Direct Synthesis of Heavy Grignard Reagents: Challenges, Limitations, and Derivatization

Article abstract of DOI:10.1002/chem.201803518

The direct synthesis of organocalcium compounds (heavy Grignard reagents) by the reduction of organyl halides with activated calcium powder succeeded in a straightforward manner for organic bromides and iodides that are bound at sp²-hybridized carbon atoms. Extension of this strategy to alkyl halides was very limited, and only the reduction of trialkylsilylmethyl bromides and iodides with activated calcium allowed the isolation of the corresponding heavy Grignard reagents. Substitution of only one hydrogen atom of the methylene moiety by a phenyl or methyl group directed this reduction toward the Wurtz-type coupling and the formation of calcium halide and the corresponding C?C coupling product. The stability of the methylcalcium and benzylcalcium derivatives in ethereal solvents suggests an unexpected reaction behavior of the intermediate organyl halide radical anions. Quantum chemical calculations verify a dependency between the ease of preparative access to organocalcium complexes and the C?I bond lengths of the organyl iodides. The bulkiness of the trialkylsilyl group is of minor importance. Chloromethyltrimethylsilane did not react with activated calcium; however, halogen-exchange reactions allowed the isolation of [Ca(CH₂SiMe₃)(thf)₃(μ-Cl)]₂. Furthermore, the metathetical approach of reacting [Ca(CH₂SiMe₃)I(thf)₄] with KN(SiMe₃)₂ and the addition of N,N,N′,N′′,N′′-pentamethyldiethylenetriamine (pmdeta) allowed the isolation of heteroleptic [CaCH₂SiMe₃{N(SiMe₃)₂}(pmdeta)]. In the reaction of this derivative with phenylsilane, the trimethylsilylmethyl group proved to be more reactive than the bis(trimethylsilyl)amido substituent.

Full text of DOI:10.1002/chem.201803518

A Journal of
Accepted Article
Title: Direct Synthesis of Heavy Grignard Reagents - Challenges,
Limitations, and Derivatization
Authors: Matthias Westerhausen, Alexander Koch, Quentin Dufrois,
Marino Wirgenings, Helmar Görls, Sven Krieck, Michel
Etienne, and Georg Pohnert
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201803518
Link to VoR: http://dx.doi.org/10.1002/chem.201803518
Supported by

10.1002/chem.201803518
Chemistry - A European Journal
Direct Synthesis of Heavy Grignard Reagents – Challenges, Limitations, and Derivatization
Alexander Koch,a Quentin Dufrois,b Marino Wirgenings,c Helmar Görls,a Sven Krieck,a Michel Etienne,b
Georg Pohnert,c Matthias Westerhausen*,a
a) Institute of Inorganic and Analytical Chemistry, Friedrich Schiller University Jena,
Humboldtstraße 8, 07743 Jena, Germany; homepage: www.ls1.uni-jena.de
e-mail: m.we@uni-jena.de, fax: +49 3641 9-48-132
b) LCC-CNRS, Université de Toulouse, CNRS, UPS, Toulouse, France
c) Institute of Inorganic and Analytical Chemistry, Friedrich Schiller University Jena,
Lessingstraße 8, 07743 Jena, Germany.
* Correspondence author
Dedicated to Professor Dietmar Stalke on the Occasion of his 60th Birthday
Abstract
The direct synthesis of organocalcium compounds (heavy Grignard reagents) via the reduction of
organyl halides with calcium powder succeeds straightforwardly for organic bromides and iodides
that are bound at sp2-hybridized carbon atoms. Extension of this strategy to alkyl halides is very
limited and only the reduction of trialkylsilylmethyl bromides and iodides with activated calcium
allows the isolation of corresponding heavy Grignard reagents. Substitution of only one hydrogen
atom of the methylene moiety by a phenyl or methyl group directs this reduction toward the Wurtz-
type coupling and the formation of calcium halide and the corresponding C-C coupling product. The
fact that methylcalcium and benzylcalcium derivatives are stable in ethereal solvents suggests an
unexpected reaction behavior of the intermediate organyl halide radical anions. Quantum chemical
calculations verify a dependency between the ease of preparative access to organocalcium
complexes and the C-I bond lengths of the organyl iodides. The bulkiness of the trialkylsilyl group is
of minor importance. Chloromethyl-trimethylsilane does not react with activated calcium, however,
halogen exchange reactions allow isolation of [Me3SiCH2Ca(thf)3(µ-Cl)]2. Furthermore, the
metathetical approach of [Me3SiCH2Ca(thf)4I] with KN(SiMe3)2 and the addition of pmdeta allow the
isolation of heteroleptic [Me3SiCH2Ca(pmdeta){N(SiMe3)2}]. In the reaction of this derivative with
phenylsilane, the trimethylsilylmethyl group proves to be more reactive than the
bis(trimethylsilyl)amido substituent.
Keywords: Grignard reaction, alkylcalcium reagents, calcium, direct synthesis, reduction reactions
Introduction
The direct syntheses of organolithium[1] and organomagnesium compounds[2] (Grignard reagents,
Scheme 1) have been well-known for more than a century and represent common text book
knowledge. The preparative procedures are straightforward, yields are commonly satisfactory and
often isolation and structure determination of these highly air- and moisture-sensitive
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
organometallics is feasible.[3] Therefore, every chemistry student is familiar with umpolung reactions,
i.e. the reaction of an organyl halide with an electropositive metal like lithium or magnesium.
Scheme 1: Preparation of organolithium and organomagnesium compounds (Grignard reagents) via
direct synthesis ( X = Cl, Br, I).
In Scheme 2 a simplified electron transfer cascade as mode for the formation of Grignard compounds
is given consisting of two single electron transfer (SET) steps from magnesium onto the organyl
halide R-X. In the bottom part of this scheme the redistribution behavior in solution (aggregation and
Schlenk equilibrium) is depicted.
Scheme 2: The Grignard reaction of magnesium with organyl halides R-X. The top row shows a
simplified picture of the electron transfer steps. The bottom part of the scheme offers a brief overview
on dinuclear species involved in the ligand scrambling reactions (Schlenk equilibrium).
In Table 1, selected physical data of the convenient metals lithium, magnesium, calcium, and
strontium are compared.[4] Calcium has a very similar electronegativity value as lithium leading to
comparable ionicity of Li-C and Ca-C bonds. The standard potentials of lithium and calcium are quite
similar whereas magnesium differs significantly with respect to these values. In this row, the largest
atomization enthalpy is found for calcium leading to the highest melting and boiling points among
these electropositive metals. The hydration enthalpies strongly depend on charge and size of ions
with increasing charge and decreasing radius leading to larger absolute values.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Table 1: Comparison of selected physical data of the s-block metals lithium, magnesium, calcium, and
strontium.[4]
Lithium
0.97
Magnesium Calcium
Strontium
0.89
5.695
11.03
+0.3
164.4
84
-1415
132
Electronegativitya
1.23
1.04
6.111
11.87
+0.3
178.2
105
1st ionization potential (eV)
2nd ionization potential (eV)
Electron affinity (eV)
5.320
75.63
-0.618
159.37
106
7.642
15.03
+0.4
147.70
129
Atomization enthalpy (kJ mol-1)
M-M bond energy (kJ mol-1)
Hydration enthalpy (kJ mol-1) b
Radius of metal ion (pm) b,c
-521
90
-1922
86
-1577
114
Standard potential E0 (V, vs NHE) b -3.040
-2.356
648.8
1105
-2.84
839
1482
-2.89
768
1380
Melting point (°C)
Boiling point (°C)
180.54
1347
a) Allred-Rochow electronegativity values. b) The values refer to monovalent lithium and divalent alkaline earth
metals. c) Radius of the hexacoordinate ion.
