Catalytic Degradation of Sulfur Hexafluoride by Rhodium Complexes

Article abstract of DOI:10.1002/anie.201505462

The development of a safe and efficient method for the degradation of SF₆ is of current environmental interest, because SF₆ is one of the most potent greenhouse gases. SF₆ is thermally and chemically extremely inert, and therefore, it has been used in various industrial applications. However, this inertness results in a major challenge for its depletion. We report on a process for a catalytic degradation of SF₆ in the homogeneous phase by using rhodium complexes as precatalysts. The SF₆ activation reactions feature mild reaction conditions, low catalyst loadings, and a high selectivity. The employment of phosphines and hydrosilanes for scavenging the sulfur and fluorine atoms of the SF₆ molecule allows the selective transformation of SF₆ into nongaseous and nontoxic compounds.

Full text of DOI:10.1002/anie.201505462

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201505462
German Edition: DOI: 10.1002/ange.201505462
SÀF Activation Very Important Paper
Catalytic Degradation of Sulfur Hexafluoride by Rhodium Complexes
Lada Zµmostnµ and Thomas Braun*
[11,13–18]
Abstract: The development of a safe and efficient method for
reaction at polyisoprene surfaces.
However, the prod-
the degradation of SF is of current environmental interest,
ucts produced by these methods are mostly gaseous, toxic, and
even corrosive, and include HF, F , SF , S F , SO F , S OF ,
6
because SF is one of the most potent greenhouse gases. SF is
6
6
2
4
2
10
2
2
2
10
thermally and chemically extremely inert, and therefore, it has
been used in various industrial applications. However, this
inertness results in a major challenge for its depletion. We
SOF , SOF , H S, SO , and SO . To date, catalytic decom-
2 4 2 2 3
positions of SF have only been achieved heterogeneously,
6
[16–18]
mainly at metal phosphates.
However, these reactions
report on a process for a catalytic degradation of SF in the
require temperatures above 800 K and give SO , SO F , and
6
3
2
2
homogeneous phase by using rhodium complexes as precata-
lysts. The SF6 activation reactions feature mild reaction
conditions, low catalyst loadings, and a high selectivity. The
employment of phosphines and hydrosilanes for scavenging
HF as major products. Therefore, it is of current interest to
develop alternative pathways for a selective degradation of
SF . Reactions mediated by transition-metal complexes
6
represent a promising approach to decompose SF6 under
the sulfur and fluorine atoms of the SF molecule allows the
mild conditions. Hitherto, the activation of SF at transition
6
6
selective transformation of SF into nongaseous and nontoxic
compounds.
metals was achieved at low-valent Ti, V, Cr, and Zr complexes
6
[19–21]
as well as at Fe and at reduced Ni complexes.
In most of
the cases the fate of the sulfur atom remained unclear. We
A
controlled degradation of sulfur hexafluoride (SF ) under
previously reported on the degradation of SF at the binuclear
6
6
[22]
mild conditions by a catalytic process can be considered as an
environmental objective of considerable importance because
an atmospheric emission of SF can thus be prevented. SF has
rhodium complex [{Rh(m-H)(dippp)} ]. In the presence of
2
HSiEt the reaction led selectively to fluorosilane, H , and the
3
2
thiolato-bridged complex [Rh (m-H)(m-SSiEt )(dippp) ],
6
6
2
3
2
been globally recognized as potent greenhouse gas with the
highest global warming potential known, which is 23500 times
which was obtained exclusively as the only sulfur-containing
product. For an efficient decomposition of SF a homogeneous
6
higher than that of CO , with an atmospheric lifetime of about
catalytic process is still desirable.
2
[
1–3]
3
200 years.
Hence, it emerged among the six most
Herein we report on the unprecedented catalytic degra-
[1]
prominent greenhouse gases included in the Kyoto Protocol.
dation of SF in a homogeneous phase by using rhodium
6
SF is widely used in a variety of industrial applications and
complexes. Phosphines and silanes were employed as scav-
engers for the sulfur and fluorine atoms. Thus, the developed
method allows the selective transformation of SF6 into
nongaseous phosphine sulfides and fluorosilanes and it
proceeds under mild conditions.
