentafluoronitrosulfane, SF5NO2
Norman Lu,*,† Joseph S. Thrasher,† Stefan von Ahsen,‡ Helge Willner,*,‡ Drahomir Hnyk,§ and
Heinz Oberhammer*,|
Department of Chemistry, UniVersity of Alabama, Tuscaloosa, Alabama 35487,
FB C Anorganische Chemie, UniVersita¨t Wuppertal, Gausstrasse 20,
D-42119 Wuppertal, Germany, Institute of Inorganic Chemistry, Academy of Sciences,
Cz-25068 Rezˇ, Czech Republic, and Institut fu¨r Physikalische und Theoretische Chemie,
UniVersita¨t Tu¨bingen, D-72076 Tu¨bingen, Germany
Received September 21, 2005
The synthesis of pentafluoronitrosulfane, SF5NO2, is accomplished either by reacting N(SF5)3 with NO2 or by the
photolysis of a SF5Br/NO2 mixture using diazo lamps. The product is purified by treatment with CsF and repeated
trap-to-trap condensation. The solid compound melts at −78 °C, and the extrapolated boiling point is 9 °C. SF5NO2
is characterized by 19F, 15N NMR, IR, Raman, and UV spectroscopy as well as by mass spectrometry. The molecular
structure of SF5NO2 is determined by gas electron diffraction. The molecule possesses C2v symmetry with the NO2
group staggering the equatorial S
−
F bonds and an extremely long 1.903(7) Å S
−
N bond. Calculated bond enthalpies
depend strongly on the computational method: 159 (MP2/6-311G++(3df)) and 87 kJ mol- (B3LYP/6-311++G-
(3df)). The experimental geometry and vibrational spectrum are reproduced reasonably well by quantum chemical
calculations.
1
Introduction
Herein we report the first successful preparation of the
long sought after pentafluoronitrosulfane, SF5NO2, and its
unambiguous characterization.
Derivatives of sulfur hexafluoride of the type SF5-X are
of great interest from a chemical and bonding point of view.
The kinetically inert and bulky SF5 group with a highly
charged sulfur atom has a strong influence on the S-X bond
and on the charge distribution of the substituent X. This has
consequences for the stability of the SF5-X molecules and
the synthetic routes. There are numerous SF5-C, SF5-N,
and SF5-O compounds that are well known,1-4 but the
synthesis of SF5-X compounds, with X being an element
of the second or higher period (except for X ) SF5, Cl, or
Br) or some simple substituent (e.g., H, I, NO, NO2, N3,
etc.), has been a challenge for preparative chemists.
Experimental Section
Synthesis of SF5NO2. The synthesis of SF5NO2 was carried out
4
by two different methods: (i) by the treatment of N(SF5)3 with
NO2 at room temperature and (ii) by the photolysis of a mixture of
SF5Br5 and NO2.
(i) The amine N(SF5)3 (0.36 g, 0.90 mmol) and NO2 (0.12 g,
2.60 mmol) were vacuum transferred at -196 °C into an FEP tube
equipped with a metal valve. Subsequently, the reaction vessel was
allowed to warm gradually to room temperature, and after 4 h, all
N(SF5)3 crystals had disappeared. The volatile products were
subjected to repeated trap-to-trap condensation in traps held at
-105, -130, and -196 °C. A few milligrams of SF5NO2 was
retained in the trap at -130 °C.
(ii) The compounds SF5Br (3.9 g, 18.8 mmol) and NO2 (0.9 g,
19.6 mmol) were transferred into a 4 L Pyrex reactor, which was
surrounded by 12 diazo lamps (TL 40 W/03). After 12 h of
irradiation, the resulting product mixture was condensed into a 300
mL stainless steel cylinder held at -196 °C. This cylinder was
then warmed to the temperature of dry ice, and all of the materials
that are volatile at this temperature were vacuum transferred into
* To whom correspondence should be addressed. E-mail: willner@
uni-wuppertal.de (H.W.), heinz.oberhammer@uni-tuebingen.de (H.O.),
normanlu@ntut.edu.tw (N.L.).
† University of Alabama.
‡ Universita¨t Wuppertal.
§ Academy of Sciences.
| Universita¨t Tu¨bingen.
(1) Tullock, C. W.; Coffmann, D. D.; Muetterties, E. L. J. Am. Chem.
Soc. 1964, 86, 357.
