CeCl3/n-BuLi: Unraveling Imamoto's Organocerium Reagent

Article abstract of DOI:10.1002/anie.202103889

CeCl₃(thf) reacts at low temperatures with MeLi, t-BuLi, and n-BuLi to isolable organocerium complexes. Solvent-dependent extensive n-BuLi dissociation is revealed by ⁷Li NMR spectroscopy, suggesting “Ce(n-Bu)₃(thf)_x<

Full text of DOI:10.1002/anie.202103889

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: CeCl3/n-BuLi: Unraveling Imamoto`s Organocerium Reagent
Authors: Tassilo Berger, Jakob Lebon, Cäcilia Maichle-Mössmer, and
Reiner Anwander
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: Angew. Chem. Int. Ed. 10.1002/anie.202103889
Link to VoR: https://doi.org/10.1002/anie.202103889

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
CeCl3/n-BuLi: Unraveling Imamoto`s Organocerium Reagent
Tassilo Berger, Jakob Lebon, Cäcilia Maichle-Mössmer, and Reiner Anwander*
Dedicated to Professor Karl W. Törnroos on the occasion of his 65th birthday
[*]
Tassilo Berger, Jakob Lebon, Dr. Cäcilia Maichle-Mössmer, Prof. Dr. Reiner Anwander
Institut für Anorganische Chemie, Eberhard Karls Universität Tübingen, Auf der Morgenstelle 18, 72076 Tübingen
E-mail: reiner.anwander@uni-tuebingen.de
Supporting information for this article is given via a link at the end of the document.
Abstract: CeCl3(thf) reacts at low temperatures with MeLi, t-BuLi, and
n-BuLi to the isolable organocerium complexes Li3Ce(CH3)6(tmeda)3,
[Li(thf)4][Ce(t-Bu)4], Li3Ce(n-Bu)6(thf)4, and Li2Ce(n-Bu)5(tmeda)2. The
n-butyl derivatives were also synthesized for the rare-earth elements
scandium, lutetium, yttrium and lanthanum, covering the entire Ln(III)
size range. Solvent-dependent extensive n-BuLi dissociation is
revealed by 7Li NMR spectroscopy, suggesting “Ce(n-Bu)3(thf)x” or
solvent-separated ion pairs like “[Li(thf)4][Ce(n-Bu)4(thf)y]” as the
dominant species of the Imamoto reagent. The stability of complexes
Li3Ln(n-Bu)6(thf)4 increases markedly with decreasing Ln(III) size.
Closer inspection of the solution behavior of crystalline Li3Lu(n-
Bu)6(thf)4 and mixtures of LuCl3(thf)2/n-BuLi in THF indicate occurring
n-BuLi dissociation only at molar ratios of <1:3. n-BuLi-depleted
complex LiLu(n-Bu)3Cl(tmeda)2 was obtained by treatment of Li2Lu(n-
Bu)5(tmeda)2 with ClSiMe3, at the expense of LiCl incorporation.
Exhaustive protonation is shown for the alcoholysis of Li3Ce(n-
Bu)6(thf)4 with neopentanol, yielding lithium-depleted heterobimetallic
= n-Bu, tBu) as effective reagents for regioselective carbon–
carbon-bond forming with various carbonyl compounds.[7] The
best results were obtained when employing equimolar mixtures at
–78 °C to –65 °C. Like for the Luche reagent
(CeCl3(H2O)7/NaBH4),[8] cerium was launched as the least
expensive rare-earth metal while a greater part of transformations
was performed with n-BuLi as an easy-to-handle (as well as the
cheapest) organolithium derivative.[9] Figure
1
depicts
characteristic features of the nucleophilic addition of such
organocerium reagents to carbonyl compounds, including smooth
and selective 1,2-addition in case of α,β-unsaturated or easily
enolizable substrate molecules,[7] functional group tolerance[10] as
well as diastereocontrol via chelate coordination.[11]
Ce3+
control of
1,4-addition
disfavored
stereochemistry
O
C
alkoxide
Li3Ce2(OCH2tBu)9(HOCH2tBu)2(thf).
Imamoto´s
Do
H
C
ketone/tertiary alcohol transformation was examined with 1,3-
diphenylpropan-2-one. [protocol: CeCl3(thf), n-BuLi (1 equiv), THF, –
35 °C, 30 min/ketone, –35 °C, 30 min] affording 99% of alcohol. The
isolation of lithium alkoxides Li4[OC(=CHPh)CH2Ph]4(thf)4 (1,3-
diphenylpropan-2-one transformation) and Li8[OCMe2(n-Bu)]6Cl2(thf)6
(acetone transformation) point to competitive enolization and LiCl as
an integral (stabilizing) part of the reagent solution, respectively.
H
deprotonation and
enolate formation
disfavored
nucleophilic attack
of LiR at carbonyl
group favored
Figure 1. Ce(III)–carbonyl coordination directs highly selective nucleophilic
addition reactions of binary CeX3/RLi (X = halogenido; R = alkyl).
The much improved selectivity compared to organolithium or
Grignard reagents was assigned to a milder basicity of the
organocerium reagent and enhanced hardness of the carbonyl
carbon atom.[4] The latter originates from the strong oxophilicity of
the trivalent cerium. Although Imamoto´s seminal organocerium
reagents embarked on a new and prosperous branch of
organolanthanide chemistry, very few studies exist that target the
structural elucidation of such bimetallic mixtures.[12] In sharp
contrast, heterobimetallic main-group organometallic reagents
has been given much greater attention, and entitled modern ate
chemistry.[13]
Imamoto-type rare-earth-metal reagents, more recently employed
for halogen–rare-earth-metal exchange reactions[14] or Zweifel
olefinations,[15] are as a rule generated in situ and have been
designated “n-Bu2LaCl•4LiCl”,[16] “n-Bu3Sm•5LiCl”,[17] or simply
“n-Bu3Ce”.[15] The formulas were derived from X-ray absorption
fine structure (EXAFS) studies[18] or Raman spectroscopy.[15]
Previous enlightening studies on the CeCl3/RLi binary system
focused mainly on the composition and activation of the cerous
chloride precursor[12,19-20] as well as the effect/effectiveness of
Introduction
The redox reagents Ceric Ammonium Nitrate (CAN
=
(NH4)2Ce(NO3)6) and SmI2(thf)2 as well as the binary alkylating
agents CeCl3/LiR (R = alkyl like CH3 or n-C4H9) constitute the
most commonly employed rare-earth-metal reagents in organic
transformations (including natural product synthesis).[1-4] The
cause of reactivity of the redox-active compounds is well
understood,[2-3] and their crystal structures revealed by X-ray
diffraction (XRD) analyses.[5],[6] In the solid state, CAN exhibits a
three-dimensional network of 8-coordinate hexanitratocerate
anions and ammonium cations interconnected by hydrogen
bonding.[5] On the other hand, samarium diiodide crystallizes as a
monomeric pentasolvate, SmI2(thf)5, from THF solution.[6] Due to
the extreme air- and moisture sensitivity and combined thermal
instability, a detailed structural investigation of organocerium
reagents has remained elusive.
In 1984, Tsuneo Imamoto et al described the use of binary
mixtures CeI3/RLi (R = Me, Et, n-Bu, secBu, Ph) and CeCl3/RLi (R
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
reagent stoichiometry.[21] It was revealed that the generally
applied thermal activation of the commercially available
heptahydrate CeCl3(H2O)7 not only generates a material of
composition [CeCl3(H2O)]n[12c] but also benefits from sonication.[20]
Importantly, organocerium addition to hydrazones was found
most effective and selective for a 1:1 stoichiometry of CeCl3/MeLi,
but the active reagent formed at different stoichiometries was
2.600(2)-2.614(2) Å[29] and 2.620(7)-2.646(8) Å,[30] respectively,
while those in formally 3-coordinate complexes Ce[CH(SiMe3)2]3
and Ce[C(SiHMe2)3]3 were detected at 2.475(7) Å[31]
2.651(2)/2.659(2)/2.672(2) Å, respectively.[32]
proposed to be
a trimethylcerium species (supported by
unreacted CeCl3).[21] Although this latter investigation “precluded
firm conclusions”, the results “do point out the fallacy of ascribing
reagent composition on the basis of mixing stoichiometry
especially at low loadings of alkyllithium”.[21]
Herein we describe the successful isolation and structural
characterization of rare-earth-metal n-butyl complexes formed in
CeCl3/n-BuLi systems devoid of ancillary ligands. NMR
7
spectroscopic studies involving the Li nucleus provide valuable
insights into the solution behavior of such binary mixtures,
pointing to the true organocerium species of the Imamoto reagent.
