Cationic Aluminium Complexes as Catalysts for Imine Hydrogenation

Article abstract of DOI:10.1002/chem.202100641

Strongly Lewis acidic cationic aluminium complexes, stabilized by β–diketiminate (BDI) ligands and free of Lewis bases, have been prepared as their B(C₆F₅)₄^? salts and were investigated for catalytic activity in imine hydrogenation. The backbone (R1) and N (R2) substituents on the ^R1,R2BDI ligand (^R1,R2BDI=HC[C(R1)N(R2)]₂) influence sterics and Lewis acidity. Ligand bulk increases along the row ^Me,DIPPBDI<^Me,DIPePBDI≈^tBu,DIPPBDI<^tBu,DIPePBDI; DIPP=2,6-C(H)Me₂-phenyl, DIPeP=2,6-C(H)Et₂-phenyl. The Gutmann-Beckett test showed acceptor numbers of: (^tBu,DIPPBDI)AlMe⁺ 85.6, (^tBu,DIPePBDI)AlMe⁺ 85.9, (^Me,DIPPBDI)AlMe⁺ 89.7, (^Me,DIPePBDI)AlMe⁺ 90.8, (^Me,DIPPBDI)AlH⁺ 95.3. Steric and electronic factors need to be balanced for catalytic activity in imine hydrogenation. Open, highly Lewis acidic, cations strongly coordinate imine rendering it inactive as a Frustrated Lewis Pair (FLP). The bulkiest cations do not coordinate imine but its combination is also not an active catalyst. The cation (^tBu,DIPPBDI)AlMe⁺ shows the best catalytic activity for various imines and is also an active catalyst for the Tishchenko reaction of benzaldehyde to benzylbenzoate. DFT calculations on the mechanism of imine hydrogenation catalysed by cationic Al complexes reveal two interconnected catalytic cycles operating in concert. Hydrogen is activated either by FLP reactivity of an Al???imine couple or, after formation of significant quantities of amine, by reaction with an Al???amine couple. The latter autocatalytic Al???amine cycle is energetically favoured.

Full text of DOI:10.1002/chem.202100641

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
Cationic Aluminium Complexes as Catalysts for
Imine Hydrogenation
Alexander Friedrich,[a] Jonathan Eyselein,[a] Holger Elsen,[a] Jens Langer,[a] Jürgen Pahl,[a]
Michael Wiesinger,[a] and Sjoerd Harder*[a]
Abstract: Strongly Lewis acidic cationic aluminium complexes,
stabilized by β–diketiminate (BDI) ligands and free of Lewis
bases, have been prepared as their B(C6F5)4 salts and were
acidic, cations strongly coordinate imine rendering it inactive
as a Frustrated Lewis Pair (FLP). The bulkiest cations do not
coordinate imine but its combination is also not an active
catalyst. The cation (tBu,DIPPBDI)AlMe+ shows the best catalytic
activity for various imines and is also an active catalyst for the
Tishchenko reaction of benzaldehyde to benzylbenzoate. DFT
calculations on the mechanism of imine hydrogenation
catalysed by cationic Al complexes reveal two interconnected
catalytic cycles operating in concert. Hydrogen is activated
either by FLP reactivity of an Al···imine couple or, after
formation of significant quantities of amine, by reaction with
an Al···amine couple. The latter autocatalytic Al···amine cycle
is energetically favoured.
À
investigated for catalytic activity in imine hydrogenation. The
backbone (R1) and N (R2) substituents on the R1,R2BDI ligand
(
R1,R2BDI=HC[C(R1)N(R2)]2) influence sterics and Lewis acidity.
Ligand bulk increases along the row Me,DIPPBDI<Me,DIPePBDI�
tBu,DIPPBDI<tBu,DIPePBDI; DIPP=2,6-C(H)Me2-phenyl, DIPeP=2,6-
C(H)Et2-phenyl. The Gutmann-Beckett test showed acceptor
numbers of: (tBu,DIPPBDI)AlMe+ 85.6, (tBu,DIPePBDI)AlMe+ 85.9,
(
Me,DIPPBDI)AlMe+ 89.7, (Me,DIPePBDI)AlMe+ 90.8, (Me,DIPPBDI)AlH+
95.3. Steric and electronic factors need to be balanced for
catalytic activity in imine hydrogenation. Open, highly Lewis
Introduction
Lewis acids are frequently used as highly robust catalysts in
many industrial applications.[1] Especially solid Lewis acids that
can be used under harsh conditions have proven to be
powerful heterogeneous catalysts in the oil refining industry. In
strong contrast, the combination of bulky molecular Lewis acids
and bases has been shown to break bonds under much milder
conditions (Scheme 1a).[2] Preventing the formation of a Lewis
acid/base pair has led to highly active mixtures that show
reactivities akin to that of transition metal complexes. This
rapidly growing field of Frustrated Lewis Pair (FLP) chemistry
developed from stoichiometric molecule activation to catalytic
transformation.[3–6] First applications of FLP’s in catalysis in-
volved the reduction of imines with H2 (Scheme 1b).[7] It was
soon realized that, provided the imine substrate is bulky, only
the Lewis acidic FLP component is needed.[8,9] Indeed, the single
action of B(C6F5)3, which is the standard Lewis acid in FLP
chemistry, suffices to catalyse imine hydrogenation (Scheme 1).