(2) Senning, A. Sulfur in Organic and Inorganic Chemistry; Dekker: New
York, 1982; Vol. 4.
(3) Gmelins Handbuch der Anorganischen Chemie; Springer-Verlag:
Berlin, 1978.
(4) Thrasher, J. S.; Nielson, J. B. J. Am. Chem. Soc. 1986, 108, 1108.
(5) Winter, R.; Terjeson, R.; Gard, G. L. J. Fluorine Chem. 1998, 89,
105.
10.1021/ic0516212 CCC: $33.50
Published on Web 01/14/2006
© 2006 American Chemical Society
Inorganic Chemistry, Vol. 45, No. 4, 2006 1783

Lu et al.
another cylinder containing 400 g of CsF. In this manner, Br2, SOF4,
and SF4 were removed from the reaction mixture. Subsequently,
trap-to-trap condensations through three traps held at -78, -130,
and -196 °C were carried out to separate SF5NO2 from impurities
such as FNO and SF6. About 90% pure SF5NO2 (0.09 g, 0.56 mmol)
was obtained in the -130 °C trap with a yield of 3%. Several
batches were again purified together by trap-to-trap condensation
under IR control. The resulting sample of about 98% purity, with
traces of S2F10 and SF4O, was then used for the following
characterization.
Analysis. Volatile materials were manipulated in a glass vacuum
line of known volume, equipped with two capacitance pressure
gauges (MKS Baratron 221 AHS-1000 and 221 AHS-10 Wilm-
ington, MA), three U-traps, and valves with PTFE stems (Young,
London, U.K.). The vacuum line was connected to an IR gas cell
(optical path length 200 mm, 0.5-mm-thick silicon windows) in
the sample compartment of an FTIR instrument. This arrangement
made it possible to follow the improvement in the purification
process. The product was stored in glass ampules under liquid
nitrogen. The ampules were opened and flame sealed again by
means of an ampule key.6
(a) NMR Spectroscopy. The 19F , 15N, and 14N NMR spectra
were recorded on a Bruker model AM 500 spectrometer at 470.507,
50.677, and 36.127 MHz for 19F, 15N, and 14N nuclei, respectively.
The samples were condensed and sealed into glass tubes (4 mm
o.d.) on a vacuum line. As the external reference/lock, CFCl3/CD2-
Cl2, K15NO3/D2O, and NH4NO3/D2O (reference peak, NO3-, was
set at 383 ppm) were used.
(b) Vibrational Spectroscopy. IR spectra of gaseous samples
were measured in the range of 4000-400 cm-1 with an FTIR
spectrometer (type 400 D, Nicolet, Madison, WI) with an optical
resolution of 2 cm-1, and 32 scans were coadded for each spectrum.
Matrix IR spectra of SF5NO2 were recorded on a Bruker IFS
66v/s FTIR instrument in reflectance mode using a transfer optic.
A DTGS detector and a KBr/Ge beam splitter were used in the
4000-400 cm-1 region. Sixty-four scans were coadded for each
spectrum using an apodized resolution of 1 cm-1. Details of the
matrix apparatus have been described elsewhere.7,8 A small amount
of pure SF5NO2 (ca. 0.1 mmol) was kept in a small U-trap at -196
°C and mounted in front of the matrix support. The trap was allowed
to reach a temperature of -65 °C while a gas stream (2-4 mmol
h-1) of argon or neon was directed over the SF5NO2 sample and
the resulting gas mixture was quenched on the matrix support at
15 or 6 K, respectively. During matrix deposition, the end of the
spray-on nozzle in front of the matrix support was heated to different
temperatures.
An FT-Raman spectrum of a solid SF5NO2 sample at -196 °C
was measured in the region of 3000-50 cm-1 at a resolution of 4
cm-1 with an FT-Raman spectrometer (RFS 100/s, Bruker, Ger-
many) by coadding 128 scans and using the 1064 nm excitation
line (500 mW) of an Nd:YAG laser. The sample was condensed
as a spot on a nickel-plated copper finger kept at -196 °C in high
vacuum. The solid sample was then excited with the laser through
a quartz window.
Figure 1. Experimental (•) and calculated (-) molecular intensities for
long (above) and short (below) nozzle-to-plate distances and residuals.
transferred into a glass cell of 10 cm optical path length equipped
with suprasil windows. The absorption cross sections (on base e)
were calculated by means of the equation σ ) 31.79TAp-1d-1
,
where σ is the cross-section in 10-20 cm2, T is the temperature in
K, A is the absorbance, p is the gas pressure in mbar, and d is the
optical path length in cm.