However, the transfer of the Grignard reaction to heavier alkaline earth metals has inherent
challenges and diverse reasons account for the lack of an early straightforward development of
organocalcium compounds of the type R-Ca-X.[5] On the one hand, these organometallics are very
sensitive toward moisture (hydrolysis) and air (oxidation) due to very polar Ca-C bonds based on the
electronegativity difference of calcium and carbon. On the other hand, the reduction of organic
halides with calcium is accompanied by side reactions such as ether degradation (formation of RH),
Wurtz-type coupling (R-R formation: R = alkyl, Wurtz reaction; R = alkyl, R’ = aryl, Wurtz-Fittig
reaction; R = aryl, Fittig reaction) or reduction of π-systems of aromatic systems (Birch-type
reduction),[6] often leading to irreproducible results of the direct synthesis with calcium. In contrast,
the reactivity of calcium metal itself (with aggravating influence of containing trace elements) is
rather low and unlike lithium (which must be handled in an argon atmosphere because it reacts with
nitrogen and oxygen yielding lithium nitride and oxide) and magnesium, activation (and occasionally
purification) is required prior to its use in Grignard-type reduction reactions. Nevertheless,
pyrophoric calcium powder still shows a limited reactivity because organyl chlorides are not suitable
as substrates for the direct synthesis. From these studies it seems clear that the synthetic difficulties
are of kinetic rather than thermodynamic origin since the heavy Grignard reagents benzylcalcium,[7]
methylcalcium[8] and n-alkylcalcium derivatives[9] are known and can safely be handled in ethereal
solvents. These challenges related to the direct synthesis initiated the development of alternative
procedures (Scheme 3) with the salt metathesis reaction of organic alkali metal complexes with
soluble calcium iodide being the favored method.[5,10]
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Scheme 3: Overview of synthetic procedures of organylcalcium complexes (L = anionic ligand like
amido, ß-diketiminato groups and others), including (i) direct synthesis (Grignard reaction) with
subsequent Schlenk equilibrium, (ii) direct metalation, (iii) organometallic metalation, (iv)
metathetical approach (A = alkali metal, in most cases potassium; X = halide or pseudohalide), (v)
transmetalation commonly yielding heterobimetallic complexes (M = Cu, Zn and others), and (vi)
hydrometallation.
In the early 1990s, the synthesis via co-condensation of organic substrate and calcium vapor as well
as determination of the molecular structure of [(dx)2Ca{CH(SiMe3)2}2] (dx = 1,4-dioxane) by the
Lappert group[11] led to a vivid revival of the organocalcium chemistry. Approximately 15 years after
this tremendous success, we developed a reliable synthesis of arylcalcium reagents via the atom-
economic direct synthesis of the organic bromides or iodides with activated calcium in common
organic solvents and determined molecular structures of arylcalcium complexes.[12] Convenient
preparative procedures were developed for an ubiquitous use of this class of heavy Grignard
reagents.[5] Alternative procedures are subject to limitations. Metalation reactions require very H-
acidic substrates (like cyclopentadiene). Transmetalation protocols generally yield metalates.
Here we study the direct synthesis and the outcome of the direct synthesis with various alkyl halides.
Furthermore, we propose an explanation for the observed challenges related to the direct synthesis
of alkylcalcium halides as well as dialkylcalcium derivatives as a consequence of the Schlenk-type
equilibrium.
Results and discussion
Limitations of the Direct Syntheses of Heavy Grignard Reagents
The reduction of organic halides with activated calcium commonly proceeds quite smoothly if the
halide is bound to an sp2-hybridized carbon atom, allowing the transfer of this procedure to alkenyl
halides.[13] Alkylcalcium derivatives prepared by reduction of alkyl bromide or iodide with activated
calcium are limited to very few representatives. As outlined above, the isolation of organocalcium
derivatives with diverse alkyl groups such as benzyl,[7] methyl[8] or n-alkyl[9] by other preparative
methods and elucidation of their molecular structures demonstrate that these organometallics are
stable in ethereal media and hydrocarbons, suggesting that side reactions during the procedure itself
are responsible for the failure of the direct synthesis of hydrocarbylcalcium halides.
Assuming a similar mechanism for the direct synthesis of heavy Grignard reagents as had been
elucidated for the classic Grignard reactions,[14] the first reaction step would be a single electron
transfer (SET) from calcium onto the organyl halide on the metal surface leading to a radical anion,
which could be desorbed into the solution (Scheme 4). This radical anion could either dissociate into
an alkyl radical and a halide anion or accept another electron leading to an alkyl anion and a halide
anion. The anions would be captured at the positively charged metal surface whereas the
electroneutral alkyl radicals could dissociate into the reaction solution where this radical had a
limited lifetime depending on the reaction rates of recombination (R-R, Wurtz-type coupling
product), solvent attack (R-H formation), and reduction (formation of the R- carbanion). Now several
reaction pathways would be feasible. The cation [CaX]+ could be solvated and split off the metal
particle. At the same time the alkyl radical could be reduced on the metal surface and then
recombine to the heavy Grignard reagent R-Ca-X. Reversal of this latter reaction sequence would
lead to reduction of the alkyl radical which then would bind to the metal, also yielding R-Ca-X which
would then be solvated by Lewis bases and liberated into the reaction solution.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Scheme 4: The reaction of calcium metal with organyl halides R-X including the different consecutive
steps forming the product of homo coupling (Wurtz-type reaction) and heavy Grignard reagent
(organocalcium halide).
In contrast to this, the general reaction of an organo halide with calcium meal leads to the isolation
of the Wurtz-type coupling products; this could be traced directly to the properties of the metal. A
key reaction step is desorption of a metal-ion out of the metal particle. This process is related to the
atomization enthalpy, which is significantly larger for calcium than for magnesium (see Table 1). In
the following, the metal ion is coordinated by solvent molecules. Solvation of magnesium and
calcium ions by water (hydration, see Table 1)[6] and tetrahydrofuran molecules[15] shows that a
larger radius expectedly and significantly reduces the absolute value of the solvation enthalpy. In
conclusion, the desorption process of calcium ions is thermodynamically less favored than for
magnesium ions. In addition, another SET step from the metal particle to the alkyl radical has to be
considered. According to the ionization enthalpy, this should be more favorable in the case of
calcium. However, this process can be related to the distance between the reaction partners (metal
particle and alkyl radical) with the consequence of a slower SET for calcium. In conclusion, the overall
process of the formation of a heavier Grignard reagent seems to be slower for calcium than for
magnesium with the consequence that side reactions like radical dimerization or radical
rearrangements gain significance.
Different reactivity patterns of magnesium and calcium have been observed in the aryl alkaline earth
metal chemistry, too. 2,4,6-Tri(tert-butyl)phenyl halides[5] and 2,4,6-triphenylphenyl halides[16] react
smoothly with magnesium whereas surprisingly different reaction patterns are observed if calcium is
used for the reduction of these bulky aryl halides. If the reduction of 2,4,6-tri(tert-butyl)phenyl halide
is performed with calcium, tunneling processes convert the intermediately formed tri(tert-
butyl)phenyl radical into the 1-[3,5-di(tert-butyl)phenyl]-1,1-dimethylethyl radical.[17] Furthermore,
lithium and magnesium are able to form dimetalated benzene with 1,4-diiodobenzene whereas the
reaction of 1,4-I2C6H4 with activated calcium selectively yields 4-iodophenylcalcium iodide.[18] The π-
systems of oligocyclic aromatic halides tend to be reduced by calcium (leading to Birch-type
reductions) much more easily than by magnesium.[6]
Therefore, we focused our studies on substitution patterns of the organyl halides that would
enhance the life time of possible radical intermediates. Based on this approach, we started our
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
investigation with the reduction of cyclopropyl iodide with activated calcium because it is generally
accepted that the carbon atoms of this alkyl group can be regarded as sp2-hybridized.[19]
Cyclopropylmagnesium derivatives[20] are known and cyclopropyl groups can be involved in stabilizing
CC agostic interactions.[21] The reaction proceeded smoothly in tetrahydropyran (THP) at 10 °C
according to Scheme 5 and yields of approximately 70 % could be achieved. This cyclopropylcalcium
halide complex (1a) is highly soluble in ethereal solvents and a Schlenk-type equilibrium converts
cyclopropylcalcium iodide 1a into di(cyclopropyl)calcium 1b and calcium diiodide. Crystallization
efforts led to precipitation of well-known [(thp)4CaI2]22 shifting the equilibrium toward the
homoleptic complexes. Despite considerable efforts, growing of single crystals of cyclopropylcalcium
derivatives suitable for X-ray diffraction experiments failed. Relevant 1H NMR data includes a
shielded H resonance at  −1.65 (br m) and two sets of resonances for the diastereotopic H in the 
0.20-0.40 (br m, 2H) and 0.50-0.75 (br m, 2H) ranges. In the 13C{1H} NMR spectrum, Ca-C resonates
at  7.9 and the C atoms are observed at  7.0.
Scheme 5: Synthesis of tetrakis(tetrahydropyran)calcium cyclopropanide iodide in tetrahydropyran,
also yielding thp complex of di(cyclopropyl)calcium via a Schlenk-type equilibrium.