6
processes owing to its unique properties, such as a low toxicity,
[4–8]
extreme inertness, and a high dielectric constant.
It is
mainly employed as a gaseous dielectric and electron-
[
6,7]
trapping agent for high-voltage power applications.
Prior
to its usage in industry, SF6 was not detected in the
atmosphere, which indicates that its presence is entirely
Initial studies showed that a solution of [Rh(H)(PEt ) ]
3
3
(1) reacts with SF at room temperature. The reaction resulted
6
anthropogenic. Note that the global SF concentration has
in the generation of F PEt accompanied by the formation of
6
2
3
grown from less than 1 ppt in 1975 to more than 8 ppt in
a black solid, which indicates the decomposition of 1.
Consequently, the reaction was performed in the presence
[
3]
2
008. Taking into account that there are often no alternative
chemicals to replace SF , increasing attention has been paid to
of a large excess of the phosphine PEt , which in principle
allows for a regeneration of 1 after a fluorination of metal-
6
3
[3,9–15]
control or even avoid its emission.
In the past decade, SF was mainly degraded or recycled
bound PEt to give F PEt . Note that, in the presence of free
6
3
2
3
[
11,13–16]
by adsorption, separation, and decomposition methods.
Because of its chemical inertness approaches for a selective
degradation of SF are extremely challenging.
PEt complex 1 is in equilibrium with the compound [Rh-
3
(H)(PEt ) ] (2), and with an excess PEt the equilibrium is
3
4
3
[
4,5,8]
[23]
Methods
shifted towards 2. The reaction of 1 with an excess SF in
6
6
for its decomposition include harsh conditions, such as high
temperatures and pressure. Many studies involve the thermal
and photoreductive decomposition of SF6 in electric dis-
charges associated with plasma etching, or a photolytic
the presence of 40 equivalents of free PEt at room temper-
3
[24]
ature gave several rhodium complexes, of which we could
[25]
identify [Rh(PEt ) ][HF ] (3), [Rh(F)(PEt ) ] (4),
and
3
4
2
3 3
[{Rh(m-F)(PEt ) } ] (5) Compound 5 can be synthesized
3
2 2
independently by treatment of 1 with NEt ·3HF at low
3
temperature. The NMR spectroscopic data of the reaction
mixture also revealed the formation of F PEt (2.1%) and the
[
*] M. Sc. L. Zµmostnµ, Prof. Dr. T. Braun
Department of Chemistry
Humboldt-Universität zu Berlin
Brook-Taylor-Strasse 2, 12489 Berlin (Germany)
E-mail: thomas.braun@chemie.hu-berlin.de
2
3
phosphine sulfide SPEt (1.4%; Table 1, entry 2; yields are
3
based on the amount of PEt added to the reaction mixture).
3
However, heating the reaction to 808C yielded selectively the
ionic complex [Rh(PEt ) ][HF ] (3) as the sole rhodium
compound after 16 h reaction time. Monitoring the reaction
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201505462.
3
4
2
1
0652
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10652 –10656

Angewandte
Chemie
Table 1: Rhodium catalyzed SF degradation.
6
[
b]
[d]
Entry Catalyst
precursor
R
HSiR’3
Conditions
Yield TON
SPR
3
[
a]
[c]
[
mol%]
[%]
[
e]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
–
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
–
–
–
–
808C, 16 h
RT, 16 h
808C, 16 h
–
0
1, 2.5
1, 2.5
1, 2.5
–
1.4
7.1
0.6
2.8
4.8
0
2.0
12
11
0.4
48
61
76
47
808C, 2 days 12
[
e]
HSiEt3
HSiEt3
HSiEt3
HSiiPr3
HSi(OEt)3 808C, 16 h
808C,16 h
RT, 16 h
808C, 16 h
808C, 16 h
–
31
1
Figure 1. P{ H} NMR spectrum of the catalytic reaction of SF with
6
1, 2.5
1, 2.5
1, 2.5
1, 2.5
1, 0.2
1, 0.2
1, 0.2
1, 0.2
1, 2.5
1, 0.2
1, 2.5
1, 0.2
1, 2.5
1, 0.2
1, 2.5
1, 0.2
6, 2.5
6, 0.2
6, 0.2
5.2
29
27
0.9
9.4
PEt ; after 16 h reaction time at 808C (Table 1, entry 3); *=P(C H F)
3
6
5
3
[
f]
(external standard).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
HSiEt3
808C, 16 h
Complex 3 was synthesized in an independent manner.
