C–F Bond Cleavage Reactions with Beryllium, Magnesium, Gallium, Hafnium, and Thorium Halides

Article abstract of DOI:10.1002/zaac.201900297

The work describes unexpected stoichiometric C–F bond cleavage reactions of beryllium, magnesium, gallium, hafnium and thorium halides with α,α,α-trifluorotoluene. The reaction of BeBr₂ / GaBr₃ or MgBr₂ / GaBr₃ mixtures as well as neat GaI₃ with α,α,α-trifluorotoluene in the presence of (OSi₂Me₄)₂ (I) yields the carbenium ion containing compounds [Ph-C(O₂Si₂Me₄)][GaX₄] (X = Br: 1, X = I/F: 2). Both compounds were successfully characterized and a defluorination type reaction under incorporation of a siloxy unit was observed. Compound 1 was also characterized by single-crystal X-ray diffraction analysis. The conversion of α,α,α-trifluorotoluene with BeI₂, HfI₄ or ThI₄ turned out to be a halodefluorination-type reaction with formation of α,α,α-triiodotoluene (3). An adequate NMR spectroscopic and the X-ray crystallographic characterization of 3 were performed for the first time.

Full text of DOI:10.1002/zaac.201900297

Journal of Inorganic and General Chemistry
DOI: 10.1002/zaac.201900297
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
C–F Bond Cleavage Reactions with Beryllium, Magnesium,
Gallium, Hafnium, and Thorium Halides
Fabian Dankert,[a] H. Lars Deubner,[b] Matthias Müller,[c] Magnus R. Buchner,[c] Florian Kraus,[b] and
Carsten von Hänisch*[a]
Dedicated to Professor Bernd Harbrecht on the Occasion of his 70th Birthday
Abstract. The work describes unexpected stoichiometric C–F bond
ination type reaction under incorporation of a siloxy unit was observed.
cleavage reactions of beryllium, magnesium, gallium, hafnium and
Compound 1 was also characterized by single-crystal X-ray diffraction
analysis. The conversion of α,α,α-trifluorotoluene with BeI2, HfI4 or
thorium halides with α,α,α-trifluorotoluene. The reaction of BeBr2
/
GaBr3 or MgBr2 / GaBr3 mixtures as well as neat GaI3 with α,α,α- ThI4 turned out to be a halodefluorination-type reaction with formation
trifluorotoluene in the presence of (OSi2Me4)2 (I) yields the carbenium of α,α,α-triiodotoluene (3). An adequate NMR spectroscopic and the
ion containing compounds [Ph-C(O2Si2Me4)][GaX4] (X = Br: 1, X = X-ray crystallographic characterization of 3 were performed for the
I/F: 2). Both compounds were successfully characterized and a defluor- first time.
F bond activation. So far, a lot of different metal centers were
Introduction
used in several reaction protocols.
Fluorinated compounds are of high importance especially as
Therein, especially transition metals and lanthanides play a
pharmaceuticals, agrochemicals and functional materials.[1–5]
major role, but also main group elements found application.[8]
Because of this, C–F bond formation and activation has
Plenty of metal centers are used and to mention a few (non-
evolved into an essential research area.[6–9]
exhaustive listing), there are examples of Pd2+,[15–18] Zr/
From the chemists’ point of view, however, this chemistry
Hf4+,[19–23] Ni2+,[24–28] Ag+,[29,30] Rh3+,[31–35] Al1+/3+ [36–38]
,
is synthetically challenging due to the fact that a C–F bond is
the strongest one in organic chemistry. Both C–F bond forma-
tion as well as C–F bond cleavage needs innovative, best pos-
sible, simple (catalytic) systems to be achieved efficiently.
With that being said, many review articles were recently pub-
lished, giving an overview about latest findings in C–F bond
activation.[6,7,10] So for example, Ichikawa and co-workers
were very recently reviewing this topic and presented a wide
ranged selection of transition metal C–F bond cleavage reac-
tions.[10] All of these reactions combine the use of metal cen-
ters which are employed for catalytic and/or stoichiometric C–
Ce3+,[39] Yb3+,[40] and Ti3+/4+ [41,42]
The mechanisms around
.
the C-F activations are quite different in these examples.
Whereas the hard ions Ti4+, Zr4+, Hf4+, Al3+, and Yb3+ mainly
act as a Lewis acids which promote C–F activation, the other
metal centers, especially Pd2+ and Ni2+ undergo more sophisti-
cated mechanisms before such a bond is about to be broken.
In general, C–F bond cleavage reactions can be divided into
two-electron reduction by transition metals, single electron re-
ductions (e.g. by low-valent metals) or fluoride abstraction
(e.g. by strong Lewis-acids, no change in oxidation number).[8]
* Prof. Dr. C. von Hänisch
E-Mail: haenisch@chemie.uni-marburg.de
* Fachbereich Chemie and Wissenschaftliches Zentrum für Material-
wissenschaften (WZMW)
Philipps-Universität Marburg
Hans-Meerwein Str. 4
35032 Marburg, Germany
* Fachbereich Chemie: Anorganische Chemie – Fluorchemie
Philipps-Universität Marburg
Hans-Meerwein Str. 4
35032 Marburg, Germany
* Fachbereich Chemie: Anorganische Chemie – Nachwuchsgruppe
Berylliumchemie
Philipps-Universität Marburg
Hans-Meerwein Str. 4
35032 Marburg, Germany
© 2020 The Authors. Published by Wiley-VCH Verlag GmbH &
Co. KGaA. · This is an open access article under the terms of the
Creative Commons Attribution-NonCommercial License, which
permits use, distribution and reproduction in any medium, provided
the original work is properly cited and is not used for commercial
purposes.