[a] A. Friedrich, J. Eyselein, H. Elsen, J. Langer, J. Pahl, M. Wiesinger, S. Harder
Inorganic and Organometallic Chemistry
Friedrich-Alexander-Universität Erlangen-Nürnberg
Egerlandstraße 1, 91058 Erlangen (Germany)
E-mail: sjoerd.harder@fau.de
Homepage: https://www.harder-research.com
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202100641
Scheme 1. (a) Activation of H2 with a sterically congested Lewis acid (LA) /
Lewis base (LB) Frustrated Lewis Pair. (b) General catalytic cycle for imine
hydrogenation by a Lewis acid.
© 2021 The Authors. Chemistry - A European Journal published by Wiley-
VCH GmbH. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and re-
production in any medium, provided the original work is properly cited.
Chem. Eur. J. 2021, 27, 1–9
1
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
Although the field of FLP catalysis rapidly evolved to hydro-
genation of numerous unsaturated substrates,[6] most investiga-
tions on imine hydrogenation are limited to variation of
substituent patterns on the borane Lewis acid catalyst.[10–13]
based on the isolation of intermediates in the catalytic cycles
and supported by DFT calculations.
Our group has been interested in molecule activation with Results and Discussion
highly Lewis acidic cationic main group metal complexes.[14–19]
We recently reported that Jordan’s cationic β-diketiminate
aluminium complex, (Me,DIPPBDI)AlMe+, can be used as a highly
Lewis acidic component in FLP activation of alkynes or CO2.[20]
Attempts to convert these stoichiometric reactions to a catalytic
protocol failed. After the first demonstration of the high
reactivity of Al-based FLP’s,[21] the field has been enormously
expanded.[22–26] By far most of these FLP applications concern
stoichiometric molecule activation and only a limited number
of catalytic procedures partially focused on FLP polymerization
have been reported.[27–30] This is inherently due to the much
higher reactivity of aluminium complexes when compared to
boron reagents, often leading to decomposition of the Al-based
Lewis acid.[21,22] However, the high reactivity of Al complexes
may also hold promise for efficient catalysis. Motivated by the
Complex syntheses and structures
Our studies focussed on a series of BDI ligands in which the
steric bulk was controlled by variation of the backbone
substituent R1 (Me or tBu) and N-substituent R2 (DIPP or DIPeP);
see Scheme 2 (DIPP=2,6-C(H)Me2-phenyl, DIPeP=2,6-C(H)Et2-
phenyl). Deprotonation of the β-diketimines with either AlMe3
or AlH3·NMe3 gave the corresponding aluminium methyl or
hydride complexes. Although aluminium methyl and hydride
complexes with the smallest ligand, Me,DIPPBDI, have been
reported,[36,37] those with the bulkier BDI ligands were hitherto
unknown. It was found that deprotonation of the β-diketimine
proligands becomes more difficult with increasing ligand bulk.
While deprotonations of Me,DIPPBDIÀ H and tBu,DIPPBDIÀ H with
AlMe3 are complete within a few hours at room temperature,
formation of the bulkier complexes needed harsher conditions:
+
high potential of cationic Al catalysts like AlEt2 in CO2 to
methane conversion,[31] we systematically investigated the
application of cationic Al catalysts in imine hydrogenation.
(
Me,DIPePBDI)AlMe2 (70 C, 12 h) and
(
tBu,DIPePBDI)AlMe2 (95 C,
°
°
While it is known that neutral Al compounds like AliBu3 catalyse
140 h). In case of Me,DIPePBDIÀ H deprotonation, we have also
[32]
°
this reaction under harsh conditions (>100 bar H2, 100 C)
been
(
able
to
isolate
the
coordination
complex
and LiAlH4 is an effective catalyst under relatively mild
Me,DIPePBDIÀ H)·AlMe3 which could be considered the first inter-
[33,34]
°
conditions (1 bar, 80 C),
we now report a series of highly
mediate along the reaction coordinate.