Figure 2. Crystal structure of Li3Ce(CH3)6(tmeda)3 (1). Atomic displacement
ellipsoids set at 50% probability. Hydrogen atoms omitted for clarity. Selected
interatomic distances [Å] and angles [°]: Ce1–C1 2.6795(19), C1–Li1 2.205(3),
Li1–N1 2.106(3); C1–Ce1–C1’ 88.05(9).
Results and Discussion
Synthesis and Solid-State Structure of the Organocerium
Derivatives Li3Ce(CH3)6(tmeda)3 and [Li(thf)4][CetBu4]. For
assessing the CeCl3/LiR salt-metathesis protocol we initially
probed the methyl derivative, since this would rule out β-H
elimination as the main decomposition pathway. Complexes of
the type Li3LnMe6(tmeda)3 featuring the entire lanthanide series
except for promethium and europium were accessed by
Schumann et al as early as 1978, via mixtures
Homoleptic anionic t-butyl complexes were previously reported
for [Li(thf)x][Ln(t-Bu)4] (Ln = Sm, Er: x = 4; Y: x = 3),[27a]
[Li(OEt2)4][Er(t-Bu)4],[22c] [Li(tmeda)2][Ln(t-Bu)4] (Ln = Tb, Lu),[22c,
27b]
and [Li(dme)3][Ln(t-Bu)4] (Ln = Tb, Er).[27c] Crystal structures
were obtained for [Li(tmeda)2][Lu(t-Bu)4][27b] and [Li(dme)3][Er(t-
Bu)4],[27c] whereas the enhanced thermal instability of derivatives
of the “lighter” rare-earth metals was pointed out. In order to test
our low-temperature set-up for organocerium derivatives prone to
b-H elimination,[33] we targeted the anionic fragment [Ce(t-Bu)4].
Although, the mixture CeCl3(thf)/t-BuLi/THF gave access to
complex [Li(thf)4][Ce(t-Bu)4] (2) at –40 °C, single crystals could be
obtained only from very concentrated, oily residues, not allowing
for decent elemental analysis (1H NMR spectrum: dtBu = 2.39 ppm,
Figure S7). Notwithstanding an XRD analysis revealed a 4-
coordinate cerium center (Figure 3), being isostructural to the
previously reported erbium[27c] and lutetium derivatives.[27b] The
Ce–C distances range from 2.501(11) to 2.544(11) Å, and are
considerably shorter than those in 6-coordinate 1 (2.6795(19) Å),
but matching those of [Li(tmeda)2][Lu(t-Bu)4] (2.32(2)-2.43(2)
Å)[27b] and [Li(dme)3][Er(t-Bu)4] (2.352(6)-2.395(6) Å),[27c] when
taking into account the Ln(III) ion size.
LnCl3/LiMe/OEt2/TMEDA
(TMEDA=
tetramethyletylene-
diamine).[22] Solid-state structures applying XRD analysis were
described for the rare-earth metals erbium[22b] and holmium.[22c]
Moreover, both the stabilizing effect of chelating tmeda, teeda
(=tetraethyletylenediamine), and dme (= dimethoxyethylene)
coligands and the enhanced instability of derivatives of the “lighter”
and larger rare-earth metals has been emphasized.[23] This is in
accord with more recent findings by Okuda et al on the stability of
Li3LnMe6(thf)x (isolable for Ln smaller than Eu), which form the
pentametallic ate complexes Li3Sc2Me9(thf)2(OEt2)3 and
Li3Ln2Me9(thf)3(OEt2)2 (Ln = Y, Tb), when crystallized from diethyl
ether solutions.[24] Such a Ln(III)-size dependency on thermal
stability is commonly observed in rare-earth-metal alkyl
chemistry[25]
and
showcased
for
derivatives
Ln(CH2SiMe3)3(thf)x,[26] [Li(dme)3][LntBu4](vide infra),[27] and
solvent-free [LnMe3]n.[28] Since the cerium derivative
Li3Ce(CH3)6(tmeda)3 (1) was only mentioned briefly as an impure
product (δCH3 = –6.4 ppm),[22-23] we re-visited its synthesis
applying a slightly modified version of the Schumann protocol.
Accordingly, cerous methyl complex 1 could be synthesized at –
10 °C in good yield (83%), and was obtained in analytically pure,
single-crystalline form. The solid-state structure of complex 1
turned out to be isostructural to the derivatives of the considerably
smaller-sized erbium and holmium (Figure 2).[22]
1
The H NMR spectrum revealed the methyl signal at δ = –4.08
ppm, reflecting a significant paramagnetic shift induced by Ce(III)
(Supporting Information, Figure S4). As expected, the Ce–C
distance of 2.6795(19) Å in 1 is considerably longer than those in
the holmium (2.563(18) Å)[22b] and erbium congeners (2.57(2)
Å).[22c] For further comparison, the Ce–C distances in 6-
coordinate Ce(CH2Ph)3(thf)3 and Ce(AlMe4)3 fall in the range
Figure 3. Crystal structure of [Li(thf)4][Ce(t-Bu)4] (2). Atomic displacement
ellipsoids set at 30% probability. Hydrogen atoms omitted for clarity. Selected
interatomic distances [Å] and angles [°]: Ce1–C1 2.524(9), Ce1–C5 2.513(8),
Ce1–C9 2.50(2), Ce1–C13 2.54(2) ; C–Ce1–C (range) 105.9(10)–111.8(8).
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
LnCl3(thf)x + 6 n-BuLi
Synthesis and Solid-State Structures of n-Butyl Derivatives
Li3Ln(n-Bu)6(thf)4 and Li2Ln(n-Bu)5(tmeda)2. Having proven the
tamable thermal instability of organocerium complexes 1 and 2
we next tackled the feasibility of the respective n-butyl derivatives.
Initially, we attempted to isolate a crystalline cerium-containing
compound from reactions of cerium chloride (thf adduct) with
various amounts of n-butyllithium in tetrahydrofuran at low
temperatures (–40 °C). Not quite unexpectedly, these endeavors
proved to be unsuccessful in the first place. Having in mind the
presumably enhanced stability of derivatives of the smaller rare-
earth metals and to better follow the metathesis reactions via
NMR spectroscopy we quickly began to focus on lutetium. Indeed,
ate complex Li3Lu(n-Bu)6(thf)4 (3Lu) (Figure 4) could be isolated
from the reaction of LuCl3(thf)2 with 3.3 equivalents of n-BuLi in n-
hexane. However, in order to accomplish complex 3Lu six
equivalents of n-BuLi must have reacted and associated per
LuCl3(thf)2. Also notably is that four thf molecules have been
accommodated in the complex despite the presence of only two
in the lutetium chloride precursor. Therefore, THF appeared to be
the limiting factor for this reaction. Optimization of the reaction
conditions gave the so far best results when anhydrous rare-
earth-metal chlorides LnCl3(thf)x (Ln = Sc, Y, La, Ce and Lu;
covering the entire Ln(III) size range) were suspended in a
mixture of n-hexane and THF and cooled to –40 °C prior to the
addition of n-BuLi. Removal of the volatiles after 30 minutes under
reduced pressure, extracting the remaining solid with n-hexane,
and concentrating the obtained solution gave crystalline Li3Ln(n-
Bu)6(thf)4 (3Ln) (Scheme 1).
THF/n-hexane,
– 3 LiCl
-40 °C, 30 min
N
thf
thf
N
3 TMEDA
n-hexane
-40 °C, 30 min
Li
Li
Ln
Ln
– n-BuLi(tmeda)
Li
Li
N
Li
thf
thf
N
4Ln Ln = Ce, Lu
3Ln Ln = Sc, Y, La, Ce, Lu
Scheme 1. Synthesis of n-butyl complexes 3Ln and 4Ln
.
The Lu–C distances range from 2.44(2) to 2.58(3) Å, matching
those in Li3Lu(CH3)6(dme)3 (2.48(4)-2.57(4) Å),[23a] Lu(AlMe4)3
(2.455(2)-2.471(2) Å),[34] and Lu(GaMe4)3 (2.465(2)-2.493(2)
Å).[35] All complexes 3Ln were characterized by NMR spectroscopy
and elemental analysis. A 45Sc NMR experiment for 3Sc showed a
broad signal pattern with a main peak centered at +502 ppm,
clearly indicating σ–bonded alkyl species in solution (Figure
S17).[36-37] For comparison, the 45Sc chemical shift of
ScMe3(THF)x was detected at +601.7 ppm.[36] Similarly, the H-
89Y HSQC NMR spectrum of 3Y revealed one low-field shifted 89
1
Y
resonance at +771 ppm (Figure S22), in accordance with a single
organoyttrium species in solution (Y(CH2SiMe3)3(thf)2: δ(89Y) =
882.7 ppm).[38] To further corroborate a similar composition of
complexes 3Ln, we probed the stabilizing effect of TMEDA for the
metal centers cerium and lutetium. Much to our delight, addition
of stoichiometric amounts of three equivalents TMEDA prior to the
crystallization of the reaction mixtures containing complexes 3Ln
afforded complexes Li2Ln(n-Bu)5(tmeda)2 (4Ln) (Ln = Ce and Lu)
(Figure 4).
