Ruthenium complexes of: P -stereogenic phosphines with a heterocyclic substituent

Article abstract of DOI:10.1039/c6dt00995f

The synthesis via phosphine-boranes of 13 new optically pure P-stereogenic diarylphosphines P(Het)PhR (Het = 4-dibenzofuranyl (DBF), 4-dibenzothiophenyl (DBT), 4-dibenzothiophenyl-S,S-dioxide (DBTO₂) and 1-thianthrenyl (TA); R = OMe, Me, i-Pr, Fc (ferrocenyl)) following the Jugé-Stephan method is described. The ligands were designed with the aim of having a heteroatom in a position capable of interacting with a metal upon coordination. The ligands and their precursors have been fully characterised, including the determination of two crystal structures of phosphine-boranes. Ru neutral complexes of the type [RuCl₂(η⁶-arene)(κP-P)] (arene = p-cymene and methyl benzoate) have been prepared and characterised, including three crystal structure determinations. Treatment of solutions of the complexes with TlPF₆ allowed the preparation of well-defined cationic complexes [RuCl(η⁶-arene)(κ²P,S-P)]PF₆ for DBT- and TA-based phosphines. The complexes possess a stereogenic Ru atom and in most of the cases they are present as a single isomer in solution. All the Ru complexes have been used in the asymmetric transfer hydrogenation of acetophenone in refluxing 2-propanol, with good activities and up to 70% ee.

Full text of DOI:10.1039/c6dt00995f

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Clavero, A.
Grabulosa, M. Rocamora, G. Muller and M. Font-Bardia, Dalton Trans., 2016, DOI: 10.1039/C6DT00995F.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/dalton

Page 1 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
Ruthenium complexes of Pꢀstereogenic phosphines with a heterocyclic substituent
Pau Clavero,a Arnald Grabulosa,*,a Mercè Rocamora,a Guillermo Muller,a and Mercè
FontꢀBardiab
aDepartament de Química Inorgànica i Orgànica, Secció de Química Inorgànica, Universitat de
Barcelona, Martí i Franquès, 1ꢀ11, Eꢀ08028, Barcelona, Spain; bDepartament de Cristal·lografia,
Mineralogia i Dipòsits Minerals, Universitat de Barcelona, Martí i Franquès, s/n, Eꢀ08028, Barcelona,
Spain
Abstract
The synthesis via phosphineꢀboranes of 13 new optically pure Pꢀstereogenic diarylphosphines
P(Het)PhR (Het = 4ꢀdibenzofuranyl (DBF), 4ꢀdibenzothiophenyl (DBT), 4ꢀdibenzothiophenylꢀ
S,Sꢀdioxide (DBTO2) and 1ꢀthianthrenyl (TA); R = OMe, Me, iꢀPr, Fc (ferrocenyl)) following
the JugéꢀStephan method is described. The ligands were designed with the aim of having a
heteroatom in a position capable of interacting with a metal upon coordination. The ligands and
their precursors have been fully characterised, including the determination of two crystal
structures of phosphineꢀboranes. Ru neutral complexes of the type [RuCl2(η6ꢀarene)(κPꢀP)]
(arene = pꢀcymene and methyl benzoate) have been prepared and characterised, including three
crystal structure determinations. Treatment of solutions of the complexes with TlPF6 allowed
the preparation of well defined cationic complexes [RuCl(η6ꢀarene)(κ2P,SꢀP)]PF6 for DBTꢀ and
TAꢀbased phosphines. The complexes possess a stereogenic Ru atom and in most of the cases
they are present as a single isomer in solution. All the Ru complexes have been used in the
asymmetric transfer hydrogenation of acetophenone in refluxing 2ꢀpropanol, with good
activities and up to 70% ee.
Introduction
The preparation of optically pure Pꢀstereogenic compounds is still a considerable challenge
despite their long history, stretching for more than a century,1, 2 and their importance as ligands
for transition metalꢀbased homogeneous catalysis.3ꢀ5 The lack of generality of most of the
known synthetic methods and the long and tedious sequences required to prepare such
1

Dalton Transactions
Page 2 of 39
View Article Online
DOI: 10.1039/C6DT00995F
compounds can be blamed for the sluggish development of this area. This makes that even
today the preparation of new ligands of this kind can be considered a valuable achievement.
During the last twenty years, however, several very promising advances have been made,4
which have allowed the synthesis of new families of ligands with superior performance in Rhꢀ
catalysed hydrogenation and other reactions, and the pace of these advances is increasing
lately.6ꢀ17 At present, most of the ligands of this kind are prepared using phosphineꢀboranes18ꢀ20
as intermediates and by asymmetric synthesis methods relying in chiral auxiliaries. Two of the
22
most important routes are those developed by Jugé, Stephan and coworkers21, furnishing
diarylphosphines, and that firstly devised by Evans and coworkers23 and much expanded by
Imamoto and coworkers24, 25 to give trialkylphosphines. Both methods are based on phosphineꢀ
boranes and employ organolithium reagents as nucleophiles or bases in at least one step.
Joining these efforts, we have described several kinds of Pꢀstereogenic monophosphines,
initially prepared by resolution of the racemic ligands26ꢀ28 and more recently by the Jugéꢀ
36, 37
Stephan29ꢀ35 or Evans32,
methods. They were initially employed in Pdꢀcatalysed
32, 34
34
hydrovinylation29,
and later to allylic substitution reactions31,
and Ruꢀcatalysed
cyclopropanation33 and transfer hydrogenation33ꢀ35, 37 reactions.
We reasoned that it would be interesting to design families of new Pꢀstereogenic
monophosphines containing heteroatoms adequately located in the ligand in order to interact
with the metal with a coordination bond or by a weaker secondary (hemilabile) interaction and
study their performance in catalysis. With these ideas in mind, a recent paper of Hayes and
coworkers38 describing the synthesis of Pꢀstereogenic monophosphinimine ligands for Znꢀ
catalysed ringꢀopening polymerisation of lactide caught our attention. In this paper the synthesis
of P(4ꢀdibenzofuranyl)MePh was described, albeit in racemic form. This phosphine was
prepared using 4ꢀlithiodibenzofuran,39 easily prepared by direct oꢀlithiation of dibenzofuran
(Scheme 1).
2