(d) Mass Spectrometry. Gas chromatography/mass spectrometry
data were obtained using a Hewlett-Packard HP 6890 series gas
chromatograph with a series 5973 mass-selective detector. The
column was a 6890 GC using a 30 m × 0.250 mm HP-1 capillary
column with a 0.25 µm stationary-phase film thickness. The flow
rate was 1 mL/min and splitless. The electron impact mass spectrum
of SF5NO2 was obtained at 70 eV. Because SF5NO2 is thermally
unstable, special conditions were used in an attempt to detect the
parent ion and the expected fragmentation pattern. For example,
the normal injector temperature of 250 °C and detector temperature
of 280 °C were lowered to 48 and 65 °C, respectively. The column
temperature was 35 °C. However, no parent ion (SF5NO2+) was
observed in the electron impact spectrum even under these
conditions.
(e) Vapor Pressure. The vapor pressure of the neat sample was
measured by using the above-mentioned vacuum line. The tem-
perature of the sample reservoir was adjusted with a series of ethanol
cold baths and measured with a Pt-100 resistance thermometer.
Occasionally, the gas phase was checked through its IR spectrum.
Very little decomposition (<1%) was detected at temperatures up
to -10 °C. Before recording the vapor pressures, we determined
the melting point of SF5NO2 in the reservoir.
(f) GED Measurements. Electron scattering intensity data for
SF5NO2 were recorded on Kodak electron image plates using a
KDG2-Diffraktograph9 at the University of Tu¨bingen, operating at
approximately 60 kV, at two nozzle-to-plate distances (25 and 50
cm). The sample was kept at -60 °C, and the inlet nozzle was at
room temperature during the experiments. Scattering data for ZnO
were recorded simultaneously and used to calibrate the electron
wavelength. Data were obtained in digital form using a microden-
sitometer at the University of Ulm. The photographic plates were
analyzed by the usual procedures.10 Averaged molecular intensities
θ
in the s ranges of 2-18 and 8-35 A-1 (s ) (4π/λ) sin /2, λ )
electron wavelength, θ ) scattering angle) are shown in Figure 1.
(g) Theoretical Calculations. The structure of the title compound
was optimized with the MP2 approximation and the B3LYP method
using 6-311++G(3df) basis sets. The calculated structure possesses
C2V symmetry with the NO2 group staggering the equatorial S-F
bonds. The calculated (MP2) barrier of the four-fold potential
(c) UV Spectroscopy. UV spectra of SF5NO2 were recorded at
room temperature with a Perkin-Elmer Lambda 900 spectrometer
with a resolution of 1 nm. Different amounts of the sample were
function for internal rotation around the S-N bond is 5.4 kJ mol-1
.
(6) Gombler, W.; Willner, H. J. Phys. E: Sci. Instrum. 1987, 20, 1286.
(7) Schno¨ckel, H.; Willner, H. In Infrared and Raman Spectroscopy:
Methods and Applications; Schrader, B., Ed.; VCH: Weinheim,
Germany, 1994; p 297.
(8) Argu¨ello, G. A.; Grothe, H.; Kronberg, M.; Willner, H.; Mack, H.-G.
J. Phys. Chem. 1995, 99, 17525.
(9) Oberhammer, H. In Molecular Structure by Diffraction Methods;
Chemical Society: London, 1976; Vol. 4, p 24.
(10) Oberhammer, H.; Gombler, W.; Willner, H. J. Mol. Struct. 1981, 70,
273.
1784 Inorganic Chemistry, Vol. 45, No. 4, 2006

Pentafluoronitrosulfane, SF5NO2
Geometry optimizations with small basis sets (6-31G(d)) predict
the S-N and S-F bonds to be too long by up to 0.14 and 0.07 Å,
respectively. All of these calculations were performed using the
Gaussian 98 package.11 Vibrational amplitudes and vibrational
corrections for all interatomic distances were derived from a
calculated force field (MP2/6-31G(d)), using the Sipachev
method.12-14
Results and Discussion
Synthesis and Thermal Properties of SF5NO2. Two
routes to SF5NO2 have been found: (i) thermal reaction of
N(SF5)3 with NO2 and (ii) photolysis of a SF5Br/NO2 mixture.
