P-Stereogenic monophosphines with the 2-p-terphenylyl and 1-pyrenyl substituents. Application to Pd and Ru asymmetric catalysis

Article abstract of DOI:10.1016/j.molcata.2014.04.026

The synthesis of five optically pure P-stereogenic monophosphines of the type PPhArR (Ar = 2-p-terphenylyl (a), 1-pyrenyl (b); R = OMe, Me, i-Pr) is described. The ligands were fully characterised and the absolute configurations of PPh(1-pyrenyl)R (3b and 5b; R = OMe and Me respectively) were confirmed by X-ray diffraction. The complexation of the monophosphines to Pd and Ru organometallic units yielded the neutral complexes [PdCl(η³-2-Me- allyl)P] (10-12) and [RuCl₂(η⁶-p-cymene)P] (16-18). Complete characterisation, including the crystal structure determination of [RuCl₂(η⁶-p-cymene)(PMePh(2-p-terphenyl))] (17a) is provided. Neutral palladium complexes appeared as mixtures of two diastereomers in solution according to NMR. The synthesis and characterisation of four cationic [Pd(η³-2-Me-allyl)(P)₂]PF₆ (13 and 14) is also described. The application of neutral Pd complexes to catalytic styrene hydrovinylation afforded moderate conversions, high chemoselectivities (>92%) to 3-phenyl-1-butene and up to 43% ee with precursor 12a. Cationic Pd complexes were tested as catalytic precursors in allylic substitution of rac-3-acetoxy-1,3-diphenyl-1-propene (rac-I), with the anion of dimethylmalonate and benzylamine as nucleophiles, obtaining full conversions and up to 80% ee in alkylation and 60% ee in amination with precursor 13a. Finally, ruthenium complexes were used as catalytic precursors in transfer hydrogenation of acetophenone, with complete conversions after several hours but low enantioselectivities.

Full text of DOI:10.1016/j.molcata.2014.04.026

Accepted Manuscript
Title: P-Stereogenic Monophosphines with the
2-p-Terphenylyl and 1-Pyrenyl Substituents. Application to Pd
and Ru Asymmetric Catalysis
`
Author: Pau Clavero Arnald Grabulosa Merce Font-Bardia
Guillermo Muller
PII:
S1381-1169(14)00164-2
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2014.04.026
MOLCAA 9088
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
28-2-2014
14-4-2014
18-4-2014
Please cite this article as: P. Clavero, A. Grabulosa, M. Font-Bardia, G. Muller, P-
Stereogenic Monophosphines with the 2-p-Terphenylyl and 1-Pyrenyl Substituents.
Application to Pd and Ru Asymmetric Catalysis, Journal of Molecular Catalysis A:
Chemical (2014), http://dx.doi.org/10.1016/j.molcata.2014.04.026
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

P-Stereogenic Monophosphines with the 2-p-Terphenylyl and 1-Pyrenyl
Substituents. Application to Pd and Ru Asymmetric Catalysis
Pau Claveroa, Arnald Grabulosaa,*, Mercè Font-Bardiab, Guillermo Mullera
aDepartament de Química Inorgànica, Universitat de Barcelona, Martí i Franquès, 1-11, E-08028,
Barcelona, Spain
bDepartament de Cristal·lografia, Mineralogia i Dipòsits Minerals, Universitat de Barcelona, Martí i
Franquès, s/n, E-08028, Barcelona, Spain
*Corresponding author: Tel.: +34 934039759; fax +34 934907725. E-mail address:
arnald.grabulosa@qi.ub.es (A. Grabulosa).
Highlights
-
-
-
-
-
P-stereogenic monophosphines with a polycyclic aromatic substituent
Neutral and cationic allylic Pd complexes: [PdCl(η3-allyl)P], [Pd(η3-allyl)P2]PF6
Neutral Ru complexes: [RuCl2(η6-p-cymene)P]
Pd-catalysed hydrovinylation of styrene and allylic substitution reactions
Ru-catalysed hydrogen transfer of acetophenone
Abstract
The synthesis of five optically pure P-stereogenic monophosphines of the type PPhArR (Ar = 2-
p-terphenylyl (a), 1-pyrenyl (b); R = OMe, Me, i-Pr) is described. The ligands were fully
characterised and the absolute configurations of PPh(1-pyrenyl)R (3b and 5b; R = OMe and Me
respectively) were confirmed by X-ray diffraction. The complexation of the monophosphines to
Pd and Ru organometallic units yielded the neutral complexes [PdCl(η3-2-Me-allyl)P] (10-12)
and [RuCl2(η6-p-cymene)P] (16-18). Complete characterisation, including the crystal structure
determination of [RuCl2(η6-p-cymene)(PMePh(2-p-terphenyl))] (17a) is provided. Neutral
palladium complexes appeared as mixtures of two diastereomers in solution according to NMR.
The synthesis and characterisation of four cationic [Pd(η3-2-Me-allyl)(P)2]PF6 (13 and 14) is
1
Page 1 of 22