Page 3 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
Scheme 1. Described preparation of racꢀP(4ꢀdibenzofuranyl)MePh.
This ligand has the heteroatom at the γ position with respect to the P atom, a feature that would
create a favoured 5ꢀmembered ring upon interaction with a transition metal. Therefore, we
started a project aiming to prepare Pꢀstereogenic phosphines bearing a heterocyclic substituent
with the following requirements: i) the ligands should have the heteroatom of the heterocycle at
γ or δ position relative to the phosphorus atom, ii) the heterocycle should be selectively lithiated
at β position, so it can be installed at the P atom by the JugéꢀStephan method and iii) the
heterocycle should be commercially available. After analysis of the literature, we concluded that
dibenzofuran (DBF), dibenzothiophene (DBT) and thianthrene (TA) met these requirements
(Scheme 2).
P
P
Ph
Ph
Ln
M
R
R
Li
Ph
Ph
A
Ph
O
A
P
R
A
A
P
+
N
H3B
Me
A
Me
1
New monodentate (P)
or
bidentate (P,A) ligands
O
S
S
S
Dibenzofuran
DBF
Dibenzothiophene
DBT
Thianthrene
TA
Scheme 2. Heterocyclic Pꢀstereogenic phosphines described in this paper.
The number of monophosphorus ligands or precursors bearing any of these substituents is very
limited. With DBF Haenel and coworkers39 first reported the preparation of 4ꢀ
diphenylphosphinodibenzofuran in the course of their studies on lithiation of DBF and DBT.
Much more recently several 4ꢀdiphenylphosphinodibenzonfuran oxides, substituted with
different moieties at the dibenzofuran fragment, have been reported because they have
45
interesting photochemical applications.40ꢀ43 Wills and coworkers44,
prepared 4ꢀ
3

Dalton Transactions
Page 4 of 39
View Article Online
DOI: 10.1039/C6DT00995F
bis(dimethylamino)phosphinodibenzofuran and condensed it at high temperature with a chiral
diamine to obtain an optically pure diazaphospholidine, a ligand that was used in Pdꢀcatalysed
allylic substitution reactions. This is the only reported example of an optically pure
monophosphorus ligand based on the DBF skeleton. Finally, Hayes and coworkers38 recently
reported the synthesis of racemic (4ꢀdibenzofuranyl)methylphenylphosphine as mentioned
before, by deprotection of its phosphineꢀborane, previously obtained by reaction of
methyllithium with (4ꢀdibenzofuranyl)menthylphenylphoshpineꢀborane. With DBT, Rauchfuss,
Rheingold and coworkers46 reported the synthesis of 4ꢀdiphenylphosphinoꢀ and 4ꢀdi(pꢀ
tolyl)phosphinodibenzothiophene and some derived Ru complexes. The crystal structure of the
former phosphine and a derived Fe complex were also described a few years later.47 4ꢀ
diphenylphosphinodibenzothiophene was also reported by Haenel and coworkers soon
afterwards.39 The only optically pure monophosphorus ligand precursor with the DBT moiety
was reported by Fiaud and coworkers,48 who attached an enantiomerically pure 2,5ꢀ
diphenylphospholane oxide moiety to the 4 position of DBT by Pdꢀcatalysed C–P bond
formation. Finally, no phosphines with the TA substituents have been described to our
knowledge. In addition, there are no examples of optically pure Pꢀstereogenic phosphines
bearing any of those heterocyclic substituents.
In this paper we describe the synthesis of series of new Pꢀstereogenic phosphineꢀboranes
containing a DBF, DBT or TA substituent employing the JugéꢀStephan method, the preparation
of several types of complexes containing [Ru(η6ꢀarene)] moieties and their application as
precatalysts to the asymmetric transfer hydrogenation of acetophenone.
Results and discussion
Ligand synthesis
The desired ligands were designed to be obtainable by the JugéꢀStephan method,21, 22 in which
the groups are sequentially introduced at the phosphorus atom via organolithium reagents.
Therefore, following slightly modified literature procedures, the selective monometallation of
DBF,38 DBT49 and TA50 was successfully accomplished by ortholithiation with nꢀbutyllithium
under different conditions (Scheme 3).
4

Page 5 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
Scheme 3. Lithiation of the heterocycles.
The solutions of the organolithiums were reacted with JugéꢀStephan’s oxazaphospholidineꢀ
borane 1 at low temperature giving aminophosphineꢀboranes 2ꢀHet in good yields as white
solids (Scheme 4).
Scheme 4. Preparation of the heterocyclic Pꢀstereogenic phosphineꢀboranes by the JugéꢀStephan method.
The acidic methanolysis of 2ꢀHet proceeded smoothly, affording phosphiniteꢀboranes 3ꢀHet as
pure pasty solids or oils after column chromatography purification. Treatment of these
compounds with excess of RLi (R = Me, iꢀPr, tꢀBu and Fc) at low temperatures was carried out
to obtain series of phosphineꢀboranes as resins or oils. It is known that this step is very sensitive
to the bulkiness of the incoming organolithium reagent.29, 51 Therefore, it is not surprising that in
the case of methyllithium the reactions were successful for all the substrates, giving the
methylphosphineꢀboranes 4ꢀHetꢀMe in good yields. Isopropyllithium reacted well with 3ꢀDBF
and 3ꢀDBT giving the desired 4ꢀHetꢀiPr phosphineꢀboranes but reaction with 3ꢀTA at –30 ºC
1
produced a compound containing two isopropyl groups. According to H and 13C NMR
spectroscopy, one of them was bound to the P atom whereas the other was not. No further
aliphatic hydrogen or carbon atoms could be detected. Addition of less than one equivalent of
isopropyllithium led to the same product with two isopropyl groups along with incomplete
conversion of the starting phosphiniteꢀborane 3ꢀTA. This fact indicates that isopropyllithium is
not able to directly attack the phosphorus atom so, it probably reacts first with the thianthrene
5

Dalton Transactions
Page 6 of 39
View Article Online
DOI: 10.1039/C6DT00995F
ring and opens it, releasing steric encumbrance at the P atom and allowing a rapid attack of a
second equivalent of isopropyllithium. Although NMR suggested that only a single
diastereomerically pure product was formed, we have been unable to clarify either its identity or
its optical purity. Interestingly, the addition of isopropyllithium to a diethyl ether solution of
thianthrene at –30 ºC did not lead to any opened product but to the full recovery of unchanged
thianthrene. Reaction of 3ꢀHet with monolithiated ferrocene worked well for Het = DBF and
DBT but not for TA, since unchanged 3ꢀTA was isolated after workup.
The introduction of the tꢀBu group is (usually)51, 52 impossible using the JugéꢀStephan method
due to steric reasons.29 In line with this finding, reaction of 3ꢀDBT and 3ꢀTA with tꢀBuLi was
unsuccessful since complex mixtures of products were obtained according to 31P NMR
spectroscopy. In contrast, under carefully controlled conditions, 3ꢀDBF reacted with tꢀBuLi to
afford the phosphine 4ꢀDBFꢀtBu, which could be isolated as an oil in 60% yield. It is possible
that the hard oxygen atom of DBF assists the nucleophilic attack of tꢀBuLi by coordination of
the Li cation.52 To take advantage of this reactivity, the triarylphosphineꢀborane 4ꢀDBFꢀDBT
was successfully prepared by reaction of 3ꢀDBF with DBTLi. A particularity of this phosphine
is that it suffers partial spontaneous deboronation and therefore the workꢀup had to be carried
out under nitrogen atmosphere to minimise the oxidation of the free phosphine. Due to this fact,
the phosphineꢀborane was not isolated but fully deprotected with morpholine (see later) to yield
the completely free phosphine, which was subsequently coordinated to ruthenium.
All the intermediates have been fully characterised by the usual techniques and the details can
be found in the experimental part. Phosphineꢀboranes 4ꢀDBFꢀFc and 4ꢀDBTꢀFc were also
characterised in the solid state by the determination of their Xꢀray crystal structures (Figure 1).
6