Scheme 1. C–F bond cleavage reactions employing silylium cations:
14]
Some eminent examples from past works, also in catalytic fashion.[11
–
The proton of the reactions depicted on the right is neutralized by different
reagents. See cited works.
Z. Anorg. Allg. Chem. 2020, 646, 1–8
1
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
However, all of them lead to deconstruction of chemically inert
C–F bonds (see cited works above for details).
For a few years now, one of our main research topics is the
synthesis and coordination ability of silicon-based ligands.
Silyl ethers and related metal···silyl ether bonding are exam-
ined for quite some time now.[43–50] The research topic is of
current interest giving insights into the Si–O bond as well as
the Lewis-basicity of siloxanes.[51,52] Siloxanes coordinate es-
pecially to cations which are chemically hard such as the early
s-block metal cations.[47,53–55] Applying the wrong reaction
conditions or using tetravalent cations such as Th4+ or Hf4+,
however, no coordination of siloxanes but C–F bond cleavage
reactions of the solvent occurs. These serendipitous C-F-cleav-
age reactions are presented herein. Putting the C–F bond cleav-
age into the context of siloxanes, the results presented herein
might be of importance regarding the recent discussion around
C–F bond activation with silylium cations. Silylium cations
mainly found applications as effective catalysts in hydrode-
fluorination[11,12,56] reactions but recently also in C–H aryl-
ation under simultaneous C–F bond activation[13,14] (see
Scheme 1 for eminent examples).
Scheme 2. Attempted pathway to obtain Si-based crown-ethers (top)
and observed C–F bond cleavage (bottom).
38.5 ppm (2) a strong low-field shift is observed for the sili-
con-based part of the cationic structure as can be seen by com-
paring these species with I [δ(29Si) = 3.6 ppm]. 19F NMR spec-
troscopy of the solution of 2 shows high field signals at
–169.72, –174.13, –175.85, –177.56, and –177.76 ppm. While
the signals at –177.56 and –174.13 ppm are assigned to com-
pound II[59] and an additional methylfluorosilane[60] respec-
tively, the other three resonances exhibit line broadening,
which is indicative for fluorine nuclei bound to quadrupolar
Results and Discussion
Our original goal was to obtain exclusively-silicon based nuclei like 69Ga or 71Ga. These chemical shifts also compare
crown ether analogues. As it is known from previous works, well with different gallium fluorides with a coordination
hybrid silicon-based crown ethers are sensitive towards Lewis number of four.[61,62] Furthermore no signal for [GaF4]–
acids and ring-opening polymerization reactions due to tem- (–149.7 ppm) could be observed via 19F NMR spec-
plate effects.[57] Thus, we applied the reaction protocol which troscopy.[62] Accordingly the signals at –169.72, –175.85, and
is represented in Scheme 2 (top). A suitable solvent here is –177.76 ppm are tentatively assigned to [GaIF3]–, [GaI2F2]–,
α,α,α-trifluorotoluene which is polar, non-coordinating, and al- and [GaI3F]–, respectively, and the anionic species of 2 is best
lows the heating of reaction mixtures up to 102 °C if necessary. described as [GaI4–xFx]–.
Therefore, this solvent was already used successfully in the
Figure 1 displays the 13C NMR spectrum of compound 1 as
synthesis of complexes with silicon based crown an example. As determined by means of single-crystal X-ray
ethers.[47,48,57] Applying the reaction conditions in Scheme 2 diffraction analysis, compound 1 crystallizes in the monoclinic
(bottom), however, we did not observe the formation of sila- space group P21/n (no. 14). The ionic structure in the solid
crown complexes. To our surprise, the sila precursor as well reveals that the [GaBr4]– ions are discrete and no direct contact
as the solvent are cleaved to form the carbenium ion containing with the C5 atom is observed. The closest Br–C5 contacts
compound [Ph-C(O2Si2Me4)][GaBr4] (1). The corresponding measure 348.4(3) pm. As a cause of CH–π interactions in the
iodide/fluoride compound [Ph-C(O2Si2Me4)][GaX4] (2, X = I/ solid state, no fully planar molecular structure is observed for
F) could be obtained by reaction of I with GaI3. In these com- the cation. These interactions measure closest C···C atom dis-
pounds the carbenium ion is embedded in a five-membered tances of 346.0(5) pm and do also form a polymeric structure
Si2O2C ring, which derives from (OSi2Me4)2 (I). As presented, motif along [100]. The two planes constituted by the aryl ring
the observed reaction can be understood as a defluorination as well as the disilane backbone are twisted against each other
type reaction with substitution of the F atoms by a Si2Me4O2 by 24.0(1)° (Figure 2).
fragment. These unusual compounds are rare examples[58] of
For the silicon-based part of the structure, quite strained O–
siloxane-stabilized carbenium ions. In the case of compound Si–Si angles are present forming this unprecedented five-mem-
1, colorless platelets were obtained from DCM/n-pentane solu- bered ring. These angles measure 88.5(1)° and 88.7(1)° which
tion so that a crystallographic characterization of such a silox- are considerably smaller than those, for example, in hybrid
ane stabilized carbenium ion could be performed for the first silicon based crown ether analogues.[44,45,48] The O1–C5 and
time. In the case of compound 2, a black oil is obtained. The O2–C5 atom distances measure 128.1(5) and 128.8(5) pm and
NMR spectroscopic investigations of all NMR active nuclei are only slightly longer than those in carboxylic acids such as
give characteristic resonances. The carbenium ionic character benzoic acid, in which the C–O atom distances measure
of these compounds is supported by a 13C NMR resonance of 124.9(1) and 125.0(1) pm.[63] Thus, a partial C–O double bond
180.3 ppm in the case of compound 1 and 182.8 ppm in the character and backbonding of the lone-pairs of the O1 and O2
case of compound 2. Furthermore, a 29Si NMR chemical shift atoms into the empty p-Orbital of the C5 atom is indicated.