HSiEt3
HSiEt3
808C, 2 days 12
808C, 5 days 15
The reaction of 1 with 2 equivalents NEt ·3HF in the presence
3
of an excess PEt gave [Rh(H) (PEt ) ][HF ] (6) in 79% yield
[
g]
3
2
3
4
2
HSiEt3
808C, 16 h
808C, 16 h
9.3
36
[26–28]
[
h]
[f]
after 2 h at room temperature (Scheme 2).
Subsequent
Me HSiEt3
Me HSiEt3
iPr3 HSiEt3
iPr
tBu
tBu HSiEt3
Ph
Ph
Et
Et
Et
Et
14
86
[
h]
treatment of 6 with neohexene resulted in the formation of 3.
The complexes 3 and 6 were characterized by NMR and IR
spectroscopy as well as by liquid-injection field desorption
ionization mass spectrometry (LIFDI-MS). LIFDI-MS data
for complex 6 revealed a peak at m/z 577.28 which can be
808C, 2 days 17
808C, 16 h 14
808C, 2 days 0.4
5.6
2.2
1.1
2.0
2.6
8.2
12
42
86
0
3
HSiEt
3
HSiEt3
[
[
[
[
i]
i]
i]
i]
808C, 16 h
808C, 2 days 0.4
808C, 16 h 6.4
808C, 2 days 1.6
808C, 16 h 29
2.9
HSiEt3
+
assigned to the molecular ion [M] and a peak at m/z 459.25
HSiEt3
+
[f]
which corresponds to the molecular ion [M -PEt ]. The
HSiEt3
HSiEt3
HSiEt3
3
31
1
808C, 2 days 8.4
808C, 5 days 17
P{ H} NMR spectrum of 6 displays two signals at d =
25.0 ppm and 9.3 ppm with the expected doublet of triplets
splitting pattern in a 1:1 ratio, consistent with a cis-config-
uration of the hydrido ligands at the metal center. The
phosphorus–rhodium coupling constants are 98.6 Hz and
86.6 Hz, typical for a rhodium center in the + III oxidation
[
e]
KHF2
HSiEt3
808C, 16 h
–
[
a] Experiments were performed under 1 atm of SF gas in [D ]toluene as
6 8
solvent. Typically, 0.9–3.1 mmol PR were used; see the Supporting
Information for experimental details. The mol% catalyst precursor based
on the amount of PR added. [b] Equivalent amounts of the hydrogen
source and PR were used. [c] Yields of SPR are based on the amount of
PR used and were determined from P NMR spectra by integration of
product resonance signals versus the external standard, using an inverse
gated H decoupling pulse sequence. [d] TON=amount of SPR (mol)/
amount of Rh catalyst (mol); The amount of SPR was calculated based
on the amount of PR used. [e] No reaction observed. [f] Complete
conversion of HSiEt to FSiEt observed. [g] Neohexene added (1 equiv
based on the amount of PEt ). [h] SPMe precipitated in part from the
reaction mixture; the yield was determined from P NMR spectra by
integration of product resonances versus the external standard to
3
[25,29]
1
3
state.
In the H NMR spectrum of 6, the resonance for the
3
3
hydrido ligands appears as a multiplet at d = À12.04 ppm. It
3
1
3
simplifies to a doublet in the phosphorus decoupled spectrum
1
1
( J
= 11.2 Hz), which confirms the mononuclear structure
H,Rh
3
À
of 6. For the [HF ] counter anion in 6 a characteristic broad
resonance is found in the H NMR spectrum at d = 12.14 ppm
3
2
1
3
19
[30–32]
3
3
and in the F NMR spectrum at d = À173.2 ppm.