The compound N(SF5)3 slowly decomposes at room
temperature,4 and the primary step in the decomposition
N(SF5)3
9
∆8 N(SF5)2 + SF5
(1)
Figure 2. 19F NMR spectrum of SF515NO2.
has been studied through matrix-isolation experiments.15 The
SF5 radicals can recombine either with each other or with
NO2:
quent reactions of the SF5 radicals in the reaction mixture
lead to SF5NO2 and side products S2F10, SF6, SF4, SF4O,
FNO, Br2, et cetera. The expected product BrNO2 is not
stable under the reaction conditions.18
SF5
SF5ONO + SF5NO2 7NO SF5 8 S2F10
(2)
(3)
2
hν
SF5Br + NO2 excess 420 nm8 SF5NO2 + side products (4)
SF5ONO f SF4O + FNO
The side products of some fluoride ion affinities are absorbed
by CsF, and the remaining compounds are separated from
SF5NO2 by repeated trap-to-trap condensation.
Pure SF5NO2 is a colorless gas that decomposes mainly
into S2F10 and NO2 (and some SF4O and FNO) at a rate of
about 3% per day at room temperature. The white solid melts
at -78 °C, and the boiling point amounts to 9 °C by
extrapolating the vapor-pressure curve
As observed in the reaction of NO2 with chlorine atoms,16
O- or N-bonded products SF5ONO or SF5NO2 can be formed.
In an attempted synthesis of SF5ONO by reacting SF5OCl
with ClNO, only the formation of Cl2, SF4O, and FNO was
found.17 Hence, SF5ONO quickly decomposes according to
eq 3.
Because SF5NO2 also decomposes under the reaction
conditions, the yield of the desired product is very low.
4
Furthermore, the difficult multistep synthesis of N(SF5)3
3788
T
ln(p/po) ) -
+ 13.33
(5)
precludes a study of its chemistry.
In contrast, starting material SF5Br5 is readily available.
Therefore, irradiation of SF5Br/NO2 mixtures with diazo
lamps (λmax ) 420 nm) has allowed the preparation of gram
quantities of the title compound. SF5Br does not absorb light
at 420 nm; hence, excited NO2 radicals must be involved in
the formation of SF5 radicals. (The threshold for the
photodissociation of NO2 is below 420 nm.) Several subse-
recorded between -75 and -10 °C.
SF5NO2 has been used as a good source of SF5 radicals.15
By low-pressure flash thermolysis of SF5NO2 with subse-
quent trapping of the products in inert gas matrixes, mainly
SF5 and NO2 are detected by IR spectroscopy along with
SF4O, FNO, SF4, and SF6.15
Spectroscopic Properties. (a) NMR Spectra. Figure 2
shows the 19F NMR spectrum of SF515NO2 with the typical
AB4 pattern and the splitting of the B4 signals (Feq) into a
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.;
Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, revision A.5; Gaussian, Inc.:
Pittsburgh, PA, 1998.
doublet due to the 15N atom. An expansion of equatorial 19
F
resonances is depicted in Figure 3. Spectra simulation yielded
2
δ(Fax) ) 46.65; δ(Feq) ) 42.92; J(FaxFeq) ) 144.1; and
2J(F15N) ) 11.6 Hz.
The quintet of the 15N NMR resonance is shown in Figure
4, which also confirms the structure of SF515NO2 with four
equatorial fluorines symmetrically bonded to the S atom. The
four equatorial fluorines couple to the 15N atom but not to
the axial fluorine. It is known that the coupling constant of
(12) Sipachev, V. A. J. Mol. Struct: THEOCHEM 1985, 121, 143.
(13) Sipachev, V. A. AdV. Mol. Struct. Res. 1999, 5, 263.
(14) Sipachev, V. A. NATO Sci. Ser., Ser. II 2002, 68, 73.
(15) Kronberg, M.; Ahsen, S. v.; Willner, H.; Francisco, J. S. Angew. Chem.,
Int. Ed. 2005, 44, 253.
1
an axial fluorine to a given nucleus is usually /10 that of
equatorial fluorines to the same nucleus. However, in the
(16) Niki, H.; Maker, P. D.; Savage, P. D.; Breitenbach, L. P. Chem. Phys.
Lett. 1978, 59, 78.
(17) Ulic, S. E.; Willner, H. Unpublished work.
(18) Scheffler, D.; Grothe, H.; Willner, H.; Frenzel, A.; Zetzsch, C. Inorg.