is observed for both compounds. With 37.1 ppm (1) and This also causes long silicon oxygen atom distances of
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
2
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
Figure 1. 13C{1H} NMR spectrum of compound 1 in CD2Cl2 (125 MHz, 300 K).
is. For this reason we performed NMR spectroscopic investi-
gations of the reaction mixture with a TMS (TMS = tetrameth-
ylsilane) capillary. By means of 29Si NMR spectroscopy we
observed a doublet of doublets at δ = 27.7 ppm and according
to our interpretation it is most likely that (Me2SiF)2 (II) had
formed, which is accompanied by the fluorophilic character
of organosilicon compounds.[68] This respective resonance was
also observed in a past work and is characteristic.[69] Also the
1
2
coupling constants of JSiF = 303.3 Hz and JSiF = 34.1 Hz
prove the finding, as these compare well with those published
before. Further resonances were also observed but could not
be assigned.
However, after workup and crystallization, compound 1 can
be obtained without impurities. In the case of using GaI3, we
were able to synthesize compound 2 without a halide salt. To
evaluate the role of the different reaction components, control
experiments were performed which included only two compo-
nents of the above mentioned reaction mixture. If MgBr2 is
reacted with siloxane I no reaction can be observed. In contrast
to this, beryllium halides react with compound I with forma-
tion of complex, multinuclear beryllium silanolato compounds,
which are related to previously described multinuclear beryl-
lium polyolates.[46] These reactions are currently investigated
in more detail and will be published in due course. The reac-
tion of MgBr2 or BeBr2 with GaBr3 leads to the formation of
insoluble colorless precipitates. Even though GaBr3 or GaI3
are only little soluble in α,α,α-trifluorotoluene, strongly col-
ored suspensions are obtained.
In case of GaBr3, 13C NMR spectroscopy shows aromatic
signals, which are shifted downfield compared to neat α,α,α-
trifluorotoluene. This indicates coordination of the CF3 group
towards the Lewis-acidic GaBr3. An analogous coordination
has been observed for PhCCl3 / AlCl3 mixtures.[70] However
no additional signals could be observed for GaI3, which is a
weaker Lewis acid than GaBr3. Furthermore, no halide ex-
change was observed between the gallium halides and the CF3-
group. While MgBr2 does not react with α,α,α-trifluorotoluene
on its own, BeBr2 and BeI2 react rapidly under the formation
of α,α,α-tribromotoluene and α,α,α-triiodotoluene (3) as appar-
Figure 2. Molecular structure of the cationic part of compound 1 in
the solid state (top) and a section of the crystal structure (bottom).
Selected bond lengths /pm: Si1–O1 177.9(3), Si2–O2 177.5(3), O1–C5
128.0(5), O2–C5 128.7(5), Si1–Si2 233.0(1). Selected bond angles /°:
O1–C5–O2 122.2(4), Si1–O1–C5 120.3(3), Si2–O2–C5 120.1(3), O1–
Si1–Si2 88.5(1), O2–Si2–Si1 88.7(1).
177.6(3) and 177.9(3) pm. In comparison to related disilanyl-
bearing systems[44,45,47,64,65] the Si–O bond is clearly elon-
gated which is accompanied by the lack of negative hypercon-
jugation interactions p(O) Ǟ σ*(Si–C).[66,67] This further ex-
plains the observed low-field shift in the 29Si NMR spectrum. ent from the 13C NMR signals for the CBr3 group at δ =
It is somehow unclear what the driving force of the reaction 35.5 ppm[71] and CI3 group at –38.8 ppm.
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
3
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
Even though the NMR spectrum of 3 was unknown, the
three iodine atoms bound to the carbon nucleus result in rela-
tivistic shielding effects, which lead to the observed high-field
shift.[72] Such stoichiometric halodefluorination reactions were
reported before.[73] Thus, we suspected a similar stepwise
mechanism. However, in the experiments with the beryllium
halides no intermediates with CF2I/CF2Br- or CFI2/CFBr2-
groups could be observed via NMR spectroscopy. We assumed
this was caused by the extremely fast reaction rates. Therefore,
we evaluated the reactivity of other hard metal iodides towards
α,α,α-trifluorotoluene. For this purpose we selected HfI4 and
ThI4, due to their easy synthesis in pure form by the reaction
of the metal powder or metal oxide with elemental iodide (see
experimental section).[74,75] Reaction monitoring via 13C NMR
spectroscopy of ThI4 in α,α,α-trifluorotoluene revealed the for-
mation of α,α,α-triiodotoluene, however, again no intermedi-
ates were observed. Simple stirring of these salts in the solvent
leads to slow conversion of the suspended MI4 crystals under
precipitation of a black powder. After workup, light yellow
platelets were obtained and investigated via SC-XRD. It turned
out that halodefluorination reactions took place for both metal
ions in seemingly similar manner as both, Hf4+ and Th4+ ions
initiate this reaction. These reactions perturb silyl ether bond-
ing when a siloxane is added but give first examples of facile
C–F bond activation. As can be seen, the fluorine atoms are
replaced with three iodine atoms from the salts.