Page 7 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
Figure 1. ORTEP representation (thermal ellipsoids drawn at 50% probability level, H atoms removed for
clarity) of 4ꢀDBFꢀFc (left) and 4ꢀDBTꢀFc (right). Distances (Å) and angles (º) for 4ꢀDBFꢀFc: P(1)–B(1),
1.913(4); P(1)–C(4), 1.779(3); P(1)–C(17), 1.816(3); P(1)–C(5), 1.812(3); O(1)–C(15), 1.388(4); O(1)–
C(16), 1.392(3); B(1)–P(1)–C(4), 114.32(16); B(1)–P(1)–C(17), 110.02(16); B(1)–P(1)–C(5),
113.67(17); C(15)–O(1)–C(16), 105.3(2). For 4ꢀDBTꢀFc: P(1)–B(1), 1.906(6); P(1)–C(10), 1.789(5);
P(1)ꢀC(11), 1.819(5); P(1)–C(17), 1.814(5); S(1)–C(27), 1.749(5); S(1)–C(28), 1.757(5); B(1)–P(1)–
C(10), 117.8(3); B(1)–P(1)–C(11), 108.0(2); B(1)–P(1)–C(17), 114.0(3); C(27)–S(1)–C(28), 91.6(2).
The crystals contain discrete molecules having the expected S absolute configuration at the P
atom. The distances and angles are in the range expected for similar compounds29, 35, 53 and very
similar for both structures. The only noticeable differences between the two structures are in the
parameters around the heteroatom: for DBF, the two O–C distances are much shorter compared
to the two S–C distances in DBT and the angle C–O–C is much wider than the C–S–C in DBT.
For both structures the heterocycle is essentially planar and with the two Cp rings of the
ferrocene being almost eclipsed, as observed in other ferrocenylphosphineꢀboranes.32, 54, 55
It is well known that the sulfur atoms of DBT and TA can be oxidised to sulfoxides (SO)50, 56ꢀ60
or sulfones (SO2).56, 57, 60ꢀ68 For this reason it was considered worth exploring the oxidation of
the ligands containing these heterocycles because the sulfoxy group of the new ligands could
interact with the metal during catalysis. Phosphineꢀborane 4ꢀDBTꢀMe was therefore treated
with a variety of oxidants such as MCPBA,58 H2O2/HAcO,60, 64ꢀ67 and CrO3/H5IO6 (Scheme 5).63
Scheme 5. Unsuccessful synthesis of 4ꢀDBTO2ꢀMe.
7

Dalton Transactions
Page 8 of 39
View Article Online
DOI: 10.1039/C6DT00995F
With the treatment with MCPBA and H2O2/HAcO it was found that partial deprotection and
oxidation of the P atom of the phosphine as well as formation of byproducts had taken place
according to 31P NMR spectroscopy. In contrast, with CrO3/H5IO6 in acetonitrile63 a single
product corresponding to the complete deprotection and oxidation, namely the trioxide 4ꢀ
DBTO3ꢀMe, could be isolated. It seems therefore that the borane protecting group can not
withstand the strongly oxidant conditions of the reaction. It was then reasoned that if oxidation
of DBT was not possible once installed at the P atom, maybe the DBTO2 fragment could be
introduced in the first step of the JugéꢀStephan method. To this end, DBT was oxidised with
hydrogen peroxide66, 67 and lithiated with nꢀBuLi (Scheme 6).
Scheme 6. Preparation of 3ꢀDBTO2.
The lithiation of DBTO2 has not been reported. After series of experiments it was found that the
best conditions consisted of adding nꢀBuLi to a solution of DBTO2 precooled at –78 ºC,
removing the cold bath immediately and stirring the mixture for 3 h at room temperature. Even
under these conditions, however, the lithiation was incomplete and not always reproducible.
Despite the rather unsatisfactory lithiation, it allowed the introduction of the oxidised
heterocycle at the P atom and following the standard method compounds 2ꢀDBTO2 and 3ꢀ
DBTO2 could be prepared. The latter compound was treated with excess of MeLi under usual
conditions but did not give the expected 4ꢀDBTO2ꢀMe but dimethylphenylphosphineꢀborane.69,
70
It is possible that the strongly electronꢀwithdrawing sulfone group weakens the P–C bond to
such an extend that can be cleaved by methyllithium even at low temperature.71 Therefore no
other phosphines with the DBTO2 were prepared. Finally, the obtained phosphineꢀboranes were
8

Page 9 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
deprotected with morpholine under standard conditions29, 53 to give the free phosphinites and
phosphines L1ꢀ13 (Scheme 7).
Scheme 7. Free phosphinites and phosphines L1ꢀ13.
The free phosphines were all airꢀsensitive, especially the tꢀBuꢀcontaining ligand L7 and hence
after deprotection the 13 ligands were immediately coordinated to Ru moieties.
Ru complexes
Neutral complexes
The ligands were used to obtain the ruthenium neutral complexes of the type [RuCl2(η6ꢀ
arene)(P)], with the arene being pꢀcymene or methyl benzoate (Scheme 8).34, 35
Scheme 8. Preparation of neutral ruthenium complexes.
The complexes were easily prepared by splitting the usually employed ruthenium pꢀcymene
dimer (D1) and for some of the ligands the much lesser used35 ruthenium methyl benzoate dimer
(D2), in dichloromethane at room temperature as previously reported for analogous
compounds.35 The products were obtained as red or brown solids that were characterised by IR,
9

Dalton Transactions
Page 10 of 39
View Article Online
DOI: 10.1039/C6DT00995F
chemical microanalysis or MS and by multinuclear NMR in solution. The data confirmed the
identity of the proposed structures and the purity of the products. Hence, single 31P resonances
were found for all the complexes and due to the chirality of the phosphorus ligand all the H and
C atoms were potentially different. In accordance, apart from the peaks corresponding to the
phosphorus ligand, 4 distinct H (4.0ꢀ6.5 ppm region) and 6 C (80ꢀ110 ppm) peaks appeared,
1
respectively, in the H and 13C{1H} NMR spectra of the pꢀcymene complexes whereas 5 H
resonances could be found for the methyl benzoate complexes. As expected, the latter
complexes also featured a singlet at approximately 3.9 ppm in the 1H NMR spectra,
corresponding to the COOMe group. Unexpectedly, for most of the methyl benzoate complexes
a pair of peaks around 53 ppm and another pair around 167 ppm can be seen in the 13C{1H}
NMR spectra, corresponding to the methylic and carbonylic carbon atoms of COOMe group.
The observation of two peaks is probably due to the presence of the two rotamers represented in
Scheme 9 in solution.
Scheme 9. Two possible isomers of the Ru complexes observed by 13C{1H} NMR spectroscopy.
Finally, a sharp band in the IR spectra of methyl benzoate complexes close to 1728 cm–1
confirms the presence of the carbonyl of the ester group. The complex Ru7 could not be
obtained satisfactorily since and an extremely broad 31P{1H} NMR spectrum resulted and
multiple peaks in the 1H NMR spectrum could be observed. This can be due to the bulkiness of
L7 precluding the efficient coordination to the Ru unit.
Single crystals, suitable for Xꢀray crystallography, could be obtained for complexes Ru5, Ru6
and Ru10 by slow diffusion of hexane into saturated solutions of the complexes in
dichloromethane. The representation of their molecular structures is given in Figure 2.
10