also described. The application of neutral Pd complexes to catalytic styrene hydrovinylation
afforded moderate conversions, high chemoselectivities (> 92%) to 3-phenyl-1-butene and up to
43% ee with precursor 12a. Cationic Pd complexes were tested as catalytic precursors in allylic
substitution of rac-3-acetoxy-1,3-diphenyl-1-propene (rac-I), with the anion of
dimethylmalonate and benzylamine as nucleophiles, obtaining full conversions and up to 80%
ee in alkylation and 60% ee in amination with precursor 13a. Finally, ruthenium complexes
were used as catalytic precursors in transfer hydrogenation of acetophenone, with complete
conversions after several hours but low enantioselectivities.
Keywords
P-stereogenic ligands; Pd allylic complexes; Ru arene complexes; hydrovinylation; allylic
substitution; transfer hydrogenation
1. Introduction
The development of new ligands for enantioselective catalysis with late transition metals has
been the main driving force for the synthesis of new chiral P(III) compounds for the last
decades, since phosphines and derivatives are known to be ideal ligands to some of the most
catalytically active metals [1]. It is well known that no ligand is ever efficient enough for all the
substrates in a particular reaction catalysed by a certain metal. Therefore, optimisation by fine-
tuning the steric and electronic properties of the ligands is usually carried out. At present, all
sources of chirality are intensively explored and hence a plethora of phosphorus ligands
containing stereogenic carbon atoms, axes and planes have been described [1, 2].
Given the configurational stability of the phosphorus atoms it has been known for over a
century [3] that the P atom itself can be a stereogenic element, giving rise to the family of P-
stereogenic ligands [4]. This type of ligands has a very long history in enantioselective
homogeneous catalysis, which started in the pioneering works of Horner [5] and Knowles [6, 7]
in Rh-catalysed enantioselective hydrogenation. More recently, the interest in P-stereogenic
ligands has been rising steadily thanks to new synthetic methodologies and applications in
catalysis [4, 8, 9].
2
Page 2 of 22

We also devoted some effort to this field with the preparation of series of mono- and
diphosphine ligands, which were used in Pd-catalysed reactions (hydrovinylation [10-13] and
allylic alkylation [14]) and in Ru-catalysed (cyclopropanation [15] and hydrogen transfer [15,
16]) enantioselective reactions. Although the optically pure monophosphine ligands were
initially obtained by resolution of the racemic mixtures by means of Pd complexes [10, 11], they
have more recently obtained by methods based on asymmetric synthesis due to its superior
versatility. Therefore, P-ligands obtained by the so-called Evans [17] and the Jugé-Stephan [18]
methods have been reported. In particular, series of ligands prepared by the Jugé-Stephan
method with the general structure depicted in Figure 1 were described [12, 14].
P
R
Ar
R = OMe, Me, i-Pr
Ar =
a
b
References 12, 14, 15
This work
PF6
Ru
R
Pd
Pd
R
R
Cl
Cl
P
P
P
R
Cl
P
Ar
Ar
Ar
Ar
Hydrovinylation
Allylic substitution
Hydrogen transfer
Figure 1. P-stereogenic monophosphines and their derived Pd and Ru complexes.
The monophosphine ligands, which contained a phenyl, a polycyclic aromatic group and a
methoxy or an alkyl group were complexed to Pd and Ru organometallic moieties. The obtained
complexes shown in the Figure were employed to the catalytic reactions mentioned before.
Some of the phosphines, in particular those bearing the 2-biphenylyl group, were found to be
competent ligands [12, 14, 16].
In this paper, the extension of these studies with new phosphines bearing the 2-p-terphenylyl (a)
and 1-pyrenyl (b) groups, rarely found in the literature, chosen with the aim of improving the
3
Page 3 of 22

performance of the previous generation of ligands is reported. To the best of our knowledge, the
only monophosphine reported to date that contains the unsubstituted 2-p-terphenylyl group is
the
racemic
phenyl(2-p-terphenylyl)phosphine
oxide
[19].
Several
achiral
hydroxyterphenylylphosphines have been described and successfully used as ligands in Pd-
catalysed cross-coupling reactions [20-22] and also extremely bulky phosphines containing a
2,3,4,5-tetraphenyl moiety have been reported and used in several Pd-catalysed reactions [23,
24]. Finally, a few achiral diphosphines, built around the terphenylyl skeleton, are also known
[25-28]. Regarding the 1-pyrenyl substituent, 1-diphenylphosphinopyrene has been described
[29].
2. Results and discussion
2.1. Ligand synthesis
From our previous studies [12, 13, 30] and from others [31, 32], it is known that standard Jugé-
Stephan [18] procedure is very well suited to prepare P-stereogenic phosphines bearing o-
monosubstituted aryl fragments. The introduction of the o-monosubstituted aryl fragment it is
carried out via organolithiums and therefore it is mandatory to have the suitable precursors to
prepare such species. In the present case, the most obvious precursors are the corresponding aryl
bromides 2-bromo-p-terphenyl (Bra) and 1-bromopyrene (Brb), which can be easily lithiated
with n-butyllithium at low temperatures (Scheme 1).
Scheme 1. Synthesis of Bra, Brb, Lia and Lib.
4
Page 4 of 22

Following a literature precedent [33], the preparation of Bra was carried out by Suzuki-Miyaura
coupling between p-biphenylylboronic acid and o-bromoiodobenzene, using [Pd2(dba)3]/PPh3 as
catalytic precursor and aqueous Na2CO3 solution as a base. After some optimisation, Bra could
be obtained as a crystalline solid in good yield. Compound Brb was also prepared with a
modified literature method [34], based on the direct bromination of pyrene with NBS in DMF.
This method is very selective towards the monobromination in position 1 and Brb could be
obtained in high yield with enough purity to be employed in the synthesis of the phosphines.
With the bromoarene precursors in hand, the Jugé-Stephan methodology was followed to have
access to the desired ligands (Scheme 2).
Scheme 2. Synthesis of the P-stereogenic ligands.
The regioselective ring opening of oxazaphospholidine-borane 1 was initially attempted
preparing the organolithiums Lia and Lib in Et2O at –78 ºC as in previous reports [12, 35]. It
was found, however, that the very low solubility of these species, probably due to the extensive
conjugation of the aryl groups employed, led to low yields of 2 and the recovery of large
quantities of unreacted 1. Therefore, Et2O was switched to THF, in which the organolithium did
not precipitate. With this modification, 1 was completely consumed and good yields of 2 could
be obtained. The ring opening was very clean with organolithium Lib but, in contrast, Lia led to
several byproducts as judged by 31P spectroscopy. Phosphamide-borane 2a featured a broad
singlet at 70.2 ppm and another two peaks were also present at 133 and 107 ppm. The signal at
higher field could be assigned to the product 9, originated by base-promoted elimination on 1
instead of the expected substitution at the P atom (Scheme 3).
5
Page 5 of 22