Chem. 1997, 36, 335.
Inorganic Chemistry, Vol. 45, No. 4, 2006 1785

Lu et al.
Figure 3. 19F NMR spectrum of SF515NO2, the expanded B4 part.
Figure 6. UV spectrum of SF5NO2 in the gas phase at 23 °C.
p) + 3a2 (Ra dp) + 5b1 (IR, Ra dp) + 6b2 (IR, Ra dp)
vibrations. They can be attributed to 8 stretching, 1 torsional,
and 12 deformation modes. Assignment to the relevant
symmetry class is assisted by the observed band contours in
the gas-phase IR spectrum and isotopic shifts as well as
through comparison with characteristic group frequencies in
related molecules and predicted band positions and the
relative IR and Raman intensities obtained by quantum
chemical calculations. Although the band positions (except
for νa NO2) are reproduced reasonably well with the MP2/
6-31G(d) method, the IR and Raman band intensities fit
better with the B3LYP/6-311G(d) method.
The SF5NO2 molecule is a nearly symmetric top rotor;
therefore, the a1-type bands should exhibit a pronounced PQR
band contour. This is true for the gas-phase bands at 1304,
801, and 597 cm-1. The assignment of characteristic NO2
vibrations νa ) 1653, νs ) 1304, and δ ) 801 cm-1 is
confirmed by 15N substitution, which causes the shifts to
occur at lower wavenumbers of 36, 16, and 8 cm-1,
respectively. The most intense IR band near 910 cm-1 in
the gas phase is split into several components in the Ne
matrix, which are assigned to the SFax stretching mode and
the two components of νa SF4eq. All other bands are more or
less coupled, and the description of modes on the basis of
the calculated displacement vectors is in part arbitrary.
(c) UV Spectrum. In the UV region, gaseous SF5NO2
shows an unstructured absorption at 276 nm and does not
absorb the radiation of the diazo lamps used in the synthesis.
The absorption may be assigned to the n f π* transition of
the NO2 chromophore. Toward shorter wavelengths, there
is increased absorption with the maximum below 200 nm.
The spectrum is depicted in Figure 6, and absorption cross
sections are shown in Table 2.
Figure 4. 15N NMR spectrum of SF515NO2.
Figure 5. IR spectrum of gaseous SF5NO2 (0.6/10.3 mbar, 20 cm optical
path length, 23 °C, upper trace) and Raman spectrum of solid SF5NO2 at
-196 °C (lower trace). Impurities - *.
case of SF515NO2, coupling between the axial fluorine and
the 15N atom was not observed.
(b) Vibrational Spectra. The gas-phase IR and solid-
phase Raman spectra of SF5NO2 are depicted in Figure 5.
All vibrational data observed in the gas phase, in a neon
matrix, and for a solid sample are listed in Table 1 and are
compared with predicted data from quantum chemical cal-
culations, and a tentative assignment of modes is also given.
The 21 fundamental vibrations of the SF5NO2 molecule
with the symmetry point group C2V transform as 7a1 (IR, Ra
(d) Mass Spectrum. The 70 eV mass spectrum of SF5-
NO2 shows the following fragment ion pattern, m/z (%,
+
+
+
ion): 127 (100, SF5 ), 108 (6.7, SF4 ), 89 (51, SF3 ), 81
(1.5, SFNO+), 70 (12, SF2 ), 64 (12, SO2 ), 51 (5.9, SF+),
+
+
+
46 (69, NO2 ). Although the molecular ion was missing, the
mass spectrum does have fragments of m/u ) 46 and 127
indicating the presence of NO2 and SF5, respectively. In
addition, the peak at m/z ) 81, which corresponds to SFNO+,
1786 Inorganic Chemistry, Vol. 45, No. 4, 2006

Pentafluoronitrosulfane, SF5NO2
Table 1. Experimental and Calculated Fundamental Wavenumbers and Band Intensities of SF5NO2
IR
gas
IR
Ne
Raman
solid
assignm. acc. to
C2V symmetry
e
f
σb