Page 11 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
Figure 2. ORTEP representations (thermal ellipsoids drawn at 50% probability level, H atoms removed
for clarity) of Ru5, Ru6 and Ru10 (from left to right). The most relevant distances and angles are given
in Table 1.
All the complexes adopt the typical pseudotetrahedral, “threeꢀlegged piano stool” geometry,
with the Ru atom in the centre of a distorted octahedron. The structures allow the confirmation
of the expected absolute configurations of the P atoms (S for the free ligands). The crystals of
complex Ru5 contain two molecules in the unit cell, whose main difference is that the pꢀ
cymene is rotated 180º around the Ru–arene centre axis. The most relevant metric parameters of
the structures are given in Table 1.
Table 1. Selected distances (Å) and angles (º) for complexes Ru5, Ru6 and Ru10.
Ru5a
2.4093(10), 2.4188(11)
2.4146(10), 2.4204(10)
2.3362(11)
2.3378(11)
2.217
Ru6
Ru10
Parameter
Ru–Cl
2.4065(12), 2.4069(13)
2.3983(11), 2.4242(11)
Ru–P
bRu–Carene
P–CPh
2.3737(13)
2.210
2.3432(11)
2.220
2.212
1.807(5)
1.824(4)
1.833(5)
1.818(4)
1.822(4)
1.815(4)
P–CHet
1.825(5)
1.836(4)
1.820(5)
1.820(5)
P–CR
1.865(5)
1.816(4)
84.32(4), 85.63(4)
86.65(4), 83.44(4)
88.53(4)
P–Ru–Cl
Cl–Ru–Cl
91.17(5), 86.33(5)
87.69(5)
85.08(4), 83.86(4)
88.14(4)
88.27(4)
aThere are two crystallographically distinct molecules in the unit cell; bAveraged value of the six η6ꢀPh Ru–C distances.
11

Dalton Transactions
Page 12 of 39
View Article Online
DOI: 10.1039/C6DT00995F
As commonly found for this type of compounds, the η6ꢀcoordinated pꢀcymene ring is located in
a way that the imaginary line defined by the two Cl atoms is approximately parallel to the line
passing through the substituted C atoms of the pꢀcymene group. It can also be seen that the
heterocyclic substituent is almost completely flat. In general, the distances and angles are in the
range expected for previously reported similar compounds.33, 34, 37, 72, 73
Cationic complexes
Neutral pꢀcymene Ru complexes were treated with thallium hexafluorophosphate (or
tetrafluoroborate in the case of Ru10) in order to abstract the chloride ligand and force the
coordination of the heteroatom of the heterocycle to the metal (Scheme 10).74ꢀ76
Scheme 10. Preparation of cationic ruthenium complexes.
Treatment of dichloromethane solutions of complexes Ru1 (δP,Ru1 = +112.7 ppm) and Ru6
(δP,Ru6 = +21.5 ppm), bearing a phosphine with the DBF group, with TlPF6 caused a rapid
precipitation of TlCl that was filtrated, the solvent removed and the crude product analysed by
NMR. In both experiments, a singlet at +113.8 and +24.0 ppm respectively in the 31P{1H}
spectra of the isolated product was observed. Since the values are almost unchanged from Ru1
and Ru6, it can be concluded that the desired complex with a fiveꢀmembered chelate ring with
the κ2P,Oꢀcoordinated phosphine did not form because a large downfield shift would be
expected.77 1H NMR spectra, however, revealed that the products were not the starting
complexes and that they contained the pꢀcymene and the phosphine moieties in 1:1 ratio. They
could correspond to dimeric species although their constitution was not further investigated. In
the case of Ru4 (δP,Ru4 = 117.0 ppm), after treatment with TlPF6, 31P{1H} NMR showed that
30% of starting material was still present but another species slightly shifted upfield (δP = 112.0
ppm) had also formed. This species could indeed correspond to the desired κ2P,Oꢀchelate since
12

Page 13 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
it is known that the ring contribution in the 31P shift is small and negative in sixꢀmembered
rings.77 Despite this, given that only partial conversion could be achieved, its synthesis was not
pursued further.
In contrast to the unsuccessful coordination of the O atom, the coordination of the S atom of the
dibenzothiophenyl and thianthryl groups could be achieved, yielding cationic complexes Ru2’,
Ru3’ and Ru9’ꢀ13’. A strong deshielding of the 31P signals (ꢀδ(Ru’ꢀRu) = 20ꢀ43 ppm)
occurred upon formation of the 5ꢀmembered ring via coordination of the S atom, as expected.77
1
Similarly, in the H NMR the peaks of the H atoms of the coordinated arene ring shifted
downfield approximately 1 ppm and in the 13C{H1} the six C resonances also shifted roughly 5
ppm downfield. These downfield shifts possibly reflect the decreased electron density of the η6ꢀ
coordinated arene ring due to the presence of a positive charge compared to the neutral Ru
complexes. The identity of the complexes was also verified by elemental analyses or high
resolution mass spectrometry as detailed in the Experimental Section. The complex Ru11*’,
bearing the methyl benzoate as coordinated arene, was also obtained by treating Ru11* with
thallium hexafluorophosphate.
An interesting aspect of the cationic Ru complexes described here is the possible formation of
two diastereomeric complexes due to the presence of a stereogenic Ru atom (Scheme 10). NMR
analysis showed a single 31P signal and a unique set of C and H signals for complexes Ru2’,
Ru3’, Ru11’ and Ru11*’ suggesting that they are present an optically pure species. In contrast,
the two isomers could be detected for the rest of the complexes, since two 31P peaks and two
sets of C and H signals were found as detailed in the Experimental Part. The ratio between
isomers was approximately 1:4 for complexes Ru9’ and Ru10’ and 1:2 for Ru12’ and Ru13’. It
seems that there is not a simple correlation between the structure of the ligand and the isomeric
ratio. Despite many attempts we were unable to obtain crystals suitable to perform Xꢀray
diffraction studies of any of the complexes in order to ascertain the absolute configuration of the
main isomer.
Ru-catalysed transfer hydrogenation
13