Scheme 3. Reaction of 1 with Lia.
This elimination reaction has been described previously in the literature [36] when bulky
organolithium reagents are employed. In the case described, it also leads to 1,4-
diphenylbenzene, which could indeed be recovered during the purification of 2a. It was found
that this compound and the other unidentified one reacted in the subsequent reactions and the
formed byproducts could not be separated from the desired intermediates. Therefore, the
optimisation of the ring-opening step was mandatory. After some experimentation, it was
discovered that by simply carrying out the reaction in more diluted conditions was enough to
maximise the formation of the desired phosphamide-borane 2a and minimise the quantities of
impurities to an acceptable level.
Following the standard procedure, acid-mediated methanolysis of 2 produced the corresponding
phosphinite-boranes 3 in good yields. Whereas 3a was a dense colourless oil, 3b was a yellow
solid that precipitated as a pure product during the methanolysis step.
Solutions of phosphinite boranes 3 in Et2O were treated at low temperature with RLi (R = Me, i-
Pr and t-Bu) to obtain monophosphine-boranes 5 and 6. As expected [12] the bulky t-Bu group
could no be introduced according to 31P NMR spectroscopy. Depending on the exact reaction
conditions, the peak of the starting phosphinite-boranes 3 (at δ around +110 ppm) and/or
multiple peaks indicating decomposition could be detected. The less bulky i-Pr group could be
successfully introduced by reaction of i-PrLi with 3a but once again starting phosphinite-borane
and/or multiple peaks were observed when 3b was employed. These results are in line with
previously published data [12] indicating that the introduction of a conjugated aromatic
substituent in the ring-opening step of the Jugé-Stephan method precludes the introduction of
the t-Bu group and sometimes even the i-Pr group. The protected phosphinite and phosphine
6
Page 6 of 22

ligands were air-stable oils (for Ar = 2-p-terphenylyl) or solids (for Ar = 1-pyrenyl), which were
characterised by the usual techniques. HPLC analysis on a chiral column of the crude products
showed that they were present as optically pure (ee > 95%) products. Single crystals, suitable to
perform X-ray diffraction studies, were obtained for 3b and 5b. The molecular structures for the
two compounds are displayed in Figure 2.
7
Page 7 of 22

Figure 2. ORTEP representation of the molecular structures of 3b (left) and 5b (right). Hydrogen atoms
have been omitted for clarity. Selected distances (Å) and angles (º) for 3b: P1–O1, 1.6005(14); P1–C17,
1.7934(19); P1–C1, 1.8081(19); P1–B1, 1.899(2); C23–O1, 1.443(2); B1–P1–O1, 115.54(10); B1–P1–
C17, 110.06(10); B1–P1–C1, 113.04(10). For 5b: P1–C23, 1.804(3); P1–C17, 1.819(2); P1–C1, 1.816(2);
P1–B1, 1.930(3); B1–P1–C23, 109.62(13); B1–P1–C17, 115.55(12); B1–P1–C1, 115.22(13).
The crystal structures confirmed that the absolute configurations of the P atom were R for 3b
and S for 5b, as expected given the enantiomer of ephedrine employed to prepare 3 [(1R, 2S)-(–
)-ephedrine] and the stereochemical course of each of the Jugé-Stephan method steps [18].
Distances and angles of both borane adducts were similar to other previously reported
phosphine-boranes [12, 16, 37]. A parameter that allows the evaluation of the steric hindrance
of the ligand is the mean value of the three BPC angles. The smaller the value, the higher the
steric hindrance of the ligand. In the present case, the values are 112.88 for 3b and 113.46 for
5b, meaning that the former is slightly more bulky than the latter. Both values are in the range
observed for other P-stereogenic phosphine-boranes [12, 37]. Borane adducts 3, 5 and 6 were
conveniently deprotected by neat morpholine at room temperature [12, 37], affording the free
phosphinites (4a and 4b) and phosphines (7a, 7b, and 8a) ready for complexation.
2.2. Neutral Pd complexes
8
Page 8 of 22

Neutral Pd allylic complexes of the type [PdCl(η3-(2-Methylallyl)(P)] (10-12), were easily
prepared by the standard method, consisting of splitting Pd dimer D1 with two equivalents of
the ligands in dichloromethane at room temperature (Scheme 4) [12].
Scheme 4. Synthesis of complexes 10-12.
The complexes were obtained as air-stable pale yellow solids. The analysis of the NMR spectra
in CDCl3 indicated that, as expected, they were present as mixtures of two diastereomeric
species, because of the presence of one chiral phosphine and the allyl moiety. Accordingly, two
sharp singlets were present in the 31P NMR spectra and duplication of the allylic peaks was
clearly observable in the 1H NMR spectra. It has to be pointed out that the presence of the chiral
phosphine ruled out the presence of any symmetry element and hence in principle H and C
atoms of the each different isomer were all different. The proportion of each isomer is indicative
of the discrimination ability of the chiral ligand in the neutral complexes. The isomeric ratio
was 1:1 for complexes bearing the phosphines with the 1-pyrenyl group (10b and 11b) and for
complex 10a. In contrast, it was around 1:2 for complexes 11a and 12a, similarly to complexes
with phosphines containing the 2-biphenylyl group [12].
2.3. Cationic Pd complexes
Cationic Pd allylic complexes of the type [Pd(η3-(2-Methylallyl)(P)2]PF6 (13-15), were easily
prepared following a previously reported method [14]. Thus, dimer D1 dissolved in
dichloromethane was treated with four equivalents of monophosphine in the presence of excess
of ammonium hexafluorophosphate (Scheme 5).
9
Page 9 of 22