The IR
3
3
spectrum of 6 showed characteristic broad bands at 2720 and
3
1
À1
1
777 cm . Both frequencies are close to those found in
À
simple bifluoride salts and can be assigned to the [HF ]
ion.
presence of free PEt . The cation [Rh(PEt ) ] was described
before.
2
determine the consumption of PMe , taking into account the generation
of F PMe . [i] No formation of F PR observed.
3
[
31–34]
Compound 3 is only stable in solution and in the
2
3
2
3
+
3
3 4
[35]
To increase the efficiency of the catalytic process, silanes
were added to the reaction mixture as a hydrogen source and
fluoride scavenger. Silanes reconvert fluorido complexes into
by NMR spectroscopy reveals that 3 is formed initially. This
suggests that complex 3 actually serves as a catalyst precursor.
[25,36–38]
In addition, the formation of SPEt (7.1%), as the only
hydrido species.
Initial reactivity studies at 1 revealed
3
sulfur containing species, and the generation of F PEt (18%)
were observed (Scheme 1, Figure 1; Table 1, entry 3). The
that the complex reacts with HSiEt to form the dihydrido
silyl complex cis-fac-[Rh(H) (SiEt )(PEt ) ] (7). However,
2 3 3 3
2
3
3
[39]
relatively large amount of SPEt3 indicates that the SÀF
in the presence of an excess PEt a reductive elimination of
3
activation is catalytic with respect to the SF molecules which
the silane occurred immediately to yield the hydrido com-
pound 2. Clearly 7 is not an active species in the catalytic
cycle, but can serve as a resting state. Complex 7 was
characterized in solution by NMR spectroscopy, as it loses
6
are converted, with a TON of 2.8 (TON = mol of SPEt /mol of
3
1
). The yield of SPEt can be increased by longer reaction
3
times. Thus, after 2 days, 12% SPEt were formed, which is
3
equivalent to a TON of 4.8 for the degradation of SF6
molecules (Table 1, entry 4).
HSiEt already in vacuo. The complex is fluxional on the
NMR time-scale as it was found for the structurally related
3
Angew. Chem. Int. Ed. 2015, 54, 10652 –10656
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10653

.
Angewandte
Communications
catalyst
.2 mol% 1 a TON of 48
was achieved after 16 h at
08C (Table 1, entry 10).
loading
of
0
8
Furthermore, a remark-
able increase in the con-
version was observed
after 2 days and 5 days
reaction time to give
TON of 61 and 76,
respectively
entries 11,12).
(Table 1,
Adding
neohexene as trapping
agent for the formed H2
has a negligible effect on
the conversion to give
Scheme 1. Catalytic Degradation of SF at [Rh(H)(PEt ) ] (1). Left: Reaction in the presence of PEt ; Table 1,
6
3
3
3
entry 3; Reaction conditions: 78 mmol 1, 3.1 mmol PEt , 1 atm SF , solvent: [D ]toluene. Right: Reaction in the
3
6
8
presence of PEt and HSiEt ; Table 1, entry 7; Reaction conditions: 32 mmol 1, 1.3 mmol PEt , 1.3 mmol HSiEt ,
SPEt in 9.3% yield with
a TON of 47 after 16 h at
3
3
3
3
3
1
atm SF , solvent: [D ]toluene.
6 8
8
08C (Table 1, entry 13).
No decomposition of SF6
was observed without
rhodium
catalyst
(Table 1, entries 1,5,25),
which indicates that the
SÀF activation and con-
Scheme 2. Independent synthesis of the precatalysts 3 and 6.
version of SF6 proceeds
at the rhodium center. In
addition, SF6 does not
[25,29]
À
2
complex cis-fac-[Rh(H) (SiPh )(PEt ) ],
which exhibits
react with KHF , the [HF ] ion being the anion in 3 and 6
2
3
3
3
2
NMR spectroscopic data that are comparable to those of 7
(Table 1, entry 25).
(
see the Supporting Information).