Dalton Transactions
Page 14 of 39
View Article Online
DOI: 10.1039/C6DT00995F
The reduction of ketones to alcohols is an extremely important transformation in organic
chemistry that can be catalytically performed by hydrogenation using hydrogen gas, or in a safer
79
way, by transfer hydrogenation, using a hydrogen donor.78,
The latter reaction has been
studied with a large number of soluble Ru(II) systems, very often chiral, to obtain
enantioenriched alcohols.80ꢀ82 The model substrate for the asymmetric transfer hydrogenation is
acetophenone and the typical conditions involve carrying out the reaction in refluxing 2ꢀ
propanol in the presence of base (Scheme 11).
O
OH
0.5% [Ru], t-BuOK
R
Ru
i-PrOH, 82 ºC
Cl
P
3 h
Cl
[Ru]
acetophenone
1-phenylethanol
Scheme 11. Ruꢀcatalysed enantioselective transfer hydrogenation of acetophenone.
Despite not being the most typical precursors, Ru complexes of the type [RuCl2(η6ꢀarene)(P)]
are easy to prepare and they are active in the reaction, as shown by us33ꢀ35, 37 and other groups.74,
83ꢀ86
The enantioselectivities of our systems with Pꢀstereogenic phosphines are, however, rather
low (up to 50% ee)33, 35, 37 so the performance of neutral and cationic Ru complexes with the
new heterocyclic phosphines was studied (Table 2).
Table 2. Results of transfer hydrogenation of acetophenone catalysed by Ru complexes.
Entrya
1
Precursor
Ru1
Ru2
Ru2’
Ru3
Ru3’
Ru4
Ru5
Ru6
Ru6*
Ru8
Time/h
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
Conversion/%b
31/79/> 99
28/69/85
40/73/82
11/28/39
8/21/29
15/32/46
92/99/> 99
35/78/> 99
80/96/> 99
25/68/83
32/75/92
54/91/> 99
10/29/36
16/32/41
19/49/75
52/96/> 99
4/9/17
ee/%c
< 5
< 5
< 5
< 5
< 5
14 (R)
< 5
< 5
< 5
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
6 (S)
13 (R)
7 (R)
< 5
Ru9
Ru9*
Ru9’
Ru10
Ru11
Ru11*
Ru11’
Ru11*’
< 5
56 (R)
70 (R)
10 (R)
70 (R)
19/48/63
14

Page 15 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
19
20
21
22
23
1/3/5
1/3/5
1/3/5
1/3/5
1/3/5
21/58/83
30 (S)
16 (S)
< 5
< 5
< 5
Ru12
Ru12’
Ru13
Ru13*
Ru13’
9/24/34
64/94/> 99
50/83/95
34/55/71
aCatalytic conditions: Ru complex (0.02 mmol) and tꢀBuOK (0.1 mmol) dissolved in 25 mL of iꢀPrOH
b
and activated at 85 ºC during 15 minutes before adding acetophenone (4.0 mmol); Conversion of
acetophenone; cEnantiomeric excess at 24 h.
The precursors were activated for 15 min in the presence of tꢀBuOK before the addition of
acetophenone to form the catalytically active rutheniumꢀhydride species.87 All were active in the
reaction, resulting in full conversion at 24 h. At shorter reaction times, however, notable
differences in activity can be seen depending on the structure of the precursor. In most of the
cases, neutral κPꢀcoordinated complexes lead to more active precursors compared to cationic
κ2P,Sꢀcoordinated counterparts (cf. for example entries 15 and 17 or 21 and 23). The complexes
with the methyl benzoate give more active systems than those with the pꢀcymene (cf. for
example entries 8 and 9 or 15 and 16), in line with previously published results with similar
systems.35 These findings suggest that η6ꢀarene decoordination or slippage (hapticity reduction)
probably occurs during the catalytic cycle. Such a process is easier for electron poor methyl
benzoate complexes compared to pꢀcymene analogues and also for neutral complexes compared
to cationic counterparts.
Finally, the enantioselection is very low for most of the precursors, as usually found with
similar monophosphorus ligands.33, 35 The precursors with L11 (entries 15ꢀ18) are moderately
enantioselective, except Ru11’ (entry 17). Interestingly, the same value of 70% ee was obtained
with complexes Ru11* and Ru11*’ (entries 16 and 18 respectively), pointing to the formation
of a common intermediate under catalytic conditions. It is worth noting that Ru11 and Ru11*
both form single cationic species in solution (vide infra), a fact that could be beneficial for the
enantioselectivity.
Conclusions
In this paper the JugéꢀStephan method has allowed the preparation of 13 of optically pure Pꢀ
stereogenic diaryl monophosphinites and monophosphines of the type PPh(Het)R (Het = 4ꢀ
15

Dalton Transactions
Page 16 of 39
View Article Online
DOI: 10.1039/C6DT00995F
DBF, 4ꢀDBT, 1ꢀTA and 4ꢀDBTO2; R = OMe, Me, iꢀPr, tꢀBu, Fc) by direct lithiation of the
heterocycle. The ligands are a valuable addition to the small number of optically pure Pꢀ
stereogenic ligands with a heterocyclic substituent.
The ligands had been designed with the idea of introducing the heteroatom (A) at a position
capable of interacting with the ruthenium centre via the formation of a favoured fiveꢀmembered
κ2P,Aꢀchelate. This coordination has been achieved for DBTꢀ and TAꢀcontaining phosphines
but not for the DBFꢀbased ligands. This is possibly due to the hard character of the oxygen
atom, showing less tendency to coordinate to the Ru atom compared to sulfur. An important
stereoselection in the formation of the stereogenic Ru atom has been observed for most of the
ligands.
The obtained complexes have been used in catalytic transfer hydrogenation of acetophenone
with the aim of comparing the performance of the new ligands with previously reported systems
based in Pꢀstereogenic PArPhR ligands (Ar = polycyclic aromatic group).33ꢀ35, 37, 87 It has been
found that the activities are similar to some of the previous generation precursors but one of the
ligands, L11, gives a considerably higher enantioselectivity.
Experimental section
General data
All compounds were prepared under a purified nitrogen atmosphere using standard Schlenk and
vacuumꢀline techniques. The solvents were purified by a solvent purification system or by
1
1
standard procedures88 and kept under nitrogen. H, 13C{1H}, and 31P{1H} and HSQC Hꢀ13C
NMR spectra were recorded using 300 and 400 MHz spectrometers using CDCl3 as solvent
unless otherwise specified. Chemical shifts are reported downfield from standards. The protons
of the BH3 of phosphineꢀboranes group appeared in the aliphatic region of the spectra as very
broad bands and have not been assigned. IR spectra were recorded in KBr and the main
absorption bands are expressed in cm–1. Highꢀresolution mass analyses (HRMS) were carried
out a timeꢀofꢀflight instrument using electrospray ionisation. Optical rotations were measured at
rt using a sodium lamp at the sodium
Dꢀline wavelength (589.592 nm). For all the
16