Scheme 5. Synthesis of complexes 13-15.
An aqueous final work-up allowed the isolation of the desired cationic complexes as air-stable
solids. In solution they slowly decomposed to Pd black and other unknown species as 31P NMR
spectroscopy evidenced. This process was faster for complexes with phosphines containing the
2-p-terphenylyl group (13a and 14a). The complexes are present as single species in solution
but the chirality of the phosphine makes each phosphine and each half of the allyl group to
appear different in the NMR spectra. For that reason, two sharp doublets (JPP = 29-55 Hz) with
31
strong roof effect were observed in the P NMR spectra and four resonances assignable to the
allylic protons could be seen. Two of them, corresponding to the anti protons, appeared as sharp
doublets (JPP = 9-13 Hz) whereas the remaining two appeared as broad singlets and
corresponded to the two syn protons. In contrast to 13 and 14, solutions of complex 15a
displayed only two peaks in the 31P NMR spectrum and several peaks belonging to the i-Pr and
1
the allylic moieties in the H NMR spectrum, which could not be assigned. It is possible the
bulkiness of the phosphine 8a precludes the appropriate coordination of two ligands to stabilise
15a. This complex was not further used in catalysis.
2.4. Ru complexes
Similarly to the Pd complexes discussed above, neutral Ru complexes of the type [RuCl2(η6-(p-
cymene)(P)] (16-18), were prepared by splitting the well-known Ru dimer D2 with two
equivalents of the appropriate phosphorus ligand (Scheme 6) [15, 38].
10
Page 10 of 22

Scheme 6. Synthesis of complexes 16-18.
The desired complexes were obtained as air-stable red solids after recrystallisation. In solution
they were conveniently characterised by NMR. They featured a singlet in the 31P NMR spectra.
1
In H and 13C NMR spectra, the chirality of the phosphine made all the H and C atoms of the
coordinated aryl moiety of the p-cymene to appear differentiated as well as the two methyl
groups of the isopropyl substituent.
Suitable monocrystals to perform X-ray diffraction studies could be obtained for 17a. Its
ORTEP representation is displayed in Figure 3.
Figure 3. ORTEP representation of 17a. Thermal ellipsoids drawn at 50% probability level. Hydrogen
atoms have been omitted for clarity. Distances (Å) and angles (º): Ru(1)–Cl(1): 2.4115(8); Ru(1)–Cl(2):
2.4170(8); Ru(1)-P(1): 2.3615(6); P(1)–C(11): 1.818(3); P(1)–C(12): 1.818(3); P(1)–C(18): 1.841(3);
C(23)–C(24): 1.491(4); C(27)–C(30): 1.487(4); P(1)–Ru(1)–Cl(1): 82.84(3); P(1)–Ru(1)–Cl(2): 85.04(3);
Cl(1)–Ru(1)–Cl(2): 89.90(3).
The complex shows the typical “piano-stool” pseudotetrahedral geometry around the Ru atom
with the p-cymene ring coordinated in η6 fashion, making the Ru to adopt a distorted octahedral
11
Page 11 of 22

coordination environment. The structure allows an additional confirmation of the absolute
configuration of the phosphorus atom of the phosphine (S, as a free ligand). The geometric
parameters of the structure are similar to other previously reported [RuCl2(η6-p-cymene)(P)]
complexes [15, 16, 39]. As expected, the two Ru-Cl distances are very similar and that the
imaginary line defined by the two Cl atoms is approximately parallel to the line defined by the
substituted carbons of the p-cymene ring. The average distance between Ru and each of the six
aromatic carbon atoms of the p-cymene ring is 2.218 Å. These distances can be divided in two
groups: the relatively short ones (those involving the three C atoms closer to the phosphine,
average distance of 2.203 Å) and the relatively long ones (those involving the three C atoms
further away from the phosphine, average distance of 2.232 Å).
2.4. Asymmetric hydrovinylation
Catalytic hydrovinylation can be viewed as the heterocodimerisation between an activated
olefin and ethylene catalysed by a transition metal complex, usually based on Ni or Pd [40, 41].
The interest in this reaction is steadily increasing, especially regarding the asymmetric version
[41-43]. The model vinyl arene substrate used was styrene with neutral Pd complexes 10-12 as
catalytic precursors (Scheme 7).
Scheme 7. Styrene hydrovinylation.
The desired hydrovinylation product, 3-phenyl-1-butene, is a chiral compound amenable to
enantioselection. The Pd catalytic species, however, are also active in isomerising it to achiral
(E/Z)-3-phenyl-2-butenes. Other typically encountered byproducts in small amounts are those
arising from the dimerization of either ethylene (1-butene) or styrene (1,3-diphenyl-1-butenes).
The main challenge of the hydrovinylation reactions is the design of ligands capable of forming
12
Page 12 of 22