Ic
Id
calcda
IIR
IRaman
1653
1304
353
201
1652
1304
916
910
907
56
21
7
1646
1315
w
1893
1314
948
945
931
793
690
658
649
580
542
523
465
438
373
362
308
305
208
200
75
312
262
34
327
384
395
15
3
0
119
4
10
<1
1
5
10
1
3
4
ν
16 b2
ν1 a1
νa NO2
νs NO2
ν SFax
ν SFeq
ν SFeq
δ NO2
γ NO2S
ν SFeq
ν SFeq
δ SF
m
w
m
ν2 a1
ν
ν
ν3 a1
ν
ν4 a1
ν8 a2
ν5 a1
909
801
1580
335
100
894
802
17 b2
11 b1
800
26
2
<1
29
5
12 b1
687
638
596
582
s
w
vw
vw
597
571
192
31
594
580
567
15
2.8
3.9
3
<1
<1
2
8
2
<1
0
22
2
ν
ν
ν6 a1
18 b2
13 b1
F NO2
δ SF
513
457
407
w
m
w
δ SF
ν
ν
ν
ν9 a2
ν7 a1
ν
ν
ν
19 b2
14 b1
20 b2
δ SF
405
4
407
1.3
3
1
0
19
1
δ SF
δ SF
δ SF
312
226
vs
w
ν SN
15 b1
21 b2
10 a2
δ NSF
δ NSF
τ N-S
<1
0
<1
2
a MP2/6-31G(d). b At band maximum in 10-20 cm2. c Integrated band intensities. d Abbreviations for strong, medium, weak, and very weak. e km‚mol-1
f
(B3LYP/6-311G(d)). Å4‚amu-1 (B3LYP/6-311G(d)).
Table 2. UV Absorption Cross Sections of SF5NO2 in the Gas Phase
λ (nm) σ (10-20 cm2) λ (nm) σ (10-20 cm2) λ (nm) σ (10-20 cm2)
Table 3. Experimental and Calculated Geometric Parameters (r/Å or
/°) for SF5NO2
MP2/
B3LYP/
200
205
210
215
220
225
230
235
240
245
1708
1147
715
420
230
119
60
31
19
15
250
255
260
265
270
276a
280
285
290
295
15
19
23
28
31
32
31
28
23
17
300
305
310
315
320
325
330
335
340
12
GED a
6-311++G(3df) 6-311++G(3df)
7.9
4.7
2.6
1.5
0.8
0.4
0.2
0.1
r(S-N)
1.903(7)
1.560(1)
0.022(14)
1.565(3)
1.543(12)
1.209(3)
90.8(2)
1.895
1.571
0.019
1.575
1.556
1.208
90.7
114.9
130.3
1.979
1.588
0.026
1.593
1.567
1.193
90.7
114.3
131.3
r(S-F)mean
∆SF ) (S-Fe) - (S-Fa)
r(S-Fe)
r(S-Fa)
r(NdO)
Fe-S-Fa
S-NdO
115.4(5)
129.2 (7)
OdNdO
a Absorption maximum.
a rh1 values; uncertainties in parentheses are 3σ values and refer to the
last digit.
gives evidence of the two fragments SF5 and NO2 being
originally bonded together.
when the SF5 group is constrained to C4V symmetry. This
constraint is justified by the ab initio calculations that predict
a difference between the equatorial F-S-F bond angles of
only 0.1°. In the final stage of the structural analysis, based
on least-squares fitting of the molecular intensities, 10
vibrational amplitudes were refined simultaneously with these
6 geometric parameters. Only three correlation coefficients
had values larger than |0.5|: SN/SNO ) -0.68, ∆SF/l2 )
-0.76, and l6/l7 ) -0.51. The final results are listed in Table
3 (geometric parameters) and Table 4 (vibrational ampli-
tudes) along with the calculated values.
By gas density measurements, the molecular mass was
determined to be 173 ( 0.5 g mol-1 (calcd 173.1).
Gas-Phase Structure. The experimental radial-distribution
curve was obtained by Fourier transformation of the molec-
ular intensities, and it is relatively rich in structural informa-
tion (Figure 7). This accounts for the fact that the experi-
mental structure is well determined on the basis of electron
diffraction data. As detailed in Table 3, just six geometric
parameters were needed to define the molecular structure
The most striking feature of the SF5NO2 structure is an
extremely long S-N bond of 1.903(7) Å. This bond is more
than 0.2 Å longer than those in pentafluorosulfenylamines,
such as SF5NF2 (1.691(5) Å),3 (SF5)2NF (1.685(5)Å),19,20 and
in the radical (SF5)2N• (1.692(4) Å).21 It is even longer than
the S-N bond in the highly strained tris(pentafluorosulfenyl)-
amine, (SF5)3N (1.829(6) Å).21 Considering the experimental
(19) Waterfeld, A.; Oberhammer, H.; Mews, R. Angew. Chem. Suppl. 1982,
834.