The halodefluorination itself was also described for other
group 13 salts such as AlX3 (X = Cl–, Br–, I–) but exclusively
α,α,α-triiodotoluene (3) has not yet been characterized prop-
erly.[73] Similar to the group of Young we observed quantitative
iodide transfer but in our case we were able to characterize the
quite air, temperature and light sensitive compound 3 by means
of NMR and IR spectroscopy. The 13C{1H} NMR spectrum
shows characteristic resonances for aromatic compounds and
a single, distinctive high field resonance for the carbon atom
substituted with the three iodine atoms at –38.8 ppm. As deter-
mined via single-crystal X-ray diffraction analysis (see Fig-
ure 3) the compound crystallizes in the orthorhombic space
group Pna21 (no. 33). A section of the crystal structure shows
that the iodine atoms as well as the aryl groups are separated.
The aryl groups are held together by weak π···π interactions.
To account π-stacking in this case is somehow controversial[76]
but as observed, the aryl planes are twisted against each other
by 43.3(1)° and the plane centroid···aryl edge distances mea-
sure 339.0(1) to 346.0(1) pm. For this reason, there might be
evidence for a weak interaction.
Figure 3. Molecular structure of the cationic part of compound 3 in
the solid (top) and excerpt of the crystal structure (bottom). Selected
bond lengths /pm: I1–C7 219.8(3), I2–C7 215.8(3), I3–C7 218.9(3).
Selected bond angles /°: I1–C7–I2 105.6(1), I1–C7–I3 107.6(1), I2–
C7–I3 107.3(1).
α,α,α-triiodotoluene (3) or α,α,α-tribromotoluene, the most
stable iodo and bromo compounds. The halodefluorination
product 3 was isolated and characterized also by means of sin-
gle-crystal X-ray diffraction analysis. Employing α,α,α-trifluo-
rotoluene, the fluorophilic silicon source I and the Lewis acid
GaI3 or the Lewis acid systems MgBr2/GaBr3 or BeBr2/GaBr3
results in the formation of the carbenium ionic compounds
[Ph-C(O2Si2Me4)][GaX4] (X = Br: 1, X = F/I: 2). The forma-
tion of these compounds can be formally understood as a si-
loxy-defluorination type reaction.
We assume the Lewis acid abstracts one fluoride from the
CF3 group. The transient [PhCF2]+ ion then reacts with silox-
ane I which acts as a fluorine atom abstractor and additionally
stabilizes the positive charge at the carbon atom. In the absence
of compound I the above mentioned halo-defluorination oc-
curs. In case of MgBr2, GaBr3 is presumably necessary to in-
crease the Lewis acidity of the magnesium ion to an extent that
C–F bond cleavage occurs, while in case of BeBr2, GaBr3 most
likely only acts as a bromide acceptor. The X-ray structure of
Conclusions
In this contribution we have presented facile stoichiometric compound 1 was determined and discrete ion pairs are present
C–F bond cleavage reactions by simple stirring of Lewis acidic in the solid state showing the carbenium ionic character of this
beryllium bromide or iodide or tetravalent transition/actinoid organosilicon compound. NMR spectroscopic investigations
metal iodides (Hf, Th) in α,α,α-trifluorotoluene. Even though are conclusive and support the assumption. The driving force
we assume this is a stepwise process at the solid liquid inter- of this reaction is presumably the formation of II. All results
face, no intermediates could be observed. This might be caused are summarized in Scheme 3 and a proposed mechanism for
by the significantly weaker C–F, C–Br, and C–I bonds in the the formation of the cation of 1 is depicted in Scheme 4. In
mixed intermediates,[77] which leads to fast conversion to the future we hope that there will be broader applications of
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
4
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
General: All work was carried out excluding moisture and air in an
our presented reactions across different protocols such as catal-
atmosphere of dried and purified argon (5.0, Praxair) using vacuum
glass lines and a glovebox (MBraun). Solvents were dried and freshly
distilled before use and subsequently stored over molecular sieves
(3 Å). Elemental iodine was sublimed in vacuo several times, the last
time from phosphorus pentoxide. Aluminum powder (Fluka, purum
Ͼ99%) was dried in vacuo at 250 °C. Th(IV) oxide (Merck, p. a.) was
dried in vacuo at 350 °C. Hafnium powder (Alfa Aesar, 99.6%) was
used without further purification. MgBr2 (Aldrich, Ͼ 99%), GaBr3
(Alfa Aesar Ͼ99%) and GaI3 (ABCR Ͼ99%) were used as received.
BeBr2 and BeI2 was synthesized according to literature procedures.[79]
Compound I has been reported before[80] but we have optimized the
reaction protocol (see below).
ysis or organofluorine chemistry in general.
For the synthesis of ThI4 and HfI4 borosilicate glass vessels were
flame-dried several times under dynamic vacuum (1ϫ10–3 mbar) be-
fore use. The glass ampoules had a length of 16 cm, an outer diameter
of 18 mm, and a wall thickness of 1.5 mm as described pre-
viously.[74,75] The top of the ampoule carries an NS14.5 inner ground
joint for filling of the starting materials and a constriction for easier
flame sealing. The second constriction at one-third of the length of the
ampoule was used to allow for easier opening of the ampoule after the
reaction.
Synthesis and Crystallization
Preparation of ThI4:[74] An ampoule was charged with 1000 mg
ThO2 (3.79 mmol), 204 + 1 mg Al powder (7.58 mmol + transport
agent) and 2883 + 256 mg I2 (7.58 mmol + transport agent) and flame
sealed under vacuum (1ϫ10–3 mbar). Heating the mixture for 5 d up
to 450 °C (source) and 350 °C (sink) yields 2297 mg canary yellow
crystals of ThI4 (3.11 mmol, 82%), which could be extracted after
transport.
Scheme 3. The C–F-bond cleavage reactions presented herein.