Page 17 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
determinations, the solvent was CH2Cl2 and the concentration was 1 g/100 mL. Transfer
hydrogenation reactions were analysed by GC with He as a carrier gas. Oxazaphospholidineꢀ
borane 1 (prepared from (1R,2S)ꢀ(–)ꢀephedrine),21 dibenzothiophene dioxide,66, 67 and Ru dimer
D289 were prepared using literature procedures whereas other reagents were used as received
from commercial suppliers.
Synthesis of the ligands
2ꢀDBF, (1R,2S)ꢀ2ꢀ{[(S)ꢀ(4ꢀdibenzofuranyl)phenylphosphanyl]methylamino}ꢀ1ꢀ
phenylpropanꢀ1ꢀolꢀborane
Dibenzofuran (1.85 g, 11.0 mmol) was dissolved in 30 mL of THF in a Schlenk flask. The
solution was cooled to –78 ºC and then 1.6 M nꢀBuLi solution in hexanes (6.9 mL, 11.0 mmol)
was added by syringe. The resulting brown solution was removed from the cold bath, left
stirring for 30 min at room temperature and then recooled to –78 ºC. At the same time
oxazaphospholidineꢀborane 1 (2.85 g, 10.0 mmol) was dissolved in 40 mL of THF and the
solution was cooled down to –78 ºC. The content of the first flask was slowly transferred to the
second Schlenk flask via cannula and the resulting mixture was stirred for 14 h. Around 30 mL
of water were added to the orange solution and THF was evaporated. The darkꢀbrown residue
was extracted with dichloromethane (3 x 30 mL) and the combined organic phases were washed
with water and dried with anhydrous sodium sulfate. The suspension was filtered and the
solvents were evaporated to dryness, leaving a yellowish pasty solid that was purified by
column chromatography (flash SiO2, from 95:5 to 80:20 of hexane/ethyl acetate). The title
product was obtained as a whitish solid. Yield: 3.52 g (77%).
1H NMR (300 MHz): 8.11 (dt, J = 7.8, 1.2, 1H), 7.96 (dm, J = 6.6, 1H), 7.78 (ddd, J = 12.3,
7.5, 1.2, 1H), 7.58 (dm, J = 8.1, 1H), 7.51ꢀ7.18 (m, Ar, 13H), 4.90 (d, 3JHH = 6.0, 1H), 4.45 (m,
1H), 2.63 (d, 3JHP = 8.1, 3H), 1.29 (d, 3JHH = 6.6, 3H). 13C{1H} NMR (101 MHz): 156.3ꢀ111.6
(C, CH, Ar), 78.6 (d, 3JCP = 6.2, CH), 58.2 (d, 2JCP = 11.0, CH), 30.9 (d, 2JCP = 4.4, CH3), 13.0
(s, CH3). 31P{1H} NMR (121 MHz): +67.5 (br, s). HRMS: calcd. for C28H27NO2P ([M] + H –
BH3), 440.1779; found, 440.1771. [α]D = +66.2º.
2ꢀDBT, (1R,2S)ꢀ2ꢀ{[(S)ꢀ(4ꢀdibenzothiophenyl)phenylphosphanyl]methylamino}ꢀ1ꢀ
phenylpropanꢀ1ꢀolꢀborane
Dibenzothiophene (2.03 g, 11.0 mmol) was dissolved in 30 mL of THF in a Schlenk flask. The
solution was cooled to –78 ºC and then 1.6 M nꢀBuLi solution in hexanes (6.9 mL, 11.0 mmol)
was added by syringe. The resulting brown solution was removed from the cold bath, left
17

Dalton Transactions
Page 18 of 39
View Article Online
DOI: 10.1039/C6DT00995F
stirring at 0 ºC for 5 h and recooled to –78 ºC. At the same time oxazaphospholidineꢀborane 1
(2.85 g, 10.0 mmol) was dissolved in 40 mL of THF and the solution was cooled down to –78
ºC. The content of the first flask was slowly transferred to the second Schlenk flask via cannula
and the resulting mixture was stirred for 14 h. Around 30 mL of water were added to the brownꢀ
yellow solution and THF was evaporated. The white residue was extracted with
dichloromethane (3 x 30 mL) and the combined organic phases were washed with water and
dried with anhydrous sodium sulfate. The suspension was filtered and the solvents were
evaporated to dryness, leaving a white pasty solid, which was purified by column
chromatography (flash SiO2, from 95:5 to 80:20 of hexane/ethyl acetate). The title product was
obtained as a white solid. Yield: 4.11 g (87%).
1H NMR (400 MHz): 8.29 (m, 1H), 8.18 (m, 1H), 7.84 (m, 1H), 7.73 (m, 1H), 7.58ꢀ7.42 (m, Ar,
9H), 7.34 (t, J = 7.6, 2H), 7.27 (t, J = 6.4, 1H), 4.96 (s, br, 1H), 4.47 (m, 1H), 2.75 (d, 3JHP = 7.6,
3H), 1.36 (d, 3JHH = 6.8, 3H). 13C{1H} NMR (101 MHz): 142.4ꢀ121.4 (C, CH, Ar), 78.9 (d, 3JCP
2
2
3
= 2.7, CH), 58.5 (d, JCP = 10.4, CH), 31.6 (d, JCP = 4.3, CH3), 11.3 (d, JCP = 5.4, CH3).
31P{1H} NMR (121 MHz): +70.1 (br, s). HRMS: calcd. for C28H27NOPS ([M] + H – BH3),
456.1551; found, 456.1540. [α]D = +52.2º.
2ꢀTA, (1R,2S)ꢀ2ꢀ{[(S)ꢀ(1ꢀthianthrenyl)phenylphosphanyl]methylamino}ꢀ1ꢀphenylpropanꢀ1ꢀ
olꢀborane
Thianthrene (600 mg, 2.8 mmol) was dissolved in 40 mL of THF in a Schlenk flask. The
solution was cooled to –78 ºC and then 1.6 M nꢀBuLi solution in hexanes (2.3 mL, 3.7 mmol)
was added by syringe. The resulting brown solution was removed from the cold bath and once
at room temperature was refluxed for 1 h, cooled to room temperature and then to –78 ºC. At the
same time the oxazaphospholidineꢀborane 1 (720 mg, 2.5 mmol) was dissolved in 40 mL of
THF and the solution was cooled down to –78 ºC. The content of the first flask was slowly
transferred to the second Schlenk flask via cannula and the resulting mixture was stirred for 14
h. Around 30 mL of water were added to the brownꢀyellow solution and THF was evaporated.
The white residue was extracted with dichloromethane (3 x 30 mL) and the combined organic
phases were washed with water and dried with anhydrous sodium sulfate. The suspension was
filtered and the solvents were evaporated to dryness, leaving a white pasty solid, which was
purified by column chromatography (flash SiO2, from 95:5 to 80:20 of hexane/ethyl acetate).
The title product was obtained as a white solid. Yield: 1.15 g (91%).
1H NMR (400 MHz): 7.67 (dt, J = 7.6, 1.2, 1H), 7.52ꢀ7.43 (m, Ar, 7H), 7.39ꢀ7.27 (m, Ar, 6H),
7.22 (td, J = 7.6, 1.6, 1H), 7.15 (td, J = 7.6, 1.6, 1H), 7.02 (dd, J = 7.6, 1.6, 1H), 4.98 (d, J = 4.4,
1H), 4.45 (m, 1H), 2.63 (d, 3JHP = 7.2, 3H), 1.32 (d, 3JHH = 6.8, 3H). 13C{1H} NMR (101 MHz):
142.4ꢀ126.2 (C, CH, Ar), 79.0 (d, 3JCP = 2.8, CH), 58.4 (d, 2JCP = 10.7, CH), 31.4 (d, 2JCP = 4.1,
3
CH3), 12.0 (d, JCP = 4.1, CH3). 31P{1H} NMR (162 MHz): +71.9 (br, s). HRMS: calcd. for
C28H27NOPS2 ([M] + H – BH3), 488.1272; found, 488.1267. [α]D = +40.4º.
18

Page 19 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
2ꢀDBTO2, (1R,2S)ꢀ2ꢀ{[(S)ꢀ(4ꢀdibenzothiophenyl dioxide)phenylphosphanyl]methylamino}ꢀ
1ꢀphenylpropanꢀ1ꢀolꢀborane
DibenzothiopheneꢀS,Sꢀdioxide (1.19 g, 5.5 mmol) was dissolved in 40 mL of THF in a Schlenk
flask. The solution was cooled to –78 ºC and then 1.6 M nꢀBuLi solution in hexanes (3.4 mL,
5.5 mmol) was added by syringe. The resulting brown solution was removed from the cold bath,
left stirring at room temperature for 3 h and recooled to –78 ºC. At the same time the
oxazaphospholidineꢀborane 1 (1.43 g, 5.5 mmol) was dissolved in 35 mL of THF and the
solution was cooled down to –78 ºC. The content of the first flask was slowly transferred to the
second Schlenk flask via cannula and the resulting mixture was stirred for 14 h. Around 30 mL
of water were added to the brownꢀyellow solution and THF was evaporated. The white residue
was extracted with dichloromethane (3 x 30 mL) and the combined organic phases were washed
with water and dried with anhydrous sodium sulfate. The suspension was filtered and the
solvents were evaporated to dryness, leaving a white solid. Yield: 1.25 g (45%).
1H NMR (400 MHz): 7.97 (m, 2H), 7.91. (dt, J = 7.6, 1.2, 1H), 7.79 (d, J = 8.0, 1H), 7.75 (d, J
= 7.6, 1H), 7.62 (td, J = 7.6, 1.2, 1H), 7.57 (m, 1H), 7.55ꢀ7.50 (m, 4H), 7.45 (m, 1H), 7.39 (d, J
= 7.6, 2H), 7.29 (td, J = 7.6, 2.0, 2H), 7.20 (tt, J = 7.2, 1.2, 1H), 5.11 (d, J = 2.8, 1H), 4.30 (m,
1H), 2.84 (d, 3JHP = 8.4, 3H), 1.24 (d, 3JHH = 6.8, 3H). 13C{1H} NMR (101 MHz): 142.4ꢀ121.2
(C, CH, Ar), 78.7 (d, 3JCP = 1.5, CH), 59.3 (d, 2JCP = 9.9, CH), 33.9 (d, 2JCP = 3.9, CH3), 9.6 (d,
3JCP = 7.2, CH3). 31P{1H} NMR (162 MHz): +73.4 (br, s). HRMS: calcd. for C28H27NO3PS ([M]
+ H – BH3), 488.1449; found, 488.1457. [α]D = +66.1º.
3ꢀDBF, (R)ꢀ(4ꢀdibenzofuranyl)methoxyphenylphosphineꢀborane
Aminophosphineꢀborane 2ꢀDBF (3.52 g, 7.7 mmol) was dissolved in 200 mL of freshly distilled
methanol, concentrated H2SO4 (0.84 mL, 1.51 g, 15.4 mmol) was carefully added and the
solution was stirred for 14 h. The solvent was removed in vacuo and the crude was purified by a
column chromatography (flash SiO2, 95:5 hexane/ethyl acetate). The title product was obtained
as a pale brown oil. Yield: 1.67 g (67%).
1H NMR (400 MHz): 8.13 (dt, J = 7.6, 1.2, 1H), 7.98ꢀ7.91 (m, 4H), 7.58 (d, J = 12.0, 1H), 7.56
3
(d, J = 12.0, 1H), 7.51ꢀ7.43 (m, 4H), 7.36 (m, 1H), 3.85 (d, JHP = 12.4, 3H). 13C{1H} NMR
2
(101 MHz): 156.1ꢀ111.6 (C, CH, Ar), 54.3 (d, JCP = 2.7, CH3). 31P{1H} NMR (121 MHz):
+106.6 (d, br, J ≈ 88). HRMS: calcd. for C19H22BNO2P ([M] + NH4), 338.1481; found,
338.1472. [α]D = –81.9º.
3ꢀDBT, (R)ꢀ(4ꢀdibenzothiophenyl)methoxyphenylphosphineꢀborane
The procedure was the same used to prepare 3ꢀDBF but starting from precursor 2ꢀDBT (2.06 g,
4.4 mmol). The desired phosphiniteꢀborane was obtained as a colourless oil. Yield: 1.19 g
(81%).
19