Note that in our previous study on employing binuclear
[
22]
The reaction of SF with PEt and HSiEt in the presence
rhodium compounds for SF activation the cleavage of the
6
3
3
6
of 2.5 mol% 1 gave selectively SPEt3 as the only sulfur-
containing product in 29% yield after 16 h at 808C, equiv-
alent to a TON of 12 based on the decomposition of SF6
molecules (Scheme 1, Figure 2; Table 1, entry 7). FSiEt3
sulfur–rhodium bonds turned out to be a major restriction to
accomplish a catalytic pathway. We attributed this to the
bridging sulfur atom between two rhodium centers at [Rh (m-
2
H)(m-SSiEt )(dippp) ], which makes the rhodium–sulfur
3
2
interaction very stable. We therefore assume that reactions
at cationic rhodium centers, such as in 3 or 6, do not lead to
binuclear sulfide species, which might hamper a catalytic
turnover. Indeed, further studies revealed that the ionic
complex 6 is also able to catalyze the degradation of SF with
6
HSiEt and PEt . The reaction proceeded also selectively to
3
3
yield H as well as F PEt , FSiEt , and SPEt as the only SF
6
2
2
3
3
3
decomposition products. By using 2.5 mol% of 6, the TON for
the SPEt3 formation after 16 h at 808C was 12 (Table 1,
entry 22), which is identical to the reaction with 1 as catalyst
precursor (Table 1, entry 7). By employing 0.2 mol% of 6
a higher TON of 86 was achieved (Table 1, entry 24). Note
that no intermediates were observed by NMR spectroscopy
during the reaction. As above for the reactions without silane,
we assume that 3 is formed initially.
3
1
1
Figure 2. P{ H} NMR spectrum of the rhodium-catalyzed reaction of
SF with PEt and HSiEt ; after 16 h reaction time at 808C (Table 1,
6
3
3
entry 7); *=P(C H F) (external standard).
6
5
3
(
> 99%) and F PEt (30%) were generated concomitantly
For a possible enhancement of the catalytic SF activation,
2
3
6
along with the formation of H . The ionic rhodium(III)
we investigated other silanes as hydrogen sources. With
2
dihydrido complex [Rh(H) (PEt ) ][HF ] (6) was detected as
2.5 mol% of 1, SF was treated with PEt and either with
2
3
4
2
6
3
the sole organometallic species. Note that the reaction
proceeded with a lower selectively by using a ratio of
HSiEt and PEt which is not 1:1. The TON is then limited
HSiiPr or HSi(OEt) . The reaction with HSiiPr (Table 1,
3
3
3
entry 8) proceeded slightly worse than with HSiEt (Table 1,
3
entry 7) to give SPEt in 27% yield with a TON of 11. By
3
3
3
by the amount of hydrosilane used, because it is entirely
using HSi(OEt) the yield of SPEt decreased dramatically to
3
3
converted into FSiEt . However, on employment of a lower
0.9% (TON 0.4; Table 1, entry 9). In addition, treatment with
3
1
0654 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10652 –10656

Angewandte
Chemie
gaseous H resulted in a low conversion (TON 1.3) as well,
Keywords: cationic complexes · rhodium · SÀF activation ·
2
consistent with the fact that H is also formed during the
sulfur hexafluoride · transition-metal catalysis
2
reaction. Moreover, we also tested a variety of different
How to cite: Angew. Chem. Int. Ed. 2015, 54, 10652–10656
Angew. Chem. 2015, 127, 10798–10802
phosphines PR (R = Me, iPr, tBu, Ph) as alternative sulfur-
3
trapping agents. The reactions were accomplished in the
presence of HSiEt , which turned out to be the best suitable
3
hydrogen source, and catalyst loadings of 2.5 mol% 1 as well
[
1] P. Forster, V. Ramaswamy, P. Artaxo, T. Berntsen, R. Betts, D. W.
Fahey, J. Haywood, J. Lean, D. C. Lowe, G. Myhre, J. Nganga, R.
Prinn, G. Raga, M. Schulz, R. Van Dorland, Changes in
Atmospheric Constituents and in Radiative Forcing, in: Climate
Change 2007: The Physical Science Basis. Contribution of
Working Group 1 to the Fourth Assessment Report of the
IPCC, Cambridge University Press, 2007.