The reduction of trimethylsilylmethyl bromide and iodide with activated calcium in tetrahydrofuran
yielded the corresponding alkylcalcium iodide (2a)[23,24] and bromide (2b) whereas Me3SiCH2Cl did not
react with calcium powder. In order to enhance the durability in ethereal solvents, the thf ligands
were exchanged by tetrahydropyran molecules (Scheme 6). The molecular structure and numbering
scheme of [Me3SiCH2Ca(thp)4Br] (3b) is depicted in Figure 1. The structure is very similar to the
homologous iodide congener [Me3SiCH2Ca(thp)4I] (3a).[23] The Ca1-C1 bond length exhibits a
characteristic value of 250.0(4) pm. Steric requirements (i.e. repulsion between thp ligands and the
bulky trimethylsilyl group) lead to a large Ca1-C1-Si1 bond angle of 143.9(2)°.
Scheme 6: Synthesis of trimethylsilylmethylcalcium bromide (X = Br, 2b) and iodide (X = I, 2a) as thf
adducts and subsequent exchange of the ligated ether by tetrahydropyran yielding
[Me3SiCH2Ca(thp)X] with X = Br (3b) and I(3a).
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Figure 1: Molecular structure and atom labeling scheme of [Me3SiCH2Ca(thp)4Br] (3b). The ellipsoids
represent a probability of 30 %, only the H atoms at C1 are shown with arbitrary radii. Selected bond
lengths (pm): Ca1-Br1 289.83(7), Ca-C1 250.0(4), Ca1-O1 239.5(2), Ca1-O2 241.3(3), Ca1-O3 239.3(2),
Ca1-O4 236.9(2), Si1-C1 181.9(4), Si1-C2 188.8(4), Si1-C3 189.2(4), Si1-C4 188.4(4); bond angles
(deg.): Br1-Ca1-C1 168.65(9), Ca1-C1-Si1 143.9(2), C1-Si1-C2 110.4(2), C1-Si1-C3 114.8(2), C1-Si1-C4
113.1(2), C2-Si1-C3 106.9(2), C2-Si1-C4 106.9(2), C3-Si1-C4 104.2(2).
Thereafter, we studied bulkier alkyl halides and the reduction of bromomethyl-triisopropylsilane with
activated calcium in THF at -40 °C. Again, high yields of 70 % were obtained. Recrystallization from a
mixture of THF and pentane yielded single crystals of [iPr3SiCH2Ca(thf)3(µ-Br)]2 (4) with bridging
bromide ions according to Scheme 7. The molecular structure and atom labeling scheme of complex
4 are depicted in Figure 2. The calcium atoms are in distorted octahedral coordination spheres with
Ca1-C1 and Ca2-C11 bond lengths of 251.2(3) and 250.5(3) pm. The bromide bridges are slightly
asymmetric with the larger Ca-Br distances trans to the alkyl substituent.
Scheme 7: Synthesis of [(iPr)3SiCH2Ca(thf)3Br]2 (4) via reduction of bromomethyl triisopropylsilane
with activated calcium in THF.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Figure 2: Molecular structure and atom labeling scheme of [iPr3SiCH2Ca(thf)3(µ-Br)]2 (4). The ellipsoids
represent a probability of 30 %, only the H atoms of the methylene groups at C1 and C11 are shown
with arbitrary radii. Selected bond lengths (pm): Ca1-Br1 303.35(7), Ca1-Br2 289.69(7), Ca2-Br1
290.09(7), Ca2-Br2 303.91(7), Ca1-C1 251.2(3), Ca1-O1 240.3(2), Ca1-O2 242.0(2), Ca1-O3 237.5(2),
Ca2-C11 250.5(3), Ca2-O4 240.0(2), Ca2-O5 244.7(2), Ca2-O6 238.1(2), Si1-C1 182.9(3), Si1-C2
191.3(3), Si1-C5 193.6(4), Si1-C8 190.8(4), Si2-C11 183.7(3), Si2-C12 193.4(4), Si2-C15 190.6(4), Si2-
C18 192.1(3); bond angles (deg.): Br1-Ca1-Br2 79.95(2), Ca1-Br1-Ca2 100.10(2), Br1-Ca2-Br2 79.79(2),
Ca1-Br2-Ca2 100.06(2), C1-Ca1-Br1 176.07(8), C1-Ca1-Br2 98.82(9), C11-Ca2-Br1 97.62(8), C11-Ca2-
Br2 175.11(8), Ca1-C1-Si1 140.0(2), Ca2-C11-Si2 142.8(2).
Substitution of the silicon atom in Me3SiCH2X by a carbon and use of neopentyl halide led to
negligible alkalinity of the reaction mixture and we were unable to isolate neopentylcalcium
complexes from such a procedure but large amounts of the Wurtz-type coupling product 2,2,5,5-
tetramethylhexane were obtained. Furthermore, substitution of trimethylsilylmethyl halide with a
methyl or tert-butyl group at the methylene moiety yielding 1-trimethylsilyl-1-iodoethane and 1-
trimethylsilyl-1-iodo-2,2-dimethylpropane led to the formation of the Wurtz-type coupling products
and significant concentrations of Grignard-type alkylcalcium complexes were not observed. In Table
2, consumption of the substrates and yields of organylcalcium derivatives from the direct synthesis of
activated calcium with organyl halides are summarized. Chloroalkanes and -arenes cannot be
reduced with activated calcium and no reaction is observed. This observation confirms that calcium is
significantly less reactive than magnesium in direct synthesis of Grignard-type reagents. This finding
is in agreement with the formation of a bis(Grignard) reagent from 1,4-diiodobenzene whereas
calcium only allows the isolation of 4-iodophenylcalcium iodide.
Table 2: Conversion of selected organyl halides into the corresponding organylcalcium reagents via a
direct synthesis of the organyl halide with activated calcium in tetrahydrofuran.
Substrate
Organylcalcium halide
Consump.
Convers.
Yield
(%) c
0
Ref.
[12]
[12,25]
[12,25]
(%) a
0
(%) b
0
Ph-Cl
Ph-Br
Ph-I
[Ph-Ca(thf)4Cl]
[Ph-Ca(thf)4Br]
[Ph-Ca(thf)4I]
71
93
70
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
4-I-C6H4-I
Mes-Br
Mes-I
[4-I-C6H4-Ca(thf)4I]
[Mes-Ca(thf)4Br]
[Mes-Ca(thf)4I]
> 95
95
< 5
86
0
20
0
19
0
0
0
[12]
[12]
[26]
[5]
2,4,6-(tBu)3C6H2-Br
2,4,6-Ph3C6H2-Br
Me-I
Me3SiCH2-Cl
Me3SiCH2-Br
Me3SiCH2-I
Me3CCH2-I
Me3SiCH(Me)-I
Me3SiCH(tBu)-I
iPr3SiCH2-Br
[2,4,6-(tBu)3C6H2-Ca(thf)4Br]
[2,4,6-Ph3C6H2-Ca(thf)4Br]
[Ca(thf)3I(µ-Me)]2
[Me3SiCH2-Ca(thf)4Cl]
[Me3SiCH2-Ca(thf)4Br]
[Me3SiCH2-Ca(thf)4I]
[Me3CCH2-Ca(thf)4I]
[Me3SiCH(Me)-Ca(thf)4I]
[Me3SiCH(tBu)-Ca(thf)4I]
[iPr3SiCH2-Ca(thf)3Br]2
> 90
> 90
> 90
0
> 80 d
0
[16]
0
0
30
70
< 5
< 5
< 5
70
13
55
0
0
0
[23,27]
> 90
> 90
> 90
> 80
12
a) Consumption refers to the substrate. b) Conversion to an organocalcium complex was estimated by an acid-
base titration of an aliquot of the reaction solution; differentiation of organylcalcium halide and
diorganylcalcium is not possible by this method. c) Yield refers to isolated crystalline compound. d) The
reduction of 2,4,6-triphenylphenyl halide with calcium in THF intermediately yields the heavy Grignard reagent
which converts into the inverse sandwich complex [{(thf)3Ca}2(µ-1,3,5-Ph3C6H3)] with calcium(I) ions (see text).