Preparation of HfI4:[74,75] An ampoule (12 mm, 1.5 mm wall thick-
ness) was charged with 125 mg Hf powder (0.700 mmol) and
355.32+10 mg I2 (2.800 mmol and 10 mg excess) and flame sealed
under vacuum (1ϫ10–3 mbar). The reactants were heated up to 300 °C
for 48 h. HfI4 (427.7 mg, 0.689 mmol, 98.4%) in the form of orange
crystals was obtained.
[(Si2Me4)O]2 (I): 2 mL of 1,1,2,2-tetramethyldichlorodisilane
(10.58 mmol) were dissolved in 50 mL of n-pentane. Subsequently,
10 mL (excess) of H2O were added. After stirring overnight, the aque-
ous phase was extracted with three 50 mL portions of n-pentane and
the organic phase was then dried with MgSO4 followed by filtration.
The solvent was removed under reduced pressure and the greasy resi-
due was sublimed in static fine-vacuum (4ϫ10–2 mbar) for several
hours. I was then obtained as large colorless blocks (1.12 g, 80%). 1H
NMR (300 MHz, CD2Cl2): δ = 0.20 [s, 24 H, Si(CH3)2] ppm. 13C{1H}
NMR (75 MHz, CD2Cl2): δ = 2.6 (s, SiCH3) ppm. 29Si NMR
(99 MHz, CD2Cl2): δ = 3.6 [s, Si(CH3)2] ppm.
[Ph-C(O2Si2Me4)]GaBr4 (1): 75 mg of I (0.38 mmol, 1.0 equiv.) were
dissolved in 5 mL α,α,α-trifluorotoluene. Subsequently, 117 mg of
GaBr3 (0.38 mmol, 1.0 equiv.) and MBr2 (0.38 mmol, 1.0 eq, M =
Be2+, Mg2+) were added. The suspension was heated to reflux twice
using a heatgun and then stirred overnight at ambient temperature. The
solvent was removed under reduced pressure and DCM (5 mL) was
added. The resulting suspension was filtered to obtain a greenish, dark
solution which was layered with 20 mL of n-pentane. Large colorless
platelets were obtained after a few days (191 mg, 80%). 1H NMR
Scheme 4. Proposed mechanism for the formation of the cation of 1.
Experimental Section
Caution! Beryllium and its compounds are regarded as toxic and car-
cinogenic. As the biochemical mechanisms that cause beryllium asso-
ciated diseases are still unknown, special (safety) precautions are (300 MHz, CD2Cl2): δ = 0.94 [s, 12 H, Si(CH3)2], 7.69–7.72 (m, 2 H,
strongly advised.[78]
CHar), 8–01–8.04 (m, 1 H, CHar), 8.34–8.37 (m, 2 H, CHar) ppm.
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
5
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
13C{1H} NMR (75 MHz, CD2Cl2): δ = –1.2 (s, SiCH3), 124.0 (s,
336-033; E-Mail: deposit@ccdc.cam.ac.uk, http://www.ccdc.cam.ac.
CHar), 130.6 (s, CHar), 134.1 (s, CHar), 140.7 (s, CHar–q), 180.8 [s, uk).
CH(SiMe2O)2] ppm. 29Si{1H} NMR (99 MHz, CD2Cl2): δ = 37.1 [s,
Crystal Data of 1: C11H17Br4GaO2Si2, monoclinic, P21/n, Z = 8,
100(2) K, a = 12.9816(6) Å, b = 16.5903(8) Å, c = 18.8676(9) Å, β =
101.723(2)°, V = 3978.7(3) Å3, ρ = 2.093 g·cm–3, multiscan absorption
correction using SADABS2016,[84] μ = 9.535 mm–1, Tmin, Tmax
0.2273, 0.7461, 2Θ range 4.276–52.998°, reflections measured
152075, independent reflections 8229 [R(int) = 0.0645], 370 param-
eters, R-index [I Ն 2σ(I)] 0.0371, wR2 (all data) 0.0953, GoF 1.049,
Δρmax, Δρmin 3.16/–2.80 e·Å3.
Si(CH3)2] ppm. IR: ν˜ = 3065 (vw), 1599 (w), 1498 (m), 1476 (s), 1424
(vs), 1313 (w), 1251 (m), 1181 (w), 1154 (w), 1097 (w), 1070 (w),
1024 (w), 908 (m), 841 (s), 806 (vs), 788 (s), 711 (s), 677 (s), 653
(m), 589 (s), 421 (w) cm–1. C11H17Br4Ga1O2Si2: calcd. C, 21.08; H,
2.73%; found C, 21.04; H, 2.89%.
=
[Ph-C(O2Si2Me4)]GaI4–xFx (2): 100 mg of I (0.38 mmol, 1.0 equiv.)
were dissolved in 5 mL α,α,α-trifluorotoluene. Subsequently, 100 mg
of GaI3 (0.38 mmol, 1.0 equiv.) were added. The suspension was
stirred overnight and freed of the solvent. Subsequently DCM (5 mL)
was added. The resulting suspension was filtered to obtain a greenish,
dark solution. After removal of the solvent, washing with several n-
pentane portions and drying the residue in vacuo for several hours
yielded compound 2 as a black oil (0.291 g, Ͼ94%). 1H NMR
(300 MHz, CD2Cl2): δ = 0.98 [s, 12 H, Si(CH3)2], 7.71–7.77 (m, 2 H,
CHar), 8–01–8.04 (m, 1 H, CHar), 8.34–8.37 (m, 2 H, CHar) ppm.