2] M. Rigby, J. Mühle, B. R. Miller, R. G. Prinn, P. B. Krummel,
L. P. Steele, P. J. Fraser, P. K. Salameh, C. M. Harth, R. F. Weiss,
B. R. Greally, S. OꢀDoherty, P. G. Simmonds, M. K. Vollmer, S.
Reimann, J. Kim, K.-R. Kim, H. J. Wang, J. G. J. Olivier, E. J.
Dlugokencky, G. S. Dutton, B. D. Hall, J. W. Elkins, Atmos.
Chem. Phys. 2010, 10, 10305 – 10320.
as 0.2 mol% 1 (Table 1, entries 14–21). Before adding SF to
6
the reaction mixtures, the exchange of the rhodium-bound
PEt ligands and the added phosphines PR (R = Me, iPr, tBu,
3
3
Ph) at the metal center was detected by NMR spectroscopy,
resulting in the formation of a number of organometallic
species as well as of PEt . All reactions gave the correspond-
3
[
ing phosphine sulfides SPR as the major sulfur-containing
3
[40]
products.
Minor amounts of SPEt3 were also obtained,
which originate from the catalyst precursor 1. FSiEt and H2
3
were formed in all cases, whereas the generation of the
fluorophosphoranes F PR was only found in the reactions
2
3
with PEt , PMe , and PiPr . Interestingly, the steric demand of
[3] G. P. Stiller, T. von Clarmann, M. Hoepfner, N. Glatthor, U.
Grabowski, S. Kellmann, A. Kleinert, A. Linden, M. Milz, T.
Reddmann, T. Steck, H. Fischer, B. Funke, M. Lopez-Puertas, A.
Engel, Atmos. Chem. Phys. 2008, 8, 677 – 695.
3
3
3
the alkyl phosphines plays
a certain role (Table 1,
entries 7,10–19). Thus, the TON increased in the order
PtBu < PiPr < PEt < PMe . The maximum TON of 86 was
3
3
3
3
[
[
4] K. Seppelt, Chem. Rev. 2015, 115, 1296 – 1306.
5] J. R. Case, F. Nyman, Nature 1962, 193, 473.
achieved in the reaction of SF with PMe and HSiEt in the
6
3
3
presence of 0.2 mol% 1 (entry 15). The reactions with PPh3
[6] E. Preisegger, R. Durschner, W. Klotz, C. A. Konig, H. Krahling,
(
Table 1, entries 20,21) do not follow this trend, presumably
C. Neumann, B. Zahn, Life Cycle Assessment—Electricity
due to the different electronic properties of the aromatic
phosphine compared to the alkyl phosphines.
In conclusion, we have shown that rhodium(I) complexes
effectively catalyze the degradation and transformation of
Supply Using SF Technology, Springer, Heidelberg, 2000.
7] L. G. Christophorou, J. K. Olthoff, R. J. VanBrunt, IEEE Elect.
Insul. Mag. 1997, 13, 20 – 24.
8] R. E. Weston, J. Phys. Chem. 1995, 99, 13150 – 13155.
9] M. Maiss, C. A. M. Brenninkmeijer, Environ. Sci. Technol. 1998,
6
[
[
[
SF . A chemical activation of a SF6 molecule generally
6
3
2, 3077 – 3086.
requires strong reductive conditions and often involves single-
[
10] K. S. Betts, Environ. Sci. Technol. 1998, 32, 487A – 488A.
[11] A. Totterdill, T. Kovµcs, J. Carlos, G. Martín, W. Feng, J. M. C.
À
electron transfer steps to give SF5 and
ion.
a
fluoride
[
5,8,19–22,41,42]
Mechanistically, we therefore suggest a pre-
Plane, J. Phys. Chem. A 2015, 119, 2016 – 2025.
[
[
12] W.-T. Tsai, J. Fluorine Chem. 2007, 128, 1345 – 1352.
13] Q. Zhuang, B. Clements, A. McFarlan, Y. Fasoyinu, Can. J.