Despite our best efforts, we could obtain crystals suitable for XRD
analysis only for the lutetium derivative 3Lu. The crystalline
material formed for the other “larger” rare-earth-metal centers
displayed only very poor diffraction behavior. Complex 3Lu
features a lutetium center surrounded by six n-butyl ligands, which
show distinct linkages to the lithium atoms. Two hydrocarbyl
ligands are unsymmetrically bridged by one alkali metal, while the
two remaining lithium atoms are unsymmetrically linked to three
n-butyl ligands each. The coordination sphere of the three lithium
atoms is completed by four THF molecules.
Figure 4. Crystal structures of Li3Lu(n-Bu)6(thf)4 (3Lu, left) Li2Lu(n-Bu)5(tmeda)2 (4Lu, middle), and Li2Ce(n-Bu)5(tmeda)2 (4Ce, right), with atomic displacement
ellipsoids set at 30% probability. Hydrogen atoms, disorders and CH atoms of the THF molecules (3Lu), and disorders of the n-butyl groups (4Lu, 4Ce) are omitted
for clarity. Selected interatomic distances [Å] and angles [°] for 3Lu: Lu1–C1 2.553(17), Lu1–C5 2.559(18), Lu1–C9 2.46(2), Lu1–C13 2.550(19), Lu1–C17 2.44(2),
Lu1–C21 2.58(3), Li1–C1 2.1555(3), Li1–C13 2.3324(3), Li2–C5 2.1928(3), Li2–C17 2.4854(3), Li2–C21 2.4623(3), Li3–C5 2.1307(3), Li3–C9 2.5339(4), Li3–C21
2.3886(3), C1–Lu1–C21 176.68(1), C5–Lu1–C13 177,55(1), C9–Lu1–C17 176.32(1), C1–Lu1–C5 88.63(1). 4Lu: Lu1–C7 2.468(2), Lu1–C11 2.5293(18), Lu1–C15
2.3797(19), Lu1–C19 2.4620(18), Lu1–C23 2.522(2), Li1–C7 2.217(4), Li1–C11 2.195(4), Li2–C19 2.224(4), Li2–C23 2.191(4), C7–Li1–C11 108.74(1), C19–Li2–
C23 105.61(1). 4Ce: Ce1–C7 2.657(3), Ce1–C11 2.700(5), Ce1–C15 2.674(3), Ce1–C25 2.664(4), Ce1–C29 2.549(3), Li1–C7 2.185(6), Li1–C11 2.222(8), Li2–C15
2.183(6), Li2–C25 2.205(6), C7–Li1–C11 112.1(3), C15–Li2–C25 114.3(3).
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
Surprisingly, besides the expected thf/tmeda donor ligand
exchange, displacement of one n-BuLi(tmeda) entity took place.
Consequently, the rare-earth-metal centers in isostructural
complexes 4Ce and 4Lu feature only five n-butyl ligands. One of
these n-butyl ligands is terminal while of the remaining ones two
are bridged by a lithium atom each, which are stabilized by one
tmeda donor each. Such n-BuLi(tmeda) displacement is in
contrast to the Schumann methyl variants Li3Ln(CH3)6(tmeda)3
including complex 1. A plausible explanation for this is the
increased steric bulk/basicity of the n-butyl versus methyl ligands.
Complexes 4Ce and 4Lu exhibit distinct Ln–C distances for the
terminal and bridging n-butyl ligands (Ce: 2.549(3) Å and
2.657(3)-2.700(5) Å; Lu: 2.3797(19) Å and 2.4620(18)-2.5293(18)
Å). Overall, only very limited structural data are available on rare-
earth-metal n-butyl complexes due to the propensity for β-H
elimination. Cyclopentadienyl-supported derivatives include
temperature, decomposition perceived visually within one hour (in
both solution and solid state). To further determine the thermal
stability of our complexes, variable temperature (VT) NMR
spectra were measured (Figures S18, S23, S26, S36 and S42).
Amazingly, β-H elimination and 1-butene formation was observed
only from +30 °C onwards, being considerably more pronounced
for 3La and 3Ce than for the respective complexes of the smaller
rare-earth metals. Because the heating was performed in 10
degree increments starting at – 40 °C and held at each
temperature for 15 minutes before measurement, it can be
concluded that these complexes are relatively stable for a short
amount of time even at ambient temperature.[33]
Crucially, both the 1H and 7Li NMR spectra of complexes 3Ln and
4Ln revealed that in solution a considerable portion of n-BuLi gets
displaced from the rare-earth-metal center. The degree of
dissociation is highly dependent on the solvent and the rare-earth
metal. In general, n-BuLi dissociation is more pronounced in THF
than in toluene. Ate complexes 3Ln of the smaller-sized rare-earth
metals yttrium, lutetium, and scandium are quite stable in toluene
solution displaying minor n-BuLi dissociation of ca. 1%, 4%, and
8% respectively (Figures S14/S19/S37). On the other hand, n-
BuLi dissociation prevails for 3Ce and 3La(ca.90%). Interestingly,
the presence of tmeda as a donor ligand in complexes 4Ln, can
constrained-geometry complexes [(η5:η1-C5Me4SiMe2NtBu)Y(µ-
1
n-Bu)]2 (Y–C: 2.542(2)/2.544(2) Å) and terminal (η 5: η
-
C5Me4SiMe2NtBu)Y(n-Bu)(dme) (Y–C: 2.435(5) Å)[39] as well as
metallocenes [(C5H4Me)2Ln(n-Bu)]2 (Ln Y: 2.551(8)/
=
2.556(11)/2.587(13) Å, Dy: 2.536(18)/2.591(18) Å).[40] Because
cerium and lutetium form the same type of complex for 4Ln, and
based on other analytical data, monolanthanide derivatives of
general formula Li3Lu(n-Bu)6(thf)4 are also proposed for the
remaining complexes 3Ln.
At this point we were once more challenged by the question why
these reactions would lead to the isolation of ate complexes
Li3Ln(n-Bu)6(thf)4 rather than the envisaged “Ln(n-Bu)xCl3-x(thf)y”
(x = 1-3). When starting from LnCl3(thf)x, which is very poorly
soluble in THF, the simplest explanation would be, that treatment
of Lewis-acidic rare-earth-metal chloride species with increasing
amounts of the strong nucleophile n-BuLi enhances its solubility,
and is therefore more likely to react faster in consecutive reactions
with n-BuLi. The formation of isolable ate complexes Li3Ln(n-
Bu)6(thf)4 is further driven by the relatively small size of the n-butyl
ligand and by switching the solvent from coordinating (THF) to
non-coordinating (n-hexane). Predominant ate complexation was
either counteract or enforce n-BuLi separation (3Ce/4Ce
:
>98%/46% versus 3Lu/4Lu: 4%/20%). For the organocerium(III)
complexes, n-BuLi ate complexation is easily detectable by
paramagnetically shifted 7Li resonances. Figure 5 depicts the 7Li
NMR spectra of complexes 1, 2, 3Ce, and 4Ce both in THF-d8 and
toluene-d8, clearly revealing (a) the great stability of the
hexamethylate complex 1, (b) the persistence of ion-separated
tert-butyl complex 2 also in solution, and (c) the beneficial effect
of tmeda (versus thf) donor ligands for intramolecular ate-complex
stabilization. The 7Li NMR spectrum of 4Ce suggests a clean
separation into [LiCe(n-Bu)4(tmeda)] and n-BuLi (signal ratio 1:1).
also
observed
for
or
Schumann´s
methyl
complexes
derivatives
Li3LnMe6(thf)x[22]
diisopropylamido
LiLn(NiPr2)4(thf)x.[41] Crucially, independent of the applied
LuCl3(thf)x/n-BuLi stoichiometry, complex 3Lu could be crystallized
as the exclusive Lu(III)-containing species, upon separation of the
solution from unreacted rare-earth-metal halide, its evaporation to
dryness, and extraction of the residue with n-hexane (XRD unit-
cell check and 1H NMR spectroscopy indicated repeatedly
formation of 3Lu). This finding suggested that the typical reaction
(preformation) conditions for a lutetium-derived Imamoto reagent,
LuCl3/n-BuLi/THF/–78 °C/30 min should form Li3Lu(n-Bu)6(thf)x as
the dominant initial lutetium species in solution. Since any
persisting equilibria in solution would impact the reactivity of the
Imamoto alkylation reagent we next took a closer look at the
solution behavior of bimetallic Li3Ln(n-Bu)6(thf)4 and the binary
system LnCl3(thf)x/n-BuLi (Ln = Ce, Lu).
