Lithium-Aluminate-Catalyzed Hydrophosphination Applications

Article abstract of DOI:10.1002/anie.201906807

Synthesized, isolated, and characterized by X-ray crystallography and NMR spectroscopic studies, lithium phosphidoaluminate iBu₃AlPPh₂Li(THF)₃ has been tested as a catalyst for hydrophosphination of alkynes, alkenes, and carbodiimides. Based on the collective evidence of stoichiometric reactions, NMR monitoring studies, kinetic analysis, and DFT calculations, a mechanism involving deprotonation, alkyne insertion, and protonolysis is proposed for the [iBu₃AlHLi]₂ aluminate catalyzed hydrophosphination of alkynes with diphenylphosphine. This study enhances further the development of transition-metal-free, atom-economical homogeneous catalysis using common sustainable main-group metals.

Full text of DOI:10.1002/anie.201906807

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Lithium-Aluminate-Catalyzed Hydrophosphination Applications
Authors: Robert E. Mulvey, Victoria Pollard, Allan Young, Ross
McLellan, Alan Kennedy, and Tell Tuttle
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201906807
Angew. Chem. 10.1002/ange.201906807
Link to VoR: http://dx.doi.org/10.1002/anie.201906807
http://dx.doi.org/10.1002/ange.201906807

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
Lithium-Aluminate-Catalyzed Hydrophosphination Applications
Victoria A. Pollard, Allan Young, Ross McLellan, Alan R. Kennedy, Tell Tuttle, and Robert E. Mulvey*
Dedication ((optional))
Abstract: Synthesized, isolated, and characterized by X-ray
crystallography and NMR spectroscopic studies, lithium
phosphidoaluminate iBu3AlPPh2Li(THF)3 has been tested as
successfully employed Al-based catalysts in hydroboration
and hydrogenation.[10] Uhl has also demonstrated that a P/Al
geminal FLP can stoichiometrically hydrophosphinate
a
catalyst for hydrophosphination of alkynes, alkenes, and
carbodiimides. Based on the collective evidence of stoichiometric
reactions, NMR monitoring studies, kinetic analysis, and DFT
calculations, a mechanism involving deprotonation, alkyne insertion
and protonolysis is proposed for the [iBu3AlHLi]2 aluminate-catalyzed
hydrophosphination of alkynes with diphenylphosphine. This study
enhances further the development of transition metal free, atom
economical homogeneous catalysis using common sustainable main
group metals.
heteroatom
substituted
nitriles,
generating
imines
incorporated into 5-atom AlCPCN heterocycles.[11] Examples
also exist of Al-catalyzed hydrophosphonylation (using P(V)
reagents)[12] However, to our knowledge no examples exist
of Al-catalyzed hydrophosphination (using P(III) reagents) of
alkynes, alkenes or carbodiimides.
Bimetallic ate complexes, which can synergistically
enhance stoichiometric reactivities over their neutral
monometallic components, are
a core theme of our
research.[13] Recently expanding this work into the catalytic
regime, we used lithium aluminates as catalysts for
hydroboration of aldehydes, ketones, imines and
acetylenes.[14] In general, the charged bimetallic species
proved more active catalysts than their neutral monometallic
components.[14b] Here, we probe the ability of our most active
Introduction
Phosphines are utilized in a range of applications spanning
agriculture, medicinal chemistry, and organocatalysis, while
their ubiquity as ligands in transition metal catalysis is
legend.[1] Hydrophosphination, adding a P–H bond across an
unsaturated C–E (E = e.g., C, N) bond, offers atom economy
to preparing phosphines.[2] Many transition (e.g., based on
Fe, Ni, Pd and Zr) [2b, 2c, 3] and rare earth (e.g., La, Sm, and
lithium aluminate, [iBu3AlHLi]2,
1
,
as
a
catalyst for
hydrophosphination of alkynes, alkenes, and carbodiimides.