Chem. Eng. 2014, 92, 32 – 35.
coordination of SF through one or more fluorine atoms at
6
a cationic rhodium species which originates from 3, because
such an interaction is facilitated at a cationic fragment. The
complexes 1, 2, 6, and 7 can serve as catalyst precursors and/or
resting states. In a subsequent step it is conceivable that the
first SÀF bond is cleaved by an inner-sphere electron transfer.
After the first SÀF bond cleavage step a rapid degradation of
[
[
14] C.-W. Hwang, C. K. Chan, J. Environ. Sci. Int. 2013, 22, 1019 –
1
027.
15] X. Song, X. Liu, Z. Ye, J. He, R. Zhang, H. Hou, J. Hazard.
Mater. 2009, 168, 493 – 500.
[16] C. T. Dervos, P. Vassiliou, J. Air Waste Manage. Assoc. 2000, 50,
37 – 141.
1
the SF (x = 1–5) moieties might occur in the coordination
x
[
17] N.-K. Park, H.-G. Park, T. J. Lee, W.-C. Chang, W.-T. Kwon,
Catal. Today 2012, 185, 247 – 252.
18] D. Kashiwagi, A. Takai, T. Takubo, H. Yamada, T. Inoue, K.
Nagaoka, Y. Takita, J. Colloid Interface Sci. 2009, 332, 136 – 144.
sphere of the metal to give initially rhodium fluorido and
sulfido species. The sulfido species are then converted into an
[
active compound to react with the next SF molecule by the
6
formation of phosphine sulfides, fluorophosphoranes, and
fluorosilanes. Compared to the reactions which proceed
heterogeneously, the developed catalytic conversions are
distinguished by the mild reaction conditions, low catalyst
[19] P. Holze, B. Horn, C. Limberg, C. Matlachowski, S. Mebs,
Angew. Chem. Int. Ed. 2014, 53, 2750 – 2753; Angew. Chem.
2
014, 126, 2788 – 2791.
[
[
[
20] B. G. Harvey, A. M. Arif, A. Gloeckner, R. D. Ernst, Organo-
metallics 2007, 26, 2872 – 2879.
21] R. Basta, B. G. Harvey, A. M. Arif, R. D. Ernst, J. Am. Chem.
Soc. 2005, 127, 11924 – 11925.
loadings, and high selectivity. Furthermore, the SF decom-
6
position products are non-gaseous and much less toxic.
22] L. Zµmostnµ, T. Braun, B. Braun, Angew. Chem. Int. Ed. 2014,
5
3, 2745 – 2749; Angew. Chem. 2014, 126, 2783 – 2787.
Acknowledgements
[23] T. Yoshida, D. L. Thorn, T. Okano, S. Otsuka, J. A. Ibers, J. Am.
Chem. Soc. 1980, 102, 6451 – 6457.
[
24] Compound 1 was used as precatalyst, because it contains
We acknowledge the DFG (Deutsche Forschungsgemein-
schaft) for financial support. We would like to thank Dr. Mike
Ahrens for the MS-LIFDI data analyses. We thank the Solvay
a defined amount of PEt , whereas 2 cannot be isolated as it
3
loses PEt in vacuo.
3
[
25] a) T. Braun, D. Noveski, B. Neumann, H.-G. Stammler, Angew.
Chem. Int. Ed. 2002, 41, 2745 – 2748; Angew. Chem. 2002, 114,
Fluor GmbH for a gift of SF6.
2
870 – 2873; b) D. Noveski, T. Braun, M. Schulte, B. Neumann,
Angew. Chem. Int. Ed. 2015, 54, 10652 –10656
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10655

.
Angewandte
Communications
H. G. Stammler, Dalton Trans. 2003, 4075 – 4083; c) D. Noveski,
T. Braun, S. Krückemeier, J. Fluorine Chem. 2004, 125, 959 – 966.
[38] a) H. Lv, Y.-B. Cai, J.-L. Zhang, Angew. Chem. Int. Ed. 2013, 52,
3203 – 3207; Angew. Chem. 2013, 125, 3285 – 3289; b) H. Lv, J.-H.
Zhan, Y.-B. Cai, Y. Yu, B. Wang, J.-L. Zhang, J. Am. Chem. Soc.
2012, 134, 16216 – 16227; c) P. Fischer, K. Goetz, A. Eichhorn, U.