(20) Waterfeld, A.; Oberhammer, H.; Mews, R. Angew. Chem., Int. Ed.
Engl. 1982, 21, 355.
Figure 7. Experimental radial distribution function and difference curve.
Interatomic distances are indicated by vertical bars. For atom numbering
see Figure 8.
(21) Weiss, I.; Thrasher, J. S.; Oberhammer, H. Unpublished work.
Inorganic Chemistry, Vol. 45, No. 4, 2006 1787

Lu et al.
Table 4. Experimental Interatomic Distances and Experimental and
Calculated Vibrational Amplitudesa
amplitude
GED
amplitude
MP2/6-31G(d)
distance
1.21
1.54-1.56
1.90
2.18
2.21
2.21
2.44
2.64
2.65
3.13
3.43
3.44
4.11
NdO
S-F
S-N
0.032(5)
0.044(2)
0.054(8)
0.047 b
0.064(2)
0.064(2)
0.082(6)
0.181(21)
0.068(8)
0.060(9)
0.132(11)
0.066 b
l1
l2
l3
0.037
0.043
0.059
0.047
0.066
0.065
0.081
0.134
0.064
0.054
0.109
0.066
0.079
O1...O2
Fe...Fe’
Fa...Fe
N...Fe
O1...Fe
S...O
l4
l4
l5
l6
l7
l8
l9
Fe...Fe”
O2...Fe
N...Fa
O...Fa
0.084(15)
l10
a Values in Å; uncertainties in parentheses are 3σ values and refer to
the last digit. b Not refined.
error limit and the systematic difference between vibra-
tionally averaged bond distances (rh1 values), derived in the
experiment, and equilibrium bond distances (re values),
derived with theoretical methods, the experimental S-N
bond length is reproduced very well by the MP2 method
with large basis sets (1.903(7) vs 1.895 Å). It is strongly
overestimated, however, by the B3LYP method (1.979 Å).
Calculated bond enthalpies depend strongly on the compu-
tational method, and values of 159 (MP2) and 87 kJ mol-1
(B3LYP) were derived. Because the B3LYP method over-
estimates the experimental S-N bond length strongly, we
expect that this method underestimates the bond enthalpy
and that the MP2 result (159 kJ mol-1) is closer to the actual
value.
It has been demonstrated that the mean S-F bond length
in SF5X compounds correlates with the electronegativity of
X:22 that is, it decreases with increasing electronegativity.
The mean S-F bond length in SF5NO2 (1.560(1) Å) is
slightly shorter than that in SF6 (1.5623(4) Å),23 indicating
the unexpected result that the NO2 group is slightly more
strongly electron withdrawing than fluorine. This can be
rationalized by the strong π-donor ability of the fluorine atom
Figure 8. Molecular model with atom numbering.
in SF5-F, which is absent in the -NO2 group. This π back-
bonding compensates for some of the electron withdrawing
action of the fluorine atom. The axial S-F bond is shorter
than the equatorial bonds by 0.022(14) Å. This result is in
agreement with the cis influence predicted by Shustorovich
and Buslaev for octahedral main group element compounds
XEL5.24 According to this effect, the E-L bonds cis to X
(equatorial bonds) are longer than the trans (axial) bond,
when the E-X bond is more covalent than the E-L bonds.
This influence is in contrast to that in transition metal
compounds where the trans influence is well established.
Both computational methods predict the mean S-F bond
length to be slightly too long by 0.011 (MP2) and 0.028 Å
(B3LYP).
Acknowledgment. We gratefully acknowledge the fi-
nancial support of the Deutsche Forschungsgemeinschaft
(DFG) and the Academy of Science of the Czech Republic
(research plan AVOZ40320502). D.H. also thanks the DAAD
for a travel grant.
IC0516212
(22) Zylka, P.; Mack, H.-G.; Schmuck, A.; Seppelt, K.; Oberhammer, H.
Inorg. Chem. 1991, 30, 59.
(23) Kelly, H. M.; Fink, M. J. Chem. Phys. 1982, 77, 1813.
(24) Shustorovich, E. M.; Buslaev, Y. Inorg. Chem. 1976, 15, 1142.
1788 Inorganic Chemistry, Vol. 45, No. 4, 2006