Quantum Chemical Investigations
To shed light on the strikingly different behavior of the studied alkyl halides, DFT calculations were
performed. The fact that the instability of alkylcalcium halides (or dialkylcalcium complexes) is not
responsible for the failure of the direct synthesis of alkylcalcium halides suggests that intermediates
behave differently in these reactions. In the case of the Grignard formation, it is well accepted that
the initial reaction step is a single electron transfer from the magnesium particle onto the alkyl halide
molecule, leading to a radical anion (see Scheme 1; for a well-balanced discussion of the Grignard
reaction see the excellent review of Garst and Soriaga[14]). It has been concluded that the key step is
the reduction of the alkyl halide and that the first electron transfer (leading to the radical anion) is
the most difficult step depending on the substrates electrophilicity.[28] Therefore, we studied these
species with the help of quantum chemical investigations. We chose the density functional theory
with the functional B97D3 and the basis set 6-311++G** as level of theory. The relaxed geometry of the
mono reduced radical anion of R-X was calculated and for comparison, we also included iodobenzene in
these studies. The computed bond lengths of the species lie in the range between 300 and 380 pm (see
Table 3); in electroneutral organic iodides the C-I bond lengths vary between 199 and 214 pm depending
on the hybridization of the carbon atom.[29]
Table 3: Calculated bond lengths (B97D3/6-311++G**) of several iodoalkanes and -arenes radical anions.
Organohalide R-I
Ph-I
Bond length of [R-I]−
302.5
329.8
cPr-I
343.1
PhCH2-I
375,1
351.0
342.3
362.9
359.7
340.1
353.7
PhC(tBu)H-I
PhC(SiMe3)H-I
Me3SiCH2-I
Me3SiCH(Me)-I
Me3SiCH(tBu)-I
(iPr)3SiCH2I
Me-I
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
366.7
Me3CCH2-I
The C-I bonds show a significant elongation upon reduction and formation of the radical anion.
Scheme 8 provides a plot of the C-I bond length as a function of the ease of formation of Grignard
reagent or Wurtz product. It is seen that the longer the C-I bond in the radical anion, the easier the
formation of the Wurtz coupling product. Slight elongation of the C-I bonds after the SET step (C-I
bond lengths below 330 pm) requires that the iodine atom is bound to an sp2 hybridized carbon
atom. These iodohydrocarbons allow the isolation of organocalcium derivatives in high yields with
iodobenzene as the ideal candidate. Therefore it is not surprising that many arylcalcium complexes
have been described in recent years. In the intermediate region (C-I bond lengths between 330 and
343 pm; depicted in orange in Scheme 8), iodoalkanes with trialkylsilyl substituents are located. They
can be reduced to the corresponding alkylcalcium complexes; however, the yields are lower due to
side reactions such as Wurtz-type coupling. If the C-I bond lengths of the radical anions lie above 343
pm, the Wurtz-type coupling is the dominant reaction and the C-C coupling product is obtained
nearly quantitatively.
Scheme 8: Plot of the iodoalkane and -arene radical anions calculated C-I bond lengths (see Table 3).
The green background symbolizes the region with straightforward Grignard-type reactions of the
organic iodide with calcium. The orange color contains radical anions which lead to formation of the
Grignard-type reagent and Wurtz coupling products. The red region shows iodoalkanes which are
reduced solely to the Wurtz coupling products.
Halide Exchange Reactions
The reduction of chloromethyltrimethylsilane with calcium powder failed. Therefore, a halogen
exchange was realized to study the influence of a lighter halogen on the NMR data and on structural
parameters. Thus, the reduction of iodomethyltrimethylsilane with calcium was performed in THF in
the presence of lithium chloride, yielding a mixture of trimethylsilylmethylcalcium halides. However,
a quantitative halide exchange was not achieved by this simple procedure. The attempt to prepare
an alkyl-arylcalcium complex via the reaction of phenylcalcium iodide with KCH2SiMe3 in
tetrahydrofuran, also containing LiCl, led to quantitative formation of [(thf)4CaPh2] and a few single
crystals of dinuclear [Me3SiCH2Ca(thf)3(µ-Cl)]2 (2c) suitable for X-ray diffraction experiments (Scheme
9). The molecular structure and atom numbering scheme of [Me3SiCH2Ca(thf)3(µ-Cl)]2 (2c) are
depicted in Figure 3.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Scheme 9: Synthesis of trimethylsilylmethylcalcium chloride via direct synthesis in LiCl-containing THF
solutions or via a metathetical approach of phenylcalcium iodide with trimethylsilylmethyl potassium
in THF in the presence of LiCl.
Figure 3: Molecular structure and atom labeling scheme of [Me3SiCH2Ca(thf)3(µ-Cl)]2 (2c). The
ellipsoids represent a probability of 30 %, only the H atoms at C1 are drawn with arbitrary radii.
Symmetry-related atoms (-x+2, -y+1, -z+1) are marked with the letter “A”. Selected bond lengths
(pm): Ca1-Cl1 273.6(1), Ca1-Cl1A 278.9(1), Ca1-C1 252.4(4), Ca1-O1 239.7(3), Ca1-O2 242.5(3), Ca1-
O3 238.9(3), Si1-C1 181.2(4), Si1-C2 189.4(5), Si1-C3 189.9(5), Si1-C4 187.3(5); bond angles (deg.):
Cl1-Ca1-Cl1A 82.36(3), Ca1-Cl1-Ca1A 97.64(3), Cl1-Ca1-C1 104.6(1), Cl1A-Ca1-C1 173.0(1), Ca1-C1-Si1
119.0(2).
Selected structural parameters are listed in Table 4 together with those of related compounds. In all
these complexes, the calcium atoms are in distorted octahedral environments. Aggregation to
dinuclear complexes of the type [R3SiCH2Ca(thf)3(µ-X)]2 reduces steric strain because a bulky ether
ligand is substituted by a bridging halide ion. This aggregation allows a Ca1-C1-Si1 bond angle of
119.0(2)° for 2c. The very bulky triisopropylsilyl group enforces aggregation (i.e. substitution of a thf
ligand by a bridging halide ion) and large Ca-C-Si angles are still inevitable but the preferred
coordination number of six can be realized. The Ca-C bond lengths vary over a very narrow range
regardless of aggregation and bulkiness of the trialkylsilyl group. A similar finding is valid for the Ca-O
distances to the ligated ether molecules. Unique properties characterize the trialkylsilylmethyl
groups. The Si-CR bond lengths to the methyl and isopropyl groups are significantly larger than the Si-
CCa distance to the calcium-bound carbon atom. This shortening is a consequence of negative
hyperconjugation, the backdonation of electron density from the carbanion into σ*(Si-CR) bonds of
the trialkylsilyl groups as discussed earlier in detail.23
Table 4: Comparison of selected structural parameters (average values, bond lengths (pm) and angles
(deg.)) of ether adducts of trialkylsilylmethylcalcium halides with hexa-coordinate calcium centers.
[R3SiCH2Ca(L)3(µ-Cl]2 [R3SiCH2Ca(L)4Br] [R3SiCH2Ca(L)4I]
2c 3b 3a
[R3SiCH2Ca(L)3(µ-Br)]2
4
Compound
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
R
L
Me
thf
Me
thp
Me
thp
iPr
thf
Ca-C
Ca-X
Ca-O
Si-CCa
Si-CR
CCa-Ca-X
Ca-CCa-Si
Ref.
252.4(4)
273.6(1), 278.9(1)
240.4
181.2(4)
188.9
104.6(1), 173.0(1)
119.0(2)
This work
250.0(4)
289.83(7)
239.3
181.9(4)
188.8
168.65(9)
143.9(2)
This work
252.7(3)
319.11(3)
239.9
183.1(3)
188.2
170.53(7)
131.2(1)
[28]
250.8
289.9, 303.6
240.4
183.3
192.0
98.2, 175.6
141.4
This work
The substitution of a halide in arylcalcium iodides via a metathetical approach with potassium
pseudohalides (like e.g. bis(trimethylsilyl)amide, diphenylphosphanide) was already investigated.[30]
Furthermore, the synthesis of bis(trimethylsilylmethyl)calcium followed a similar procedure via the
reaction of trimethylsilylmethylcalcium halide with KCH2SiMe3.[27] A similar approach was studied for
trimethylsilylmethylcalcium iodide to elucidate the influence of an alkyl group. Thus [Me3SiCH2-
Ca(thf)4I] was reacted with potassium bis(trimethylsilyl)amide in tetrahydrofuran yielding the
corresponding heteroleptic complex [Me3SiCH2-Ca(thf)3{N(SiMe3)2}] (5a) according to Scheme 10.
Substitution of all thf ligands by tridentate N,N,N',N'',N''-pentamethyldiethylenetriamine (pmdeta)
allowed crystallization of [Me3SiCH2-Ca(pmdeta){N(SiMe3)2}] (5b) with penta-coordinate calcium
centers.
Scheme 10: Metathetical approach for the synthesis of [Me3SiCH2-Ca(thf)3{N(SiMe3)2}] (5a) and
subsequent ligand exchange yielding [Me3SiCH2-Ca(pmdeta){N(SiMe3)2}] (5b).