13C{1H} NMR (75 MHz, CD2Cl2): δ = –0.85 (s, SiCH3), 123.8 (s,
CHar), 130.8 (s, CHar), 134.4 (s, CHar), 141.2 (s, CHar–q), 181.0 [s,
C(SiMe2O)2] ppm. 29Si{1H} NMR (99 MHz, CD2Cl2): δ = 38.5 [s,
Si(CH3)2] ppm. IR: ν˜ = 2958 (vw), 2896 (vw), 1598 (w), 1498 (m),
1474 (m), 1420 (vs), 1315 (w), 1256 (m), 1182 (vw), 1152 (w), 1024
(w), 1001 (w), 836 (m), 793 (m), 752 (m), 712 (m), 673 (m), 590 (w)
cm–1.
Crystal Data of 3: C7H5I3, orthorhombic, Pna21, Z = 8, 100(2) K, a
= 11.5191(5) Å, b = 25.2908(10) Å, c = 6.7947(3) Å, V = 1979.48(15)
Å3, ρ = 3.153 g·cm–3, multiscan absorption correction using SAD-
ABS2016,[84] μ = 9.406 mm–1, Tmin, Tmax = 0.5708, 0.7462, 2Θ range
4.784–63.174°, reflections measured 92174, independent reflections
6607 [R(int) = 0.0383], 182 parameters, R-index [I Ն 2σ(I)] 0.0158,
wR2 (all data) 0.0286, GoF 1.138, Δρmax, Δρmin 0.69/–0.65 e·Å3.
Acknowledgements
This work was financially supported by the Deutsche Forschungsge-
meinschaft (DFG). F.D. thanks the X-ray Department (M. Marsch, R.
Riedel, and Dr. S. Ivlev), Philipps-Universität, Marburg, for measure-
ment time and their kind advice.
α,α,α-Triiodotoluene (3): 75 mg of MI4 (M = Hf4+, Th4+) were dis-
solved in 2 mL of α,α,α-trifluorolouene (excess). The suspension was
stirred for 5 h followed by filtration. The clear, yellow-red solution
was concentrated in vacuo and stored at –24 °C. Large pale-yellow
platelets were obtained overnight. The conversion was quantitative. 1H
NMR (300 MHz, CD2Cl2): δ = 7.11–7.14 (m, 2 H, CHar), 7.30–7.33
(m, 1 H, CHar), 7.94–7.96 (m, 2 H, CHar) ppm. 13C{1H} NMR
(75 MHz, CD2Cl2): δ = –38.8 (s, CI3), 127.0 (s, CHar), 128.0 (s, CHar),
128.0 (s, CHar), 151.4 (s, Car-q) ppm. IR: ν˜ = 1586 (w), 1572 (w),
1519 (m), 1479 (m), 1435 (w), 1387 (w), 1329 (w), 1306 (w), 1294
(w), 1234 (w), 1185 (w), 1154 (w), 1104 (w), 1076 (w), 1027 (w), 984
(w), 916 (w), 840 (w), 818 (m), 802 (s), 760 (vs), 680 (m), 652 (s),
585 (m) 571 (vs), 458 (w) cm–1.
Keywords: Lewis acids; C–F bond cleavage; Carbenium
ions; Actinides; Silicon; Fluorine; Defluorination
References
[1] T. Fujiwara, D. O’Hagan, J. Fluorine Chem. 2014, 167, 16–29.
[2] Y. Zhou, J. Wang, Z. Gu, S. Wang, W. Zhu, J. L. Acenã, V. A.
Soloshonok, K. Izawa, H. Liu, Chem. Rev. 2016, 116, 422–518.
[3] J. Wang, M. Sánchez-Roselló, J. L. Aceña, C. Del Pozo, A. E.
Sorochinsky, S. Fustero, V. A. Soloshonok, H. Liu, Chem. Rev.
2014, 114, 2432–2506.
[4] P. Shah, A. D. Westwell, J. Enzyme Inhib. Med. Chem. 2007, 22,
527–540.
[5] G. Theodoridis, Chapter 4 Fluorine-Containing Agrochemicals:
An Overview of Recent Developments, in Advances in Fluorine
Science 2C, Elsevier 2006.
[6] Q. Shen, Y. G. Huang, C. Liu, J. C. Xiao, Q. Y. Chen, Y. Guo, J.
Fluorine Chem. 2015, 179, 14–22.
[7] T. A. Unzner, T. Magauer, Tetrahedron Lett. 2015, 56, 877–883.
[8] T. Stahl, H. F. T. Klare, M. Oestreich, ACS Catal. 2013, 3, 1578–
1587.
[9] D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308–319.
[10] T. Fujita, K. Fuchibe, J. Ichikawa, Angew. Chem. Int. Ed. 2019,
58, 390–402.
Control Experiments: For a 13C NMR monitoring of the respective
C–F bond cleavage reactions we were placing 30 mg of the respective
salt in a J. Young NMR tube. An excess of 0.58 mL of α,α,α-trifluoro-
toluene was subsequently added. We allowed reacting the salt with the
solvent for 2 d. After this period we employed a measurement at a
Bruker AV III 500 MHz spectrometer using the α,α,α-trifluorotoluene
as an internal standard.
Crystallography: Single crystal X-ray diffraction experiments were
carried out with a Bruker D8 Quest diffractometer at 100(2) K with
Mo-Kα radiation (λ = 0.71073). All structures were solved by Direct
Methods (SHELXT) and refined with full-matrix-least-squares cycles
against F2 using SHELXL-2015 on OLEX2 platform.[81–83] Visualiza-
tion of all structures was performed with the Diamond software pack-
age Version 4.6.0. All Ellipsoid plots represent the 50% probability
level.