Lewis acidic cationic β-diketiminate (BDI) Al catalysts. Similar as
in the recently reported imine hydrogenation with cationic Zr
catalysts,[35] the advantage of cationic (BDI)Al+ catalysts is the
facile control over electronics and sterics by tuning the ligand
through variation of substituents. A potential mechanism is
Crystal structures of the aluminium hydride complex
(
tBu,DIPPBDI)AlH2, the aluminium methyl complexes (tBu,DIPPBDI)
AlMe2, (Me,DIPePBDI)AlMe2 and (tBu,DIPePBDI)AlMe2, as well as the
intermediate (Me,DIPePBDIÀ H)·AlMe3, are shown in Figure 1. Like in
the previously reported structures of (Me,DIPPBDI)AlMe2 and
(
Me,DIPPBDI)AlH2,[36,37] the (BDI)AlR2 complexes exhibit Al centres
Figure 1. Crystal structures of neutral Al complexes; H atoms are partially omitted for clarity.
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
2
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
Scheme 2. Synthesis of cationic aluminium complexes.
with a tetrahedral coordination geometry and in all cases Al
resides out of the BDI ligand plane. Calculation of the buried
volume for comparable aluminium alkyl complexes (Scheme 2)
show that ligand bulk increases along the row Me,DIPPBDI<
Me,DIPePBDI�tBu,DIPPBDI<tBu,DIPePBDI.
crystallization could be enforced by scratching the glass walls
with a spatula. Thus, the cations (Me,DIPPBDI)AlH+, (Me,DIPePBDI)
AlMe+, (tBu,DIPPBDI)AlMe+ and (tBu,DIPePBDI)AlMe+ were isolated in
the form of their borate salts as off-white microcrystalline solids
in yields of 65–99%. All complexes were fully characterized by
NMR methods and elemental analysis and for [(tBu,DIPPBDI)AlMe+]
The neutral (BDI)AlR2 complexes were converted into (BDI)
À
À
À
AlR+ cations by reaction with [Ph3C+][B(C6F5)4 ] in the polar but
[B(C6F5)4 ] and [(tBu,DIPePBDI)AlMe+][B(C6F5)4 ] crystals structures
could be determined (Figure 2). These reveal charge-separated
species with a trigonal planar coordination geometry for the Al
centres. In contrast, the previously reported complex [(Me,DIPPBDI)
weakly coordinating solvent chlorobenzene (Scheme 1). Like
observed previously in the syntheses of comparable (BDI)Mg+
cations[14–17] or (BDI)Zn+ cations,[18] the colour change from dark
orange towards pale yellow or colourless indicated completion
of the reaction. Purification of the borate salts, however, was
challenging and afforded the laborious development of individ-
ual procedures for each individual complex. All crystallization
attempts were hampered by the formation of clathrates which
is typical for these type of complexes.[19,20,36,38–41] In some cases
-
AlMe+][B(C6F5)4 ][38] features an Al centre with an additional
Al···FC6F4B(C6F5)3 contact. All cationic Al complexes dissolve
1
moderately in bromobenzene-d5 and H, 13C, 11B and 19F NMR
spectra are in agreement with the highly symmetric species as
observed in the solid state structures.
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
3
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
87) are known to be strong Lewis acids and therefore located in
the top quarter of this scale.
The highest AN of 95.3 was observed for the cation
(
Me,DIPPBDI)AlH+. This is significantly higher than the AN of 89.7
for Jordan’s cation
(
Me,DIPPBDI)AlMe+ which we previously
reported.[20] The much higher Lewis acidity of the hydride vs.
the methyl complex could be explained by the electron
releasing properties of the Me group (Hammett parameters:
σm =À 0.07, σp =À 0.17)[44] but partially also could be related to
steric factors. The better accessibility of the metal centre in
(
Me,DIPPBDI)AlH+ may lead to stronger complexation of Et3PO.
Indeed, the AN for the most sterically congested cations
(
tBu,DIPPBDI)AlMe+ (85.6) and (tBu,DIPePBDI)AlMe+ (85.9) is consid-
erably lower than that for (Me,DIPPBDI)AlMe+ (89.7). Since the AN
for (Me,DIPePBDI)AlMe+ is 90.8, also electronic factors could be
important: tBu substituents in the backbone are much more
electron releasing than the Me substituents.
Catalytic imine hydrogenation
All cationic Al complexes were investigated as Lewis acidic
catalysts for imine hydrogenation in a chlorobenzene/C6D6
mixture (2/1); see ESI for further details (Figures S52–S71). It was
found that the cationic Al complexes with DIPeP substituents at
N were fully inactive. Considerable activity was observed for the
cation (tBu,DIPPBDI)AlMe+ but the somewhat less bulky cation
Figure 2. Crystal structures of the cations in [(tBu,DIPPBDI)AlMe+][B(C6F5)4À ] and
[(tBu,DIPePBDI)AlMe+][B(C6F5)4À ]; H atoms are omitted for clarity.
(
Me,DIPPBDI)AlMe+ showed only slight activities (Table 2). In
contrast, a similar hydride complex (Me,DIPPBDI)AlH+, which was
found to be the most Lewis acidic in the series, showed no
activity. These results clearly demonstrate that imine hydro-
genation with Lewis acidic cationic Al complexes requires a fine
balance between sterics and electronics.
Attempts to prepare similar (BDI)AlCl+ cations, in which the
Al centre should be considerably more Lewis acidic, failed.
Reaction of (tBu,DIPPBDI)Al(Me)Cl, prepared by deprotonation of
the β-diketimine with AlMe2Cl, with the trityl salt [Ph3C+]
Within the FLP concept, bulky ligands at Al are required. If
the Al centre is accessible for imine coordination, a stable (BDI)
Al+···imine complex is formed which does not react with H2.
Indeed, NMR studies show strong complexation between
À
[B(C6F5)4 ] gave a mixture of (tBu,DIPPBDI)AlCl2 and [(tBu,DIPPBDI)
À
AlMe+][B(C6F5)4 ]. The in situ generated cation (tBu,DIPPBDI)AlCl+ is
presumably too Lewis acidic to be isolated and after abstraction
of another Cl from the starting material (tBu,DIPPBDI)Al(Me)Cl the
cation (tBu,DIPPBDI)AlMe+ is formed.
(
Me,DIPPBDI)AlH+ and PhC(H)=NtBu to give a tightly bound
complex that is fully unreactive towards H2 (Figure S56). Also
°
The Lewis acidity of various cationic Al complexes was
determined by the Gutmann-Beckett method (Table 1). Therein
the perturbation of the 31P NMR shift of Et3PO coordinated to
the Lewis acid of interest is converted into an acceptor number
(AN) that ranges from hexane (AN=0) to SbCl5 (AN=100).[42,43]
Benchmark Lewis acids like B(C6F5)3 (AN=77.1) and AlCl3 (AN=
heating this complex to 80 C did not lead to insertion of the
imine in the AlÀ H bond. The Al cations protected by a BDI
ligand with DIPeP-substituents at N do not form a complex with
PhC(H)=NtBu (Figure S57–S58). Also (tBu,DIPPBDI)AlMe+ does not
form a complex with PhC(H)=NtBu but its combination reacts
smoothly with H2 (Figure S52–S53).
Within the FLP concept, a strongly Lewis acidic metal centre
is required for H2 activation. However, if the metal’s Lewis
acidity is too strong, the strongly bound Al hydride complex
formed after H2 activation is not hydridic enough to react with
the imine. The most active cation, (tBu,DIPPBDI)AlMe+, incorporates
the perfect balance between moderate sterics and moderate
Lewis acidity.
Table 1. Quantification of the Lewis acidity of different cationic Al
À
complexes (as B(C6F5)4 salt) with the Gutmann-Beckett method (based on
31P NMR shifts in bromobenzene-d5).
Cation
Acceptor Number (AN)[a]
(
(
(
(
(
tBu,DIPPBDI)AlMe+
tBu,DIPePBDI)AlMe+
Me,DIPPBDI)AlMe+
Me,DIPePBDI)AlMe+
Me,DIPPBDI)AlH+
85.6
85.9
89.7
90.8
95.3
At low H2 pressure (1.5–6 bar) the cation (tBu,DIPPBDI)AlMe+
performed well in the hydrogenation of PhC(H)=NtBu, the
benchmark substrate in imine hydrogenation (Table 2). Catalyst
loadings could be lowered to 5 mol% and temperatures to
[a] AN=2.21×[δ31P(Et3PO complex)À 41]
25 C. The performance of (tBu,DIPPBDI)AlMe+ is comparable to
°
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
4
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
Table 2. Imine hydrogenation catalysed by cationic aluminium complexes with either (tBu,DIPPBDI)AlMe+ or (Me,DIPPBDI)AlMe+ cations. All reactions were
performed in chlorobenzene/C6D6 (2/1; v/v).
°
Entry
[Al]
mol%
R1
R2
H2 [bar]
T [ C]
t [h]
Conv.
1
2
3
4
5
6
7
8
tBu,DIPP
Me,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
Me,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
tBu,DIPP
10
10
10
10
5
5
5
10
10
10
5
Ph
Ph
Ph
Ph
Ph
Ph
Ph
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
iPr
1.5
1.5
1.5
6
60
60
25
25
25
60
80
60
60
25
60
80
60
60
25
60
25
60
60
60
25
25
25
60
25
80
80
3.5
90
66
19
90
240
66
3
40
8
66
16
120
66
160
37
144
90
66
17.5
210
120
17
>99%
98%
>99%
98%
88%
95%
6
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
6
>99%
>99%
95%
>99%
>99%
>99%
>99%
>99%
20%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
30%
p-ClÀ C6H4
p-ClÀ C6H4
p-ClÀ C6H4
p-ClÀ C6H4
p-ClÀ C6H4
p-MeÀ C6H4
p-MeÀ C6H4
p-MeÀ C6H4
Mes
Mes
Mes
Mes
tBu
tBu
tBu
tBu
tBu
tBu
tBu
nPr
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
5
10
10
10
10
10
10
5
10
10
10
5
6
1.5
1.5
6
1.5
1.5
6
1.5
1.5
1.5
1.5
1.5
1.5
iPr
iPr
iPr
tBu
tBu
tBu
tBu
10
10
5
16
17
144
140
>99%
>99%
67%
10
28%
that of B(C6F5)3 or to that of B/P or Zr/P FLP’s.[7,8,35] The catalyst
tolerates (sp2)CÀ Cl in (p-ClÀ C6H4)C(H)=NtBu as a functional
group and, due to the beneficial electronic effect of a para-Cl
substituent (σp = +0.23),[44] its hydrogenation is significantly
faster (entries 8–12). Reduction of (p-MeÀ C6H4)C(H)=NtBu is
considerably slower (σp =À 0.17)[44] but full conversion could be
Mechanism: experimental and theoretical
investigations
The requirement that for any reactivity all three, the catalyst,
imine and H2, need to be present simultaneously, implies a FLP
type mechanism similar to that proposed by Stephan and co-
workers (Scheme 1b).[8] After the catalytic imine hydrogenation,
°
reached at 60 C (Table 2, entries 13–15). A bulky mesityl
substituent at C also retards conversion (entries 16–19) which is
related to steric hindrance impeding hydride transfer to C. Alkyl
substituents on the imine C slow down conversion by electron
release, making the imine C less electrophilic. Consequently,
long reaction times are needed for hydrogenation of tBuC(H)=
NtBu, tBuC(H)=NiPr, or nPrC(H)=NtBu (entries 20–27). No
conversion was found for PhC(H)=NPh, a substrate with a
conjugated (activated) C=N bond. This is likely due to formation
of intermediate PhCH2N(Ph)À which is stabilized by resonance.
Also the imines CF3C(H)=NtBu, MesC(H)=NMes, iPrC(H)=NtBu
or ketimines could not be converted due to a combination of
steric or electronic factors. Although the catalyst could also not
reduce ketones with H2, it was found that using 10 mol%
À
the original catalyst [(tBu,DIPPBDI)AlMe+][B(C6F5)4 ] could be
successfully recycled by crystallization. We also have been able
to crystallize borate salts of iminium and ammonium cations
(see Figure S80–S82). Therefore, we propose that the reaction
starts with the iminium catalytic cycle in Scheme 1b but with
increasing amine concentrations a switch to an autocatalytic
ammonium cycle is predicted (Scheme 3). The possibility that
amine product and Lewis acid form an active FLP was already
suggested by Klankermeyer[9] and later verified by DFT calcu-
lations in the group of Papai.[49] These interconnected cycles of
FLP activation with either imine or amine as the Lewis base are
now a generally accepted working hypothesis in FLP-catalysed
imine hydrogenation.[13,50–52]
(
tBu,DIPPBDI)AlMe+ converts benzaldehyde to benzylbenzoate at
°
°
60 C in quantitative yield (10 mol% cat., 60 C, 4 days),
irrelevant whetherH2 is present or not (Figure S59). This trans-
formation, known as the Tishchenko reaction, is traditionally
catalysed by Al alkoxides.[45–47] The ability of cationic Al
complexes to mediate this reaction was demonstrated by the
group of Venugopal just recently.[48]
Comprehensive DFT calculations on catalytic hydrogenation
of PhC(H)=NtBu with the catalyst (tBu,DIPPBDI)AlMe+ (the borate
anion was neglected for simplicity) have been performed at the
B3PW91/def2TZVP level of theory with solvent correction using
the PCM method for PhCl. Scheme 4 shows the energy profile
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
5
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
Scheme 3. Interconnected catalytic cycles for imine hydrogenation with [(tBu,DIPPBDI)AlMe+][B(C6F5)4À ].
Scheme 4. Energy profile for imine hydrogenation with catalyst (tBu,DIPPBDI)AlMe+; the non-coordinating anion B(C6F5)4À has been neglected for simplicity
(B3PW91/def2TZVP, PCM=PhCl, relative ΔH values at 298 K and 1 bar are given in kcalmolÀ 1).
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
6
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
for the imine cycle and the integrated autocatalytic pathway
(ΔH in kcalmolÀ 1).
The catalytic activity of these cationic Al complexes depends
strongly on steric and electronic effects which require a fine
balance. While the open, highly Lewis acidic cation (Me,DIPPBDI)
AlH+ strongly coordinates imines, rendering it essentially inert
for FLP activation of H2, the most shielded cation (tBu,DIPePBDI)
AlMe+ of lower Lewis-acidity does not bind imines but also not
activate H2. High activities were observed for (tBu,DIPPBDI)AlMe+
which efficiently reduced various imines.
Isolation of iminium and ammonium reaction intermediates
suggests that two catalytic cycles operate in concert. Hydrogen
is activated either by FLP reactivity of an Al···imine couple or,
after formation of significant quantities of amine, by reaction
with an Al···amine couple. DFT calculations show that the latter
autocatalytic Al···amine cycle is energetically the most favour-
able pathway. The most important message of this work is that
small changes in the ligand environment of cationic Al
complexes can have major consequences for successful FLP
catalysis.
Combination of (tBu,DIPPBDI)AlMe+, imine and H2 does not
lead to any complex formation (I1). In transition I2* the cation
and imine cooperate in breaking the HÀ H bond which needs an
activation energy of +17.1 kcal/mol. Formation of (tBu,DIPPBDI)
Al(Me)H and the iminium cation is exothermic by À 5.0 kcal/mol.
The activation energy for hydride attack at the iminium cation
is +18.7 kcal/mol which is slightly higher (I3!I4*) than that for
H2 cleavage. Due to steric congestion in complex I5, the release
of amine product is exothermic by À 11.6 kcal/mol.
The presence of amine opens up the autocatalytic cycle in
which Al and amine activate H2. Taking the Al-amine complex
from the former cycle as a starting point, the activation energy
is only +2.6 kcal/mol. Starting from the separate Al cation and
amine, also only+14.3 kcal/mol is required to reach transition
state A2*. Note that the transition states for H2 activation with
Al/imine and Al/amine are quite different (selected transition
states are shown in Scheme 3; all other calculated structures are
shown in Figure S83). Whereas the Al/imine transition state I2*
°
°
is close to linear (Al···HÀ H: 161.3 , HÀ H···N 174.7 ) and late on Supporting Information
the reaction coordinate (HÀ H: 0.846 Å), the Al/amine transition
°
°
state A2* is bent (Al···HÀ H: 108.5 , HÀ H···N 164.4 ) and early on
the reaction coordinate (HÀ H: 0.780 Å). Subsequent proton
transfer from the ammonium cation to the imine is a low
energy process with an activation energy of +5.8 kcal/mol. This
is followed by hydride!iminium attack which needs an
activation energy of +18.9 kcal/mol.
Experimental data, crystallographic details including ORTEP
1
plots, H, 11B, 19F and 13C NMR spectra, details for the catalysis
and DFT calculations including XYZ-files.
Acknowledgements
The energy profile shows that the rate determining step in
both cycles is nucleophilic hydride!iminium attack. The resting
states in the catalytic cycles are therefore the iminium and
ammonium borate salts. This explains why these could be
successfully crystallized from reaction mixtures during catalysis.
Cleavage of the HÀ H bond is more efficient with the Al/amine
FLP which means that the autocatalytic cycle becomes more
important with reaction progress. Similar conclusions were
drawn from calculational studies on imine hydrogenation with
B(C6F5)3.[49,52]
We acknowledge the Deutsche Forschungsgemeinschaft for
support (HA3218/10-1) amd thank Mrs. C. Wronna and A. Roth
(University of Erlangen-Nürnberg) for CHN analyses and J.
Schmidt and Dr. C. Färber (University of Erlangen-Nürnberg) for
assistance with the NMR analyses. Open access funding enabled
and organized by Projekt DEAL.
Conflict of Interest
The authors declare no conflict of interest.
Conclusion
Keywords: Aluminium · cations · imine · hydrogenation · DFT
We present a detailed experimental and computational study
on the catalytic transformation of imines with low pressure of
dihydrogen at ambient temperatures mediated by cationic Al
complexes. Cationic β-diketiminate Al complexes are readily
available in good yields by reaction of (BDI)AlR2 (R=Me or H)
[1] A. Corma, H. García, Chem. Rev. 2003, 103, 4307–4366.
[2] J. S. J. McCahill, G. C. Welch, D. W. Stephan, Angew. Chem. Int. Ed. 2007,
46, 4968–4971; Angew. Chem. 2007, 119, 5056–5059.
[3] L. J. Hounjet, D. W. Stephan, Org. Process Res. Dev. 2014, 18, 385–391.
[4] M. Alcarazo, Synlett 2014, 25, 1519–1520.
[5] D. W. Stephan, Acc. Chem. Res. 2015, 48, 306–316.
[6] J. Lam, K. M. Szkop, E. Mosaferi, D. W. Stephan, Chem. Soc. Rev. 2019, 48,
3592–3612.
[7] P. A. Chase, G. C. Welch, T. Jurca, D. W. Stephan, Angew. Chem. Int. Ed.
2007, 46, 8050–8053; Angew. Chem. 2007, 119, 8196–8199.
[8] P. A. Chase, T. Jurca, D. W. Stephan, Chem. Commun. 2008, 2, 1701–
1703.
[9] D. Chen, J. Klankermayer, Chem. Commun. 2008, 2130–2131.
[10] G. Erös, K. Nagy, H. Mehdi, I. Pápai, P. Nagy, P. Király, G. Tárkányi, T.
Soõs, Chem. Eur. J. 2012, 18, 574–585.
À
with [Ph3C+][B(C6F5)4 ]. β-Diketiminate ligands of increasing
bulk have been used: Me,DIPPBDI<Me,DIPePBDI�tBu,DIPPBDI<tBu,DIPePB-
À
DI. Crystal structures of [(tBu,DIPPBDI)AlMe+][B(C6F5)4 ] and
À
[(tBu,DIPePBDI)AlMe+][B(C6F5)4 ] revealed charge separate cation-
anion pairs with planar trigonal coordination geometries for Al.
Quantification of the Lewis acidity with the Gutmann-Beckett
method gave a wide span of acceptor numbers ranging from
AN=85.6 for (tBu,DIPPBDI)AlMe+ to AN=95.3 for the cationic Al
hydride (Me,DIPPBDI)AlH+.
[11] J. S. Reddy, B.-H. Xu, T. Mahdi, R. Fröhlich, G. Kehr, D. W. Stephan, G.
Erker, Organometallics 2012, 31, 5638–5649.
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
7
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100641
Chemistry—A European Journal
[12] D. J. Scott, M. J. Fuchter, A. E. Ashley, Angew. Chem. Int. Ed. 2014, 53,
10218–10222; Angew. Chem. 2014, 126, 10382–10386.
[13] S. Tussing, L. Greb, S. Tamke, B. Schirmer, C. Muhle-Goll, B. Luy, J.
Paradies, Chem. Eur. J. 2015, 21, 8056–8059.
[14] J. Pahl, H. Elsen, A. Friedrich, S. Harder, Chem. Commun. 2018, 54, 7846–
7849.
[34] H. Elsen, J. Langer, G. Ballmann, M. Wiesinger, S. Harder, Chem. Eur. J.
2021, 27, 401–411.
[35] S. R. Flynn, O. J. Metters, I. Manners, D. F. Wass, Organometallics 2016,
35, 847–850.
[36] C. E. Radzewich, M. P. Coles, R. F. Jordan, J. Am. Chem. Soc. 1998, 120,
9384–9385.
[15] A. Friedrich, J. Pahl, H. Elsen, S. Harder, Dalton Trans. 2019, 48, 5560–
[37] C. Cui, H. W. Roesky, H. Hao, H.-G. Schmidt, M. Noltemeyer, Angew.
Chem. Int. Ed. 2000, 39, 1815–1817; Angew. Chem. 2000, 112, 1885–
1887.
5568.
[16] K. Thum, A. Friedrich, J. Pahl, H. Elsen, J. Langer, S. Harder, Chem. Eur. J.
2021, 27, 2513–2522.
[38] C. E. Radzewich, I. A. Guzei, R. F. Jordan, J. Am. Chem. Soc. 1999, 121,
[17] A. Friedrich, J. Pahl, J. Eyselein, J. Langer, N. van Eikema Hommes, A.
Görling, S. Harder, Chem. Sci. 2021, 12, 2410–2418.
[18] A. Friedrich, J. Eyselein, J. Langer, S. Harder, Organometallics 2021, 40,
8673–8674.
[39] J. Pahl, A. Friedrich, H. Elsen, S. Harder, Organometallics 2018, 37, 2901–
2909.
448–457.
[40] J. Pahl, S. Brand, H. Elsen, S. Harder, Chem. Commun. 2018, 54, 8685–
[19] S. Brand, H. Elsen, J. Langer, W. A. Donaubauer, F. Hampel, S. Harder,
Angew. Chem. Int. Ed. 2018, 57, 14169–14173; Angew. Chem. 2018, 130,
14365–14369.
8688.
[41] V. A. Dodonov, A. G. Morozov, R. V. Rumyantsev, G. K. Fukin, A. A.
Skatova, P. W. Roesky, I. L. Fedushkin, Inorg. Chem. 2019, 58, 16559–
16573.
[20] T. E. Stennett, J. Pahl, H. S. Zijlstra, F. W. Seidel, S. Harder, Organo-
metallics 2016, 35, 207–217.
[21] G. Ménard, D. W. Stephan, Angew. Chem. Int. Ed. 2012, 51, 8272–8275;
Angew. Chem. 2012, 124, 8397–8400.
[22] C. Appelt, H. Westenberg, F. Bertini, A. W. Ehlers, J. C. Slootweg, K.
Lammertsma, W. Uhl, Angew. Chem. Int. Ed. 2011, 50, 3925–3928;
Angew. Chem. 2011, 123, 4011–4014.
[23] W. Uhl, C. Appelt, J. Backs, H. Westenberg, A. Wollschläger, J. Tannert,
Organometallics 2014, 33, 1212–1217.
[42] U. Mayer, V. Gutmann, W. Gerger, Monatsh. Chem. 1975, 106, 1235–
1257.
[43] M. A. Beckett, G. C. Strickland, J. R. Holland, S. K. Varma, Polymer 1996,
37, 4629–4631.
[44] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[45] W. C. Child, H. Adkins, J. Am. Chem. Soc. 1923, 47, 789–807.
[46] F. J. Villani, F. F. Nord, J. Am. Chem. Soc. 1947, 69, 2605–2607.
[47] I. Lin, A. R. Day, J. Am. Chem. Soc. 1952, 74, 5133–5135.
[48] R. Kannan, R. Chambenahalli, S. Kumar, A. Krishna, A. P. Andrews, E. D.
Jemmis, A. Venugopal, Chem. Commun. 2019, 55, 14629–14632.
[49] T. A. Rokob, A. Hamza, A. Stirling, I. Pápai, J. Am. Chem. Soc. 2009, 131,
2029–2036.
[24] S. Styra, M. Radius, E. Moos, A. Bihlmeier, F. Breher, Chem. Eur. J. 2016,
22, 9508–9512.
[25] N. Aders, L. Keweloh, D. Pleschka, A. Hepp, M. Layh, F. Rogel, W. Uhl,
Organometallics 2019, 38, 2839–2852.
[26] D. Pleschka, M. Uebing, M. Lange, A. Hepp, A. L. Wübker, M. R. Hansen,
E. U. Würthwein, W. Uhl, Chem. Eur. J. 2019, 25, 9315–9325.
[27] M. A. Courtemanche, J. Larouche, M. A. Légaré, W. Bi, L. Maron, F. G.
Fontaine, Organometallics 2013, 32, 6804–6811.
[28] M. G. M. Knaus, M. M. Giuman, A. Pöthig, B. Rieger, J. Am. Chem. Soc.
2016, 138, 7776–7781.
[50] S. Tussing, K. Kaupmees, J. Paradies, Chem. Eur. J. 2016, 22, 7422–7426.
[51] J. Paradies, Eur. J. Org. Chem. 2019, 283–294.
[52] T. Privalov, Eur. J. Inorg. Chem. 2009, 2229–2237.
[53] P. H. M. Budzelaar, A. B. van Oort, a G. Orpen, Eur. J. Inorg. Chem. 1998,
1485–1494.
[54] T. X. Gentner, B. Rösch, G. Ballmann, J. Langer, H. Elsen, S. Harder,
Angew. Chem. Int. Ed. 2019, 58, 607–611; Angew. Chem. 2019, 131, 617–
621.
[29] Y. Zhang, G. M. Miyake, M. G. John, L. Falivene, L. Caporaso, L. Cavallo,
E. Y. X. Chen, Dalton Trans. 2012, 41, 9119–9134.
[30] C. Appelt, J. C. Slootweg, K. Lammertsma, W. Uhl, Angew. Chem. Int. Ed.
2013, 52, 4256–4259; Angew. Chem. 2013, 125, 4350–4353.
[31] M. Khandelwal, R. J. Wehmschulte, Angew. Chem. Int. Ed. 2012, 51,
7323–7326; Angew. Chem. 2012, 124, 7435–7439.
[55] B. Rösch, T. X. Gentner, J. Eyselein, A. Friedrich, J. Langer, S. Harder,
Chem. Commun. 2020, 56, 11402–11405.
[32] J. A. Hatnean, J. W. Thomson, P. A. Chase, D. W. Stephan, Chem.
Commun. 2014, 50, 301–303.
[33] H. Elsen, C. Färber, G. Ballmann, S. Harder, Angew. Chem. Int. Ed. 2018,
57, 7156–7160; Angew. Chem. 2018, 130, 7274–7278.
Manuscript received: February 19, 2021
Accepted manuscript online: March 29, 2021
Version of record online: ■■■, ■■■■
Chem. Eur. J. 2021, 27, 1–9
www.chemeurj.org
8
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
A series of strongly Lewis acidic
cationic aluminium complexes, stabi-
lized by β–diketiminate (BDI) ligands
and free of Lewis bases, have been
A. Friedrich, J. Eyselein, H. Elsen, J.
Langer, J. Pahl, M. Wiesinger, S. Harder*
1 – 9
̄
Cationic Aluminium Complexes as
Catalysts for Imine Hydrogenation
prepared as their B(C6F5)4 salts. Their
catalytic activity in imine hydrogena-
tion depends strongly on the substitu-
ents in the BDI ligand. DFT calcula-
tions on the mechanism reveal two
interconnected catalytic cycles.
Hydrogen is activated either by
Al···imine or Al···amine Frustrated-
Lewis-Pairs.