Molecular structure and atom labeling scheme of [Me3SiCH2-Ca(pmdeta){N(SiMe3)2}] (5b) are
depicted in Figure 4. Despite a smaller coordination number of calcium the Ca1-C1 bond length
remains unchanged and a typical value of 253.2(2) pm is observed. The Ca1-N1 distance of 232.5(2)
pm is slightly elongated compared to [(dme)Ca{N(SiMe3)2}2] (227.1(3) pm, dme = 1,2-
dimethoxyethane)[41] and [(thf)2Ca{N(SiMe3)2}2] (229.4(3) and 230.9(3) pm)[42] with tetra-coordinate
metal centers in distorted tetrahedral coordination spheres. The Ca-Namide bond lengths in
[(tmeda)Ca{N(SiMe3)2}2] are elongated to 231.5(1) pm[43] verifying steric strain induced by the bulky
tmeda ligand. Despite small N1-Si2/3 bond lengths only a slightly enlarged Si2-N1-Si3 bond angle of
124.5(1)° is observed in derivative 5.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Figure 4: Molecular structure and atom labeling scheme of [Me3SiCH2-Ca(pmdeta){N(SiMe3)2}] (5b).
The ellipsoids represent a probability of 30 %, only the hydrogen atoms at C1 are drawn with arbitrary
radii. Selected bond lengths (pm): Ca1-N1 232.5(2), Ca1-C1 253.2(2), Ca1-N2 263.1(2), Ca1-N3
255.8(2), Ca1-N4 258.5(2), C1-Si1 182.7(2), Si1-C2189.1(3), Si1-C3 188.6(3), si1-C4 188.1(3), N1-Si2
169.2(2), N1-Si3 168.9(2), Si2-C5 187.2(2), Si2-C6 187.0(3), Si2-C7 189.0(3), Si3-C8 188.1(2), Si3-C9
188.7(2), Si3-C10 188.4(2); bond angles (deg.): N1-Ca1-C1 115.64(7), Ca1-C1-Si1 142.2(1), Ca1-N1-Si2
122.48(9), Ca1-N1-Si3 112.47(9), Si2-N1-Si3 124.5(1).
Furthermore, the pmdeta adduct of trimethylsilylmethylcalcium bis(trimethylsilyl)amide 5b was
combined with phenylsilane in [D6]benzene as depicted in Scheme 11 to elucidate whether the alkyl
or the amido group is preferably transferred to the silicon atom. NMR studies revealed that the
known hexanuclear hydrido amido complex [(pmdeta)Ca6H9{N(SiMe3)2}3] was formed. The nature of
this complex had already been authenticated by Harder and coworkers via NMR experiments and
crystal structure determination.[34] Obviously, the transfer of the alkyl groups is favored whereas the
bulky amido anions and the pmdeta ligands effectively shield the Ca6H9 calcium hydride cage
(Scheme 11).
Scheme 11: Reaction of [Me3SiCH2-Ca(pmdeta){N(SiMe3)2}] (5b) with excess of phenylsilane yielding
[(pmdeta)Ca6H9{N(SiMe3)2}3]. The shaded balls represent the hydrogen ions of the Ca6H9 calcium
hydride cage (L = pmdeta). The lines clarify the octahedral arrangement of the calcium atoms.
Conclusion
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
The discrepancy between the inertness of calcium metal and the enormous reactivity of its
organocalcium compounds (heavy Grignard reagents) has so far hampered a vast development of the
organic chemistry of this alkaline earth metal in contrast to the importance and broad application of
classic Grignard reagents. On the one hand, a high atomization energy of the metal (20 % larger than
for Mg) and low absolute solvation energies of the ions (22 % smaller than for Mg2+) synergistically
disfavor the Grignard reaction with calcium. Due to a larger radius of calcium (Ca: 197.4, Ca2+: 114;
Mg: 159.9, Mg2+: 86 pm)[26] a closer approach of the organyl halide onto the surface of the alkaline
earth metal particle for the SET reaction is possible for the lighter congener. These factors disfavor
the formation of a heavy Grignard reagent of the type RCaX. On the other hand, the enormous
reactivity of organylcalcium derivatives renders side reactions like ether degradation and Wurtz-type
coupling reactions more detrimental.
However, the straightforward synthesis of arylcalcium halides and diarylcalcium established a
promising heavy Grignard chemistry of calcium. Activation of calcium is required to overcome the
inertness of the metal. The direct synthesis is an adequate strategy to reduce organyl halides with
the halogen atom bound to a sp2-hybridized carbon atom and reduction of alkenyl- and cyclopropyl
bromides and iodides with calcium yields the corresponding organocalcium complexes. The
reduction of alkyl halides is significantly more intricate and only the reaction of trialkylsilylmethyl
bromide and iodide with calcium lead to the formation of alkylcalcium halides. The substitution of
only a single hydrogen atom of the methylene fragment by a methyl group precludes the formation
of the corresponding heavy Grignard reagent and only the Wurtz-type coupling product forms. The
direct synthesis is also inapplicable for the synthesis of benzyl-, methyl- and neopentylcalcium
compounds even though methylcalcium and benzylcalcium complexes areperfectly stable in ethereal
solutions.
Quantum chemical calculations verify a very strong influence of the C-I bond lengths in organyl iodide
radical anions varying between approximately 300 and 380 pm with iodobenzene showing the
smallest C-I bond length and alkyl iodides having the largest values. Radical anions with C-I bond
lengths below 330 pm smoothly form the heavy Grignard reagents R-Ca-I via a second SET step. For
radical anions with intermediate C-I distances the formation of the heavy Grignard reagent competes
with the Wurtz-type C-C coupling reaction whereas all radical anions with long C-I bonds yield the C-C
coupling product besides calcium iodide. This finding can be interpreted in the sense that the second
SET step is delayed and the organyl iodide radical anions with long (weak) C-I bonds dissociate
yielding the organyl radicals (which recombine to R-R) and the iodide anions (which form CaI2).
Organyl iodide radical anions with short C-I bonds are significantly less prone to dissociation and a
second SET step yields carbanions and iodide ions leading to the formation of the heavy Grignard
reagent R-Ca-I.
The calcium ions prefer an octahedral coordination sphere. Enhancement of steric strain leads to
substitution of an ether ligand by a bridging halide, maintaining the coordination number of six. This
behavior is evident for the trimethylsilylmethylcalcium halides. The bromide and iodide ions show
large distances to the calcium ions of 289.8 and 319.1 pm, respectively, whereas the short Ca-Cl bond
increases steric repulsion within the complex. This steric pressure leads to dissociation of one ether
ligand and dimerization via bridging halide ions. Enhancement of steric repulsion by replacement of
the Me3SiCH2 group by the bulkier iPr3SiCH2 substituent induces dissociation of one ether ligand as
well.
The challenges related to the atom economic direct synthesis of heavy Grignard reagents are related
to the retardation of the second SET reaction. One strategy for future investigations is doping of the
calcium prior to use with other (electropositive) metals. Addition of electron transfer catalysts could
also alleviate this hindrance.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Experimental Section
Materials and methods: All manipulations were carried out in an inert nitrogen or argon atmosphere
using standard Schlenk techniques, if not otherwise noted. Tetrahydrofuran, tetrahydropyran, and
pentane were dried over KOH and subsequently distilled over sodium/benzophenone in a nitrogen
atmosphere prior to use. PMDETA and all used alkyl halides were dried over CaH2. Deuterated
solvents were dried over sodium, distilled, degassed, and stored under nitrogen over sodium. 1H,
13C{1H}, and 29Si{1H} NMR spectra were recorded on Bruker Avance 400 and Fourier 300
spectrometers. Chemical shifts are reported in parts per million relative to SiMe4 as an external
standard referenced to the solvents residual proton signal (1H and 13C{1H}). Bromocyclopropane was
purchased from Alfa Aesar. Benzyl bromide was purchased from Merck. Me3CCH2I and KN(SiMe3)2
were purchased from Sigma Aldrich. The starting materials [(thp)4Ca(CH2SiMe3)I],[30] PhCH(tBu)Br,[35]
PhCH(SiMe3)Br,[36] tBuCHI2,[37] iPr3SiCHBr2,[38] Me3SiCH2Br,[39] and Me3SiCH(Me)I[40] were synthesized
according to literature protocols. KCH2SiMe3 was prepared by the metal-exchange reaction of
LiCH2SiMe3 with KOtBu in pentane. LiCH2SiMe3 was synthesized from Li ribbons and ClCH2SiMe3.
Purity was controlled by NMR spectroscopic measurements, by determination of the alkalinity of an
aliquot of the reaction solution and by complexometric titration of the Ca content with EDTA.