Radius, Organometallics 2012, 31, 1374 – 1383; d) J. A. Panetier,
S. A. Macgregor, M. K. Whittlesey, Angew. Chem. Int. Ed. 2011,
50, 2783 – 2786; Angew. Chem. 2011, 123, 2835 – 2838; e) T. F.
Beltrµn, M. Feliz, R. Llusar, J. A. Mata, V. S. Safont, Organo-
metallics 2011, 30, 290 – 297; f) M. F. Kühnel, D. Lentz, Angew.
Chem. Int. Ed. 2010, 49, 2933 – 2936; Angew. Chem. 2010, 122,
2995 – 2998; g) S. P. Reade, M. F. Mahon, M. K. Whittlesey, J.
Am. Chem. Soc. 2009, 131, 1847 – 1861; h) J. Vela, J. M. Smith, Y.
Yu, N. A. Ketterer, C. J. Flaschenriem, R. J. Lachicotte, P. L.
Holland, J. Am. Chem. Soc. 2005, 127, 7857 – 7870; T. Braun, J.
Izundu, A. Steffen, B. Neumann, H.-G. Stammler, Dalton Trans.
2006, 5118 – 5123.
+
[
26] The ion [Rh(H) (PEt ) ] was described in literature with
2
3 4
À
À
À
[
OH] , [ClO ] , and [BPh ] as counter anions, but the NMR
data were not discussed.
4
4
[27,28]
[
[
[
27] C. Cocevar, A. Camus, G. Mestroni, J. Organomet. Chem. 1972,
5, 389 – 395.
28] T. Yoshida, T. Okano, Y. Ueda, S. Otsuka, J. Am. Chem. Soc.
981, 103, 3411 – 3422.
29] T. Braun, F. Wehmeier, K. Altenhoener, Angew. Chem. Int. Ed.
007, 46, 5321 – 5324; Angew. Chem. 2007, 119, 5415 – 5418.
3
1
2
[
[
30] No splitting of the signals was observed up to 183 K.
31] N. A. Jasim, R. N. Perutz, S. J. Archibald, Dalton Trans. 2003,
2
184 – 2187.
[
[
[
[
[
[
32] B. K. Bennett, R. G. Harrison, T. G. Richmond, J. Am. Chem.
Soc. 1994, 116, 11165 – 11166.
33] G. PØrez-Hernµndez, J. Gonzalez-Vazquez, L. Gonzalez, J. Phys.
Chem. A 2012, 116, 11361 – 11369.
34] The presence of polyfluoride anions of the formula [F(HF)n]
[39] M. Teltewskoi, Dissertation, Humboldt-Universität zu Berlin,
2014.
[40] P NMR spectroscopic data of the reactions with 2.5 mol% 1 in
À
31
(
n > 1) might be also possible.
35] J. J. Gambaro, W. H. Hohman, D. W. Meek, Inorg. Chem. 1989,
8, 4154 – 4159.
36] L. Zµmostnµ, M. Ahrens, T. Braun, J. Fluorine Chem. 2013, 155,
32 – 142.
37] a) T. Ahrens, J. Kohlmann, M. Ahrens, T. Braun, Chem. Rev.
015, 115, 931 – 972; b) M. F. Kuehnel, D. Lentz, T. Braun,
Angew. Chem. Int. Ed. 2013, 52, 3328 – 3348; Angew. Chem.
013, 125, 3412 – 3433; c) T. Braun, F. Wehmeier, Eur. J. Inorg.
Chem. 2011, 613 – 625.
the presence of PMe (Table 1, entry 14) and PtBu (Table 1,
3
3
entry 18) showed signals for small amounts (5.1% and 4.5%) of
unidentified minor products.
2
[41] J. Pola, P. Engst, M. Horak, Collect. Czech. Chem. Commun.
1979, 44, 2092 – 2095.
[42] D. Sevenard, P. Kirsch, A. A. Kolomeitsev, G.-V. Rçschenthaler
1
(Merck KGaA), DE 10220901, 2004.
2
2
Received: June 14, 2015
Published online: July 16, 2015
1
0656 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10652 –10656