catalytic precursors leading to high chemo- and enantioselectivities of the hydrovinylation
product [41]. With Ni and Pd systems, the ligand, usually with P as a donor atom, has to be
monodentate due to the particularities of the catalytic cycle [41] a feature that makes more
challenging the design of efficient ligands.
Previous research carried out with monophosphines PArPhR (Ar = 1-naphthyl, 9-phenanthryl,
2-biphenylyl; R = OMe, Me, i-Pr) led to some selective catalysts towards 3-phenyl-1-butene in
the hydrovinylation of styrene [12, 13, 30]. In this paper, these results can be conveniently
compared to those obtained with the catalytic precursors based on the 2-p-terphenylyl and 1-
pyrenyl phosphines presented here. The results are given in Table 1.
Table 1. Styrene hydrovinylation catalysed with 10-12 complexes.
Entrya Precursor Time/h Conv.b/% Codimersc/%
Select.d/%
> 99.9
92.6
95.0
3.7
> 99.9
95.8
TOFe/h–1
15
ee/%
34 (R)
5 (S)
19 (S)
–
1
2
3
10a
10b
11a
11b
11b
12a
6
6
6
6
1
6
9.1
43.2
23.1
> 99.9
16.9
62.6
9.1
41.9
22.8
97.5
16.9
62.6
68
38
–
165
106
4
5
3 (S)
43 (S)
6f
aCatalytic conditions: 25 ºC, ≈ 15 bar of initial pressure of ethylene in 15 mL of CH2Cl2. Ratio styrene/Pd
b
c
d
= 1000/1; Conversion of starting styrene; Total amount of codimers; Percentage of 3-phenyl-1-
e
f
butene/codimers; TOF values calculated from the total amount of codimers formed; This result had
already been reported in a review [41].
As expected, all neutral allylic complexes are active in catalytic hydrovinylation with very
different activities depending on the substituents at the P atom, as clearly evidenced by the TOF
values at 6 h. The order of increasing activity is 10a < 11a < 11b < 12a < 11b. In fact, precursor
11b leads to a very fast catalytic system so in order to avoid extensive isomerisation of 3-
phenyl-1-butene the reaction time had to be shortened (cf. entries 4 and 5). The catalytic
systems are very selective towards the primary α-hydrovinylation product, namely 3-phenyl-1-
butene, which is higher than 92% at moderate conversions (entries 2 and 6). Regarding the
enantioselectivity, only the systems containing the 2-p-terphenylyl substituent in the phosphine
lead to some enantioenriched 3-phenyl-1-butene (entries 1, 3 and 6). As no secondary
interaction can be envisaged with this kind of ligands, the moderate enantioselectivities obtained
can be explained by pure steric differentiation [12]. In this regard, the linearity of the 2-p-
13
Page 13 of 22

terphenylyl group in the monophosphorus ligands is clearly better to the planarity of the 1-
pyrenyl substituent.
The comparison these results with previously published data with similar phosphines [12, 13,
30] show that overall the activities are somewhat lower for the phosphines discussed here. The
selectivities are maintained whereas the enantioselectivities of the precursors with phosphines
bearing the 2-p-terphenylyl group are very similar to those of the precursors with phosphines
bearing the 2-biphenylyl group [12, 30]. These results show that the introduction of an extra
phenyl ring at the end of the 2-biphenylyl moiety does not improve the performance of the
ligands, pointing out that probably it situates in a position too far away from the catalytically
active Pd centre to have any remarkable effect. The fact that the 1-pyrenyl-containing phosphine
ligands generate precursors that give to racemic 3-phenyl-1-butene (compared to up to the 22%
ee obtained with the 1-naphthyl-containing phosphines [12]), demonstrates that this substituent
is not adequate for the asymmetric hydrovinylation of styrene.
2.5. Pd-catalysed allylic substitution
Allylic substitution reactions can be defined as the substitution of a leaving group, located in an
allylic position, by an incoming nucleophile, catalysed by a catalytic precursor very often based
on Pd. Its asymmetric version has been thoroughly studied and applied to numerous natural
product syntheses [44] and it is commonly used to test the performance of new bidentate chiral
ligands [45]. One of the model substrates in this reactions is racemic trans-1,3-diphenyl-2-
propen-1-acetate (1,3-diphenylallyl acetate), rac-I, and two of the nucleophiles of reference are
the anion derived from dimethyl malonate and benzylamine (Scheme 8).
Scheme 8. Allylic substitution reactions on rac-I.
Cationic Pd complexes (usually prepared in situ) of the type [Pd(η3-allyl)(PP)]X (PP = chiral
diphosphine, two monodentate phosphines or in general a bidentate ligand) are the usual
14
Page 14 of 22

precursors in this type of transformation. In order to compare [14] the effect of the modification
of the phosphine substituents on the catalytic outcome, the performance of complexes 13 and 14
in asymmetric allylic substitutions was explored. The results are given in Table 2.
15
Page 15 of 22

Table 2. Results of asymmetric allylic substitutions of rac-I with complexes 13 and 14.
Entrya
Nucleophile
DMM
Precursor
13a
Conversion/%b
> 99
ee/%c
80 (S)
< 5
1
2
3
4
5
6
7
8
DMM
DMM
DMM
14a
13b
14b
13a
14a
13b
14b
> 99
> 99
> 99
> 99
> 99
> 99
> 99
< 5
7 (S)
60 (R)
< 5
< 5
< 5
Benzylamine
Benzylamine
Benzylamine
Benzylamine
aCatalytic conditions for allylic alkylations with dimethyl malonate: Pd complex (0.01 mmol), rac-I (1
mmol), dimethyl malonate (3 mmol), BSA (3 mmol) and KOAc (1 mg) in 4 mL of CH2Cl2 at rt for 24 h;
for allylic substitutions with benzylamine: Pd complex (0.01 mmol), rac-I (1 mmol) and benzylamine (3
b
mmol) in 4 mL of CH2Cl2 at rt for 24 h; Conversion percentage expressed as rac-I consumption,
determined by NMR and HPLC; cEnantiomeric excesses determined by HPLC.
Regardless of the substituents of the phosphine, all the cationic complexes were active both in
alkylation and amination of rac-I, leading to quantitative conversions to the substituted product
after 24 h of reaction time. The enantioselectivities were found to be very dependent on the
ligand. It can be seen that phosphines containing the 1-pyrenyl substituent (entries 3, 4, 7 and 8)
are clearly not a good choice for allylic substitutions on rac-I, since they give very low chiral
induction in the products. More interesting results are obtained with precursors with phosphines
PPh(2-p-terphenylyl)R (entries 1, 2, 5, 6). For both substitution reactions racemic products are
obtained for R = Me (entries 2 and 6) whereas good enantioselectivities are achieved for R =
OMe (entries 1 and 5). These results are slightly better but follow the same trend than those
previously published with 2-biphenylyl containing monophosphines in allylic alkylation [14].
2.6. Ru-hydrogen transfer
In hydrogen transfer reactions, a carbonyl from a ketone is reduced to the parent alcohol by a
hydrogen donor, usually an alcohol, under Ru or Ir catalysis [46, 47]. These reactions are
interesting because they avoid using hydrogen gas or dangerous metallic hydrides to reduce the
ketone. In the asymmetric version [46, 48-50], the model substrate is undoubtedly
acetophenone, dissolved in isopropanol as hydrogen donor (Scheme 9).
16
Page 16 of 22

Scheme 9. TH of actophenone.
Although they are not the most typically used precursors, complexes of the type [RuCl2(η6-p-
cymene)P] (P = monophosphorus ligand) in the presence of a base, are known to be active in
hydrogen transfer reactions [15, 16, 38, 51]. Following previous studies with analogous
complexes, the performance of complexes 16-18 was studied. The results are given in Table 3.
Table 3. Hydrogen transfer of acetophenone catalysed by complexes 16-18.
Entrya
Precursor
16a
Time/h
1/3/5/24
1/3/5/24
1/3/5/24
1/3/5/24
1/3/5/24
Conversion/%b
9/29/48/99
11/29/41/99
12/29/42/99
43/72/85/99
8/24/39/83
ee/%c
< 5
< 5
12 (R)
9 (S)
6 (R)
1
2
3
4
5
17a
18a
16b
17b
aCatalytic conditions: Ru complex (0.02 mmol) and t-BuOK (0.1 mmol) dissolved in 25 mL of i-PrOH
b
and activated at 82 ºC during 15 minutes before adding acetophenone (4.0 mmol); Conversion of
starting acetophenone; cEnantiomeric excess at 24 h.
All catalytic precursors are active in the reaction at reflux of isopropanol in the present of
excess of t-BuOK, after activating the system for 15 minutes in the absence of acetophenone.
Full conversions were obtained in all cases after 24 of reaction time except for 17b (entry 5).
Interestingly, the conversions at the same reaction times are all very similar except for 16b
(entry 4), which is considerably faster than the rest. This result is in contrast to previously
published series of similar ligands, for which phosphines always led to faster precursors
compared to the parent phosphinites [15]. On the enantioselectivities side, very low
enantioselectivities were obtained for the five precursors. In this case, the elongation of the 2-
biphenylyl substituent is very detrimental to enantioselectivity since precursor containing the
phosphine P(i-Pr)Ph(2-biphenylyl) [15] leads to a 45% ee under the same conditions but
precursor containing the phosphine P(i-Pr)Ph(2-p-terphenylyl) leads to only 12% ee (entry 3). A
possible explanation is that the long 2-p-terphenylyl moiety forces the phosphine to direct the
17
Page 17 of 22

substituents far away from the Ru centre as it is observed in the crystal structure of 17a
described above.
3. Conclusions
In conclusion, the successful preparation of P-stereogenic phosphines containing a bulky 2-p-
terphenylyl or 1-pyrenyl group by the Jugé-Stephan method has been described. The new
ligands have been used to prepare and characterise three families of organometallic complexes:
neutral Pd allylic complexes (10-12), cationic Pd allylic complexes (13 and 14) and neutral Ru
p-cymene complexes (16-18). Neutral Pd complexes have been used in catalytic
hydrovinylation of styrene with good chemo- and regioselectivity results for all the ligands and
a moderate enantioselectivity (43% ee) with 12a. Cationic Pd complexes have been used in
catalytic allylic substitutions on rac-I, reaching full conversions both for alkylations and
aminations, with up to 80% ee in allylic alkylation and up to 60% in amination with complex
13a. Finally, Ru complexes have been used in hydrogen transfer reactions with good activities
but very low diastereoselectivities.
The best enantioselectivities have been invariably obtained with ligands containing the 2-p-
terphenylyl group. In spite of this, the best enantioselectivities are only marginally better than
those with phosphines containing the 2-biphenylyl group and the overall catalytic performance
of the presented ligands is not significantly better than previous results with phosphines with
smaller substituents [12, 14, 15].
4. General procedures for catalytic runs
4.1. Hydrovinylation reactions
Hydrovinylation reactions were carried out in a stainless-steel autoclave fitted with an external
jacket connected to an ethanol bath and the temperature controlled using a thermostat to ± 0.5
ºC. The internal temperature was controlled with a thermocouple. The Pd precursor (0.01
mmol), styrene (1.04 g, 20 mmol) and AgBF4 (2.1 mg, 0.011 mmol) were dissolved in 15 mL of
dichloromethane and stirred for 10 min, protected from light. After filtering off the AgCl
formed, the solution was quickly placed, by syringe, into the autoclave, which had been purged
by successive vacuum/nitrogen cycles and thermostated to 25 ºC. Ethylene was admitted until a
pressure of around 15 bar was reached. After the allotted time, the autoclave was slowly
depressurized and 10 mL of a 10% aqueous NH4Cl solution was added. The mixture was stirred
18
Page 18 of 22