[11] R. Panisch, M. Bolte, T. Müller, J. Am. Chem. Soc. 2006, 128,
9676–9682.
[12] V. J. Scott, R. Çelenligil-Çetin, O. V. Ozerov, J. Am. Chem. Soc.
2005, 127, 2852–2853.
[13] O. Allemann, S. Duttwyler, P. Romanato, K. K. Baldridge, J. S.
Siegel, Science 2014, 346, aaa4609–aaa4609.
[14] H. F. T. Klare, ACS Catal. 2017, 7, 6999–7002.
[15] M. R. Cargill, G. Sandford, A. J. Tadeusiak, D. S. Yufit, J. A. K.
Howard, P. Kilickiran, G. Nelles, J. Org. Chem. 2010, 75, 5860–
5866.
[16] S. Sabater, J. A. Mata, E. Peris, Nat. Commun. 2013, 4, 1–7.
[17] N. A. Jasim, R. N. Perutz, A. C. Whitwood, T. Braun, J. Izundu,
B. Neumann, S. Rothfeld, H. G. Stammler, Organometallics 2004,
23, 6140–6149.
Crystallographic data (excluding structure factors) for the structures in
this paper have been deposited with the Cambridge Crystallographic
Data Centre, CCDC, 12 Union Road, Cambridge CB21EZ, UK. Copies
of the data can be obtained free of charge on quoting the depository
numbers CCDC-1961377 (1) and CCDC-1961376 (3) (Fax: +44-1223-
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
6
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
[18] R. T. Thornbury, F. D. Toste, Angew. Chem. Int. Ed. 2016, 55, [51] J. S. Ritch, T. Chivers, Angew. Chem. Int. Ed. 2007, 46, 4610–
4613.
11629–11632.
[52] M. Fugel, M. F. Hesse, R. Pal, J. Beckmann, D. Jayatilaka, M. J.
Turner, A. Karton, P. Bultinck, G. S. Chandler, S. Grabowsky,
Chem. Eur. J. 2018, 24, 15275–15286.
[53] F. Dankert, C. von Hänisch, Inorg. Chem. 2019, 58, 3518–3526.
[54] F. Dankert, J. Heine, J. Rienmüller, C. von Hänisch, Cryst-
EngComm 2018, 20, 5370–5376.
[19] R. R. Schrock, J. Adamchuk, K. Ruhland, L. P. H. Lopez, Orga-
nometallics 2003, 22, 5079–5091.
[20] B. M. Kraft, R. J. Lachicotte, W. D. Jones, J. Am. Chem. Soc.
2000, 122, 8559–8560.
[21] M. Fujiwara, J. Ichikawa, T. Okauchi, T. Minami, Tetrahedron
Lett. 1999, 40, 7261–7265.
[22] W. D. Jones, Dalton Trans. 2003, 3, 3991–3995.
[23] B. L. Edelbach, A. K. F. Rahman, R. J. Lachicotte, W. D. Jones,
Organometallics 1999, 18, 3170–3177.
[24] J. Zhou, J. H. J. Berthel, M. W. Kuntze-Fechner, A. Friedrich,
T. B. Marder, U. Radius, J. Org. Chem. 2016, 81, 5789–5794.
[25] A. D. Sun, J. A. Love, Org. Lett. 2011, 13, 2750–2753.
[26] T. Schaub, M. Backes, U. Radius, J. Am. Chem. Soc. 2006, 128,
15964–15965.
[55] J. Pahl, H. Elsen, A. Friedrich, S. Harder, Chem. Commun. 2018,
54, 7846–7849.
[56] C. Douvris, C. M. Nagaraja, C. H. Chen, B. M. Foxman, O. V.
Ozerov, J. Am. Chem. Soc. 2010, 132, 4946–4953.
[57] F. Dankert, C. Donsbach, J. Rienmüller, R. Richter, C.
von Hänisch, Chem. Eur. J. 2019, 25, 15934–15943..
[58] W. Kirmse, M. Guth, S. Steenken, J. Am. Chem. Soc. 1996, 118,
10838–10849.
[59] K. G. Sharp, P. A. Sutor, J. Am. Chem. Soc. 1977, 99, 8046–8048.
[60] S. G. Frankiss, J. Phys. Chem. 1967, 71, 3418–3421.
[61] B. Neumüller, Coord. Chem. Rev. 1997, 158, 69–101.
[62] C. Bour, J. Monot, S. Tang, R. Guillot, J. Farjon, V. Gandon,
Organometallics 2014, 33, 594–599.
[63] R. Feld, M. S. Lehmann, K. W. Muir, J. C. Speakman, Z. Kristal-
logr., Cryst. Mater. 1981, 157, 215–231.
[64] F. Dankert, K. Reuter, C. Donsbach, C. von Hänisch, Inorganics
2018, 6, 15.
[65] K. Reuter, S. S. Rudel, M. R. Buchner, F. Kraus, C. von Hänisch,
Chem. Eur. J. 2017, 23, 9607–9617.
[66] F. Weinhold, R. West, J. Am. Chem. Soc. 2013, 135, 5762–5767.
[67] F. Weinhold, R. West, Organometallics 2011, 30, 5815–5824.
[68] J. Horstmann, M. Niemann, K. Berthold, A. Mix, B. Neumann,
H.-G. Stammler, N. W. Mitzel, Dalton Trans. 2017, 46, 1898–
1913.
[27] A. Steffen, M. I. Sladek, T. Braun, B. Neumann, H. G. Stammler,
Organometallics 2005, 24, 4057–4064.
[28] P. Fischer, K. Götz, A. Eichhorn, U. Radius, Organometallics
2012, 31, 1374–1383.