Synthesis of iodocyclopropane: In an argon filled glovebox, a Li ribbon (1.5 g, 0.22 mol) was cut into
small pieces and placed in a Schlenk flask with 150 mL of diethyl ether. A solution of
bromocyclopropane (10.0 g, 0.0826 mol) in 150 mL of Et2O was added dropwise at 0 °C over a period
of 3 hours. Then the slurry was stirred at r.t. overnight to afford a cloudy suspension also containing
unreacted Li chunks. The suspension was filtered and the resulting solution was cooled down to –78
°C. A solution of I2 (20 g, 0.079 mol) in Et2O (150 mL) was added dropwise till the persistence of a
brown colour. The solution was warmed to r.t. and treated with water and then with an aqueous
Na2SO3 solution. The organic phase was separated and dried over anhydrous MgSO4. Et2O was
removed by distillation at atmospheric pressure. Then CaH2 was added and the slurry was stirred
overnight. The brown residue was purified by trap-to-trap distillation to afford a colourless liquid
(8.94 g, 64%) iodocyclopropane. The analytical data are in accordance with literature values.[41]
Synthesis of Me3SiCH(tBu)I (1-iodo-1-trimethylsilyl-2,2-dimethylpropane): A solution of 2.5M nBuLi
(10.6 mL, 26.5 mmol) in hexane was added at -78 °C to a solution of HN(iPr)2 (3.7 mL, 26.5 mmol) in
20 mL of a mixture of THF and Et2O (1:1). The solution was stirred for 15 min and then cooled to –115
°C and then 17 g of tBuCHI2 (13 mmol) were added, imminently leading to an orange-red suspension.
After 15 minutes, 1.7 mL of Me3SiCl were added at the same temperature. The slurry was stirred for
additional 30 min and then warmed to r.t. The reaction mixture was quenched with water and Et2O.
The combined organic phase was washed with water and then dried with anhydrous Na2SO4. The
solvent was removed in vacuum and CaH2 was added to the residue. The product was obtained by
trap to trap condensation, yielding 3.1 g of a colorless liquid (yield: 89 %, purity (GC): 95 %)
containing small amounts of Me3CCH2I (5 %). 1H NMR (CDCl3, 400.13 MHz): δ = 0.27 (s, 9H), 1.19 (s,
9H), 3.38 (s, 1H). 13C{1H} NMR (CDCl3, 100.6 MHz): δ = 1.9, 30.9, 35.5, 44.4. 29Si{1H} NMR (CDCl3, 79.5
MHz) δ = 4.0. IR: 2955 m, 2897 w, 2867 w, 1474 w, 1461 w, 1392 w, 1364 m, 1260 m, 1249 s, 1224 s,
1144 w, 1085 w, 1059 w, 1017 w, 935 w, 907 w, 836 s, 765 w, 751 w, 686 w, 617 w, 580 w. HR-MS
(EI-MS): [M-Me]+ C7H16ISi+: calc.: 255.0060, found: 255.0044.
Synthesis of iPr3SiCH2Br: iPr3SiCHBr2 (1.015 g, 3.07 mmol) was dissolved in 15 mL of THF and cooled to
–115 °C. A 2.5M hexane solution of nBuLi (1.23 mL, 3.07 mmol) was slowly added to this solution.
The yellow solution was stirred for 5 min and then quenched with 5 mL of MeOH at this temperature.
After warming to r.t., water and Et2O were added and the organic phase isolated and dried with
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
anhydrous Na2SO4. The solvent was removed in vacuum and CaH2 was added to the clear colorless
liquid. The product was obtained by trap-to-trap condensation, yielding 750 mg of a colorless liquid
of iPr3SiCH2Br (97 %). 1H NMR (CDCl3, 300.13 MHz): δ = 1.06 (d, 18H, J = 6.32 Hz), 1.15 (m, 3H, J = 6.32
Hz), 2.58 (s, 2H). 13C{1H} NMR (CDCl3, 100.6 MHz): δ = 18.5, 12.2, 10.8. 29Si{1H} NMR (CDCl3, 79.5
MHz): δ = 4.81. IR: 2941 s, 2890 m, 2965 s, 1461 s, 1383 m, 1366 w, 1247 w, 1134 w, 1068 w, 1016 w,
1000w, 918 w, 881 s, 717 s, 663 m, 617 m 594 m, 570 m. HR-MS (EI-MS): [M-iPr]+ C7H16BrSi+: calc.:
207.0199, found: 207.0197.
General procedure for the reaction of activated calcium with alkyl halide: Activated calcium (20 mg,
0.5 mmol, 1.1 equiv.) was suspended in 3 mL of THF and cooled to –78 °C prior to addition of alkyl
halide (0.45 mmol, 1 equiv.). The reaction mixture was stirred for 1h at –78 °C and then slowly
warmed to r.t. Then all solids were removed by filtration. This solution was analysed by acidimetric
titration. For this purpose 1 mL was hydrolyzed, 3 mL of Et2O were added and the phases were
separated. The organic layer was dried with anhydrous Na2SO4 and directly analysed by GC/MS.
Synthesis of cyclopropylcalcium iodide (1a): Iodocyclopropane (1.53 g, 9.13 mmol) was added at 10 °C
to a suspension of Ca* (366 mg, 9.13 mmol) in 20 mL of tetrahydropyran. The slurry was shaken for
two hours. Then the reaction mixture was warmed to r.t. and unreacted calcium was removed by
filtration. Acidimetric titration of an aliquot of the resulting red solution gave a conversion of 70 %.
Repeated concentration of the solution by evaporation under vacuum and cooling to –40 °C afforded
several crops of colorless crystals of [(thp)4CaI2]. Once no more precipitate was obtained, the mother
liquor was evaporated to dryness and washed several times with pentane. The brown solid residue
consisted of a mixture of cyclopropyl calcium iodide and dicyclopropyl calcium (400 mg). Due to the
composition as a mixture, reliable elemental analysis were not obtained. 1H NMR ([D8]THF, 400.13
MHz): δ = -1.65 (bm, 1H, Hα), 0.20-0.40 (br m, 2H, Hβ/β’), 0.50-0.75 (br m, 2H, Hβ/β’), 1.49 (m, thp),
1.60 (m, thp), 3.53 (t, J = 5.14 Hz, thp). 13C{1H} NMR ([D8]THF, 125.8 MHz): δ = 6.98 (Cβ), 7.89 (Cα),
24.6 (thp), 27.6 (thp), 69.1 (thp).
Synthesis of [(thp)4CaBr(CH2SiMe3)] (3b): Activated calcium (482 mg, 12.02 mmol, 1.1 equiv.) was
suspended in 20 mL of THF. The suspension was cooled to –78 °C before 1.9 g of BrCH2SiMe3 (10.9
mmol, 1 equiv.) were added. The reaction mixture was shaken for 2h at –78 °C. Then all solids were
removed by filtration yielding 15 mL of a 0.219 M solution of [(thf)4CaBr(CH2SiMe3)] (3a) in THF (30 %
yield by acidimetric titration). The solvent was removed and the residue was dissolved in 5 mL of
THP. This procedure was repeated 3 times. Finally, the remaining white solid was dissolved in 5 mL of
THP and 15 mL of pentane. The solution was cooled to – 40 °C, yielding crystals of
[(thp)4CaBr(CH2SiMe3)] (3b). The crystalline solid was collected, yielding 781 mg of 3b (1.4 mmol, 13
%), which easily loses coordinated thp during the isolation. Therefore, no reliable combustion
analysis data could be obtained. 1H NMR ([D8]THF, 400.13 MHz): δ = -1.83 (s, 2H), -0.13 (s, 9H), 1.53
(m, 16 H), 1.63 (m, 8H), 3.57 (t, 16 H, J = 5.14 Hz). 13C{1H} NMR ([D8]THF, 100.6 MHz): δ = 4.5, 5.1,
23.5, 26.6, 68.1. Ca content (%): calc. for [(thp)3.5CaBr(CH2SiMe3)]: 7.8, found: 8.0.
Synthesis of [iPr3SiCH2Ca(thf)3(µ-Br)]2 (4): Activated calcium (403 mg, 10.05 mmol, 1.4 equiv.) was
suspended in 20 mL of THF. The suspension was cooled to –78 °C before 1.7 g of iPr3SiCH2Br (6.8
mmol, 1 equiv.) were added. The reaction mixture was shaken for 2h at –40 °C. Then all solids were
removed by filtration, yielding 18 mL of a clear 0.263M solution of 4 in THF (70 % yield by acidimetric
titration). The solution was concentrated to 5 mL and then 20 mL of pentane were added. Cooling to
–40 °C yielded a few crystals of 4. Additional 20 mL of pentane were added and the reaction mixture
was stored for one week at –40 °C. The crystalline solid was collected, yielding 413 mg of
[iPr3SiCH2Ca(thf)3(µ-Br)]2 (0.4 mmol, 12 %), which easily loses coordinated THF during handling and
isolation. 1H NMR ([D8]THF, 400.23 MHz): δ = -2.26 (br s, 2H), 0.71 (sept, 3H, 7.36 Hz), 0.92 (d, 18 H,
7.36 Hz), 1.71 (m, 12 H), 3.57 (m, 12 H). 13C{1H} NMR ([D8]THF, 100.6 MHz): δ = -10.5, 11.9, 17.5, 23.5,
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
65.4. 29Si{1H} NMR ([D8]THF, 79.5 MHz): δ = 10.6. Ca content (%): calc. for [iPr3SiCH2Ca(thf)2.3(µ-Br)]2:
8.7, found: 8.6.
Synthesis of [Me3SiCH2Ca(thf)3(µ-Cl)]2 (2c): KCH2SiMe3 (358 mg, 2.76 mmol, 1 equiv.) , containing LiCl
and/or KCl from the direct synthesis of LiCH2SiMe3 from ClCH2SiMe3 and lithium, were dissolved in 5
mL of THF at 0 °C. To this solution, 14.2 mL of a 0.195M solution of [(thf)4CaPhI] (2.76 mmol) in THF
were added dropwise. The white suspension was stirred for 30 min. Afterwards, all solids were
removed by filtration, leading to an orange solution. The volume was reduced to 7 mL and additional
7 mL of pentane were added. The solution was stored at –40°C, leading to quantitative precipitation
of [(thf)4CaPh2] beneath a few crystals of [Me3SiCH2Ca(thf)3(µ-Cl)]2, which only allow the
determination of the crystal structure as well as the 1H-NMR spectrum of this compound. The 13C{1H}
NMR spectrum shows only the singlet of the methyl-groups. The analysis of the latter solution
verified that only chloride counterions are present and no other halides. Due to cocrystallization of
[(thf)4CaPh2] no analytically pure product was obtained. 1H NMR ([D8]THF, 300.13 MHz): δ = -1.85 (s,
2H), -0.14 (s, 9H), 1.80 (m, 12 H), 3.64 (m, 12H). 13C{1H} NMR ([D8]THF, 75.4 MHz): δ = 4.5, 25.4, 67.2.
Synthesis of [(pmdeta)Ca{N(SiMe3)2}(CH2SiMe3)] (5b): KN(SiMe3)2 (293 mg, 1.47 mmol, 1 equiv.) was
dissolved in 5 mL of THF. To this solution 21 mL of a 0.07M tetrahydropyran solution of
[(thp)4Ca(CH2SiMe3)I] (3a, 1.47 mmol, 1 equiv.) were added at 0 °C. The suspension was stirred for
half an hour. Then 0.33 mL of PMDETA were added. All volatiles were removed in vacuum and the
residue was extracted with 20 mL of pentane. All solids were removed by filtration and the volume of
the colorless solution was reduced to 5 ml. Storing at –40 °C led to precipitation of 180 mg of
crystalline 5b (28 %). 1H NMR ([D6]benzene, 300.13 MHz): δ = -1.37 (s, 2H), 0.37 (s, 18H), 0.42 (s, 9 H),
1.38 (m, 4H), 1.64 (s, 3H), 2.00 (m, 4H), 2.13 (s, 12H). 13C{1H} NMR ([D6]benzene, 100.6 MHz): δ = 6.0,
6.2, 6.7, 45.7, 47.7, 55.5, 56.9. Ca contente (%): calc. for [(pmdeta)Ca{N(SiMe3)2}(CH2SiMe3)]: 8.7,
found: 8.5.
Structure Determinations: The intensity data for the compounds were collected on a Nonius
KappaCCD diffractometer using graphite-monochromated Mo-K radiation. Data were corrected for
Lorentz and polarization effects; absorption was taken into account on a semi-empirical basis using
multiple-scans.[42-44] The crystal structures were solved by Direct Methods (SHELXS[45]) and refined by
2
full-matrix least squares techniques against Fo (SHELXL-97[45] and SHELXL-2014[46]). The hydrogen atoms
bonded to the methylene-groups were located by difference Fourier synthesis and refined isotropically.
All other hydrogen atoms were included at calculated positions with fixed thermal parameters. All non-
hydrogen, non-disordered atoms were refined anisotropically.[45,46] The crystal of 2c contains large
voids, filled with disordered solvent molecules. The size of the voids are 271 Å3/unit cell. Their
contribution to the structure factors was secured by back-Fourier transformation using the SQUEEZE
routine of the program PLATON[47] resulting in 101 electrons/unit cell. Crystallographic data as well as
structure solution and refinement details are summarized in Table S1 in the Supporting Information.
XP (SIEMENS Analytical X-ray Instruments, Inc.)[48] was used for structure representations.
Computational details: All DFT calculations presented in this study were performed using the Gaussian
16 program.[49] The geometry optimizations of the studied mono radical anions were performed at
B97D3 level of theory with the basis set 6-311++G** for all atoms except iodine which was described by
the relativistic core potential MWB and the corresponding basis set. Frequency analysis showed no
imaginary frequencies.
Supporting Information available
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
NMR spectra, crystallographic and structure refinement details. Crystallographic data (excluding
structure factors) has been deposited with the Cambridge Crystallographic Data Centre as
supplementary publication CCDC-1825811 for 3b, CCDC-1825812 for 4, CCDC-1825813 for 2c, and
CCDC-1825814 for 5b. Copies of the data can be obtained free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK [E- mail: deposit@ccdc.cam.ac.uk].
The authors declare no conflict of interest.
Acknowledgement
Dr. Alexander Koch is grateful to the Fonds der Chemischen Industrie im Verband der Chemischen
Industrie (FCI/VCI, Frankfurt, Germany; fund 510259) and to the Graduate Academy of the Friedrich
Schiller University Jena for generous Ph.D. grants. Dr Quentin Dufrois is grateful to the French
Ministry for Higher Education and Research MENESR for a Ph.D. grant. We acknowledge the support
by the NMR service platform (http://www.nmr.uni-jena.de) of the Faculty of Chemistry and Earth
Sciences of the Friedrich Schiller University Jena. M.E. thanks Université Paul Sabatier for a sabbatical
leave.
References
[1] a) B. J. Wakefield: Organolithium Methods, Academic Press: London, 1988; b) A.-M. Sapse, P. v. R.
Schleyer (eds.): Lithium Chemistry: A Theoretical and Experimental Overview, Wiley: New York,
1995; c) J. Clayden: Organolithiums: Selectivity for Synthesis, Pergamon: Amsterdam, 2002; d) Z.
Rappoport, I. Marek (eds.): The Chemistry of Organolithium Compounds, Wiley: Chichester, 2004;
e) R. Luisi, V. Capriati (eds.): Lithium Compounds in Organic Synthesis: From Fundamentals to
Applications, Wiley-VCH: Weinheim, 2014.
[2] a) B. J. Wakefield: Organomagnesium Methods in Organic Synthesis, Academic Press: London,
1995; b) G. S. Silverman, P. E. Rakita (eds.): Handbook of Grignard Reagents, CRC Press: Boca
Raton, 1996; c) H. G. Richey (ed.): Grignard Reagents: New Developments, Wiley: Chichester,
2000; d) Z. Rappoport, I. Marek (eds.): The Chemistry of Organomagnesium Compounds, Wiley:
Chichester, 2008; e) S. Krieck, M. Westerhausen in T. P. Hanusa (ed.): The Lightest Metals: Science
and Technology from Lithium to Calcium, Wiley: Chichester, 2015; pp. 213-229.
[3] a) T. Kottke, D. Stalke, J. Appl. Cryst. 1993, 26, 615-619; b) D. Stalke, Chem. Soc. Rev. 1998, 27,
171-178.
[4] A. F. Holleman, E. Wiberg, N. Wiberg: Lehrbuch der Anorganischen Chemie (Holleman-Wiberg), W.
de Gruyter: Berlin, 102. ed., 2007.
[5] a) M. Westerhausen, Angew. Chem. Int. Ed. 2001, 40, 2975-2977; b) M. Westerhausen, M.
Gärtner, R. Fischer, J. Langer, L. Yu, M. Reiher, Chem. Eur. J. 2007, 13, 6292-6306; c) M.