for 10 min in order to quench the catalyst. The organic layer was separated, dried with Na2SO4,
filtered through a plug of SiO2 and subjected to GC analysis.
4.2. Allylic substitutions with dimethyl malonate
Under a nitrogen atmosphere, the appropriate Pd precursor (0.01 mmol), trans-1,3-diphenyl-2-
propen-1-acetate (rac-I, 1 mmol), dimethyl malonate (3 mmol), BSA (3 mmol) and KOAc (1
mg) were dissolved, in this precise order, in 4 mL of dichloromethane. The flask was covered
with an aluminium foil and left stirring for the allotted time. In order to quench the reaction,
diethyl ether (20 mL) and aqueous 10% ammonium chlorides solution (20 mL) were added.
After extraction, the organic phase was dried with anhydrous sodium sulfate, filtered and the
1
solvents removed in vacuo. The crude was analysed by H NMR in order to estimate the
conversion. The crude was dissolved in dichloromethane and passed trough a column of silica to
remove the metallic impurities. The eluent was removed in vacuo and the residue was analysed
by NMR and HPLC.
4.3. Allylic substitutions with benzylamine
Under a nitrogen atmosphere, the Pd precursor (0.01 mmol), trans-1,3-diphenyl-2-propen-1-
acetate (rac-I, 1 mmol) and benzylamine (3 mmol) were dissolved, in this precise order, in 4
mL of dichloromethane. The flask was covered with an aluminium foil and left stirring for the
allotted time. In order to quench the reaction, diethyl ether (20 mL) and aqueous 10%
ammonium chloride solution (20 mL) were added. After extraction, the organic phase was dried
with anhydrous sodium sulfate, filtered and the solvents removed in vacuo. The eluent was
removed in vacuo and the residue was analysed by NMR and HPLC.
4.4. Transfer hydrogenation reactions
A 50 mL schlenk flask was charged with the ruthenium precursor (0.02 mmol) and potassium
tert-butoxide (11.2 mg, 0.1 mmol) and was purged with three vacuum/argon cycles. Under a
gentle flow of argon, 25 ml of 2-propanol were added and the flask heated to reflux (85 ºC) for
15 minutes. After that time acetophenone (0.468 L, 4.0 mmol) was rapidly added to start the
catalytic run. The reaction was monitored at the allotted times by taking aliquots of around 0.1
mL and analysing them by GC.
Acknowledgements
The authors thank the Ministerio de Educación y Ciencia (MEC, grant number CTQ2010-
15292) for financial support of this work. We also thank Dr. Sebastian Gischig for the crystal
structure determination of 3b performed at the Department of Chemistry, ETH Hönggerberg,
Zürich, Switzerland.
19
Page 19 of 22

References
[1] A. Börner, Ed., Phosphorus Ligands in Asymmetric Catalysis: Synthesis and Applications,
Wiley-VCH, Weinheim, 2008.
[2] P.C.J. Kamer, P.W.N.M. van Leeuwen, Ed., Phosphorus(III) Ligands in Homogeneous
Catalysis: Design and Synthesis, Wiley, 2012.
[3] J. Meisenheimer, L. Lichtenstadt, Chem. Ber., 44 (1911) 356-359.
[4] A. Grabulosa, P-Stereogenic Ligands in Enantioselective Catalysis, Royal Society of
Chemistry, Cambridge, 2011.
[5] L. Horner, H. Siegel, H. Büthe, Angew. Chem. Int. Ed., 7 (1968) 942.
[6] W.S. Knowles, M.J. Sabacky, Chem. Commun., (1968) 1445-1446.
[7] W.S. Knowles, Angew. Chem. Int. Ed., 41 (2002) 1998-2007.
[8] K.M. Pietrusiewicz, M. Zablocka, Chem. Rev., 94 (1994) 1375-1411.
[9] A. Grabulosa, J. Granell, G. Muller, Coord. Chem. Rev., 251 (2007) 25-90.
[10] J. Albert, J.M. Cadena, J.R. Granell, G. Muller, J.I. Ordinas, D. Panyella, C. Puerta, C.
Sañudo, P. Valerga, Organometallics, 18 (1999) 3511-3518.
[11] J. Albert, R. Bosque, J. Magali Cadena, S. Delgado, J.R. Granell, G. Muller, J.I. Ordinas,
M. Font Bardia, X. Solans, Chem. Eur. J., 8 (2002) 2279-2287.
[12] A. Grabulosa, G. Muller, J.I. Ordinas, A. Mezzetti, M.A. Maestro, M. Font-Bardia, X.
Solans, Organometallics, 24 (2005) 4961-4973.
[13] A. Grabulosa, A. Mannu, G. Muller, T. Calvet, M. Font-Bardia, J. Organomet. Chem., 696
(2011) 2338-2345.
[14] A. Grabulosa, G. Muller, R. Ceder, M.A. Maestro, Eur. J. Inorg. Chem., (2010) 3372-3383.
[15] A. Grabulosa, A. Mannu, A. Mezzetti, G. Muller, J. Organomet. Chem., 696 (2012) 4221-
4228.
[16] R. Aznar, A. Grabulosa, A. Mannu, G. Muller, D. Sainz, V. Moreno, M. Font-Bardia, T.
Calvet, J. Lorenzo, Organometallics, 32 (2013) 2344-2362.
[17] A.R. Muci, K.R. Campos, D.A. Evans, J. Am. Chem. Soc., 117 (1995) 9075-9076.
[18] S. Jugé, M. Stephan, J.A. Laffitte, J.P. Genêt, Tetrahedron Lett., 31 (1990) 6357-6360.
[19] Y. Kuninobu, T. Yoshida, K. Takai, J. Org. Chem., 76 (2011) 7370-7376.
[20] S. Ishikawa, K. Manabe, Chem. Lett., 36 (2007) 1302-1303.
[21] S. Ishikawa, K. Manabe, Tetrahedron, 66 (2010) 297-303.
[22] M. Yamaguchi, H. Katsumata, K. Manabe, J. Org. Chem., 78 (2013) 9270-9281.
[23] T. Iwasawa, T. Komano, A. Tajima, M. Tokunaga, Y. Obora, T. Fujihara, Y. Tsuji,
Organometallics, 25 (2006) 4665-4669.
[24] Y. Akai, T. Yamamoto, Y. Nagata, T. Ohmura, M. Suginome, J. Am. Chem. Soc., 134
(2012) 11092-11095.
[25] A. Velian, S. Lin, A.J.M. Miller, M.W. Day, T. Agapie, J. Am. Chem. Soc., 132 (2010)
6296-6297.
[26] S. Lin, M.W. Day, T. Agapie, J. Am. Chem. Soc., 133 (2011) 3828-3831.
[27] S. Lin, D.E. Herbert, A. Velian, M.W. Day, T. Agapie, J. Am. Chem. Soc., 135 (2013)
15830-15840.
[28] S. Suseno, K.T. Horak, M.W. Day, T. Agapie, Organometallics, 32 (2013) 6883-6886.
[29] J. Hu, J.H.K. Yip, D.-L. Ma, K.-Y. Wong, W.-H. Chung, Organometallics, 28 (2009) 51-
59.
[30] L. Rodríguez, O. Rossell, M. Seco, A. Grabulosa, G. Muller, M. Rocamora,
Organometallics, 25 (2006) 1368-1376.
[31] S. Rast, M. Stephan, B. Mohar, Tetrahedron Lett., 53 (2012) 6815-6818.
[32] B. Mohar, A. Cusak, B. Modec, M. Stephan, J. Org. Chem., 78 (2013) 4665-4673.
[33] S. Moon, H. Yoon, Y. Kang, Acta Cryst., E66 (2010) o743.
[34] R.H. Mitchell, Y. Lai, R.V. Williams, J. Org. Chem., 44 (1979) 4733-4735.
[35] U. Nettekoven, P.C.J. Kamer, P.W.N.M. van Leeuwen, M. Widhalm, A.L. Spek, M. Lutz,
J. Org. Chem., 64 (1999) 3996-4004.
20
Page 20 of 22

[36] M. Stephan, D. Sterk, B. Modec, B. Mohar, J. Org. Chem., 72 (2007) 8010-8018.
[37] R.M. Stoop, A. Mezzetti, F. Spindler, Organometallics, 17 (1998) 668-675.
[38] A. Grabulosa, A. Mannu, E. Alberico, S. Denurra, S. Gladiali, G. Muller, J. Mol. Catal. A:
Chem., 363-364 (2012) 49-57.
[39] J. Wolf, K. Thommes, O. Briel, R. Scopelliti, K. Severin, Organometallics, 27 (2008)
4464-4474.
[40] T.V. RajanBabu, Synlett, (2009) 853-885.
[41] G. Muller, A. Grabulosa, M. Rocamora, R.M. Ceder, Catal. Sci. Technol., 3 (2013) 1446-
1464.
[42] T.V. RajanBabu, Chem. Rev., 103 (2003) 2845-2860.
[43] T.V. RajanBabu, C.R. Smith, 5.16 Reduction: Enantioselective Hydrovinylation of
Alkenes, in: E.M. Carreira, Y. Hisashi (Eds.) Comprehensive Chirality, Elsevier, Amsterdam,
2012, pp. 355-398.
[44] B.M. Trost, M.L. Crawley, Chem. Rev., 103 (2003) 2921-2943.
[45] B.M. Trost, D.L. van Vranken, Chem. Rev., 96 (1996) 395-422.
[46] G. Zassinovich, G. Mestroni, S. Gladiali, Chem. Rev., 92 (1992) 1051-1089.
[47] D. Klomp, U. Hanefeld, J.A. Peters, Transfer Hydrogenation Including the Meerwein-
Ponndorf-Verley Reduction, in: J.G. de Vries, C.J. Elsevier (Eds.) Handbook of Homogeneous
Hydrogenation, Wiley-VCH, Weinheim, 2007, pp. 585-630.
[48] R. Noyori, S. Hashiguchi, Acc. Chem. Res., 30 (1997) 97-102.
[49] S. Gladiali, E. Alberico, Chem. Soc. Rev., 35 (2006) 226-236.
[50] A.J. Blacker, Enantioselective Transfer Hydrogenation, in: J.G. de Vries, C.J. Elsevier
(Eds.) Handbook of Homogeneous Hydrogenation, Wiley-VCH, Weinheim, 2007, pp. 1215-
1244.
[51] L. Wang, Q. Yang, H.-Y. Fu, H. Chen, M.-L. Yuan, R.-X. Li, Appl. Organometal. Chem.,
25 (2011) 626-631.
21
Page 21 of 22

Page 22 of 22