Solution Behavior of n-Butyl Complexes 3Ln and 4Ln Probed
by NMR spectroscopy. All crystallized n-butyl complexes are
stable, when stored as solids at –40 °C. However, when dissolved
in any solvent, complexes 3La, 3Ce and 4Ce had fully decomposed
after 24 h. In contrast, but not unexpectedly, the complexes of the
smaller rare-earth metals are stable in solution at –40 °C for up to
one week. Warming complexes 3Ln and 4Ln to ambient
Figure 5. 7Li NMR spectra (194.37 MHz, 233 K) of complexes 1, 2, 3Ce, and 4Ce
recorded in THF-d8 or toluene-d8.
,
These findings have important implications for the composition of
the “active” n-BuLi-derived Imamoto reagent. When used as a 1:1
mixture of CeCl3(thf) and n-BuLi in THF at temperatures of –35 °C,
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
the formation of ate complex Li3Ce(n-Bu)6(thf)4 (3Ce) seems highly
unfavored. This can be concluded from the 7Li NMR spectrum of
3Ce, which does indicate only a small portion of lithium and
paramagnetic cerium(III) in close proximity (δLi = 81.5 ppm). An
equimolar mixture of CeCl3(thf) and n-BuLi formed in situ in THF-
d8 at –45 °C did not reveal any paramagnetically shifted 7Li NMR
resonance (Figure S33). The latter seem to appear only at ratios
conditions). As in case of n-BuLi, the two distinct signals for the
lutetium-bounded n-Bu ligands might represent lutetium
complexes of distinct aggregation “LixLu(n-Bu)3+x(thf)y”(x = 0-3).
7
The dissociation of n-BuLi in 3Lu is also corroborated by the Li
NMR spectrum (trace III/right) showing the characteristic pattern
of n-BuLi[47-48] and a very broad signal at 0 ppm, indicative of Lu–
n-Bu–Li moieties and rapid n-BuLi exchange. Interestingly, when
examining in situ formed solutions of LuCl3(thf)2/n-BuLi (traces IV-
VII), free n-BuLi was observed only when more than three
equivalents of n-BuLi were used per lutetium. This is supported
by the respective 7Li NMR spectra, which also suggests the
formation of free LiCl in case of <3 equivalents of n-BuLi. For
better comparison the 1H and 7Li spectra of n-BuLi–LiCl mixtures
are shown (trace I). Moreover, addition of LuCl3(thf)2 to ate
complex Li3Lu(n-Bu)6(thf)4 (3Lu) results in complete consumption
of free n-BuLi and formation of LiCl (trace VIII). This implies that
dissociated n-BuLi engages in “normal” ligand exchange with
added LuCl3(thf)2. However, when the solvent is removed the in-
situ formed complexes scramble to the complex 3Lu, LuCl3 and
LiCl (upon crystallization from n-hexane).
Based on these observations, it can be hypothesized about the
following scenario. The initial intermediate/transient product of the
LuCl3(thf)2/x n-BuLi reaction is certainly “Ln(n-Bu)Cl2”. However,
as soon as this heteroleptic complex is formed, its better solubility
in THF (compared to LnCl3) will imply a more rapid reaction
(compared to LnCl3) with the remaining n-BuLi in solution. Thus,
ate complexes of the type “LixLu(n-Bu)3+x(thf)y”, and ultimately but
not exclusively Li3Ln(n-Bu)6(thf)4 (3Ln), represent the dominant
rare-earth-metal species in solution. As revealed by NMR
spectroscopies, 3Lu is labile in solution and engages in a
dissociation equilibrium with n-BuLi; the displaced n-BuLi should
react further with LuCl3. Therefore, likely reaction products
depending on the LuCl3(thf)2/n-BuLi ratio are “Lu(n-Bu)3” and
“LixLu(n-Bu)3+x(thf)y” with x≤2 for a ratio of 1:3 and smaller, and
species “Ln(n-Bu)3” and “LixLu(n-Bu)3+x(thf)y” with x ranging from
1 to 3, for a ratio of larger than 1:3. A shortage of THF solvent via
extraction of the reaction products into n-hexane leads to ate
complex Li3Lu(n-Bu)6(thf)4 (3Lu) as the only isolable (crystalline)
species. Performing the reactions for longer time periods under
<
1:3. For comparison, the intramolecular ate complex
LiCe[N(SiHMe2)2]4(thf) (in C6D6/1,2-difluorobenzene)[42] and
tmeda-adduct 4Ce (in THF-d8) display Li chemical shifts of 84.4
and 81.3 ppm. The H NMR spectrum in THF-d8 of 3La featuring
7
1
the similarly sized lanthanum center lends further support to this
assumption, as the n-butyl resonances appear in an approximate
1:1 ratio (Figure S24). Moreover, 1:1 mixtures of CeCl3(thf) and
n-BuLi, obtained in THF at –45 °C contain a substantial amount
(40-50%) of unreacted CeCl3(thf) upon preformation for 30
minutes. Incomplete transmetalation in CeCl3/RLi mixtures has
been pointed out previously at various occasions.[12a,21] Therefore,
unlike the smaller rare-earth metals, which favor intramolecular
ate complexation even in solution, the organocerium species
prevailing under Imamoto conditions are most likely “Ce(n-
Bu)3(thf)x” or solvent-separated ion pairs like “[Ce(n-
Bu)4(thf)y][Li(thf)4]”, similar to tert-butyl complex 2. The formation
of heteroleptic species “Ce(n-Bu)xCly(thf)z” (x = 1,2; x+y = 3)
seems very unlikely due to the persistence of ligand redistribution
forming homoleptic complexes and/or the favorable occurrence of
β-H elimination. Non-ate mixed hydrocarbyl/halide Ln(III)
complexes have been structurally authenticated for phenyl and
benzyl derivatives but have remained elusive for alkyl ligands
capable of β-H elimination. Representative examples include
(C6H5)GdCl2(THF)4,[43] Ln(CH2Ph)2I(thf)3 (Ln = Y, Er),[44] and ion-
separated [YMeI(py)5][I].[24]
The dissociation behavior was further investigated in a series of
NMR experiments (Figure 6), comparing complex 3Lu to the
reactions of LuCl3(thf)2 with various amounts of n-BuLi, and n-
BuLi itself. The 1H NMR spectrum of crystalline ate complex
Li3Lu(n-Bu)6(thf)4 (3Lu) shows two signal sets for metal-bonded
CH2 groups (Figure 6, trace III/left). The Lu–CH2 moieties
resonate at about –0.5 ppm, and hence are significantly shifted to
lower field compared to the characteristic pattern of n-BuLi at –
1.0 to –1.5 ppm (cf., trace II/left).[45-46] The two signal sets are
clearly indicative of n-BuLi dissociation in THF solution (Imamoto
otherwise
identical conditions, resulted
in
extensive
decomposition.
Figure 6. 1H NMR (500.13 MHz, THF-d8, 193 K) (left) and 7Li NMR spectra (194.37 MHz, THF-d8, 193 K) (right) of LuCl3(thf)2 with x n-BuLi (x = 1, 3, 6, 12) compared
to 3Lu, 3Lu+2 LuCl3(thf)2, n-BuLi, and n-BuLi+LiCl.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
Solution Behavior of n-Butyl Complexes 3Ln and 4Ln Probed
by Derivatization Reactions. The preferred dissociation of n-
BuLi from ate complexes Li3Ln(n-Bu)6(thf)4 (3Ln) of the larger rare-
earth metals was further revealed by the reaction of crystalline 3Ce
with LuCl3(thf)2 in THF, affording 3Lu in low crystalline yields of
23% (XRD unit-cell check). Interestingly, the dissociated n-BuLi
seems to exert a stabilizing effect on the organocerium species,
since the solution turned brown rather quickly, upon addition of
LuCl3(thf)2. In order to enforce the formation of n-BuLi-depleted
“Ln(n-Bu)3” we searched for reactions, which would possibly
convert any dissociated n-BuLi selectively and ideally into
products not affecting the isolation of putative “Ln(n-Bu)3”. Luckily,
such a reaction path could be observed for the treatment of 4Lu
with trimethylsilyl chloride. The reaction occurred to be slow at –
40 °C, but it produced a minor amount of crystalline LiLu(n-
Bu)3Cl(tmeda)2 (5) (Scheme 2, Figure 7). On various other
occasions, when trying to precipitate LiCl or extract “Ln(n-Bu)3”
from the LiCl-containing residue, only progressive decomposition
could be observed. Again, depending on the size of the rare-
earth-metal center, complete decomposition took place in a few
minutes (“Ce(n-Bu)3”) or several hours (“Y(n-Bu)3”). This clearly
shows the stabilizing effect of LiCl in these reactions making THF
an ideal solvent.
lithium to lutetium. Striking are the distinct Lu–C distances of the
terminal n-butyl ligands of 2.367(3) and 2.421(3) Å. While the
shorter distance approximately matches the terminal one of
complex 3Lu (2.3797(19) Å), the longer distance is almost as long
as that of the bridging n-Bu ligand (2.447(3) Å). This might
indicate a trans influence of the weakly coordinated chlorido
ligand. In accordance with the crystal structure, the 1H NMR
spectrum of 5 display two distinct signal sets for the n-butyl
ligands in a 2:1 ratio (Figure S48).
Unsurprisingly, the n-butyl ligands of complexes 3Ln get easily
protonated in the presence of alcoholic substrates. As an example,
treatment of Li3Ce(n-Bu)6(thf)4 (3Ce) with six equiv. of neopentanol
resulted in the crystallization of the heterobimetallic cluster
Li3Ce2(OCH2tBu)9(HOCH2tBu)2(thf) (6, Scheme 3, Figure S1). On
the basis of the crystal structure and Ce–O distances, the
connectivity of 6 can be assigned as Li3Ce2(µ3-OCH2tBu)3(µ2-
OCH2tBu)4(OCH2tBu)2(HOCH2tBu)2(thf).
The
µ2-bridging
neopentoxy ligands involve one lithium and cerium each, the µ3-
bridging ones connect two cerium atoms with one lithium, and the
terminal ones are coordinated to one cerium center (Ce1). The
coordination sphere of the lithium atoms is saturated with one thf
and two alcohol donor molecules.
N
N
ClSiMe3
n-hexane
-40 °C, 4 h
N
Li
N
N
Cl
Lu
Li
Lu
– LiCl
– (n-Bu)SiMe3
N
Li
N
N
4Lu
5
Scheme 3. Alcoholysis of cerium complex 3C with neopentanol.
Scheme 2. Reactivity of lutetium complex 4Lu with ClSiMe3: “lithium depletion”.
The alcohol donors of 6 engage in hydrogen bonding with one
terminal (O3) and one µ 2-bridging neopentoxy ligand (O10). Due
to the distinct coordination modes, the Ce–O distances span a
wide range of 2.196(2) to 2.608(2) Å, but match those of other
cerous alkoxides.[49] Overall, the solid-state structure of 6 features
a completely asymmetric complex with all five metal centers
displaying different coordination environments. A similar structure
was reported for the yttrium neopentoxide Li3Y2(µ3-OCH2tBu)(µ3-
HOCH2tBu)(µ2-OCH2tBu)5(OCH2tBu)3(HOCH2tBu)2.[50]
The
formation and structural characterization of Li3Ce2 complex 6
clearly reflects lithium depletion compared to the Li3Ce precursor
3Ce, and pictures the intricacy of 3Ce in solution. Complex 6 shows
intricate solution behavior itself as evidenced by at least six
signals in the 7Li NMR spectrum, including paramagnetically
shifted ones (Figure S52).
Figure 7. Crystal structure of LiLu(n-Bu)3Cl(tmeda)2 (5) with atomic
displacement ellipsoids set at 50% probability. Hydrogen atoms, and disorders
of the n-butyl groups are omitted for clarity. Selected interatomic distances [Å]
and angles [°] for 5: Lu1–C1 2.421(3), Lu1–C5 2.367(3), Lu1–C18 2.447(3),
Lu1–Cl1 2.7191(9), Lu1–N1 2.635(3), Lu1–N2 2.540(2), Li1–C18 2.287(6), Li1–
Cl1 2.308(5), C18–Lu1–Cl1 89.26(9), C18–Li1–Cl1 104.6(2).
Reactivity of n-Butyl Complexes 3Ln toward An Enolizable
Ketone. In this study, 1,3-diphenylpropan-2-one was employed
as a test molecule (Table 1), since its selective conversion into
the respective tertiary alcohol has been already shown in the
original work by Imamoto.[7] Back then the following protocols
were applied: [Ce, I2, THF, 0 °C (CeI3 formed in situ)/n-BuLi (1
equiv.), –65 °C, 30 min/ketone, –65 °C, 3 h] and [“CeCl3“ (later
determined as CeCl3(H2O)), n-BuLi (1 equiv.), THF, –78 °C, 1
h/ketone, –65 °C, 3 h], yielding the alcohol in 98% and 96%,
respectively.
In the solid state, the lutetium atom of complex 5 adopts a
distorted octahedral coordination geometry, involving three n-
butyl ligands, one chlorido ligand and a chelating tmeda molecule.
The chlorido and one n-butyl ligand bridge to the lithium atom
forming a four-membered ring. The coordination of the lithium
atom is completed by the second tmeda molecule. Noteworthy, in
the course of this reaction one tmeda ligand was transferred from
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 1. Overview of the Ketone Reduction
Ph
OH
1) cerium compound
O
O
2) additive(s)
+
Ph
Ph
Ph
Ph
Ph
solvent, –40 °C, 30 min
entrya
compound
equiv. ketone
solvent
additives
yield alcohol [%]
recovered ketone [%]
1
2
CeCl3(thf)
CeCl3(thf)
3Ce
0.77
6
THF
THF
THF
THF
Et2O
Et2O
Et2O
toluene
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
1 n-BuLi
99
77
70
88
54b
78
89
76
74
89
80
90
62b
79
73
74
44
75
62
50
0
20
13
3
46b
15
10
11
16
8
6 n-BuLi
3
6
-
4
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Ce
3Lu
3Lu
3Y
3Lu
1
-
5
6
-
6
3
-
7
1
-
-
8
6
9
6
3 LiCl
5 CeCl3(thf)
10
11
12
13
14
15
16
17
18
19c
20
6
6
5 “Ce turbo chloride” to ketone
14
1
6
5 “Ce turbo chloride” after ketone
6
5 Sc(OTf)3 after ketone
38b
21
16
23
56
12
38
50
6
5 AlCl3 after ketone
6
3 TMEDA
6
-
6
3 TMEDA
6
-
6
-
-
1
1 n-BuLi
a –40 °C. b Ratios determined by NMR spectroscopy. c Ambient temperature.
In our hands, the slightly modified version [CeCl3(thf), n-BuLi (1
equivalent), THF, –35 °C, 30 min/ketone, –35 °C, 30 min] gave
99% of alcohol (entry 1), clearly documenting the efficiency of the
Imamoto transformation. The importance of using CeCl3/n-BuLi in
a 1:1 stoichiometry is revealed by the reactivity of crystalline or in
situ formed complex 3Ce (entries 2 and 3). The yield of alcohol was
significantly decreased at the expense of ketone enolization
caused by dissociated n-BuLi (for the exclusive reaction behavior
of n-BuLi, see entry 20). Conducting the reaction in diethyl ether
gave also the best results when CeCl3(thf)/n-BuLi was used in a
1:1 stoichiometry (entries 5-7). Surprisingly, complex 3Ce
performed similarly in THF and toluene (entry 8).
To comprehend why the alcohol yield is lower in case of 3Ce than
for CeCl3(thf)/n-BuLi (ratio 1:1) (entries 3 and 4 versus 1), various
additives were tested. While the addition of LiCl did not affect the
reaction outcome (entry 9), additional CeCl3(thf) did markedly
increase the alcohol yield (entry 10). This effect was even
significantly enhanced, if “cerium turbo chloride” was added to the
organocerium compound subsequently to the ketone (entry 12.
“Cerium turbo chloride” is the combination of CeCl3 with two
equivalents of LiCl and completely soluble in THF. It was prepared
according to the method reported by the Knochel group[51] As
revealed by NMR spectroscopy in case of lutetium (Figure 6, trace
VIII), dissociated n-BuLi will react with added LnCl3 along with
ligand scrambling and provide for the further supply of cerium-
bonded n-butyl. Therefore the limited performance of complex 3Ce
with 6 equivalents of ketone results from progressing n-BuLi
dissociation and its changed (reduced/non-selective) reactivity
toward the ketone. For ate complexes 3Ln of the smaller-sized
yttrium and lutetium, which are more stable in solution, the
mediocre performance in the 6-equivalent reaction might be
attributable to the formation of sterically demanding alkoxy
ligands in putative “Li3Ln[OC(CH2Ph)2n-Bu]6”. Overall, any effect
of the rare-earth-metal size is not apparent since ate complexes
3Y and 3Lu showed a performance very similar to that of the cerium
congener 3Ce (entries 16 and 18). The importance of performing
the transformation in the presence of rare-earth metals, was also
revealed when employing scandium triflate Sc(OTf)3 or AlCl3 as
Lewis acids, instead of additional CeCl3 (entry 13 and 14). Both
reactions led to decreased alcohol yields and increased ketone
recovery (via enolization). This and the distinct outcome of the
ketone transformation when changing the order of ketone and
“cerium turbo chloride” (entries 11 and 12) suggest an
intramolecular ketone activation/alkylation scenario akin to a four-
membered transition state (1,2-addition), rather than the
participation of several metal complexes (Figure 8/A and B).
Moreover, favorable ketone coordination/adduct formation[52-53]
and alkoxide formation have been shown previously.[54]
Mn+
L
Ce3+
O
C
O
R
R
R
C
R
Ce3+
A
B (Mn+–L = Lewis-acidic
metal–ligand fragment)
Figure 8. Proposed “intramolecular” reaction involving
a four-membered
transition state of the ketone reduction (A) rather than a multimolecular reaction
path like B.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
The impact of TMEDA was examined for in situ formed 4Ce and
4Lu (entries 15 and 17), indicating a pronounced n-BuLi(tmeda)
dissociation only in case of the lutetium reaction. Finally,
performing the ketone transformation with 3Lu at ambient
temperature resulted not unexpectedly in a significant drop of
alcohol formation, but also documents the relative Ln(III)-size-
dependent stability of the complexes at ambient temperature for
a short time.
Finally, the isolation and crystallization of lithium alkoxide co-
products gave further insights into the Imamoto alkylation
scenario. Prolonging the reaction time of the “incomplete”
transformation of six equivalents of 1,3-diphenylpropan-2-one
with 3Ce (Table 1, entry 3) to three weeks, led to the crystallization
of the enolization product lithium 1,3-diphenylprop-1-en-2-ate
Li4[OC(=CHPh)CH2Ph]4(thf)4 (7) (Figure S2). The dissociated,
significantly more stable n-BuLi (compared to 3Ce) acts as a base,
deprotonating unreacted ketone at the α-position of the carbonyl
moiety. This competitive reaction path reflects the main part of the
recovered ketone, as listed in Table 1, since aqueous work up will
involve keto-enol tautomerism. Complex 7 features a common
structural motif in lithium alkoxide complexes[55-56] with the lithium
and alkoxy oxygen atoms occupying alternating positions of a
cube. The lithium atoms are saturated with one THF molecule
each.
Unfortunately, single-crystalline products were not accessible.
The formation of such species is further corroborated by the
solution behavior of pre-isolated crystalline ate complex Li3Ce(n-
Bu)6(thf)4 (3Ce), displaying extensive n-BuLi dissociation in THF
solution. Note that the product obtained from the reaction of
CeCl3•2LiCl (cerium turbo chloride) with three equivalents of n-
BuLi in THF at –30 °C was previously analyzed as “n-Bu3Ce” by
Raman spectroscopy (absence of significant CeCl3 and n-BuLi
Raman lines).[15] Activation of the carbonylic substrate and its
transformation (here reduction to the alkoxy moiety) takes place
at the same cerium center, and can proceed three times. Such
1,2-addition reactions have been proven for yttrium methyls
[YMe3]n,[57] [YMe2(thf)5][BPh4], and [YMe(thf)6][BPh4]2,[24] as well
as neosilyls [Y(CH2SiMe3)2(thf)4][A] (A = BPh4, Al(CH2SiMe3)4)
[58]
and [Y(CH2SiMe3)(thf)5][BPh4]2
employing fluorenone and
benzophenone. Aqueous work-up will lead to the alcoholic
product and a recyclable inorganic cerium compound.
Conclusion
Seamless low-temperature synthesis and crystallization
techniques give access to isolable organocerium complexes
Li3Ce(CH3)6(tmeda)3, [Li(thf)4][Ce(t-Bu]4], Li3Ce(n-Bu)6(thf)4, and
Li2Ce(n-Bu)5(tmeda)2. 1H/7Li NMR spectroscopic studies on in situ
formed solutions of equimolar mixtures of CeCl3(thf) and n-BuLi
in THF at –35 °C suggest non-isolable “Ce(n-Bu)3(thf)x” or
solvent-separated ion pairs like “[Li(thf)4][Ce(n-Bu)4(thf)y]” as
effective organocerium species in respective Imamoto-type
alkylation reactions of carbonylic substrates. This hypothesis is
corroborated by the solution behavior of ate complex Li3Ce(n-
Bu)6(thf)4, which displays extensive displacement of n-BuLi. As
revealed for the benchmark substrate molecule 1,3-
diphenylpropan-2-one, the prevailing dissociation of n-BuLi
results in decreased regioselectivity of the ketone/alcohol
transformation. In contrast, ate complexes of the type
Li3Ce(CH3)6(thf)x could be likely species in methyllithium derived
Imamoto-type reagents as revealed by the high stability of
Li3Ce(CH3)6(tmeda)3 in THF solution. Not unsurprisingly, the
effective composition of such CeCl3/RLi reagents is highly
dependent on the rare-earth metal, the hydrocarbyl group, and
the solvent. Finally, the presence of THF-soluble co-product LiCl
in in situ formed reagents is proposed to adopt an active role by
exerting a stabilizing effect on the organocerium species.
Compound LiLu(n-Bu)3Cl(tmeda)2 features a structural snapshot
of the likely involvement of LiCl in the soluble part of Imamoto-
type transformations.
Preliminary tests with CeCl3(thf)/n-BuLi mixtures in the molar ratio
1:3 and acetone as the substrate led to the crystallization of the
lithium alkoxide Li8[OCMe2(n-Bu)]6Cl2(thf)6 (8) (Figure S3). The
incorporation of LiCl into the cluster core unambiguously
documents its integral part of the reagent solution. This is also
revealed by the 7Li NMR spectra depicted in Figure 6 (right, traces
VI-VIII). Dissolved LiCl might also associate with the
organocerium species (as detected for complex 5) thus exerting
a stabilizing effect.
Proposed Formation and Reactivity of the Imamoto
Organocerium Reagent (Scheme 4). Equimolar amounts of
CeCl3 and n-BuLi, when combined in THF at low temperatures (–
40 °C) afford the cerous n-butyl complex “Ce(n-Bu)3(thf)x” or
solvent-separated ion pairs like “[Ce(n-Bu)4(thf)y][Li(thf)4]”, as
suggested by (a) 1H and 7Li NMR spectroscopies and (b)
considerable amounts of unreacted cerium chloride. In particular,
7Li NMR spectroscopy indicates complete transmetalation
(complete consumption of n-BuLi) and the absence of adjacent
cerium and lithium centers.
Li3Ce(n-Bu)6(thf)4 (3Ce
)
THF
- n-BuLi
n-hexane
+ n-BuLi
O=CR2
Fact is that a large number of organometallics-promoted organic
transformations (including the in situ formation of the multi-
component organometallic reagent) is routinely performed at low
temperatures (–78 °C ® 0 °C). It can be safely assumed that the
application of modern cryotechniques will continue to promote a
better understanding of the reagent´s formation, composition, and
effectiveness, ultimately leading to both optimized reagents and
conditions for substrate conversion.
(
enolization)
THF
–40 °C
30 min
3 CeCl3(thf)
+
3 n-BuLi
“Ce(n-Bu)3(thf)x“ [+ 2 CeCl3(thf), 3 LiCl]
or “[Li(thf)4][Ce(n-Bu)4(thf)x’]“
O=CR2
THF
–40 °C, 30 min
“Ce[OCR2(n-Bu)](n-Bu)2(thf)y“
aqueous
NH4Cl or
Na2CO3
Acknowledgements
“Ce[OCR2(n-Bu)]3(thf)z“
(
HOCR2(n-Bu))
We thank Dr. Klaus Eichele for his help with the NMR
measurements.
Scheme 4. The Imamoto organocerium reagent in action.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
C. Alcaraz, U. Groth, Angew. Chem. Int. Ed. Engl. 1997, 36, 2480-2482;
Angew. Chem. 1997, 109, 2590-2592.
Conflict of interest
[12] a) A. Shakoor, K. Jacob, K.-H. Thiele, Z. Anorg. Allg. Chem. 1985, 521,
57-60; b) K. Jacob, K.-H. Thiele, Z. Anorg. Allg. Chem. 1986, 536, 147-
152; c) W. J. Evans, J. D. Feldman, J. W. Ziller, J. Am. Chem. Soc. 1996,
118, 4581-4584; d) W. J. Evans, M. A. Ansari, J. D. Feldman, R. J.
Doedens, J. W. Ziller, J. Organomet. Chem. 1997, 545-546, 157-162; e)
D. A. Conlon, D. Kumke, C. Moeder, M. Hardiman, G. Hutson, L. Sailer,
Adv. Synth. Catal. 2004, 346, 1307-1315; f) J. E. Kim, D. S. Weinberger,
P. J. Carroll, E. J. Schelter, Organometallics 2014, 33, 5948-5951; g) J.
E. Kim, A. V. Zabula, P. J. Carroll, E. J. Schelter, Organometallics 2016,
35, 2086-2091.
The authors declare no conflict of interest.
Keywords: Carbonyl Alkylation • Cerium • n-Butyl • Lithium • 7Li
NMR spectroscopy
[1]
a) H. B. Kagan, J. L. Namy, Tetrahedron 1986, 42, 6573-6614; b) B. M.
Trost, I. Fleming, T. Imamoto in Comprehensive organic synthesis, Eds.,
Pergamon, Oxford, 1991, Chapter 1.8; c) G. A. Molander, Chem. Rev.
1992, 92, 29-68; d) T. Imamoto, Lanthanides in organic synthesis,
Academic Press, London, 1994.
[13] a) R. E. Mulvey, Organometallics 2006, 25, 1060-1075; b) R. E. Mulvey,
F. Mongin, M. Uchiyama, Y. Kondo, Angew. Chem. Int. Ed. 2007, 46,
3802-3824; Angew. Chem. 2007, 119, 3876-3899; c) R. E. Mulvey, Acc.
Chem. Res. 2009, 42, 743-755; d) T. X. Gentner, R. E. Mulvey, Angew.
Chem. Int. Ed. 2021, 60, 9247-9262; Angew. Chem. 2021, 133, 9247-
9262.
[2]
a) H. Tse-Lok, Synthesis 1973, 1973, 347-354; b) V. Nair, J. Mathew, J.
Prabhakaran, Chem. Soc. Rev. 1997, 26, 127-132; c) A. K. Das, Coord.
Chem. Rev. 2001, 213, 307-325; d) V. Nair, L. Balagopal, R. Rajan, J.
Mathew, Acc. Chem. Res. 2004, 37, 21-30; e) V. Nair, A. Deepthi,
Tetrahedron 2009, 65, 10745-10755; f) V. Sridharan, J. C. Menéndez,
Chem. Rev. 2010, 110, 3805-3849.
[14] A. Music, D. Didier, Synlett 2019, 30, 1843-1849.
[15] A. Music, C. Hoarau, N. Hilgert, F. Zischka, D. Didier, Angew. Chem. Int.
Ed. 2019, 58, 1188-1192; Angew. Chem. 2019, 131, 1200-1204.
[16] A. D. Benischke, L. Anthore-Dalion, G. Berionni, P. Knochel, Angew.
Chem. Int. Ed. 2017, 56, 16390-16394; Angew. Chem. 2017, 129,
16608-16612.
[3]
a) J. L. Namy, P. Grirad, H. B. Kagan, Nou. J. Chim. 1977, 1, 5-7; b) H.
B. Kagan, J. Alloys Compd. 2006, 408-412, 421-426; c) K. C. Nicolaou,
S. P. Ellery, J. S. Chen, Angew. Chem. Int. Ed. 2009, 48, 7140-7165;
Angew. Chem. 2009, 121, 7276-7301; d) M. Szostak, D. J. Procter,
Angew. Chem. Int. Ed. 2012, 51, 9238-9256; Angew. Chem. 2012, 124,
9372-9390; e) M. Szostak, N. J. Fazakerley, D. Parmar, D. J. Procter,
Chem. Rev. 2014, 114, 5959-6039; f) X. Just-Baringo, D. J. Procter, Acc.
Chem. Res. 2015, 48, 1263-1275; g) Á. Péter, D. J. Procter, CHIMIA
2020, 74, 18-22; h) R. S. Miller, J. M. Sealy, M. Shabangi, M. L. Kuhlman,
J. R. Fuchs, R. A. Flowers, J. Am. Chem. Soc. 2000, 122, 7718-7722; i)
E. Prasad, R. A. Flowers, J. Am. Chem. Soc. 2002, 124, 6895-6899; j) P.
R. Chopade, E. Prasad, R. A. Flowers, J. Am. Chem. Soc. 2004, 126,
44-45; k) J. A. Teprovich Jr., M. N. Balili, T. Pintauer, R. A. Flowers II,
Angew. Chem. Int. Ed. 2007, 46, 8160-8163; Angew. Chem. 2007, 119,
8308-8311; l) R. A. Flowers Ii, Synlett 2008, 2008, 1427-1439; m) K. A.
Choquette, D. V. Sadasivam, R. A. Flowers, J. Am. Chem. Soc. 2010,
132, 17396-17398; n) D. V. Sadasivam, J. A. Teprovich, D. J. Procter, R.
A. Flowers, Org. Lett. 2010, 12, 4140-4143; o) T. V. Chciuk, W. R.
Anderson, R. A. Flowers, J. Am. Chem. Soc. 2016, 138, 8738-8741; p)
S. Maity, R. A. Flowers, J. Am. Chem. Soc. 2019, 141, 3207-3216.
a) T. Imamoto, Pure Appl. Chem. 1990, 62, 747-752; b) T. Imamoto,
Organocerium Reagents, Vol. 1, Chiba University, Japan, 1992; c) H.-J.
Liu, K.-S. Shia, X. Shang, B.-Y. Zhu, Tetrahedron 1999, 55, 3803-3830;
d) G. Bartoli, E. Marcantoni, M. Marcolini, L. Sambri, Chem. Rev. 2010,
110, 6104-6143; e) T.-L. Ho, in Fieser and Fieser's Reagents for Organic
Synthesis, 2017.
[17] L. Anthore-Dalion, A. D. Benischke, B. Wei, G. Berionni, P. Knochel,
Angew. Chem. Int. Ed. 2019, 58, 4046-4050; Angew. Chem. 2019, 131,
4086-4090.
[18] B. Wei, D. Zhang, Y.-H. Chen, A. Lei, P. Knochel, Angew. Chem. Int. Ed.
2019, 58, 15631-15635; Angew. Chem. 2019, 131, 15777-15782.
[19] V. Dimitrov, K. Kostova, M. Genov, Tetrahedron Lett. 1996, 37, 6787-
6790.
[20] a) N. Greeves, L. Lyford, Tetrahedron Lett. 1992, 33, 4759-4760; b) N.
Greeves, L. Lyford, J. E. Pease, Tetrahedron Lett. 1994, 35, 285-288.
[21] S. E. Denmark, J. P. Edwards, O. Nicaise, J. Org. Chem. 1993, 58, 569-
578.
[22] a) H. Schumann, J. Müller, Angew. Chem. Int. Ed. Engl. 1978, 17, 276;
Angew. Chem. 1978, 90, 307; b) H. Schumann, J. Pickardt, N. Bruncks,
Angew. Chem. Int. Ed. Engl. 1981, 20, 120-121; Angew. Chem. 1981,
93, 127; c) H. Schumann, J. Mueller, N. Bruncks, H. Lauke, J. Pickardt,
H. Schwarz, K. Eckart, Organometallics 1984, 3, 69-74.
[4]
[23] a) H. Schumann, H. Lauke, E. Hahn, J. Pickardt, J. Organomet. Chem.
1984, 263, 29-35; b) H. Schumann, J. Less Common Met. 1985, 112,
327-341.
[24] M. U. Kramer, D. Robert, S. Arndt, P. M. Zeimentz, T. P. Spaniol, A.
Yahia, L. Maron, O. Eisenstein, J. Okuda, Inorg. Chem. 2008, 47, 9265-
9278.
[5]
[6]
T. A. Beineke, J. Delgaudio, Inorg. Chem. 1968, 7, 715-721.
W. J. Evans, T. S. Gummersheimer, J. W. Ziller, J. Am. Chem. Soc. 1995,
117, 8999-9002.
[25] M. Zimmermann, R. Anwander, Chem. Rev. 2010, 110, 6194-6259.
[26] a) M. F. Lappert, R. Pearce, J. Chem. Soc., Chem. Commun. 1973, 126-
126; b) S. Bambirra, M. W. Bouwkamp, A. Meetsma, B. Hessen, J. Am.
Chem. Soc. 2004, 126, 9182-9183; c) H. Schumann, D. M. M.
Freckmann, S. Dechert, Z. Anorg. Allg. Chem. 2002, 628, 2422-2426.
[27] a) A. L. Wayda, W. J. Evans, J. Am. Chem. Soc. 1978, 100, 7119-7121;
b) H. Schumann, W. Genthe, E. Hahn, J. Pickardt, H. Schwarz, K. Eckart,
J. Organomet. Chem. 1986, 306, 215-225; c) W. Noh, G. S. Girolami,
Polyhedron 2007, 26, 3865-3870.
[7]
a) T. Imamoto, T. Kusumoto, M. Yokoyama, J. Chem. Soc., Chem.
Commun. 1982, 1042-1044; b) T. Imamoto, T. Kusumoto, Y.
Tawarayama, Y. Sugiura, T. Mita, Y. Hatanaka, M. Yokoyama, J. Org.
Chem. 1984, 49, 3904-3912; c) T. Imamoto, Y. Sugiura, J. Organomet.
Chem. 1985, 285, C21-C23; d) T. Imamoto, Y. Sugiura, N. Takiyama,
Tetrahedron Lett. 1984, 25, 4233-4236; e) T. Imamoto, N. Takiyama, K.
Nakamura, T. Hatajima, Y. Kamiya, J. Am. Chem. Soc. 1989, 111, 4392-
4398.
[28] a) H. M. Dietrich, G. Raudaschl-Sieber, R. Anwander, Angew. Chem. Int.
Ed. 2005, 44, 5303-5306; Angew. Chem. 2005, 117, 5437-5440; b) L. C.
H. Gerber, E. Le Roux, K. W. Törnroos, R. Anwander, Chem. Eur. J.
2008, 14, 9555-9564.
[8]
[9]
a) J. L. Luche, J. Am. Chem. Soc. 1978, 100, 2226-2227; b) J.-L. Luche,
L. Rodriguez-Hahn, P. Crabbé, J. Chem. Soc., Chem. Commun. 1978,
601-602; c) A. L. Gemal, J. L. Luche, J. Am. Chem. Soc. 1981, 103,
5454-5459.
[29] A. J. Wooles, D. P. Mills, W. Lewis, A. J. Blake, S. T. Liddle, Dalton Trans.
2010, 39, 500-510.
a) H. Li, Synlett 2012, 23, 1407-1408; b) T. L. Rathman, J. A.
Schwindeman, Org. Process Res. Dev. 2014, 18, 1192-1210.
[30] G. Occhipinti, C. Meermann, H. M. Dietrich, R. Litlabø, F. Auras, K. W.
Törnroos, C. Maichle-Mössmer, V. R. Jensen, R. Anwander, J. Am.
Chem. Soc. 2011, 133, 6323-6337.
[10] For examples, see: a) T. Sato, R. Kato, K. Gokyu, T. Fujisawa,
Tetrahedron Lett. 1988, 29, 3955-3958; b) G. Bartoli, E. Marcantoni, L.
Sambri, M. Tamburini, Angew. Chem. Int. Ed. Engl. 1995, 34, 2046-2048;
Angew. Chem. 1995, 107, 2163-2164.
[31] A. G. Avent, C. F. Caro, P. B. Hitchcock, M. F. Lappert, Z. Li, X.-H. Wei,
Dalton Trans. 2004, 1567-1577.
[32] A. Pindwal, S. Patnaik, W. C. Everett, A. Ellern, T. L. Windus, A. D.
Sadow, Angew. Chem. Int. Ed. 2017, 56, 628-631; Angew. Chem. 2017,
129, 643-646.
[11] For examples, see: a) H. Fujioka, M. Fuji, Y. Okaichi, T. Yoshida, H.
Annoura, Y. Kita, Y. Tamura, Chem. Pharm. Bull. 1989, 37, 602-605; b)
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
[33] W. J. Evans, J. H. Meadows, A. L. Wayda, W. E. Hunter, J. L. Atwood, J.
Am. Chem. Soc. 1982, 104, 2015-2017.
[34] M. Zimmermann, N. Å. Frøystein, A. Fischbach, P. Sirsch, H. M. Dietrich,
K. W. Törnroos, E. Herdtweck, R. Anwander, Chem. Eur. J. 2007, 13,
8784-8800.
[35] M. Zimmermann, R. Litlabø, K. W. Törnroos, R. Anwander,
Organometallics 2009, 28, 6646-6649.
[36] D. Barisic, D. Diether, C. Maichle-Mössmer, R. Anwander, J. Am. Chem.
Soc. 2019, 141, 13931-13940.
[37] A. Mortis, D. Barisic, K. Eichele, C. Maichle-Mössmer, R. Anwander,
Dalton Trans. 2020, 49, 7829-7841.
[38] B. R. Elvidge, S. Arndt, P. M. Zeimentz, T. P. Spaniol, J. Okuda, Inorg.
Chem. 2005, 44, 6777-6788.
[39] P. Voth, S. Arndt, T. P. Spaniol, J. Okuda, L. J. Ackerman, M. L. H. Green,
Organometallics 2003, 22, 65-76.
[40] R. Grindell, B. M. Day, F.-S. Guo, T. Pugh, R. A. Layfield, Chem.
Commun. 2017, 53, 9990-9993.
[41] T. Spallek, O. Heß, M. Meermann-Zimmermann, C. Meermann, M. G.
Klimpel, F. Estler, D. Schneider, W. Scherer, M. Tafipolsky, K. W.
Törnroos, C. Maichle-Mössmer, P. Sirsch, R. Anwander, Dalton Trans.
2016, 45, 13750-13765.
[42] U. Bayer, L. Bock, C. Maichle-Mössmer, R. Anwander, Eur. J. Inorg.
Chem. 2020, 2020, 101-106.
[43] G. Lin, Z. Jin, Y. Zhang, W. Chen, J. Organomet. Chem. 1990, 396, 307-
313.
[44] D. P. Mills, A. J. Wooles, J. McMaster, W. Lewis, A. J. Blake, S. T. Liddle,
Organometallics 2009, 28, 6771-6776.
[45] J. F. McGarrity, C. A. Ogle, J. Am. Chem. Soc. 1985, 107, 1805-1810.
[46] O. Tai, R. Hopson, P. G. Williard, Org. Lett. 2017, 19, 3966-3969.
[47] a) G. Fraenkel, A. M. Fraenkel, M. J. Geckle, F. Schloss, J. Am. Chem.
Soc. 1979, 101, 4745-4747; b) D. Reed, D. Barr, R. E. Mulvey, R. Snaith,
J. Chem. Soc., Dalton Trans. 1986, 557-564; c) A. C. Jones, A. W.
Sanders, M. J. Bevan, H. J. Reich, J. Am. Chem. Soc. 2007, 129, 3492-
3493.
[48] a) G. Fraenkel, M. Henrichs, M. Hewitt, B. M. Su, J. Am. Chem. Soc.
1984, 106, 255-256; b) B. Qu, D. B. Collum, J. Am. Chem. Soc. 2006,
128, 9355-9360; c) G. Barozzino-Consiglio, M. Rouen, H. Oulyadi, A.
Harrison-Marchand, J. Maddaluno, Dalton Trans. 2014, 43, 14219-
14228.
[49] T. J. Boyle, L. J. Tribby, S. D. Bunge, Eur. J. Inorg. Chem. 2006, 4553-
4563.
[50] T. J. Boyle, M. L. Neville, J. M. Sears, R. Cramer, ChemistrySelect 2016,
1, 473-481.
[51] A. Krasovskiy, F. Kopp, P. Knochel, Angew. Chem. Int. Ed. 2006, 45,
497-500; Angew. Chem. 2006, 118, 511-515.
[52] H. J. Heeres, M. Maters, J. H. Teuben, G. Helgesson, S. Jagner,
Organometallics 1992, 11, 350-356.
[53] A. R. Crozier, K. W. Törnroos, C. Maichle-Mössmer, R. Anwander, Eur.
J. Inorg. Chem. 2013, 409-414.
[54] U. Bayer, R. Anwander, Dalton Trans. 2020, 49, 17472-17493.
[55] E. Weiss, Angew. Chem. Int. Ed. Engl. 1993, 32, 1501-1523; Angew.
Chem. 1993, 105, 1565.
[56] For recent examples of lithium enolates, see: a) K. J. Kolonko, D. J.
Wherritt, H. J. Reich, J. Am. Chem. Soc. 2011, 133, 16774-16777; b) J.
Guang, Q. P. Liu, R. Hopson, P. G. Williard, J. Am. Chem. Soc. 2015,
137, 7347-7356; c) J. Jermaks, E. H. Tallmadge, I. Keresztes, D. B.
Collum, J. Am. Chem. Soc. 2018, 140, 3077-3090.
[57] H. M. Dietrich, C. Meermann, K. W. Törnroos, R. Anwander,
Organometallics 2006, 25, 4316-4321.
[58] Y. Nakajima, J. Okuda, Organometallics 2007, 26, 1270-1278.
10
This article is protected by copyright. All rights reserved.

10.1002/anie.202103889
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Seamless low-temperature syntheses, crystallization and spectroscopies provide structural details of the heterobimetallic reagent
routinely used for regioselective carbon–carbon-bond forming with carbonyl compounds. 7Li NMR spectroscopy takes on an
important role here, tracking down n-BuLi dissociation.
11
This article is protected by copyright. All rights reserved.