Yb)[3]
metal
catalysts
have
been
used
for
Results and Discussion
hydrophosphination, while solvent and catalyst-free
hydrophosphinations can also be thermally induced under
certain circumstances.[4] Developing catalysts based on main
group metals is currently trending in homogeneous catalysis
including hydrophosphination.[5] K, Ca and Mg complexes
have been reported as hydrophosphination catalysts for
alkene, alkyne and carbodiimide substrates with diphenyl
phosphine (HPPh2), forming alkyl phosphines, vinyl
phosphines and phosphaguanidines, respectively.[3a, 3b, 6] Tin
First, we reacted phenylacetylene with HPPh2 and 10 mol%
of dimeric [iBu3AlHLi]2 in C6D6 at 70 °C. This reaction gave a
very low 5% conversion after 20 h and it was proposed a
higher boiling point solvent would be required to allow the
catalysis to proceed. Moving to d8-toluene at 110 °C yielded
72% conversion (1:8 E/Z ratio, Fig. 1) within 20 h, to the anti-
Markovnikov product consistent with syn addition of H–P
across the C≡C bond. Changing the solvent to polar d8-THF
(65 °C) lowered the E/Z selectivity to 1:3. We propose that in
d8-THF solution the resulting lithium aluminium phosphide
(vide infra) exists in an equilibrium, as observed by the
presence of three separate distinct signals for the iso-butyl
ligands in the 1H NMR spectrum. Using CD2Cl2 (40 °C)
poisoned the catalyst giving no product. Satisfied this system
is a capable hydrophosphination catalyst, a range of alkynes
were screened (Table 1). Internal alkynes, diphenylacetylene
and 1-phenyl-1-propyne reacted faster than terminal
alkynes. However, more challenging unactivated alkynes, 1-
hexyne and 3-hexyne did not react, in common with other
reports of main group catalysed hydrophosphination.[7b]
To gain more insight, a 1:1:3 stoichiometric reaction
between [iBu3AlHLi]2, HPPh2, and THF, completely
consumed the aluminate giving a solution that deposited
complexes
are
also
capable
of
catalysing
hydrophosphination reactions,[2c, 7] though Cp*2SnCl2 (Cp* =
pentamethylcyclopentadienyl) required an H2 atmosphere to
inhibit competing phosphine dehydrocoupling of HPPh2.[7b, 8]
Attractive industrially due to its high natural abundance and
low toxicity, aluminium is gaining prominence in this main
group catalysis enlightenment.[9] Work by Roesky, Wright,
Cowley/Thomas, Harder, Stephan, and others, have
Dr V. A. Pollard, A. Young, Dr R. McLellan, Dr A. R. Kennedy, Prof.
T. Tuttle, Prof. R. E. Mulvey
WestCHEM, Department of Pure and Applied Chemistry
University of Strathclyde
Glasgow, G1 1XL (UK)
E-mail: r.e.mulvey@strath.ac.uk
crystals
of
the
lithium
(isolated yield, 41%) (Fig. 1a).
results from deprotonation of HPPh2
aluminium
phosphide,
Supporting information for this article is given via a link at the end of
the document
iBu3AlPPh2Li(THF)3,
Phosphidoaluminate
2,
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
c)
a)
b)
P
Al
Li
O
Figure 1. a) Synthesis of iBu3AlPPh2Li(THF)3, 2; b) Molecular structure of 2, H atoms and disordered THF molecules omitted for clarity, thermal ellipsoids drawn at
40% probability; c) Depiction of E-, Z-stereoisomers and α-regioisomers arising from hydrophosphination of alkyne substrates.
by
1
, and importantly, implies this process is the first step in
(Fig. 1b) is
catalysis. Again attempted catalysis with unactivated
1-hexyne or 3-hexyne and HPPh2 by proved unsuccessful.
catalytic hydrophosphination. Crystalline
2
2
monomeric, with three THF molecules solvating Li. 1H DOSY
Deaggregation aside, the coordination shell surrounding a
metal cation can play a key role in modulating the Lewis
acidity of the metal, thereby providing a potential route to
modify reactivity. Thus, we explored the effect of the Lewis
donor on hydrophosphination of PhC≡CPh. A range of Lewis
donor additives were added to the hydrophosphination
NMR studies confirm it remains monomeric in d8-toluene
solution.[15] Alternatively,
2 can be made by co-complexing
LiPPh2 with iBu3Al/THF.
Aware that dehydrocoupling can compete with
hydrophosphination, a control reaction between HPPh2 and
10 mol% of
2
in d8-toluene was heated at 110 °C for 20 h.
reactions of diphenylacetylene catalysed by 1. Adding either
Less than 15% of HPPh2 had undergone dehydrocoupling to
form 1,1,2,2–tetraphenyl diphosphine (determined by 31P
NMR spectra) signifying that this is unlikely to be a significant
two equivalents (with respect to the catalyst) of bidentate
donor TMEDA (N,N,N’,N’-tetramethylethylenediamine) or
one equivalent of 12-crown-4 result in quantitative
conversions in 1 hour, the same time as when 3 equivalents
of THF are used. The molecular structure of the
organometallic compound in the presence of polydentate 12-
crown-4 was determined via X-ray crystallography as the
contacted ion pair structure, iBu3AlPPh2Li(12-crown-4), with
the phosphorus atom bridging the Al and Li centres (see
ESI). Unfortunately, due to poor quality data no geometric
parameters can be discussed, however the structure
provides unequivocal proof of atomic connectivity. The use
of isolated iBu3AlHLi(PMDETA) also results in quantitative
problem in this system. Subsequently
2 was tested as a
catalyst for hydrophosphination of alkynes, under the
previously optimised conditions (10 mol% [Al], d8-toluene,
110 °C), (Table 1). For phenylacetylene a 95% conversion
(1:3 E/Z ratio) of the anti-Markovnikov product was obtained
after 20 h (cf. 72% using
1), albeit with reduced E/Z
stereoselectivity. By contrast, Waterman’s tin catalyst
Cp*2SnCl2 is poorly active for PhC≡CH (10 mol% catalyst,
18 h, 65 °C, 4% yield).[7b] Using
2, hydrophosphination is
much faster with internal alkynes than terminal alkynes, with
a 99% yield (10:1 E/Z ratio) for diphenylacetylene being
product formation within 1 h (tridentate PMDETA =
obtained within just 1 h (
1-propyne fully converts to the anti-Markonikov vinyl
phosphine product within 1 h. The catalytic activity of with
1
takes 5 hours). Similarly, 1-phenyl-
N,N,N’,N’’N’’-pentamethyldiethylenetriamine). This complex
was crystallographically characterised (Fig. 2), but all organic
2
ligands exhibit significant disorder which precludes a
PhC≡CPh compares favourably with the β-diketiminato
calcium amide catalyst DIPPNacNacCa(HMDS)(THF), which
required extended reaction times (10 mol% catalyst, 75 °C,
13 h, 94% yield).[6c] However, [Ca(PPh2)2(THF)4] catalytically
hydrophosphinated diphenylacetylene after 2 h at room
temperature.[6e] Cui used an imino-amidinate ligated Ca
catalyst for quantitative hydrophosphination of 1-phenyl-1-
propyne after 5 h at 60 °C, using 5 mol% [Ca].[3a]
discussion of geometric parameters beyond atomic
connectivity.
Adding a catalytic amount (30 mol%) of THF to 10 mol%
of
within the same time as that using pre-formed
that deaggregation of dimeric by THF is advantageous in
1
resulted in hydrophosphination of diphenylacetylene
2
, suggesting
1
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 1. Hydrophosphination of alkynes, alkenes and carbodiimides using 1 –
6 as catalysts.
Alkynes[a]
a)
b)
c)
Al
Li
Catalyst; yield (%); E/Z/α-isomer ratio; time (h)
N
1 72%, 1:8:0, 20 h
2 95% (78%), 1:3:0,
20 h
1 98%, 10:1, 5 h
2 99% (80%), 10:1,
1 h
1 99%, 1:1:0, 6 h
2 99% (79%), 1:1:0, 1 h
3 95%, 1:3:0, 10 h
3 98%, 10:3, 3 h
3 86%, 1:3:0, 20 h
4 72%, 1:8:8, 20 h
5 62%, 1:12:4, 20 h
6 62%, 1:12:4, 20 h
Figure 2. Molecular structure of iBu3AlHLi(PMDETA). Thermal ellipsoids are
drawn at 40% probability, and disordered iBu groups, disordered PMDETA, and
hydrogen atoms, except hydride, have been removed for clarity.
Alkenes[b]
d)
e)
f)
Interestingly, the E/Z-isomer ratio is dependent on the donor
used. PMDETA and 12-crown-4 are less selective (E/Z 2:1;
and 5:1 respectively, versus 10:1 with 3 THF), whereas 2
TMEDA donors result in enhanced E/Z-selectivity of 19:1.
Adding one equivalent of bulky tetradentate Me6-TREN,
takes 3 h for quantitative conversion (E/Z 10:1). Adding two
equivalents of bidentate dppe (diphenylphosphinoethane)
results in conversion in 5 hours, albeit with good E/Z
selectivity (16:1). Interestingly it appears that when two
bidentate donors are added, TMEDA or dppe, marked
improvements in selectivity occur. Finally, adding three
1 89%,6 h
2 84% (60%), 6 h
3 72%, 6 h
2 84%, 20 h
2 86%, 18 h
g)
h)
2 87%, 20 h
2 93%, 4 h
equivalents of PPh3 to preformed
2 results in both slower
catalysis (1.5 h) and poorer selectivity (E/Z 4:1) than those
observed with the THF variant, indicating the phosphine
Lewis donor may inhibit the hydrophosphination process.
Next, the more challenging hydrophosphination of
Carbodiimides[c]
i)
j)
alkenes was examined using
2 (Table 1). Styrene undergoes
hydrophosphination in 6 h, at 110 °C, yielding 84% of the
anti-Markovnikov product. Halo-substituted styrenes are also
tolerated (Table 1, entries e-f). 4-Vinyl anisole undergoes
hydrophosphination to the alkyl phosphine product in 87%
yield after 20 h at 110 °C. Bulkier substrates such as α-
methyl styrene, trans-β-methyl styrene, and the less
2 99% (80%), 0.25 h
2 86%, 20 h
General conditions: 0.6 mmol substrate, 0.5 mmol HPPh2, d8-toluene.
Conversions based against NMR internal standard
1H
hexamethylcyclotrisiloxane. E/Z/α stereoselectivity based on 31P NMR
spectra. Selected isolated yields in parenthesis. [a]/[b] 10 mol% [Al] catalyst,
110 °C [c]: 5 mol% [Al] catalyst, RT.
activated
alkene
1-hexene
did
not
undergo
hydrophosphination with
2
as the catalyst. Similar failures
with both Ca and Sn based catalysts have been noted for
these substrates.[6c, 7b] Hydrophosphination of vinyl boronic
acid pinacol ester (vinyl Bpin) achieved a 93% yield after 4 h
at 110 °C, producing linear phosphine boronic ester
Ph2P(CH2)2Bpin. To our knowledge this is the first time
Ph2P(CH2)2Bpin has been made via a hydrophosphination
Phosphidoaluminate
2 is also an able catalyst for
hydrophosphination of carbodiimides at room temperature.
Thus, using 5 mol% catalyst loading (Table 1, entries i-j),
diisopropylcarbodiimide is converted fully to the
phosphaguanidine product within 15 minutes; while bulkier
dicyclohexylcarbodiimide required 20 h, to achieve 86%
conversion. Hill reports quantitative yields for diisopropyl–
and dicyclohexyl–carbodiimides within 1 h and 4 h,
route,
since
earlier
published
methods
required
hydroboration of diphenyl vinyl phosphine.[16]
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
despite several attempts, we tentatively assign them as two
Table 2. Effect of Lewis donor upon hydrophosphination catalysis of
isomers resulting from insertion of diphenylacetylene into 2.
diphenylacetylene.
Subsequent addition of HPPh2 and further heating at 110 °C
allows for product formation [8.9 ppm (E-isomer)
and -7.3 ppm (Z-isomer)] and regeneration of
2 (-49.3 ppm).
We rationalise the intermediate (at 3.4 ppm) reacts onwards
to form the E-stereoisomer as it is consumed faster than the
other intermediate. Significantly there are no resonances
corresponding to LiPPh2 in the spectrum (-52 ppm) further
Lewis donor additive
None [iBu3AlHLi]2, 1
1 + 3 equiv. THF
Time (h)
Yield(%) / (E:Z ratio)
98 / 10:1
5
1
reinforcing the view that
2 is the catalytically active species.
We propose a catalytic cycle (Scheme 1) that begins by
deprotonation of HPPh2 by [iBu3AlHLi]2, releasing H2 and
99 / 10:1
iBu3AlHLi(PMDETA)
1 + 1 equiv. 12-crown-4
1 + 2 equiv. TMEDA
1 + 1 equiv. Me6-TREN
1 + 2 equiv. dppe
1
99 / 2:1
forming compound 2. Next, a substrate molecule inserts into
the Al–P bond. Subsequent protonolysis by a second
equivalent of HPPh2 accesses the hydrophosphinated
1
99 / 5 : 1
1
99 / 19:1
product whilst regenerating the active catalytic species 2.
Since after the facile room temperature deprotonation
step, alkyne insertion and protonolysis are the other key
3
95 / 10:1
5
95 / 16:1
steps, we performed
a deuterium labelling study to
investigate the cycle further. Catalytic hydrophosphination
between PhC≡CPh and DPPh2 favoured formation of the E-
stereoisomer and deuterium was incorporated into the vinyl
iBu3AlPPh2Li(THF)3
2 + 3 equiv. PPh3
1
99 / 10:1
1.5
99 / 4:1
General conditions: 0.6 mmol substrate, 0.5 mmol HPPh2, d8-toluene, 10
mol% [Al] catalyst, 110 °C. Conversions based against 1H NMR internal
standard hexamethylcyclotrisiloxane. E/Z/α stereoselectivity based on 31P
NMR spectra.
respectively, using 2 mol% Ca(HMDS)2 as catalyst, also at
room temperature. Significantly longer reaction times were
seen when using DIPPNacNacCa(HMDS)(THF) as a catalyst
(1.5 mol%) (iPr, 6 h, 99%: Cy, 28 h, 85%).[6d] KHMDS is also
found to be a good catalyst for carbodiimides requiring low
catalyst loadings and short reaction times,[6a] while a sodium
magnesiate
carbodiimides.[6f]
Attempting to pinpoint the active catalyst, four
compounds LiPPh2 ), iBu3Al ), iBu2AlH and
iBu2AlPPh2 (
),[17] were screened for catalytic viability using
also
catalyses
hydrophosphination
of
(
3
(
4
(5)
Scheme 1. Proposed reaction mechanism for hydrophosphination of
diphenylacetylene by pre-catalyst 1, showing formation of active species 2.
6
PhC≡CH as a model substrate (Table 1). Using LiPPh2 as a
catalyst yields 86% conversion to the vinyl phosphine after
20 h, with anti-Markovnikov regioselectivity, similar to that of
2
phosphine product, Ph(Ph2P)C=C(D)Ph, as confirmed by H
NMR spectra and GC-MS (see ESI). Also, in a stoichiometric
reaction between [iBu3AlHLi]2 and DPPh2, HD was detected
2
. Compounds
stereo-selectivities as well as lower yields for PhC≡CH
hydrophosphination. Interestingly when is used as catalyst
4-6 afford different product regio- and
1
1
in the H NMR spectrum (triplet at 4.45 ppm, J = 42.8 Hz),
confirming the initial deprotonation step.
4
(72%; 1:8:8 E/Z/α) the major isomer products are the Z- anti-
Markovnikov isomer, and equally the Markovnikov [α –
A kinetic isotope effect experiment (KIE) was conducted
for hydrophosphination of diphenylacetylene by recording
the reaction profile in duplicate for HPPh2 and DPPh2 at
isomer; Ph(Ph2P)C=CH2]. In contrast,
2 does not give any
appreciable α–isomer, suggesting is not disproportionating
2
100 °C, in d8-toluene, with 10 mol% of 2. By monitoring the
in solution at high temperature into LiPPh2 and iBu3Al. In
order to ascertain whether LiPPh2 is implicated in the
catalytic profile we conducted another stoichiometric reaction
consumption rate of phosphine by 31P NMR, rates were
obtained, and in each case the overall reaction rate is
pseudo–first order. From these rates a KIE of 1.38 ± 0.13
was determined (see ESI). This is a small value, compared
with other literature reports, and suggest that cleavage of the
(see ESI). Monitoring reaction of
toluene, by 31P NMR shows that after 2 h at 110 °C full
consumption of occurs with concomitant growth of two new
signals at 3.4 and -16.5 ppm. Unable to isolate these species
2 and PhC≡CPh, in d8-
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
a)
1b.2)
c)
1.2
1
1
0.8
0.6
0.4
0.2
0
1
0.8
0.6
0.4
0.2
0
0.8
0.6
0.4
0.2
0
0
1000 2000 3000 4000 5000
Σ[HPPh2]-1Δt
0
500
1000
1500
2000
0
100
200
300
400
t[cat]1
Σ[alkyne]1Δt
5 mol% 7.5 mol% 10 mol% 12.5 mol%
1M HPPh2
0.9M HPPh2 0.8M HPPh2
1 M alkyne 1.2 M alkyne 1.3 M alkyne 1.4 M alkyne
1.1M HPPh2 0.7M HPPh2
Figure 3. Variable Time Normalisation Analysis (VTNA) plots illustrating the order in a) catalyst (first order); b) phosphine (inverse first order); c) diphenylacetylene
(first order).
P–H bond is only involved to a minor extent in the rate
determining step.[7b] This also indicates that alkyne insertion
barrier for the formation of the catalyst was challenging to
isolate as a result of the complex potential energy surface
associated with the large dimer species. However, a bond
scan along the coordinate associated with the formation of
H2 provided an indicative barrier (∆E*) of ~38 kcal/mol, which
would be achievable under the reaction conditions and lead
into
2
is rate determining, which given the rather congested
and bulky nature of the alkyne is unsurprising.
structure of
2
Next, we conducted a kinetic analysis of the reaction
using the Variable Time Normalisation Analysis (VTNA)
method reported by Burés, allowing us to obtain valuable
mechanistic detail under synthetically relevant conditions to
three half-lives (see Fig. 3 and ESI).[18] The reaction order
with respect to [catalyst] was determined by conducting
reactions using different catalyst concentrations, while
keeping [alkyne] and [phosphine] constant. These data
showed the reaction rate increases with increasing [catalyst],
and that the order in catalyst is 1. This situation is consistent
irreversibly to
contrast to the induction step, the first step in the catalytic
cycle (adding diphenylacetylene to ) is mildly endergonic for
2 given the exothermic nature of this step. In
2
both the E and Z isomers of the intermediate shown in
Scheme 1. However, the E isomer is more stable in the
intermediate state of the reaction (∆G = 6.9 kcal/mol), with
the Z-isomer (∆G = 8.3 kcal/mol) being further destabilised
by 1.4 kcal/mol, relative to the E-isomer. Finally, generation
of the product and reformation of the catalytic species occurs
in an exergonic reaction. In this step, the formation of the Z-
isomer (∆G = -25.3 kcal/mol) is favoured over the E-isomer
(∆G = -20.4 kcal/mol). The reversal of the relative stabilities
of the isomers in the intermediate state versus the product
state suggests that the formation of the intermediate is
deterministic for the final product distribution, which favours
the experimentally determined E-isomer. The rate-limiting
step for the reaction could not be located as the calculation
of transition states proved elusive for these bulky
compounds. However, the relative stabilities of the
intermediates and products determined for this pathway
indicate that the mechanism proposed is achievable under
the reaction conditions employed.
with the reaction proceeding via
a monomeric rate
determining step during the reaction. Variation of [phosphine]
under synthetically relevant conditions revealed that
increasing concentration of phosphine inhibits the reaction,
giving a phosphine order of -1. This inhibition likely results
from pre-coordination of the phosphine, blocking off the
alkyne for insertion. Also, we have already established that
bulky mono- and bidentate phosphines slow down reactivity
in our Lewis donor study (vide supra). Lastly, variation of
[alkyne] revealed a first order dependence in [alkyne],
indicating that the alkyne is involved in the rate limiting step.
Finally, in order to reinforce our experimental insight, we
turned to DFT calculations. Run on the full system with the
internal alkyne, diphenylacetylene, used as the model
substrate, the calculations were performed at the B3LYP-
D3/[19] 6-311G(d,p)[20] level of theory employing a continuum
solvent with the dielectric constant of toluene within the
Conclusion
IEFPCM model.[21] The relative stability of the formation of
from [iBu3AlHLi]2 with 2 HPPh2 and 6 THF molecules was
initially investigated. Formation of the catalyst is
thermodynamically favourable despite the entropic penalty
associated with THF coordination, with calculated
∆G = -63.8 kcal/mol (∆H = -130.8 kcal/mol). The activation
2
Previously lithium aluminates have been shown to be active
catalysts for hydroboration of aldehydes, ketones, imines and
acetylenes. This new study extends the catalytic chemistry of
these bimetallic main group compounds by reporting the first
example of Al-catalysed hydrophosphination of alkynes, alkenes
(2)
a
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
and carbodiimides, using the lithium aluminate (pre)catalyst
[iBu3AlHLi]2. A mechanism is proposed for the alkyne catalysis,
elucidated by stoichiometric reactions, thought to proceed via
formation of the crystallographically defined lithium aluminium
phosphide, iBu3AlPPh2Li(THF)3, 2, followed by insertion of the
alkyne into the Al-P bond, then protonolysis of a second
equivalent of the phosphine to generate the vinylphosphine
product and regenerate the catalyst. While intuitively the
formation of an anionic aluminium centre saturated by four anionic
ligands as in an aluminate might be expected to have insufficient
Lewis acidity to engage in hydrophosphination processes, it is
clear from the different results obtained using a number of Lewis
donor solvent molecules that the presence of the lithium helps to
circumvent this apparent handicap so pointing to bimetallic
synergistic behaviour.
Acknowledgements
Thanks go to Dr Alberto Hernán-Gómez and Dr David Nelson for
insightful comments on the kinetic data, and to the EPSRC (DTP
award EP/M508159 to V. A. P.) for funding. Computational results
were obtained using the EPSRC-funded ARCHIE-WeSt High
Performance Computer (www.archie-west.ac.uk) (EPSRC grant
no. EP/K000586/1). The data set underlying this research can be
located at https://doi.org/10.15129/b7fc7e13-84f5-430c-b3ef-
1e3ba380c003.
Keywords:
Aluminate
•
Homogeneous
Catalysis
•
Hydrophosphination • Lithium • Phosphine
[1]
[2]
K. B. Dillon, F. Mathey, J. F. Nixon, Phosphorus: The carbon copy, John
Wiley and Sons, Chichester, 1998.
a) D. S. Glueck, in C-X Bond Formation (Ed.: A. Vigalok), Springer Berlin
Heidelberg, Berlin, Heidelberg, 2010, pp. 65-100; b) V. Koshti, S.
Gaikwad, S. H. Chikkali, Coord. Chem. Rev. 2014, 265, 52-73; c) C. A.
Bange, R. Waterman, Chem. -Eur. J. 2016, 22, 12598-12605.
a) H. Hu, C. Cui, Organometallics 2012, 31, 1208-1211; b) B. Liu, T.
Roisnel, J.-F. Carpentier, Y. Sarazin, Chem. -Eur. J. 2013, 19, 13445-
13462; c) M. R. Douglass, T. J. Marks, J. Am. Chem. Soc. 2000, 122,
1824-1825.
Experimental Section
Full experimental characterisation and synthetic procedures are described
in the supporting information.
[3]
Synthesis of iBu3AlPPh2Li(THF)3, 2: method a) To a stirred solution of
[iBu3AlHLi]2 (0.412 g; 1 mmol) in hexane (10 mL) was added HPPh2
(0.34 mL; 2 mmol) and the reaction stirred 1 h. THF (0.5 mL; 6 mmol) was
added then the volatiles were removed. The residue was taken up in
hexane (5 mL) and toluene (1 mL). Subsequent cooling to – 30 °C yielded
the desired product as pale-yellow crystals. Crystalline yield 0.494 g;
0.82 mmol; 41 %. method b) To a stirred solution of HPPh2 (0.17 mL; 1
mmol) in hexane (5 mL) was added dropwise nBuLi (0.63 mL; 1.6
M/hexane; 1 mmol) and the resulting bright yellow suspension stirred for 1
h. Addition of iBu3Al (1 mL; 1 M/hexane; 1 mmol) generated a clear pale-
yellow solution, which was stirred for 1 h. THF (0.3 mL; 3 mmol) was added
and the pale-yellow solution cooled at – 30 °C overnight. Crystalline yield
0.150 g; 0.25 mmol; 24%. 1H NMR (400.1 MHz, d8-toluene, 300 K): δ 0.48
(d of d, J = 6.93 Hz, 2.88 Hz, 6H, iBu CH2); 1.31 (d, J = 6.29 Hz, 18H, iBu
CH3); 1.41 (m, 12H, THF CH2); 2.26 (m, 3H, iBu CH); 3.44 (m, 12H, THF
CH2); 7.00 (m, 2H [overlapping solvent], Ph); 7.17 (m, 4H [overlapping
solvent], Ph); 7.14 (m, 4H, Ph) ppm. 31P NMR (104.2 MHz, d8-toluene,
300 K): δ - 49.2 ppm. 13C{1H} NMR (151 MHz, d8-toluene, 300 K): δ 25.4
(THF CH2); 25.5 + 25.6 (iBu CH2); 28.3 (iBu CH); 29.5 (iBu CH3); 68.6
(THF CH2); 124.9 (Ar C–H); 127.3 (d, J = 6.17 Hz, Ar C–H); 134.0 (d, J =
13.04 Hz, Ar C–H); 144.1 (d, J = 13.08 Hz, ipso Ar) ppm. 7Li NMR
(155.5 MHz, d8-toluene, 300 K): δ 0.21 (s) ppm. 27Al NMR: no signal was
observed.
[4]
[5]
[6]
Y. Moglie, M. J. González-Soria, I. Martín-García, G. Radivoy, F. Alonso,
Green Chem. 2016, 18, 4896-4907.
a) P. P. Power, Nature 2010, 463, 171; b) M. S. Hill, D. J. Liptrot, C.
Weetman, Chem. Soc. Rev. 2016, 45, 972-988.
a) W.-X. Zhang, M. Nishiura, Z. Hou, Chem. Commun. 2006, 3812-3814;
b) N. T. Coles, M. F. Mahon, R. L. Webster, Chem. Commun. 2018, 54,
10443-10446; c) M. R. Crimmin, A. G. M. Barrett, M. S. Hill, P. B.
Hitchcock, P. A. Procopiou, Organometallics 2007, 26, 2953-2956; d) M.
R. Crimmin, A. G. M. Barrett, M. S. Hill, P. B. Hitchcock, P. A. Procopiou,
Organometallics 2008, 27, 497-499; e) T. M. A. Al-Shboul, H. Görls, M.
Westerhausen, Inorg. Chem. Commun. 2008, 11, 1419-1421; f) M. De
Tullio, A. Hernán-Gómez, Z. Livingstone, W. Clegg, A. R. Kennedy, R.
W. Harrington, A. Antiñolo, A. Martínez, F. Carrillo-Hermosilla, E. Hevia,
Chem. -Eur. J. 2016, 22, 17646-17656; g) J. Pahl, T. E. Stennett, M.
Volland, D. M. Guldi, S. Harder, Chem. -Eur. J. 2019, 25, 2025-2034.
a) K. A. Erickson, L. S. H. Dixon, D. S. Wright, R. Waterman, Inorganica
Chim. Acta 2014, 422, 141-145; b) J. P. W. Stelmach, C. A. Bange, R.
Waterman, Dalton Trans. 2016, 45, 6204-6209.
[7]
[8]
[9]
R. L. Melen, Chem. Soc. Rev. 2016, 45, 775-788.
a) W. Li, X. Ma, M. G. Walawalkar, Z. Yang, H. W. Roesky, Coord. Chem.
Rev. 2017, 350, 14-29; b) G. I. Nikonov, ACS Catal. 2017, 7, 7257-7266;
c) S. Dagorne, R. Wehmschulte, ChemCatChem 2018, 10, 2509-2520;
d) C. Weetman, S. Inoue, ChemCatChem 2018, 10, 4213-4228.
General Catalytic Reaction: The desired catalyst loading was added to
0.5 mL of d8-toluene solution (unless alternative solvent specified)
containing the substrate precursor (0.6 mmol) and HPPh2 (0.5 mmol,
0.09 mL). The reaction mixture was transferred to a sealed J. Young’s tap
NMR tube and the reaction was regularly monitored by 1H and 31P NMR
spectroscopy until the formation of the products was completed as
determined by integration versus an internal capillary standard
[10] a) Z. Yang, M. Zhong, X. Ma, S. De, C. Anusha, P. Parameswaran, H.
W. Roesky, Angew. Chem. Int. Ed. 2015, 54, 10225-10229; b) Z. Yang,
M. Zhong, X. Ma, K. Nijesh, S. De, P. Parameswaran, H. W. Roesky, J.
Am. Chem. Soc. 2016, 138, 2548-2551; c) R. J. Less, H. R. Simmonds,
D. S. Wright, Dalton Trans. 2014, 43, 5785-5792; d) A. Bismuto, M. J.
Cowley, S. P. Thomas, ACS Catal. 2018, 8, 2001-2005; e) A. Bismuto,
S. P. Thomas, M. J. Cowley, Angew. Chem. Int. Ed. 2016, 55, 15356-
15359; f) H. Elsen, C. Färber, G. Ballmann, S. Harder, Angew. Chem. Int.
Ed. 2018, 57, 7156-7160; g) J. A. Hatnean, J. W. Thomson, P. A. Chase,
D. W. Stephan, Chem. Commun. 2014, 50, 301-303.
(hexamethylcyclotrisiloxane).
For
alkynes
and
alkenes,
the
hydrophosphination catalysis was performed at 110 °C with 10 mol% [Al]
catalyst loading. For carbodiimides the hydrophosphination catalysis was
performed at room temperature with 5 mol% [Al] catalyst loading. The
yields reported are based on 1H NMR and 31P relative to the internal
standard. In all cases, the bulk of the NMR solution can be attributed to
either product compounds or starting materials. Isolated yields are
provided for example substrates, isolated via either recrystallization
methods or column chromatography, as reported in the ESI.
[11] L. Keweloh, N. Aders, A. Hepp, D. Pleschka, E.-U. Würthwein, W. Uhl,
Dalton Trans. 2018, 47, 8402-8417.
[12] a) P. Merino, E. Marqués-López, R. P. Herrera, Adv. Synth. Catal. 2008,
350, 1195-1208; b) S. M. Härling, B. E. Fener, S. Krieck, H. Görls, M.
Westerhausen, Organometallics 2018, 37, 4380-4386.
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
[13] a) R. E. Mulvey, S. D. Robertson, Top Organomet. Chem. 2014, 47, 129-
158; b) S. D. Robertson, M. Uzelac, R. E. Mulvey, Chem Rev 2019.
[14] a) L. E. Lemmerz, R. McLellan, N. R. Judge, A. R. Kennedy, S. A. Orr,
M. Uzelac, E. Hevia, S. D. Robertson, J. Okuda, R. E. Mulvey, Chem. -
Eur. J. 2018, 24, 9940-9948; b) V. A. Pollard, M. A. Fuentes, A. R.
Kennedy, R. McLellan, R. E. Mulvey, Angew. Chem. Int. Ed. 2018, 57,
10651-10655; c) V. A. Pollard, S. A. Orr, R. McLellan, A. R. Kennedy, E.
Hevia, R. E. Mulvey, Chem. Commun. 2018, 54, 1233-1236.
[15] R. Neufeld, D. Stalke, Chem. Sci. 2015, 6, 3354-3364.
[16] a) S. G. Thangavelu, K. E. Hocker, S. R. Cooke, C. N. Muhoro, J.
Organomet. Chem. 2008, 693, 562-566; b) B. R. Nichols, N. G.
Akhmedov, J. L. Petersen, B. V. Popp, Dalton Trans. 2018, 47, 8456-
8465.
[17] S. A. Sangokoya, W. T. Pennington, G. H. Robinson, D. C. Hrncir, J.
Organomet. Chem. 1990, 385, 23-31.
[18] a) J. Burés, Angew. Chem. Int. Ed. 2016, 55, 2028-2031; b) J. Burés,
Angew. Chem. Int. Ed. 2016, 55, 16084-16087; c) C. D. T. Nielsen, J.
Burés, Chem. Sci. 2019, 10, 348-353.
[19] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098-3100; b) A. D. Becke, J.
Chem. Phys. 1993, 98, 5648-5652; c) C. Lee, W. Yang, R. G. Parr, Phys.
Rev. B 1988, 37, 785-789; d) S. H. Vosko, L. Wilk, M. Nusair, Can. J.
Phys. 1980, 58, 1200-1211; e) P. J. Stephens, F. J. Devlin, C. F.
Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98, 11623-11627; f) R.
H. Hertwig, W. Koch, Chem. Phys. Lett. 1997, 268, 345-351; g) S.
Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[20] a) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639-5648;
b) R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys.
1980, 72, 650-654; c) M. J. Frisch, J. A. Pople, J. S. Binkley, J. Chem.
Phys. 1984, 80, 3265-3269.
[21] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999-3094.
This article is protected by copyright. All rights reserved.

10.1002/anie.201906807
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents (Please choose one layout)
Layout 1:
RESEARCH ARTICLE
Text here
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
Layout 2:
RESEARCH ARTICLE
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
Alluring Aluminium: catalytic hydrophosphination of unsaturated organic
substrates is usually associated with transition metals, but here aluminium is found
to be an attractive catalytic partner
This article is protected by copyright. All rights reserved.