[29] W. W. Ji, E. Lin, Q. Li, H. Wang, Chem. Commun. 2017, 53,
5665–5668.
[30] T. Fujita, Y. Watabe, S. Yamashita, H. Tanabe, T. Nojima, J. Ichi-
kawa, Chem. Lett. 2016, 45, 964–966.
[31] M. Aizenberg, D. Milstein, Science 1994, 265, 359–361.
[32] V. V. Grushin, Acc. Chem. Res. 2010, 43, 160–171.
[33] T. Braun, F. Wehmeier, Eur. J. Inorg. Chem. 2011, 613–625.
[34] M. Aizenberg, D. Milstein, J. Am. Chem. Soc. 1995, 117, 8674–
8675.
[35] R. P. Hughes, D. C. Lindner, J. Am. Chem. Soc. 1997, 119, 11544–
11545.
[36] J. Terao, M. Nakamura, N. Kambe, Chem. Commun. 2009, 6011–
6013.
[37] C. Bakewell, A. J. P. White, M. R. Crimmin, Angew. Chem. Int.
Ed. 2018, 57, 6638–6642.
[69] D. A. Stanislawski, R. West, J. Organomet. Chem. 1981, 204,
307–314.
[70] U. S. Racherla, T. Daniel, P. R. Rajamohanan, N. R. Ayyangar, J.
Am. Chem. Soc. 1989, 111, 7659–7661.
[71] Q. Sun, X. Yu, M. Bao, M. Liu, J. Pan, Z. Zha, L. Cai, H. Ma,
C. Yuan, X. Qiu et al., Angew. Chem. Int. Ed. 2018, 57, 4035–
4038.
[72] R. V. Viesser, L. C. Ducati, C. F. Tormena, J. Autschbach, Phys.
Chem. Chem. Phys. 2018, 20, 11247–11259.
[73] K. K. K. Goh, A. Sinha, C. Fraser, R. D. Young, RSC Adv. 2016,
6, 42708–42712.
[38] N. Suzuki, T. Fujita, K. Y. Amsharov, J. Ichikawa, Chem. Com-
mun. 2016, 52, 12948–12951.
[39] T. Fujita, M. Takazawa, K. Sugiyama, N. Suzuki, J. Ichikawa,
Org. Lett. 2017, 19, 588–591.
[40] A. M. Träff, M. Janjetovic, L. Ta, G. Hilmersson, Angew. Chem.
Int. Ed. 2013, 52, 12073–12076.
[41] T. Akiyama, K. Atobe, M. Shibata, K. Mori, J. Fluorine Chem.
2013, 152, 81–83.
[42] M. F. Kuehnel, P. Holstein, M. Kliche, J. Krüger, S. Matthies, D.
Nitsch, J. Schutt, M. Sparenberg, D. Lentz, Chem. Eur. J. 2012,
18, 10701–10714.
[74] H. L. Deubner, S. S. Rudel, F. Kraus, Z. Anorg. Allg. Chem. 2017,
643, 2005–2010.
[75] B. Krebs, D. Sinram, J. Less-Common Met. 1980, 76, 7–16.
[76] C. R. Martinez, B. L. Iverson, Chem. Sci. 2012, 3, 2191.
[77] G. Glockler, J. Phys. Chem. 1959, 63, 828–832.
[78] D. Naglav, M. R. Buchner, G. Bendt, F. Kraus, S. Schulz, Angew.
Chem. Int. Ed. 2016, 55, 10562–10576.
[79] M. Müller, F. Pielnhofer, M. R. Buchner, Dalton Trans. 2018, 47,
12506–12510.
[80] T. Takano, N. Kasai, M. Kakudo, Bull. Chem. Soc. Jpn. 1963, 36,
[43] K. Reuter, G. Thiele, T. Hafner, F. Uhlig, C. von Hänisch, Chem.
Commun. 2016, 52, 13265–13268.
[44] K. Reuter, M. R. Buchner, G. Thiele, C. von Hänisch, Inorg.
Chem. 2016, 55, 4441–4447.
[45] K. Reuter, F. Dankert, C. Donsbach, C. von Hänisch, Inorganics
2017, 5, 11.
585–590.
[46] M. R. Buchner, M. Müller, F. Dankert, K. Reuter, C. von Hänisch,
Dalton Trans. 2018, 47, 16393–16397.
[47] F. Dankert, C. Donsbach, C.-N. Mais, K. Reuter, C. von Hänisch,
[81] G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem. 2015,
71, 3–8.
[82] G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv. 2015,
71, 3–8.
Inorg. Chem. 2018, 57, 351–359.
[83] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H.
Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341.
[84] Bruker, SADABS, Bruker AXS Inc., Madison, Wisconsin, USA,
2016.
[48] F. Dankert, K. Reuter, C. Donsbach, C. von Hänisch, Dalton
Trans. 2017, 46, 8727–8735.
[49] S. Harder, B. Freitag, P. Stegner, J. Pahl, D. Naglav, Z. Anorg.
Allg. Chem. 2015, 641, 2129–2134.
[50] B. Freitag, P. Stegner, K. Thum, C. A. Fischer, S. Harder, Eur. J.
Inorg. Chem. 2018, 2018, 1938–1944.
Received: November 15, 2019
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
7
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
F. Dankert, H. L. Deubner, M. Müller, M. R. Buchner,
F. Kraus, C. von Hänisch* ................................................ 1–8
C–F Bond Cleavage Reactions with Beryllium, Magnesium,
Gallium, Hafnium, and Thorium Halides
Z. Anorg. Allg. Chem. 2020, 1–8
www.zaac.wiley-vch.de
8
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim