Facile Reversible Benzophenone Insertion into Rare-Earth Metal Pyrazolate Complexes

Article abstract of DOI:10.1002/ejic.201700635

Treatment of the homoleptic Ce^IV pyrazolate complex [Ce(Me₂pz)₄]₂ (Me₂pz = 3,5-dimethylpyrazolato) with benzophenone (bp) led to the formation of an Me₂pz-substituted diphenylmethoxy-(N,O)-chelating ligand (pdpm), possibly metal-templated through initial coordination of bp to the cerium atom and subsequent bp insertion into the Ce–N(Me₂pz) bond. This coordination/insertion process was shown to be reversible, leading to a complex sequence of equilibria involving multiple degrees of insertion/de-insertion and association/dissociation. The dependency on temperature and the amount of bp of all equilibria was revealed, with insertion/association of bp being favored at low temperatures and de-insertion/dissociation preferentially occurring at elevated temperatures. Such sets of equilibria were also observed for the treatment of trivalent complexes [Ln(Me₂pz)₃(thf)]₂ (Ln = La, Ce, Lu) with bp. Through structural analysis, the trivalent complexes were shown to be less effective in the bp-to-pdpm conversion than the Ce^IV derivative, giving direct evidence of how an increase in rare-earth Lewis acidity aids in ketone anchorage and concomitant conversion. The observed equilibria seem to also apply to the more illustrious organocerium systems. The conversion of bp into the corresponding tertiary alcohol by the routinely employed reagent CeCl₃/nBuLi is the most selective when termination of the reaction by hydrolysis is performed at lower temperatures, with a reagent ratio bp/CeCl₃/nBuLi of 1:1:1.

Full text of DOI:10.1002/ejic.201700635

A Journal of
Accepted Article
Title: Facile Reversible Benzophenone Insertion into Rare-Earth Metal
Pyrazolate Complexes
Authors: Daniel Werner, Glen Deacon, Peter Junk, 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: Eur. J. Inorg. Chem. 10.1002/ejic.201700635
Link to VoR: http://dx.doi.org/10.1002/ejic.201700635

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
Facile Reversible Benzophenone Insertion into Rare-Earth Metal
Pyrazolate Complexes
Daniel Werner,a Glen B. Deacon,b Peter C. Junk,c and Reiner Anwander*a
Abstract: Treatment of the homoleptic CeIV pyrazolate complex [Ce(Me2pz)4]2 (Me2pz = 3,5-dimethylpyrazolato) with benzophenone (bp) led to the
formation of a Me2pz-substituted diphenylmethoxy-(N,O)-chelating ligand (pdpm), possibly metal-templated via initial coordination of bp to cerium
and subsequent bp insertion into the Ce–N(Me2pz) bond. This coordination/insertion process was shown to be reversible, leading to a complex
sequence of equilibria involving multiple degrees of insertion/de-insertion and association/dissociation. The dependency on temperature and
amount of bp of all equilibria was revealed, with insertion/association of bp being favored at low temperatures and de-insertion/dissociation
occurring preferably at elevated temperatures. Such set of equilibria was also observed for the treatment of trivalent complexes [Ln(Me2pz)3(thf)]2
(Ln = La, Ce, Lu) with bp. Through structural analysis, the trivalent complexes were shown to be less effective in bp to pdpm conversion than the
CeIV derivative, giving direct evidence of how an increase in rare-earth Lewis acidity aids in ketone anchorage and concomitant conversion. The
observed equilibria seem to apply to more illustrious organocerium systems alike: the conversion of bp into the corresponding tertiary alcohol by the
routinely employed reagent “CeCl3/n-BuLi” is most selective when the terminating hydrolysis of the reaction mixture was performed at lower
temperatures, and with a reagent ratio of 1:1:1.
Introduction
at CeIV. In contrast to paramagnetic CeIII-based systems[9] the
diamagnetic nature of CeIV also permits detailed NMR
spectroscopic investigations providing deeper insight into rare-
earth metal based insertion processes.[1b,d] Benzophenone (bp)
was selected as the ketone, not only because it does not engage
in any obvious redox chemistry with cerium,[10] but also because
its role in insertion reactions is of wider interest, especially in f-
block chemistry.[11]
We found that addition of bp to a solution of [Ce(Me2pz)4]2
generates a complex series of equilibria (Scheme 1), involving
assoication/dissociation of bp into the coordination sphere of
cerium, and insertion/de-insertion of bp into the cerium–N(Me2pz)
bond. Such equilibria are not exclusive to the teravalent
oxidation state and could be observed for the trivalent
complexes [Ln(Me2pz)3(thf)]2 (Ln = La, Ce, Lu) as well. The
equilibria were further shown to be relevant to the “CeCl3/n-
BuLi”-mediated conversion of bp into the corresponding tertiary
alcohol.
With the ability to readily access both trivalent and tetravalent
oxidation states, cerium stands out from the other rare-earth
elements (Ln), especially regarding its versatile application in
organic transformations.[1] Aside from the ubiquitous use of ceric
ammonium nitrate (CAN) as a strong single electron oxidant,[1f, 2]
CeIII reagents have also shown to be highly effective in the
conversion of ketones to alcohols (prominent reagents involve
CeCl3/LiR mixtures where R = alkyl anion).[3] It is generally
accepted that the Lewis-acidic CeIII center draws the ketone into
coordination which weakens the C=O bond, thus encouraging
nucleophilic attack by co-ligands.[4] However, this reasoning
does not explain why cerium outperforms, other harder trivalent
rare-earth elements of smaller ionic radii,[3a,3h] even if the
increase in performance is only slight. Considering that
treatment of CeIII complexes with 1,4-benzoquinone can give
CeIV hydroquinolate complexes[5] via two single electron transfer
processes (which have been recently shown to be reversible),[6]
it is conceivable that (ordinary) ketone anchorage at Ce(III)
occurs similarly. Nevertheless, based on the Lewis acidity
criterion alone, isolable CeIV complexes would be superior
reagents to anchor ketones in such transformations. However,
as CeIV complexes are generally strongly oxidizing agents of
varied stability,[5d, 6-7] the presence of reducing co-ligands applied
in these transformations (e.g. alkyl, hydrido, halogenido), would
cause spontaneous reduction of the CeIV ion making such
investigations infeasible.
Results and Discussion
Initial identification of the equilibria sequence
Treatment of 1 with one equivalent of bp in C6D6 caused a color
change from dark red to a red-orange. Analysis by 1H NMR
spectroscopy at 25 °C indicated a mixture of species, namely
[Ce(Me2pz)4](bp) (1·bp), anticipated [Ce(Me2pz)3(pdpm)] (2,
pdpm = (3,5-dimethylpyrazol-1-yl)diphenylmethanolate), and a
We recently introduced di-alkyl-substituted pyrazolato
ligands to CeIV chemistry,[6] isolating the dimeric complex
[CeIV(Me2pz)4]2 (1, Me2pz = 3,5-dimethylpyrazolato). Complex 1
was remarkably stable under an argon atmosphere, and as
pyrazolate/pyrazoles are known nucleophiles,[8] it was
envisioned that this complex would provide an excellent platform
to study ketone anchorage and subsequent nucleophilic attack
di-pdpm species [Ce(Me2pz)2(pdpm)2] (3) in
a
ratio of
approximately 1.4:23:1 (Scheme 1, Figure S1). Apparently, the
formation of 2 involved nucleophilic attack of a Me2pz ligand on
the carbonyl group, generating this new “N,O” ligand – a result of
bp coordination to cerium and concomitant bp insertion into the
Ce-N bond (Scheme 1, 1a & 1i). Considering that 1·bp is also
present, the insertion process appears reversible (Scheme 1, 1ii)
at ambient temperature, generating the first set of equilibria. It is
not definite that bp is coordinated to 1, but the resonance of the
ortho hydrogen atoms appear further downfield than the
expected chemical shift of non-coordinated bp. Moreover, the
C(4)–H resonance for the Me2pz ligands is distinct from that of
the dimer [Ce(Me2pz)4]2 (6.11 ppm vs 5.84 ppm[6]), making
monomerized “[Ce(Me2pz)4](bp)” an appropriate transient
species.
[a]
Prof. Dr. R. Anwander and Dr. D. Werner, Institut für Anorganische
Chemie, University of Tübingen (EKUT) Auf der Morgenstelle 18,
72076, Germany, E-mail: reiner.anwander@uni-tuebingen.de,
http://uni-tuebingen.de/syncat-anwander
[b]
[c]
Prof. Dr. G. B. Deacon, School of Chemistry, Monash University,
Clayton, Victoria 3800 (Australia)
Prof. Dr. P. C. Junk, College of Science & Engineering, James Cook
University, Townsville, Queensland 4811 (Australia)
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
Scheme 1. Equilibria occurring from the reactions between [Ce(Me2pz)4]2 (1) and benzophenone (bp): cerium-bp association (1a-3a) and dissociation (1d-3d), and bp
insertion (1i-3i) and de-insertion (1ii-3ii) into the Ce–N(Me2pz) bond.
In further support of this scenario, the C=O resonance for 1·bp
65 °C, the integrations for 3 increased along with resonances for
dimeric [Ce(Me2pz)4]2 (1), while those of 2 decreased, indicating
was not detected in the 13C NMR spectrum (Figure S2), which
can be explained by a rapid coordination/dissociation (Scheme 1, that once the source of bp is exhausted, the equilibrium
1a/1d) of bp to cerium. [12] The formation of 3 from this reaction
mixture was unexpected and interesting, and even occurred in
the presence of an excess of 1. This suggested another set of
association/insertion equilibria, even if only to a minor extent (a
2:3 ratio of approximately 1:30, Scheme 1, 2a-2i/2d-2ii).[13] The
sequence (1à1·bp, 1·bpà2 etc) comes to a halt. However, at
this point the system can undergo a ligand exchange process to
generate the two more stable species 1 and 3, rather than
transient compounds (1·bp, 2, and 2·bp).[14] This is further
evident as despite the observation that 2 was the major species
in solution at ambient temperature, attempted crystallization of 2
from an n-hexane solution failed giving only crystalline 3 and 1.
The solid-state structure of 1 was different from that of the
toluene solvate reported previously,[6] featuring Me2pz ligands
with different bridging modes, namely 1κ(N,Nʹ):2κ(N) and
1κ(N,Nʹ):2κ(N,Nʹ) as opposed to two 1κ(N,Nʹ):2κ(N,Nʹ) Me2pz
ligands in 1·toluene (Figure S7). The structural elucidation of 3
confirmed an 8-coordinate cerium atom featuring two cisoid-
η2(N,Nʹ) Me2pz ligands and two N,O-chelating pdpm ligands
(Figure 1). Analysis of the pdpm bond lengths confirm the
reduction of the C=O bond of bp to a single bond (1.378(3) Å, c.f.
bp: C=O: 1.23(1)Å).[15] Furthermore, short bond lengths of
formation of
3
likely occurs through
a
transient
[Ce(Me2pz)2(pdpm)2](bp) species, (2·bp, Scheme 1), but this
compound was not observed under the prevailing conditions.
The reaction mixture, and associated equilibria, are indeed
complicated. To further elucidate this system, the deciding
parameters were sought by altering the reaction solvent,
reaction temperature, and varying the equivalents of bp used.
Underpinning the equilibria sequence
It was observed that choice of solvent dramatically influenced
the association/dissociation of the pdpm ligand. For example,
when bp is added to 1 in THF-d8, no formation of pdpm was
observed (Figure S4). This is likely due to saturation of the
coordination sphere of cerium by THF, inhibiting bp coordination
and subsequent insertion into the Ce–N(Me2pz) bond. Therefore,
the equilibria were only studied in non-coordinating solvents. If
the C6D6 reaction mixture was heated, the resonances for 3
decreased (giving 2 and 1·bp, pushing the equilibrium to the left,
Scheme 1), and were completely removed at 45 °C. Continued
heating above 75 °C, caused complete bp de-insertion (Figure
S5), with the broadened signals indicating that a degree of bp
association and dissociation to the coordination sphere of
cerium is likely occurring. In contrast to heating, cooling the
reaction mixture to 10 °C gave an increase in 2 at the expense
of 1·bp, giving a total ratio of 1·bp:2:3 of 1.4:23:1.
2.173(2)
[Ce(tBuO)2(tBuOH)2(NO3)2]: Ce-O(tBuO): 2.026(5)/2.023(5) Å, Ce-
O(tBuOH)
2.521(5)/2.529(5) Å).[16] Nevertheless, nitrogen-
Å
indicate
alkoxide
formation,
c.f.,
:
functionalized di-phenylmethoxide ligands are rare, c.f.
(NMe2)Ph2CO in dimeric [{Li((NMe2)Ph2CO)(thf)}2], which was
obtained
by
treatment of phenyllithium
with
N,Nʹ-
dimethylbenzamide.[17] Complex 3 appeared to be the most
stable derivative in the solid state, as when 1 was treated with
two, or even four equivalents of bp, crystallization at ambient
temperature only gave 3.
Despite the formation of a favorable CeIV alkoxide bond,
and a five-membered ring, either addition of another equivalent
of bp to 2 (in C6D6, Figure S8, S9), or dissolution of crystals of 3
in toluene-d8 (Figure S10), caused rapid de-insertion (Scheme 1,
2ii) and dissociation (Scheme 1, 2d) of bp from the second pdpm
ligand, supporting the proposed complicated sequence of
equilibria. Again, 2 appeared the major species at ambient
temperature, with a ratio for complexes 2:3 of 4.5:1, and at these
temperatures only trace amounts for 1·bp were observed
(Figure S8).
To further examine the equilibria at even lower
temperature, C6D6 was displaced by toluene-d8 (giving the same
total volume). It was revealed that upon cooling to -15 °C the
resonance attributable to 1·bp decreased, and resonances for
both 2 and 3 increased (Figure S6). At -35 °C, all 1·bp was
consumed, leaving a 2:3 ratio of 2:1. Upon further cooling to -
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
presence of an innocent standard (FeCp2). The reaction
mixtures were then incrementally cooled to -80 °C, and analysed
by 1H NMR spectroscopy. In either case, complex 3 appeared to
be the major species at lower temperatures, however,
resonances attributable to two other species could be detected
as well. From their respective integrations, one set of
resonances was ascribed to targeted 4, but the other species
had a pdpm:Me2pz ratio of 1:3, with the presence of one
equivalent of coordinated “bp” at 7.60 ppm, next to the
resonance of free bp (7.65 ppm). This suggested the formation
of a [Ce(Me2pz)3(pdpm)](bp)-type species (2·bp, as shown in
Scheme 1, 2a), which is surprising, considering it is not formed
when a solution of crystalline 3 is cooled to -80 °C. When more
equivalents of bp were added, the resonances attributable to
2·bp increased, and with 10 equivalents of bp, 2·bp was the
major species at -80 °C (Figure S23). We had a closer look at
the reaction mixture of 3 and two equivalents of bp, to possibly
explain why 2·bp is favored in this system over the dissolution of
crystalline 3 at the same temperature. Warming the reaction
mixture from -80 °C to -70 °C showed the decrease in both 3
and bp (Figure 2, orange and blue lines), whereas 2·bp and 4
increased only slightly (Figure 2, green and yellow lines). Thus,
as not all of 3 is being converted to 4 or 2·bp, and as the
integrations for bp also decreased, there is likely another
intermediate species present such as “[Ce(Me2pz)2(pdpm)2(bp)]
(3·bp, not shown in Figure 2)”. “3·bp” could be attributable to the
broadened features in the NMR spectrum, due to the rapid
dissociation/association of bp. It isn’t until the solution is warmed
beyond -70 °C that resonances for 4 significantly increase,[19]
and those of 2·bp decrease. Therefore, between -80 and -70 °C,
bp can coordinate to cerium, but only above -70 °C bp activation
can occur. It is further noteworthy that resonances of 1 were not
observed, indicating that ligand exchange processes are absent
in the presence of bp.
Figure 1. Molecular structure of [Ce(Me2pz)2(pdpm)2] (3). Ellipsoids are shown
at 50% probability level, hydrogen atoms removed for clarity. Selected bond
lengths (Å) and angles (°):Ce1–O1 2.173(2), Ce1–N1 2.564(2), Ce1–N3
2.402(2), Ce1–N4 2.371(2), C6–O1 1.378(3) Å; O1-Ce1-N1 63.78(5),
Cent(N3/4)-Ce1-Cent(N3ʹ/4ʹ): 93.66(7).
Thus, in the absence of 1, and as formation/disassembly of
transient 2·bp is too rapid at ambient temperature, the remaining
bp remains unbound to cerium. This is supported by both 1H and
13C NMR spectroscopic analysis of the C6D6 solution, with the
ortho-H resonance for the liberated bp at 7.69 ppm matching
that of free bp (c.f. 7.56 ppm in 1∙bp), and the C=O resonance
detected at 196.1 ppm (Figure S9, further comparisons:
[Cp*2YCl(bp)]: 13C NMR: C=O δ = 206.6 ppm; vs free bp: 196.5
ppm).[18]
Applying single crystals of 3 dissolved in toluene-d8, the
temperature-dependent equilibria were examined between -
80 °C to 80 °C by 1H NMR spectroscopy. At 10 °C in toluene-d8,
the ratio of complexes 2:3 was approximately 1:1 (Figure S10),
highlighting the influence of the additional equivalent of bp on
the equilibria sequence. When cooled to -80 °C, only the
resonances of 3 were identified (Figure S11, S12), while heating
to >80° C afforded complete bp de-insertion from the Ce–N
bonds, but there appeared to be some degree of bp association
to the cerium center (Figures S13 and S14). Furthermore, there
were obvious color changes observed when the sample was
held at different temperatures, corresponding to different
positions of the equilibria. At lower temperatures, the solution
lightened in color, giving either an orange solution
(corresponding to the color of crystals of 3), or light yellow when
more equivalents of bp were added. At higher temperatures, the
solution was dark red, indicating the absence of pdpm ligands.
This was also evident by UV-vis spectroscopy (Figure S15),
where at 80 °C the reaction mixture of 3 in toluene gave an
almost identical spectrum to that of neat 1 at the same
temperature, further supporting the complete de-insertion of bp
and re-formation of the Ce–N(Me2pz) bonding at 80 °C.
Figure 2. 1H NMR integrations of complexes 2, 2·bp, 3, 4 and bp from the
reaction mixture between 3 and two equivalents of bp at varied temperatures
(integrations were calculated relative to FeCp2, for easier comparisons,
integrations shown for bp are presented with one less equivalent e.g. at -40 °C
shown ratio = 0.2, actual ratio = 1.2 etc.).
Considering complex 3 was isolated from the reaction
between
1 and four equivalents of bp, it appears that
bp/Me2pzàpdpm conversion by 1 is limited to two bp units upon
crystallization. Due to the high stability of 2, neither an additional
association/insertion (Scheme 1, 3a/3i) nor ligand re-
arrangement to form [Ce(Me2pz)(pdpm)3] (4) is feasible at these
temperatures. It was initially assumed that this was due to steric
constraints at the CeIV center, however, we were curious if either
putative [Ce(Me2pz)(pdpm)3] (4) or even perhaps [Ce(pdpm)4]
(5) could form in solution at lower temperatures in the presence
of additional equivalents of bp. Thus, reaction mixtures
containing 3, and either one (Figure S16-19), or two (Figure
S20-21) equivalents of bp were dissolved in toluene-d8, in the
The entire equilibria sequence can be rationalised by two distinct
processes (Figure 2), one involving the coordination of bp to
cerium (i.e. the bp association processes, 1à1·bp, 2à2·bp,
3à3·bp), and the other involving the bp/Me2pzàpdpm
conversion (i.e. the bp insertion processes, 2·bp 3, 3·bp 4).
à à
However, although the 3rd equivalent of bp allows formation of 4
at low temperature, it also plays an additional role which
undermines the insertion process. When more equivalents of bp
are present, the likelihood of bp coming into the coordination
sphere of cerium increases (i.e. 3 + bp à 3·bp), and upon
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
coordination, it can either form pdpm (i.e. 3·bpà4) by insertion,
or induce a de-insertion exchange process for one pdpm ligand
(Scheme 2, 3·bpà2·bp + bp). This is assumed rapid and
unnoticeable above -40 °C, but if bp coordination and the
induced bp de-insertion (i.e. 3 + bp à 2·bp + bp) are faster than
the insertion of the new incoming bp unit (i.e. 2·bp à 3), then
the association/dissociation equilibrium becomes not only
observable (Scheme 2ii), but more pronounced in the presence
of higher amounts of bp. However, examination of this equilibria
is limited, as when the sample is heated beyond -60 °C, the bp
insertion equilibria proceeds (slowly removing 2·bp and forming
4, Scheme 2iii), and above -50 °C the typical temperature
dependent behavior of these equilibria dominates (Scheme 1
aiii/diii). It is further noteworthy that the formation of the pdpm
ligand appears to occur only in the presence of a metal center
(templating effect), as treatment of Me2pzH with bp resulted in
no reaction (Figure S24).
Comparisons with trivalent complexes
To further determine the capacity of CeIV in ketone conversion,
and if such equilibria is restricted to the tetravalent oxidation
state, we compared the activation ability of the corresponding
trivalent pyrazolate complexes species toward bp. Accordingly,
[Ln(Me2pz)3(thf)] (Ln = La (6a), Lu (6b))[20] and [Ce(Me2pz)3] (7)[6]
were treated with either three (Ce) or two (La, Lu) equivalents of
bp. Unfortunately, the complexes 6 and 7 were poorly soluble in
non-coordinating solvents, and thus were treated with bp in THF
prior to crystallization from n-hexane (Ln
= La, Ce), or
cyclohexane (Lu). Indeed, bp insertion did also occur for all LnIII
derivatives affording dimeric species [Ln2(Me2pz)3(pdpm)3] (Ln =
La (8La), Ce (8Ce); the solid-state structures are shown in Figure
3 (Ln = Ce), and Figure S28 (Ln = La)) and [Lu(Me2pz)2(pdpm)]
(9Lu), with pdpm:Me2pz ratios of 1:1 and 1:2, respectively. The
pdpm ligands are located exclusively in bridging positions, with
either three bridging pdpm and six terminal Me2pz ligands (8La,
8Ce) or two bridging pdpm and Me2pz ligands (μ-η1:η1) and two
terminal Me2pz ligands (9Lu). Attempted H NMR spectroscopic
1
Ph
Ph
(i)
N
N
investigations for 8La and 8Ce were difficult because of their poor
solubility. But it was observed that both complexes rapidly
liberate bp, indicative of similar equilibria sequences. In the case
of La, complete bp de-insertion and dissociation occurred at
ambient temperature giving [La(Me2pz)3] as determined by
elemental analysis. It was only complex 9Lu featuring the hardest
Lewis acid center of the LnIII derivatives that retained its
constitution in solution, although poorly soluble. The 1H
resonance for the pdpm (C(4)–H of the Me2pz component)
ligand in complex 9Lu appeared at an identical value to that of 2,
suggesting that the pdpm ligand of 2 may also be bridging as
these resonances are distinct from the equivalent pdpm C–H
resonance observed in mononuclear 3.
Ph
Ph
>–40 °C
N
N
O
O
O
O
Ce
Ce
+
+
Ph
Ph
Ph
Ph
(Me2pz)2
(Me2pz)2
pdpm
pdpm
(3)
(3)
(ii)
–60 °C - –50 °C
>–60 °C
–80 °C - –70 °C
(iii)
Ph
>–40 °C
Ph
Ph
Ph
N
N
N
N
O
O
Ph
Ph
O
Ph
Ce
Ce
(Me2pz)2
Me2pz
Ph
pdpm
N
N
pdpm
(4)
+
(2·bp)
O
The decreased pdpm:Me2pz ratios observed for the
trivalent species give direct evidence of increased reactivity of
the CeIV cation in ketone anchorage, and ergo emphasize the
importance of Lewis acidity on ketone anchorage. However, The
difference between LaIII/CeIII and LuIII can be rationalized by the
Scheme 2. Possible path for the formation of 2·bp and 4 from 3, and the bp
induced de-insertion and exchange. (i): bp de-insertion and exchange in the
pdpm ligand; (ii) bp coordination induced pdpm de-insertion and the
consequential dissociation/association equilibrium; (iii) formation of 4.
Figure 3. Molecular structure of [Ce2(Me2pz)3(pdpm)3] (8Ce, left) and [Lu(Me2pz)2(pdpm)]2 (9Lu, right). Ellipsoids shown at 50% probability level, hydrogen atoms
and phenyl rings of pdpm were removed for clarity. Selected bond lengths (Å) and angles (°) for 8Ce: Ce1-O1: 2.441(3), Ce1-O2: 2.373(3), Ce1-O3: 2.481(3), Ce1-
N1: 2,615(4), Ce1-N3: 2.586(4), Ce1-N7: 2.474(5), Ce1-N8: 2.456(4), Ce2-O1: 2.380(3), Ce2-O2: 2.543(3), Ce2-O3: 2.502(3), Ce2-N5: 2.595(4), Ce2-N9:
2.587(4), Ce2-N10: 2.435(4), Ce2-N11: 2.470(5), Ce1-O1-Ce2: 98.20(12), Ce1-O2-Ce2: 95.62, Ce1-O3-Ce2: 93.99(11)Ce1-O1-C6: 109.6(3), O1-Ce1-N1:
62.88(5). Selected bond lengths (Å) and angles for 9Lu (°): Lu1–O1 2.194(2), Lu1–N1 2.390(2), Lu1–N3 2.419(2), Lu1–N4ʹ 2.392(2), Lu1–N5 2.291(2), Li1–N6
2.297(2), C6–O1 1.386(3), Lu1–O1–Lu1ʹ 94.785(4), N1-Lu1-O1: 68.46(7)
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
following: the larger ions can accommodate more bp units into
their coordination sphere, but their weaker Lewis acidity
(compared to Lu) makes their associated bp units more labile in
solution. Thus, upon crystallization, when these larger ions are
starved of donor solvent molecules, another bp unit can
coordinate to form pdpm, but it immediately undergoes de-
insertion and is dissociated upon dissolution. For the smaller
lutetium ion, the increased Lewis acidity enables retention of one
pdpm ligand in solution. Crucially, the LuIII center is still not hard
enough to anchor and convert two bp units, thus being less
effective than the higher oxidation state species CeIV.
Nucleophilic attack on carbonyl groups by Me2pz– is known, but
it has been only observed from ligand degradation processes.[21-
mixture containing a reagent ratio of 1:3:3 (CeCl : LiBu : bp,
respectively, method c) at lower temperatures (as at ambient
temperatures organocerium decompositions prevail, method d,
see Experimental section), since this mixture should favorably
form “[Ce(dppo)3]”. Thus, CeCl3 and bp (ratio of 1:1), were
stirred at -78 °C in THF, and a slight excess of n-BuLi was
added. Although the reaction was not as selective as other
systems,[24] the lower-temperature hydrolysis favored Hdppo
formation. In all reactions bp was the limiting reagent, thus as it
was recovered in varied amounts after hydrolysis, supporting
that a dppo/n-Bu de-insertion/dissociation equilibrium is indeed
present. Although this is only a preliminary analysis and the
experiment does contain problems (such as the necessity to use
THF as a solvent, which will induce a degree of dissociation),
the results support the presence of a dissociation/de-insertion
equilibrium in these systems as well as “CeCl2(n-Bu)”-type
components as active species in 1:1:1 ratio reaction mixtures.
22]
Nevertheless, although experimentation was limited due to
the poor solubility of complexes 6 and 7, it appears that such
complex sequences of equilibria are not exclusive to the
tetravalent oxidation state, and therefore may be relevant to
other systems involving such insertion processes.
Do similar equilibria exist in the “CeCl3/n-BuLi” reaction
mixture?
Table 1. Summary of the reactions between CeCl3, bp and n-BuLi, and the
ratio of the obtained organic products (Hdppo: Hdpm : bp).
To determine if such bp/L+bp equilibria (L = monoanionic
ligand) are restricted to only the pyrazolate system, we revisited
the commonly used CeCl3/LiR mixture on the conversion of bp to
1,1-diphenylpentan-1-ol (Hdppo, after hydrolysis). Although the
reaction between n-BuLi and Bp is well known for over 40 years
(giving a mixture of Hdppo and diphenylmethanol (Hdpm) upon
hydrolysis),[23] to the best of our knowledge it has not yet been
performed in the presence of CeCl3 reagents (though CeI3 was
shown to selectively convert (PhCH2)2CO to the tertiary
alcohol).[3d] Accordingly, similar association/dissociation and
insertion/de-insertion phenomena within the system CeCl3/n-
Ratio
CeCl3/ bp/n-BuLi
Thydrolysis/
Method
Hdppo
Hdpm
bp
°C
a
b
1:1:1
1:1:1
1:3:3
-55a
0b
44
10
15
3.6
3
1
1
1
c
-55a
3.6
a
b
Stirred for 40 min at -65 – 55 °C in THF before hydrolysis. Slowly warmed
to 0 °C in THF and then hydrolysed.
Conclusions
BuLi/bp”
would
involve
equilibria
like
CeCl2(n-
Bu)+bp↔CeCl2(dppo) (dppo= 1,1-diphenylpentan-1-olate, see
Scheme 3, Table 1), dependent on temperature and reagent
ratio parameters.
In conclusion, the current rare-earth metal-templated Ph2CO
association/insertion de-insertion/dissociation equilibria appear
new, as well as the temperature and oxidation state dependence
upon successful activation. In solution, association of pyrazolato
ligands and Ph2CO into a NNO tertiary alkoxy ligand (3,5-
dimethylpyrazol-1-yl)diphenylmethoxy, pdpm) is favored at low
temperatures, while dissociation occurs preferably at ambient
temperature and higher. Furthermore, this is the first reported
example of direct comparisons between the reactivity of trivalent
and tetravalent cerium for the same organic transformation, at
least to the best of our knowledge, manifesting compellingly the
superior performance of CeIV. The feasibility of detailed NMR
spectroscopic investigations gave insights into the behavior of
rare-earth metal based insertion reactions, an area with very
limited understanding of what is occurring in the transition
states[25] or reaction mixtures[26] involved. At the same time, our
observations were shown to relate to more illustrious
organocerium systems like the classical CeCl3/n-BuLi-promoted
ketone transformations. There, similar equilibria might prevail
and product (=alcohol) selectivity get enhanced by a terminating
hydrolysis at lower temperature, which ensures maximal ketone
anchorage and insertion into the Ce–X bond. performed at lower
temperatures, and with a reagent ratio of 1:1:1.
Ph
Ph
THF
nBu
CeCl3 + x n-BuLi + x bp
x = y = 3
rt
O
nBu
x = y = 1
THF
Ce
nBu
bpn
nBu
Ph
Ph
Ph
Ph
nBu
high T
O
O
Ph
Ph
nBu
low T
Ce
Ce
O
Cl
Cl
Cl
Cl
2
Ce
-55 °C
hydrolysis
method a
rt
nBu
hydrolysis
method b
bp
- 55 °C
hydrolysis
method c
OH
OH
nBu
Ph
Ph
y n-BuH
y bp
x Ph
Ph
+
+ z
Ph
Ph
H
nBu
O
- 60 °C
3
Hdppo
Hdpm
Ce
Scheme 3.
A
simplified “CeCl3/n-BuLi/bp” system; the penultimate
organocerium complex prior hydrolysis is considerably more complicated than
presented.
Assuming that such equilibria exist, we expected the
hydrolysis of a reaction mixture containing a reagent ratio of
1:1:1 at lower temperatures (method a), should outperform
ambient temperature hydrolysis (method b, see Scheme 3).
However, if no equilibria is present in this system, then the
highest selectivity should be achieved by hydrolysis of a reaction
Experimental
General experimental procedures and instrumentation
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
All manipulations were performed using glovebox (MBraun 200B; <0.1
ppm O2, <0.1 ppm H2O) or Schlenk line techniques under an atmosphere
of purified argon gas in oven dried glassware. Solvents (THF, n-hexane,
and toluene) were purified by Grubbs columns (MBraun SPS, solvent
purification system) and stored in a glovebox. Benzophenone, n-BuLi
(2.5 M in hexanes), and Me2pzH were purchased from Sigma Aldrich and
used as received. CeCl3 was purchased from Sigma-Aldrich and was
converted to CeCl3(thf)1.05 through Soxhlet extraction. [Ce(Me2pz)4]2 (1),
[Ce(Me2pz)3] (7), and [Ln(Me2pz)3(thf)]2 (Ln = La (6a), Lu (6b)) were
synthesized according to published procedures.[6,20] The NMR spectra of
air and moisture-sensitive compounds were performed in pre-dried (over
NaK) benzene-d6 (C6D6), toluene-d8 or THF-d8 with J. Young valve NMR
spectroscopy tubes. Analyses were performed at 25 oC with either a
Bruker AVII+400 (1H: 400.13 MHz, 1 3C: 100.16 MHz), Bruker-Avance II
500 (1H: 500.13 MHz; 13C: 125.76 MHz), or a Bruker DRX250 (1H: 250
MHz, 13C 63 MHz) machine. Infrared spectra were recorded on a Nicolet
6700 FTIR spectrometer (ṽ = 4000-600 cm-1) using a DRIFT chamber
with dry KBr/sample mixtures and KBr windows, or as Nujol mulls (in CsI
plates). Magnetic susceptibilities and μeff were calculated by the Evans
method.[27] UV-Vis measurements were performed on a Perkin-Elmer
Lambda 35 UV/Vis spectroscopy instrument as dilute toluene solutions.
Elemental analysis (C, H, N) were performed on the bulk sample with an
Elementar Vario Micro cube. Reported yields are given for compounds
after a satisfactory elemental analysis was obtained from the bulk sample.
toluene) (see Figure S4). To the NMR tube, bp (0.0030 g, 0.016 mmol)
was added. No color change was observed. 1H NMR (THF-d8, 300 K, 400
MHz, integrations for bp were lower due to unequal stoichiometry): δ =
2.29 (br s, ~26 H, toluene + Me2pz(CH3)), 6.09 (br s, ~4 H, C(4)), 7.16 (m,
~4 H, toluene + bp), 7.46 (m, ~2.6 H, bp), 7.51 (m, ~1.5 H, bp), 7.76 ppm
(m ~2.6 H, bp), (see Figure S4, bottom). Method d: attempted
crystallization of [Ce(Me2pz)3(pdpm)] (2): [Ce(Me2pz)4]·1/8toluene (0.0106
g, 0.0194 mmol) and bp (0.0035 g, 0.019 mmol) were dissolved in
toluene and stirred for several minutes. The red/orange reaction mixture
was evaporated to a light oil and n-hexane was added. Upon storage at
ambient temperature, fractional crystallization gave single crystals of
[Ce(Me2pz)2(pdpm)2] (identified by unit cell comparison with the genuine
sample, see below, α = 10.97, β = 19.03, γ = 19.70), and after separation
from the crystals of 2, crystallization from the supernatant solution gave
crystals of [Ce(Me2pz)4] (1).
Experimental details concerning reactions between [Ce(Me2pz)4]2(1)
and
two
equivalents
of
bp,
and
characterization
of
[Ce(Me2pz)2(pdpm)2] (3).
Method a: NMR scale reaction: [Ce(Me2pz)4]·1/8toluene (0.0045 g, 0.0084
mmol) was dissolved in C6D6 (0.4 mL) and bp (0.0035 g, 0.019 mmol, in
slight excess) was added, causing a color change from red to red/orange.
1H NMR spectroscopic analysis indicated the formation of
[Ce(Me2pz)3(pdpm)] (2), [Ce(Me2pz)2(pdpm)2] (3), and trace 1·bp. 1H
NMR (C6D6, 400 MHz, 300 K, integrations were made relative to one
equivalent of 2, giving a ratio of 2:3 = 4:1 respectively, see Figure S8 for
assignment): δ = 1.31(s, 3 H, 2), 1.35 (s, 1.5 H, 3), 2.04 (s, 3 H, 2), 2.11
(s, 3 H, toluene + 3), 2.19 (s, 24 H, 1·bp(trace) + 2 + 3), 5.33 (s, 1 H, 2),
5.47 (s, 0.5 H, 3), 6.09 (s, 4 H, 2 + 3 + 1·bp), 7.03-7.09 (m, ~26 H, bp + 2
+ 3 + toluene), 7.31 (br s, 2 H, 3), 7.42 (d, 4 H, 2), 7.69 ppm (d, 8 H, bp in
slight excess). Refer to Figure S9 for 13C NMR spectra, which showed
the C=O resonance at 195.87 ppm. Method b: large scale reaction
between [Ce(Me2pz)4]2 and two equivalents of bp: [Ce(Me2pz)4]2 (1) was
synthesized in situ from [Ce{N(SiHMe2)2}4] and four equivalents of
Me2pzH (giving: 0.1114 g, 0.2140 mmol of 1) in a toluene solution.
Benzophenone (0.0789 g, 0.4333 mmol) was added and the solution
changed color from red to red/orange. Volatiles were removed by
evaporation of the reaction mixture to dryness (in vacuo), and the
resulting bright orange powder was washed with cyclohexane. The
washings were kept and allowed to slowly evaporate, giving crystals of
[Ce(Me2pz)2(pdpm)2] (3). The supernatant solution was discarded. The
aforementioned bright orange powder was evaporated to dryness (in
vacuo), giving [Ce(Me2pz)2(pdpm)2] (3) (0.1420 g, 75%). 1H NMR
(toluene-d8, 500 MHz, 283.15 K, See Figure S10 for partial spectra): δ =
1.31 (s, 3 H, CH3, 2), 1.35 (s, 6 H, CH3, 3), 2.02 (s, 3 H, CH3, 2), 2.13 (br
s, 12 H, CH3, 3), 2.18 (br s, ~24 H, 1·bp (trace) + 2 + 3), 5.30 (s, ~1.2 H, 2),
5.44 (s, 2 H, 3), 6.06 (br s, 5.7 H, Me2pz – 1·bp(trace) + 2 + 3), 7.02-7.12
(Ar-H pdpm, bp), 7.25 (m, ~8 H, o-Ar-H-1), 7.36 (m, ~4.7 H, o-Ar-H(2)),
7.64 ppm (m, 4 H, o-Ar-H(bp); 1H NMR (toluene-d8, 500 MHz, 193.15 K,
see Figure S11 for assignments): δ = 1.20 (s, 6 H, CH3), 2.20 (br s, 12 H,
CH3), 2.40 (br s, 6 H, CH3), 5.26 (s, 2 H, C-H(pdpm)), 6.21 (s, 2 H, C-H),
7.00 (s, 12 H, bp-mH/pH), 7.26 ppm (s, 8 H, bp-oH); 13C NMR (toluene-d8,
125.76 MHz, 193.15 K, see Figure S12 for assignments, FeCp2 was
added as a standard for other experiments): δ = 13.0, 13.6, 102.9, 107.0,
111.1, 138.4, 143.8, 145.1 ppm; 1H NMR (toluene-d8, 500 MHz, 353.15 K,
see Figure S13, trace amounts of 2 were not included): δ = 2.15 (br s, 24
H, CH3, 1/1·bp), 5.98 (br s, 4 H, 1/1·bp), 7.10 (br s, 8 H, Ar-H (m-bp)),
7.41-7.58 ppm (br s, 12 H, Ar-H(o/p-bp); elemental analysis calcd (%) for
the bright orange powder (3), C46H48CeN8O2 (885.06 g mol-1): C 62.43, H
5.47, N 12.66; found: C 62.94, H 6.23, N 12.95. UV vis (toluene, R.T.):
λmax = 416 nm (ε = 7500 Mol-1 cm-1), refer to Figure S15. Method c:
isolation of 3 from treatment of [Ce(Me2pz)4]2 with four equivalents of bp:
[Ce(Me2pz)4]·1/8toluene (0.0200 g, 0.0376 mmol) and bp (0.0292 g,
0.1602 mmol) were combined in toluene. The solution was concentrated
by exposure of the reaction mixture to vacuum, and cyclohexane was
added, giving orange block crystals of [Ce(Me2pz)2(pdpm)2] (3). The
crystals were repeatedly washed with minimal amounts of cyclohexane
and evaporated to dryness (in vacuo). Crystal yield (0.0240 g, 72%); IR
(Nujol): ν = 3344 (br, vw), 3163 (w), 1666 (vw, trace bp), 1529 (vw), 1512
Experimental details
Note: Dimeric complexes [Ce(Me2pz)4]2·1/4toluene (1), [Ce(Me2pz)3(thf)]2,
were treated as a monomeric species (e.g., [Ce(Me2pz)4]·1/8toluene) for
stoichiometric calculations. From one synthesis, complex 1 was obtained
as [Ce(Me2pz)4]2·toluene, and was used as [Ce(Me2pz)4]·1/2toluene for
stoichiometric calculations as indicated below.
Experimental
[Ce(Me2pz)4]·1/8toluene
identification of [Ce(Me2pz)3(pdpm)] (2).
details
concerning
the
treatment
(bp)
of
and
(1)
with benzophenone
Method a: by 1H NMR spectroscopy at varied temperature:
[Ce(Me2pz)4]·1/8toluene (0.0120 g, 0.0225 mmol) was dissolved in C6D6
(0.5 mL), bp (0.0043 g, 0.0235 mmol) was added, and the solution
changed color to red/orange. FeCp2 (0.0019 g, 0.0102 mmol) was added
1
as an internal standard to assist in ratio determination. H NMR analysis
indicated an equilibrium between [Ce(Me2pz)3(pdpm)] (2, see Figure S1)
and 1·bp, with trace amounts of [Ce(Me2pz)2(pdpm)2] (3). 1H NMR (C6D6,
500 MHz, note: trace impurities and resonances for 3 are not included,
integrals are set to one equivalent of FeCp2, see Figure S1 for
assignments): δ = 1.31 (s, 10 H, CH3 – pdpm (2)), 2.05 (s, 10 H, CH3 –
pdpm(2)), 2.20 (s, ~73 H, CH3 – Me2pz(1-3)), 4.00 (s, 10 H, FeCp2), 5.34
(s, 3.36 H, C-H – pdpm (2)), 6.11 (br s, ~11.67 H, -C(4)-H-Me2pz (1-3)),
6.96-7.15 (m, aromatic m/p bp/pdpm resonances), 7.43 (m, ~13.7 H, o-Ar
– pdpm (3)), 7.59 ppm (br s, 1.38 H, o-Ar – bp); 13C NMR (C6D6, 300 K,
63 MHz, was performed on a fresh sample, without FeCp2 added, only
resonances for 2 were identified): δ = 13.0 (CH3pdpm), 13.4 (CH3Me2pz),
104.0 (-OC(Ph)2(Me2pz)), 107.7 (C(3/5)pdpm), 112.6 (C(3/5)Me2pz), 130.3
(Ar-C), 132.1 (Ar-Cpdpm), 138.3 (Ar-Cpdpm), 139.6 (Ar-Cpdpm), 145.0 (Ar-
Cpdpm), 145.2 (C4Me2pz), 146.9 ppm (C4pdpm) (see Figure S2). The reaction
was cooled to 10 °C, then heated to 75 °C and monitored by 1H NMR
spectroscopy at varied temperature (Figure S5). This reaction mixture
was evaporated to dryness (in vacuo), leaving a red oil. Then, toluene-d8
was added and the reaction mixture was analysed by 1H NMR
spectroscopy at varied temperatures (0 °C - –65 °C, Figure S6). Method
b: by 1H NMR with an excess of [Ce(Me2pz)4]2 (1): [Ce(Me2pz)4]
·1/4toluene (0.0178 g, 0.0327 mmol), FeCp2 (0.0033 g, 0.0177 mmol),
and bp (0.0050 g, 0.0274 mmol) were dissolved in toluene-d8 (giving a
total volume of 0.67 mL). The mixture was added to an NMR tube and
analysed by 1H NMR spectroscopy showing the presence of 3 (see
Figure S3). Method c: treatment of [Ce(Me2pz)4]2 (1) (in excess) with one
equivalent of bp in THF-d8: [Ce(Me2pz)4]·1/2toluene (0.0115 g, 0.0203
mmol) was dissolved in THF-d8. 1H NMR (THF-d8, 400 MHz, 300.15 K): δ
= 2.32 (br s, 26 H, 1 + toluene), 6.09 (br s, 4 H, 1), 7.16 ppm (m, 2 H,
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
(w), 1304 (s), 1275 (s), 1224 (w), 1169 (m), 1154 (m), 1070 (m), 1025
(m), 966 (m), 938 (m), 891 (m), 844 (w), 763 (m), 722 (s), 700 (w), 684
(w) cm-1; χmol = 4.49x10-4 cm3 mol-1, μeff = 1.03 B.M (ΔHz: 6, 0.009 mol L-
1); elemental analysis calcd. (%) for C46H48CeN8O2 (885.06 g mol-1): C
62.43, H 5.47, N 12.66; found: C 62.06, H 5.20, N 12.73. Method d:
variable temperature 1H NMR spectrum: a 1H NMR tube was charged
with crystalline 3 (0.0116 g, 0.0131 mmol) and FeCp2 (0.0018 g, 0.0097
mmol), and then the mixture was dissolved in toluene-d8 (giving a total
volume of 0.58 mL). The sample was analysed by 1H NMR spectroscopy
at temperatures between –30 and +60 (see Figure S14).
produced a colorless poorly soluble white powder which was identified by
elemental analysis as [La(Me2pz)3] with some degree of n-hexane of
crystallization (best results were obtained with 0.4 equivalents of n-
hexane per La). Elemental analysis calcd. (%) for (C17.4H26.6,LaN6): C
45.56, H 5.84, N 18.32; found: C 45.20, H 5.73, N 18.68.
Treatment of [Ce(Me2pz)3] (7) with three equivalents of benzophenone
and isolation of [Ce2(Me2pz)3(pdpm)3]·n-hexane )(8Ce). [Ce(Me2pz)3]
(0.0373 g, 0.0877 mmol) and bp (0.0460 g, 0.2524 mmol) were combined
in THF with stirring. The solution was evaporated to dryness (in vacuo),
and n-hexane was added, giving colorless block crystals of
Attempted isolation of [Ce(Me2pz)(pdpm)3] (4).
[Ce2(Me2pz)3(pdpm)3]·n-hexane. Yield = 0.0280 g
(46%); 1H NMR
(benzene-d6, 300 K, 400 MHz, performed on crystals prior to drying, due
to paramagnetic broadening integrations were unreliable): δ = -30.29 (s),
-10.41 (s,), -7.97 (br s), -7.42 (s), 0.86 (m, n-hexane), 1.21 (m, n-hexane),
2.75 (br s), 4.63 (br s), 6.95 (br s – bp), 7.54 (br s -bp), 7.93 (s, bp), 9.98
(s), 13.51 (br s), 15.79 ppm (br s); IR (Nujol): ν = 3110 (vw), 3087 (w),
3061 (w), 3023 (m), 1667 (vw, trace bp impurity), 1544 (s), 1516 (m),
1349 (m), 1331 (m), 1317 (w), 1277 (w), 1237 (m), 1224 (m), 1202 (w),
1169 (m), 1089 (s), 1070 (s), 1039 (vs), 1008 (w), 963 (m), 943 (m), 913
(w), 897 (m), 884 (m), 786 (m), 776 (s), 761 (s), 737 (s), 700 (s), 669 (w)
cm-1; elemental analysis calcd. (%) for C144H158Ce4N24O6 (2881.47 g mol-
1, loss of lattice n-hexane): C: 60.02, H 5.53, N 11.67; found: C 59.89, H
4.66, N 11.73. Note that attempted reaction with THF-d8 did not lead to
pdpm formation: 1H NMR experiment: [Ce(Me2pz)3] (0.0250 g, 0.0588
mmol) was dissolved in THF-d8 (~0.4 mL) and bp (0.0107 g, 0.0587
mmol) was added. 1H NMR (THF-d8, 300 K, 400 MHz): δ = 6.40 (s, 18 H,
Me2pz-CH3), 7.07 (m, 4 H, bp(3,5)), 7.81 (m, 2 H, bp(4)), 7.98 (br s, 4 H,
1
Method a: 1H NMR scale reaction of 3 with one equivalent of bp: to a H
NMR tube containing crystalline 3 (0.0072 g, 0.0081 mmol), FeCp2
(0.0010 g, 0.0053 mmol), and bp (0.0015 g, 0.0082 mmol), toluene-d8
was added and the sample analysed by 1H NMR at variable
temperatures,
where
resonances
attributable
to
complexes
[Ce(Me2pz)3(pdpm)] (2) and [Ce(Me2pz)2(pdpm)2] (3) were detected at
ambient temperature and resonances for [Ce(Me2pz)1(pdpm)3] (4) and a
bp coordinated [Ce(Me2pz)2(pdpm)(bp)] (2·bp) species were identified at
lower temperatures. The spectra are displayed in Figure S16, S17, and
S18. 13C NMR (125.15 MHz, 353.2 K, toluene-d8): δ = 13.2, 112.6, 130.3,
132.0, 145.11, 180.9, 195.1 (br) (see Figure S19). Method b: attempted
crystallization of [Ce(Me2pz)(pdpm)3] (3): [Ce(Me2pz)4]·1/8toluene (0.0164
g, 0.0327 mmol) and bp (0.0165 g, 0.0905 mmol) were dissolved in
toluene and stirred for several minutes. The light orange reaction mixture
was evaporated to a light oil and n-hexane was added. Upon storage at -
35°C, the light yellow solution gave orange crystals of 2 (identified by unit
cell comparisons with the genuine sample, α = 11.09, β = 19.16, γ =
19.88), after separation from the crystals of 2, crystallization from the
supernatant solution at -35°C gave more crystals of 2. Method c, 1H NMR
scale reaction of 2 with two equivalents of bp: To a 1H NMR tube
containing crystalline 2 (0.0085 g, 0.0096 mmol), FeCp2 (0.0028 g,
0.0151 mmol), and bp (0.0032 g, 0.0175 mmol), toluene-d8 was added.
The sample was analysed by 1H NMR at variable temperatures (see
Figures S20 + S21). At ambient temperature, 2 was the major species in
a ratio of 1:2 of 1:2(See Figure S20). At lower temperatures, resonances
for [Ce(Me2pz)(pdpm)3] (3) were identified (see Figure S20). UV vis
spectroscopic analysis was also performed on 2 + 2 equivalents of bp in
toluene at varied temperatures (See Figure S22). Method d: 1H NMR-
scale reaction of 3 with ten equivalents of bp: to a 1H NMR tube
containing crystalline 3 (0.0092 g, 0.0103 mmol), FeCp2 (0.0040 g,
0.0214 mmol), and bp (0.0175 g, 0.0960 mmol), toluene-d8 was added.
The reaction mixture was analysed at varied temperatures, and the
spectra are displayed in Figure S23. At ambient temperatures the only
Ce containing species was [Ce(Me2pz)2(pdpm)2] (3), at lower
temperatures the presence of 4, 3 and 2·bp was observed.
bp(2,6)), 11.95 ppm (s,
materials.[6,20]
3 H, C(4)H), corresponding to the starting
Treatment of [Lu(Me2pz)3(thf)] (6b) with bp and isolation of
[Lu(Me2pz)2(pdpm)] (9Lu): [Lu(Me2pz)3(thf)] (0.0335 g, 0.0629 mmol), and
bp (0.0220 g, 0.1207 mmol) were stirred in C6D6. The resulting slurry of
white powder was evaporated to dryness (in vacuo), and THF was added.
The mixture was evaporated to dryness (in vacuo), and cyclohexane was
added, causing immediate precipitation of
a white solid. Storage
overnight led to crystallization of [Lu(Me2pz)2(pdpm)] (9Lu) as colorless
crystals. The crystals were washed with toluene and n-hexane giving 9Lu
as a white powder (0.0180 g, 44%). 1H NMR (400 MHz, C6D6, 300 K): δ =
1.04 (s, 3 H, pdpm), 1.58 (s, 3 H, pdpm), 2.27 (br s, ~13 H, Me2pz +
toluene), 5.34 (s, 1 H, pdpm), 6.67 (br s, ~4 H, pdpm (C(4)–H + Ar-H2,6),
6.93 (m, ~6.8 H, pdpm(Ar-H3,4,5) + toluene), 7.71 ppm (d, 2 H, Ar-H2,6).
Elemental analysis calcd. (%) for C56H62Lu2N12O2 (1285.11 g mol-1): C
52.34, H 4.86, N 13.08; found: C 51.62, H 4.99, N 12.79.
Control reaction
Treatment of Me2pzH with bp: Me2pzH (0.0342 g, 0.3557 mmol) and bp
(0.0656 g, 0.3600 mmol) were combined in C6D6 and analysed by 1H
NMR spectroscopy showing only starting materials. Addition of
LuCl3(THF)3 (0.0097 g, 0.0195 mmol) and heating did not give any
significant changes to the NMR spectrum.
Reaction of trivalent species [Ln(Me2pz)3(thf)]2 (Ln = La, Ce, Lu) with
bp.
Treatment of [La(Me2pz)3(thf)]2 (6a) with two equivalents of bp and
isolation of [La2(Me2pz)3(pdpm)3]2 (8La): [La(Me2pz)3(thf)] (0.0306 g,
0.0616 mmol) and bp (0.0220 g, 0.1207 mmol) were dissolved and
stirred in THF. The colorless solution was concentrated to an oil and
cyclohexane was added. Colorless crystals of [La2(Me2pz)3(pdpm)3]·1/4 n-
hexane (8La·1/4 n-hexane) formed at ambient temperature, and were
suitable for X-ray crystallographic analysis. The crystals were dried in
vacuum causing the crystals to fracture, leaving a white powder of 8La·n-
hexane (0.0240 g, 56%). Upon dissolution in toluene or n-hexane, a
white poorly soluble powder formed. The 1H NMR spectrum of crystalline
8La·n-hexane was complicated as not all the powder dissolved, and
multiple species were present in solution, however, resonances for free
bp were identified (Figure S24). Elemental analysis performed on further
dried powder of 8La indicated loss of n-hexane (C138H144La2N24O6, loss of
lattice n-hexane upon further drying prior analysis): C 59.39, H 5.20, N
12.04; found C 58.97, H 5.33, N 12.17. The reaction was repeated with
one equivalent of bp instead of two and stirred in THF. When the
colorless solution was concentrated to an oil, addition of n-hexane
General procedures for reactions between CeCl3, benzophenone,
and n-BuLi.
Two different ratios of reactants (CeCl3, bp, n-BuLi) were employed,
either 1:1:1 (methods a and b) or 1:3:3 (methods c and d). At ambient
temperature CeCl3 and bp were stirred in THF for approximately one
hour. The reaction mixture was cooled to -78 °C and BuLi was added
dropwise. The reaction was either stirred for 40 minutes at –65 - –55 °C
(methods a and c) or allowed to warm to 0 °C (methods b and d), where
for method d decomposition was noted by a color change to dark brown.
For methods a and c hydrolysis was performed in the following manner:
at -55 °C a cold mixture of isopropanol and toluene was added, then after
a few minutes, cold water was added and the light yellow solution
changed to colorless and a white precipitate formed. The reaction was
warmed to -20 °C and acetic acid was added (1 mL). For method b,
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
hydrolysis was performed at 0 °C where first ice cold isopropanol and
toluene was added, followed by ice cold water, then after 10 minutes
acetic acid. After hydrolysis, the mixtures were evaporated to dryness
and diethyl ether and water was added. The ether layer was extracted,
dried over MgSO4, evaporated to an oil (in vacuo), CDCl3 was added and
a small amount was extracted analysed by 1H NMR spectroscopy giving
the ratio of products listed in Table 1.
diphenylmethanol : bp of 3.6:3:1. Method d: 1:3:3 ratio, attempted
hydrolysis at ambient temperature. CeCl3(thf)1.05 (0.0800 g, 0.2483 mmol),
bp (0.1341, 0.7360 mmol), and n-BuLi (0.75 mmol) was added. The
solution turned light green upon n-BuLi addition and slowly changed color
to light yellow then upon warming the mixture turned to orange, then
brown, suggesting decomposition occurred. The solution was filtered,
concentrated, and stored at -35 °C however no crystallization occurred.
Method a: 1:1:1 ratio, hydrolysis at low temperature. CeCl3(thf)1.05
(0.2650 g, 0.0823 mmol), bp (0.1487, 0.8160 mmol), and n-BuLi (0.83
mmol) were used. The solution turned light green upon n-BuLi addition
and quickly changed color to light yellow. 1H NMR and 13C NMR
X-ray crystallography: All compounds were examined on a 'Bruker
APEX-II CCD' diffractometer at 100.15 K (3, 8Ce, 8La)or 150.15 K (1, 9Lu),
mounted on a fibre loop in paratone crystallography oil. Absorption
corrections were completed using Apex II program suite.[28] Structural
solutions were obtained by charge flipping (3, 8La, 8Ce) methods or direct
methods (1, 9Lu) and refined using full matrix least squares methods
against F2 using SHELX2013,[29] within the OLEX 2 graphical interface.[30]
A list of the parameters are found in Table 2. Notes: 8Ce: Three n-hexane
molecules within the lattice, two of which were half occupancy and
disordered, one requiring DFIX commands, one carbon atom (C16s) on
one half n-hexane was not modelled with hydrogen atoms, hence 1H
atom is missing from the formula. ISOR command used on one
disordered carbon atom on a phenyl ring of pdpm. 8La: Crystals were
slightly twinned giving large residual q peaks close to the metal centre.
ISOR command used on NPD atoms. SQUEEZE[31] used on solvent
accessible voids to remove low occupancy hexane within the cell.
spectroscopy indicated
a product ratio of 1,1,-diphenyl-1-pentanol
(dpbo) : diphenylmethanol : bp of 44:15:1. Method b: 1:1:1 ratio,
hydrolysis at ambient temperature. CeCl3(thf)1.05 (0.4099 g, 1.272 mmol),
bp (0.2325 g, 1.257 mmol), and n-BuLi (1.275 mmol) were used. The
solution turned light green upon n-BuLi addition and quickly changed
color to light yellow. Warming to 0 °C showed no sign of color change. 1H
NMR and 13C NMR spectroscopy indicated a product ratio of 1,1,-
diphenyl-1-pentanol (dpbo) : diphenylmethanol : bp of 10:3.6:1. Method
c: 1:3:3 ratio, hydrolysis at low temperature. CeCl3(thf)1.05 (0.1000 g,
0.3104 mmol), bp (0.1639, 0.8999 mmol), and n-BuLi (1.000 mmol) were
employed. The solution turned light green upon n-BuLi addition and
slowly changed color to light yellow. 1H NMR and 13C NMR spectroscopy
indicated
a
product ratio of 1,1,-diphenyl-1-pentanol (dpbo)
:
Table 2. X-ray crystallographic parameters for complexes 1-9Lu
.
Sample code
1
3
8Ce
8La
9Lu
CCDC number
1541213
C40H56Ce2N16
1041.25
Monoclinic
C2/c
1441020
C46H48CeN8O2
885.04
Orthorhombic
Pbcn
1441021
C150H171Ce4N24O6
2966.58
1533910
C141H151La4N24O6
2833.5
1533909
C56H62Lu2N12O2
1285.12
Monoclinic
P21/n
Formula
Molecular weight (g/mol)
Crystal system
Triclinic
Triclinic
Space group
P-1
P-1
a (Å)
43.071(4)
10.1234(9)
23.158(2)
90
11.0246 (3)
19.0945 (6)
19.8727 (6)
90
14.7756(13)
19.1565(17)
26.950(2)
99.1604(13)
94.8490(13)
98.1482(13)
7410.0(11)
2
12.8034(5)
21.8008(8)
25.2194(9)
69.5714(16)
85.3375(18)
87.9783(17)
6574.6(4)
2
11.4786(18)
19.187(3)
12.588(2)
90
b (Å)
c (Å)
α (°)
β (°)
115.6215(12)
90
90
109.600(3)
90
γ (°)
90
V (Å3)
9104.6(14)
8
4183.4 (2)
4
2611.7(7)
2
Z
F(000)
4192.0
1.519
1816
3038.0
2882.0
1280.0
Dx (mg m-3)
1.405
1.330
1.431
1.634
μ = (mm-1)
2.022
1.14
1-.266
1.338
3.812
Data/restraints/parameters
Goodness-of-fita on F2[a]
Final R indexes [I>=2σ (I)][b,c]
58970/0/537
1.05
5227/0/262
1.10
36771/19/1682
1.032
28282/6/1599
1.143
5743/0/331
1.020
r1 = 0.025, wR2
0.061
=
r1
=
0.0563, wR2
=
r1
=
0.0525, wR2
=
r1
=
0.0205,
r1 = 0.029, wR2 = 0.06
0.1238
0.1085
wR2 = 0.0406
[d]
[a]GOF = [Σw(F02 - Fc2)2 / (n0 - np)]1/2
.
[b]R1 = Σ(||F0| - |Fc||) / Σ|F0|, F0 > 4σ(F0). [c]wR2 = {Σ[w(F02 - Fc2)2 / Σ[w(F02)2]}1/2
.
ISOR commands were used to remove APD
atoms generated from the poor crystal qualty.
Additional details: Electronic Supplementary Information (ESI)
available: showing spectra of reported complexes, CCDC: 1541213 (1),
1441016 (3), 1533910 (8La), 1441021 (8Ce), 1533909 (9Lu). For ESI and
crystallographic data in CIF or other electronic format see DOI:XXXXX
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
[12] For labile bp ligands featuring non-detectable 13C C=O resonances, see
also: a) D. J. H. Emslie, W. E. Piers, M. Parvez, R. McDonald,
Organometallics 2002, 21, 4226-4240; b) M. R. MacDonald, J. W. Ziller,
W. J. Evans, Inorg. Chem. 2011, 50, 4092-4106.
Acknowledgements:
Foundation (Grant AN 238/16-1) and the ARC (DP16010640) for
funding.
We thank the German Science
[13] These ratios are determined from the integration of the ortho-position
hydrogen atoms of both bp (7.56 ppm) and the pdpm ligands in 2 (7.41
ppm) and 3 (7.31 ppm). The ratio of 2:3 can also be determined from
the C(4)–H resonance of the pdpm ligand which for 2 occurs at 5.33
and in 3 at 5.48 ppm.
Keywords: alkoxide • benzophenone • cerium • pryrazolate •
rare earth elements, ketone insertion
[1]
a) J.-L. Luche, L. Rodriguez-Hahn, P. Crabbe, J. Chem. Soc., Chem.
Commun., 1978, 601-602; b) H. B. Kagan, J. L. Namy, Tetrahedron
1986, 42, 6573-6614; c) C. R. Johnson, B. D. Tait, J. Org. Chem. 1987,
52, 281-283; d) L. A. Paquette, J. L. Romine, H. S. Lin, J. Wright, J. Am.
Chem. Soc. 1990, 112, 9284-9292; e) G. A. Molander, Chem. Rev.
1992, 92, 29-68; f) T. Imamoto, Lanthanides in Organic Synthesis,
Academic Press, Ltd, 1994; g) V. Sridharan, J. C. Menéndez, Chem.
Rev. 2010, 110, 3805-3849 h) N. A. Piro, J. R. Robinson, P. J. Walsh,
E. J. Schelter, Coord. Chem. Rev. 2014, 260, 21-36.
[14] The C(4)–H resonances of the Me2pz ligands were broad and occur in
the same region for each complex, and therefore ratios were readily
determined for 2 and 3 by their individual pdpm resonance.
[15] E. B. Fleischer, N. Sung, S. Hawkinson, J. Phys. Chem. A. 1968, 72,
4311-4312.
[16] W. J. Evans, T. J. Deming, J. M. Olofson, J. W. Ziller, Inorg. Chem.
1989, 28, 4027-4034.
[17] M. Adler, M. Marsch, N. S. Nudelman, G. Boche, Angew. Chem. 1999,
111, 1345-1347; Angew. Chem. Int. Ed. 1999, 38, 1261-1263.
[18] a) W. J. Evans, C. H. Fujimoto, M. A. Johnston, J. W. Ziller,
[2]
[3]
a) K. Binnemans, in Handbook on the Physics and Chemistry of Rare
Earths, Vol. Volume 36 (Eds.: J.-C. G. B. Karl A. Gschneidner, K. P.
Vitalij), Elsevier, 2006, pp. 281-392; b) V. Nair, A. Deepthi, Chem. Rev.
2007, 107, 1862-1891.
Organometallics 2002, 21, 1825-1831; b) sbds database
-
http://sdbs.db.aist.go.jp (National Institute of Advanced Industrial
Science and Technology, 15.03.2016).
a) A. L. Gemal, J. L. Luche, J. Am. Chem. Soc. 1981, 103, 5454-5459;
b) T. Imamoto, T. Kusumoto, M. Yokoyama, J. Chem. Soc., Chem.
Commun. 1982, 1042-1044; c) T. Imamoto, Y. Sugiura, N. Takiyama,
Tetrahedron Lett. 1984, 25, 4233-4236; d) T. Imamoto, T. Kusumoto, Y.
Tawarayama, Y. Sugiura, T. Mita, Y. Hatanaka, M. Yokoyama, J. Org.
Chem. 1984, 49, 3904-3912; e) T. Imamoto, N. Takiyama, K.
Nakamura, T. Hatajima, Y. Kamiya, J. Am. Chem. Soc. 1989, 111,
4392-4398; f) N. Greeves, L. Lyford, Tetrahedron Lett. 1992, 33, 4759-
4760; g) G. Bartoli, E. Marcantoni, L. Sambri, M. Tamburini, Angew.
Chem. 1995, 107, 2163-2164; Angew. Chem. Int. Ed. 1995, 34, 2046-
2048; h) H.-J. Liu, K.-S. Shia, X. Shang, B.-Y. Zhu, Tetrahedron 1999,
55, 3803-3830; i) L. Kürti, B. Czakó, Strategic Applications of Named
Reactions in Organic Synthesis: Background and Detailed
Mechanisms., Elsevier Academic Press, 2005.
[19]
Attempts at crystallization of 4 at low temperatures failed as complex 3
rapidly crystallized, pulling the equilibrium towards formation of 3 in
solution.
[20] G. B. Deacon, R. Harika, P. C. Junk, B. W. Skelton, D. Werner, A. H.
White, Eur. J. Inorg. Chem. 2014, 2412-2419.
[21] a) A. Domingos, N. Marques, A. Pires de Matos, I. Santos, M. Silva,
Organometallics 1994, 13, 654-662; b) E. C. B. Alegria, L. s. M. D. R. S.
Martins, M. Haukka, A. J. L. Pombeiro, Dalton. Trans. 2006, 4954-4961.
[22] A similar insertion equilibrium between a TiIV imide complex and bp has
been discussed in minor detail on the basis of NMR spectroscopic data:
A. E. Guiducci, C. L. Boyd, P. Mountford, Organometallics 2006, 25,
1167-1187.
[23] N. Hideo, Y. Hisashi, N. Hitosi, Chem. Lett. 1972, 1, 1167-1168.
[24] M. Hatano, S. Suzuki, K. Ishihara, J. Am. Chem. Soc. 2006, 128, 9998-
9999.
[4]
[5]
E. Marcantoni, M. Petrini, in Comprehensive Organic Synthesis II
(Second Edition), Elsevier, Amsterdam, 2014, pp. 344-364.
a) A. Sen, H. A. Stecher, A. L. Rheingold, Inorg. Chem. 1992, 31, 473-
479; b) I. J. Casely, S. T. Liddle, A. J. Blake, C. Wilson, P. L. Arnold,
Chem. Commun. 2007, 5037-5039; c) J. R. Robinson, C. H. Booth, P. J.
Carroll, P. J. Walsh, E. J. Schelter, Chem. Eur. J. 2013, 19, 5996-6004;
d) D. Werner, G. B. Deacon, P. C. Junk, R. Anwander, Chem. Eur. J.
2014, 20, 4426-4438.
[25] G. Bartoli, C. Cimarelli, E. Marcantoni, G. Palmieri, M. Petrini, J. Chem.
Soc., Chem. Commun. 1994, 715-716.
[26] a) W. J. Evans, J. D. Feldman, J. W. Ziller, J. Am. Chem. Soc. 1996,
118, 4581-4584; b) C. Alcaraz, U. Groth, , Angew. Chem. 1997, 109,
2590-2592; Angew. Chem. Int. Ed. 1997, 36, 2480-2482.
[27] a) Evans, D. F., J. Chem. Soc. 1959, 2003; b) Schubert, E. M., J. Chem.
Educ. 1992, 69, 62.
[6]
[7]
D. Werner, G. B. Deacon, P. C. Junk, R. Anwander, Dalton Trans. 2017,
46, 6265-6277.
[28] W. Kabsch, J. Appl. Cryst. 1993, 26, 795-800.
a) W. J. Evans, T. J. Deming, J. W. Ziller, Organometallics 1989, 8,
1581-1583; b) O. Eisenstein, P. B. Hitchcock, A. G. Hulkes, M. F.
Lappert, L. Maron, Chem. Commun. 2001, 1560-1561; c) P. B.
Hitchcock, A. G. Hulkes, M. F. Lappert, Inorg. Chem. 2004, 43, 1031-
1038.
[29] G. M. Sheldrick, Acta Crystallogr. Sect. A. 2008, A64, 112.
[30] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H.
Puschmann, J. Appl. Cryst. 2009, 42, 339-341.
[31] A. Spek, Acta Cryst. 2015, 71, 9-18.
[8]
a) V. J. Arán, M. Kumar, J. Molina, L. Lamarque, P. Navarro, E. García-
España, J. A. Ramírez, S. V. Luis, B. Escuder, J. Org. Chem. 1999, 64,
6135-6146; b) R. Kowalczyk, J. Skarzewski, Pol. J. Chem. 2007, 81,
1987-1999.
[9]
J. E. Kim, A. V. Zabula, P. J. Carroll, E. J. Schelter, Organometallics
2016, 35, 2086-2091.
[10]
A. R. Crozier, K. W. Törnroos, C. Maichle-Mössmer, R. Anwander, Eur.
J. Inorg. Chem. 2013, 2013, 409-414.
[11] a) Z. Hou, K. Takamine, O. Aoki, H. Shiraishi, Y. Fujiwara, H.Taniguchi,
J. Chem. Soc., Chem. Commun. 1988, 668-670; b) Z. Hou, H.
Yamazaki, K. Kobayashi, Y. Fujiwara, H. Taniguchi, J. Chem. Soc.,
Chem. Commun. 1992, 722-724; c) Z. Hou, A. Fujita, Y. Zhang, T.
Miyano, H. Yamazaki, Y. Wakatsuki, J. Am. Chem. Soc. 1998, 120,
754-766; d) 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; e) X. Li, J. Hong, R. Liu, L. Weng, X. Zhou,
Organometallics 2010, 29, 4606-4610; f) S. J. Kraft, P. E. Fanwick, S. C.
Bart, Organometallics 2013, 32, 3279-3285; g) L. A. Solola, A. V.
Zabula, W. L. Dorfner, B. C. Manor, P. J. Carroll, E. J. Schelter, J. Am.
Chem. Soc. 2017, 139, 2435-2442.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201700635
European Journal of Inorganic Chemistry
Entry for the Table of Contents
FULL PAPER
Daniel Werner, Glen B. Deacon, Peter
C. Junk, and Reiner Anwander*
Page No. – Page No.
Facile Reversible Benzophenone
Insertion into Rare-Earth Metal
Pyrazolate Complexes
Tetravalent [Ce(Me2pz)4] is superior to trivalent complexes [Ln(Me2pz)3(thf)]2 (Ln =
La, Ce, Lu) when it comes to an effective metal-templated benzophenone
conversion at low temperature.
This article is protected by copyright. All rights reserved.