Westerhausen, M. Gärtner, R. Fischer, J. Langer, Angew. Chem. Int. Ed. 2007, 46, 1950-1956; d) M.
Westerhausen, Z. Anorg. Allg. Chem. 2009, 635, 13-32; e) M. Westerhausen, J. Langer, S. Krieck, C.
Glock, Rev. Inorg. Chem. 2011, 31, 143-184; f) M. Westerhausen, J. Langer, S. Krieck, R. Fischer, H.
Görls, M. Köhler, Top. Organomet. Chem. 2013, 45, 29-72; g) M. Westerhausen, A. Koch, H. Görls,
S. Krieck, Chem. Eur. J. 2017, 23, 1456-1483.
[6] M. Köhler, J. Langer, R. Fischer, H. Görls, M. Westerhausen, Chem. Eur. J. 2013, 19, 10497-10500.
[7] a) F. Feil, S. Harder, Organometallics 2000, 19, 5010-5015; b) V. Knapp, G. Müller, Angew. Chem.
Int. Ed. 2001, 40, 183-186; c) S. Harder, F. Feil, A. Weeber, Organometallics 2001, 20, 1044-1046;
d) S. Harder, F. Feil, Organometallics 2002, 21, 2268-2274; e) F. Feil, C. Müller, S. Harder, J.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Organomet. Chem. 2003, 683, 56-63; f) S. Harder, S. Müller, E. Hübner, Organometallics 2004, 23,
178-183; g) A. Torvisco, K. Ruhlandt-Senge, Inorg. Chem. 2011, 50, 12223-12240; h) C. Müller, S.
Krieck, H. Görls, M. Westerhausen, Inorg. Chem. 2015, 54, 2100-2102.
[8] B. M. Wolf, C. Stuhl, C. Maichle-Mössmer, R. Anwander, J. Am. Chem. Soc. 2018, 140, 2373-2383.
[9] A. S. S. Wilson, M. S. Hill, M. F. Mahon, C. Dinoi, L. Maron, Science 2017, 358, 1168-1171.
[10] a) T. P. Hanusa, Coord. Chem. Rev. 2000, 210, 329-367; b) J. S. Alexander, K. Ruhlandt-Senge, Eur.
J. Inorg. Chem. 2002, 2761-2774; c) W. D. Buchanan, D. G. Allis, K. Ruhlandt-Senge, Chem.
Commun. 2010, 46, 4449-4465; d) A. Torvisco, K. Ruhlandt-Senge, Top. Organomet. Chem. 2013,
45, 1-28.
[11] F. G. N. Cloke, P. B. Hitchcock, M. F. Lappert, G. A. Lawless, B. Royo, J. Chem. Soc., Chem.
Commun. 1991, 724-726.
[12] M. Gärtner, H. Görls, M. Westerhausen, Synthesis 2007, 725-730.
[13] M. Köhler, H. Görls, J. Langer, M. Westerhausen, Chem. Eur. J. 2014, 20, 5237-5239.
[14] J. F. Garst, M. P. Soriaga, Coord. Chem. Rev. 2004, 248, 623-652.
[15] A. Koch, S. Krieck, H. Görls, M. Westerhausen, Inorganics 2016, 4, 39.
[16] a) S. Krieck, H. Görls, L. Yu, M. Reiher, M. Westerhausen, J. Am. Chem. Soc. 2009, 131, 2977-
2985; b) S. Krieck, H. Görls, M. Westerhausen, J. Am. Chem. Soc. 2010, 132, 12492-12501.
[17] a) J. Meisner, J. Kästner, Angew. Chem. Int. Ed. 2016, 55, 5400-5413; b) J. Meisner, M. N.
Markmeyer, M. U. Bohner, J. Kästner, Phys. Chem. Chem. Phys. 2017, 19, 23085-23094.
[18] J. Langer, M. Köhler, R. Fischer, F. Dündar, H. Görls, M. Westerhausen, Organometallics 2012, 31,
6172-6182.
[19] A. de Meijere, Angew. Chem. Int. Ed. Engl. 1979, 18, 809-826.
[20] Q. Dufrois, S. Mallet-Ladeira, C. Dinoi, M. Etienne, Eur. J. Inorg. Chem. 2018, 2387-2392.
[21] M. Etienne, A. S. Weller, Chem. Soc. Rev. 2014, 43, 242-259.
[22] J. Langer, S. Krieck, R. Fischer, H. Görls, M. Westerhausen, Z. Anorg. Allg. Chem. 2010, 636, 1190-
1198.
[23] M. Köhler, A. Koch, H. Görls, M. Westerhausen, Organometallics 2016, 35, 242-248.
[24] A. Koch, S. Krieck, H. Görls, M. Westerhausen, Organometallics 2017, 36, 2811-2817.
[25] R. Fischer, M. Gärtner, H. Görls, M. Westerhausen, Organometallics 2006, 25, 3496-3500.
[26] R. Fischer, M. Gärtner, H. Görls, M. Westerhausen, Angew. Chem. Int. Ed. 2006, 45, 609-612.
[27] A. Koch, M. Wirgenings, S. Krieck, H. Görls, G. Pohnert, M. Westerhausen, Organometallics 2017,
36, 3981-3986.
[28] D. T. Sawyer, A. Sobkowiak, J. L. Roberts: Electrochemistry for Chemists, 2nd ed., Wiley: New York,
1995.
[29] J. March: Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, 3rd ed., Wiley, New
York, 1985; p. 19.
[30] M. Gärtner, H. Görls, M. Westerhausen, Organometallics 2007, 26, 1077-1083.
[31] M. Westerhausen, W. Schwarz, Z. Anorg. Allg. Chem. 1991, 604, 127-140.
[32] M. Westerhausen, M. Hartmann, N. Makropoulos, B. Wieneke, M. Wieneke, W. Schwarz, D.
Stalke, Z. Naturforsch. 1998, 53b, 117-125.
[33] C. Glock, H. Görls, M. Westerhausen, Inorg. Chim. Acta 2011, 374, 429-434.
[34] B. Maitland, M. Wiesinger, J. Langer, G. Ballmann, J. Pahl, H. Elsen, C. Färber, S. Harder, Angew.
Chem. Int. Ed. 2017, 56, 11880-11884.
[35] A. J. Fry, T. A. Powers, J. Org. Chem. 1987, 52, 2498-2501.
[36] H. Andringa, Y. A. Heus-Kloos, L. Brandsma, J. Organomet. Chem. 1987, 336, C41-C43.
[37] A. Pross, S. Sternhell, Aust. J. Chem. 1970, 23, 989-1003.
[38] D. L. Usanov, H. Yamamoto, Org. Lett. 2012, 14, 414-417.
[39] J. L. Speier, J. Am. Chem. Soc. 1951, 73, 826-827.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
[40] O. Gebauer, H. Gayer, U. Heinemann, S. Herrmann, S. Hillebrand, H.-L. Elbe, R. Ebbert, U.
Wachendorff-Neumann, P. Dahmen, K.-H. Kuck, Patent WO 2004-EP14592.
[41] D. E. Applequist, D. F. O‘Brien, J. Am. Chem. Soc. 1963, 85, 743-748.
[42] R. Hooft, COLLECT, Data Collection Software; Nonius B.V., Netherlands, 1998.
[43] Z. Otwinowski, W. Minor in C. W. Carter, R. M. Sweet (eds.): Methods in Enzymology, Vol. 276,
Macromolecular Crystallography, Part A, pp. 307-326, Academic Press: New York, 1997.
[44] SADABS 2.10, Bruker-AXS Inc., 2002, Madison, WI, USA.
[45] G. M. Sheldrick, Acta Cryst. 2008, A64, 112-122.
[46] G. M. Sheldrick, Acta Cryst. 2015, C71, 3-8.
[47] A. L. Spek, Acta Cryst. 2015, C71, 9-18.
[48] XP, Siemens Analytical X-Ray Instruments Inc., 1990, Karlsruhe, Germany; 1994; Madison, WI,
USA.
[49] Gaussian 16, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J.
R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V.
Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F.
Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A.
Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J.
Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K.
Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene,
C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, and D.
J. Fox, Gaussian, Inc., Wallingford CT, 2016.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803518
Chemistry - A European Journal
Table of Contents Entry
The success of the direct synthesis of heavy Grignard reagents, i.e. organocalcium iodides, strongly
depends on the C-I bond length of the organyl iodide radical anion that forms after the initial SET
reaction, yielding either R-Ca-I and/or Wurtz-type C-C coupling products.
This article is protected by copyright. All rights reserved.