Dalton Transactions
Page 20 of 39
View Article Online
DOI: 10.1039/C6DT00995F
1H NMR (400 MHz): 8.32 (d, J = 8.0, 1H), 8.17 (m, 1H), 8.09 (dd, J = 13.2, 7.6, 1H), 7.82ꢀ7.76
(m, 3H), 7.60 (td, J = 7.2, 2.0, 1H), 7.53ꢀ7.40 (m, 5H), 3.86 (d, 3JHP = 12.4, 3H). 13C{1H} NMR
2
(101 MHz): 141.1ꢀ121.5 (C, CH, Ar), 54.2 (d, JCP = 2.3, CH3). 31P{1H} NMR (121 MHz):
+110.6 (d, br, J ≈ 89). HRMS: calcd. for C19H22BNOPS ([M] + NH4), 354.1253; found,
354.1252. [α]D = –78.0º.
3ꢀTA, (R)ꢀMethoxyphenyl(1ꢀthianthrenyl)phosphineꢀborane
The procedure was similar to that used to prepare 3ꢀDBF but starting from precursor 2ꢀTA
(1.15 g, 2.3 mmol) and stirring for 3 days. The desired phosphiniteꢀborane was obtained as a
white pasty solid. Yield: 447 mg (53%).
1H NMR (400 MHz): 7.91 (ddd, J = 11.2, 7.6, 1.2, 1H), 7.74ꢀ7.69 (m, 3H), 7.52 (td, J = 7.2,
1.2, 1H), 7.46ꢀ7.37 (m, 4H), 7.21 (td, J = 7.6, 1.6, 1H), 7.12 (td, J = 7.6, 1.2, 1H), 6.96 (dd, J =
3
7.6, 1.2, 1H), 3.80 (d, JHP = 12.4, 3H). 13C{1H} NMR (101 MHz): 140.0ꢀ127.0 (C, CH, Ar),
54.1 (d, 2JCP = 2.5, CH3). 31P{1H} NMR (162 MHz): +109.3 (d, br, J ≈ 83). HRMS: calcd. for
C19H22BNOPS2 ([M] + NH4), 386.0973; found, 386.0976. [α]D = –10.5º.
3ꢀDBTO2, (R)ꢀ(4ꢀdibenzothiophenyl dioxide)methoxyphenylphosphineꢀborane
The procedure was the same used to prepare 3ꢀDBF but starting from precursor 2ꢀDBTO2 (1.00
g, 3.0 mmol). The desired phosphiniteꢀborane was obtained as a white solid. Yield: 433 mg
(59%).
1H NMR (400 MHz): 7.93ꢀ7.78 (m, 6H), 7.68ꢀ7.62 (m, 2H), 7.56 (td, J = 7.6, 0.8, 1H), 7.52ꢀ
3
7.42 (m, 3H), 3.99 (d, JHP = 12.0, 3H). 13C{1H} NMR (101 MHz): 139.4ꢀ121.3 (C, CH, Ar),
55.4 (d, 2JCP = 2.0, CH3). 31P{1H} NMR (162 MHz): +112.0 (d, br, J ≈ 73). HRMS: calcd. for
C19H22BNO3PS ([M] + NH4), 386.1151; found, 386.1156. [α]D = –291.1º.
4ꢀDBFꢀMe, (S)ꢀ(4ꢀdibenzofuranyl)methylphenylphosphineꢀborane
Methoxyphosphineꢀborane 3ꢀDBF (673 mg, 2.1 mmol) was dissolved in 30 mL diethyl ether,
and the solution was cooled down to –30 ºC. A 1.6 M MeLi solution in diethyl ether (2.6 mL,
4.2 mmol) was added by syringe and the mixture was stirred for 1 h before slowly warming it
up to room temperature. About 15 mL of water were added and the mixture was extracted with
diethyl ether (3 x 10 mL), the combined organic phases are washed with 20 mL of water and
dried with anhydrous sodium sulfate. After filtration, the solvent was removed in vacuo and the
crude product was purified by column chromatography (flash SiO2, 95:5 hexane/ethyl acetate).
The title product was obtained as a colourless oil. Yield: 523 mg (82%).
1H NMR (400 MHz): 8.11 (d, J = 7.6, 1H), 8.00ꢀ7.95 (m, 2H), 7.85ꢀ7.80 (m, 2H), 7.59ꢀ7.55 (m,
2
2H), 7.52ꢀ7.33 (m, 5H), 2.23 (d, JHP = 10.8, 3H). 13C{1H} NMR (101 MHz): 156.9ꢀ111.6 (C,
1
CH, Ar), 11.1 (d, JCP = 41.6, CH3). 31P{1H} NMR (121 MHz): +7.7 (d, br, J ≈ 81). HRMS:
calcd. for C19H22BNOP ([M] + NH4), 322.1532; found, 322.1530. [α]D = +140.8º.
20

Page 21 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
4ꢀDBFꢀiPr, (S)ꢀ(4ꢀdibenzofuranyl)isopropylphenylphosphineꢀborane
The procedure was the same used to prepare 4ꢀDBFꢀMe. Starting from 3ꢀDBF (1.15 g, 3.6
mmol) and 0.7 M iꢀPrLi solution in pentane (15.2 mL, 10.8 mmol) the desired phosphiniteꢀ
borane was obtained as a colourless oil. Yield: 897 mg (75%).
1H NMR (400 MHz): 8.15ꢀ7.97 (m, 5H), 7.66 (d, J = 8.0, 1H), 7.53 (td, J = 7.6, 1.2, 1H), 7.45ꢀ
7.38 (m, 5H), 3.67 (m, 1H), 1.29 (dd, 3JHP, 3JHH = 17.2, 7.2, 3H), 1.16 (dd, 3JHP, 3JHH = 17.2, 6.8,
1
3H). 13C{1H} NMR (101 MHz): 156.7ꢀ111.7 (C, CH, Ar), 22.6 (d, JCP = 37.9, CH), 17.1 (d,
2
2JCP = 2.8, CH3), 16.8 (d, JCP = 2.9, CH3). 31P{1H} NMR (121 MHz): +26.1 (d, br, J ≈ 77).
HRMS: calcd. for C21H26BNOP ([M] + NH4), 350.1845; found, 350.1842. [α]D = +228.9º.
4ꢀDBFꢀtBu, (S)ꢀ(tertꢀbutyl)(4ꢀdibenzofuranyl)phenylphosphineꢀborane
The procedure was the same used to prepare 4ꢀDBFꢀMe. Starting from 3ꢀDBF (732 mg, 2.1
mmol) and 1.6 M tꢀBuLi solution (1.5 mL, 2.3 mmol) the desired phosphiniteꢀborane was
obtained as a colourless oil. Yield: 440 mg (60%).
1H NMR (400 MHz): 8.17ꢀ8.12 (m, 2H), 7.98 (td, J = 7.6, 0.8, 1H), 7.90ꢀ7.85 (m, 2H), 7.49ꢀ
3
7.43 (m, 4H), 7.42ꢀ7.36 (m, 3H), 1.41 (d, JHP = 14.8, 9H). 13C{1H} NMR (101 MHz): 155.4ꢀ
111.7 (C, CH, Ar), 31.8 (d, 1JCP = 31.3, C), 27.8 (d, 2JCP = 3.0, CH3). 31P{1H} NMR (121 MHz):
+36.7 (d, br, J ≈ 70). HRMS: calcd. for C22H28BNOP ([M] + NH4), 364.2002; found, 350.2014.
[α]D = +82.3º.
4ꢀDBFꢀFc, (S)ꢀ(4ꢀdibenzofuranyl)ferrocenylphenylphosphineꢀborane
Ferrocene (2.5 g, 13.4 mmol) was dissolved in 20 mL of THF in a Schlenk flask. The solution
was cooled to 0 ºC, 1.6 M tꢀBuLi solution in pentane (16.7 mL, 26.9 mmol) was added by
syringe and the mixture left stirring for 2 h. At this point 40 mL of hexane were added and the
solution was cooled down to –78 ºC, which caused the precipitation of FcLi. The solid was
filtered under nitrogen, washed with hexane and dried in vacuo. In parallel, 3ꢀDBF (2.15 g, 6.7
mmol) was dissolved in 20 mL of THF and the solution was cooled down to –78 ºC. Solid FcLi
was rapidly added to that solution and the mixture was left stirring for 14 h. About 15 mL of
water were added and most of the THF was removed in vacuo. The mixture was extracted with
dichloromethane (3 x 10 mL), the combined organic phases are washed with 20 mL of water
and dried with anhydrous sodium sulfate. After filtration, the solvent was removed in vacuo and
the red crude product was purified by column chromatography (flash SiO2, 70:30
hexane/dichloromethane) and recrystallised in dichloromethane/hexane. The title product was
obtained as an orange solid. Yield: 1.80 g (56%).
1H NMR (400 MHz): 8.13 (d, J = 7.6, 1H), 8.04 (dd, J = 12.8, 7.6 1H), 7.96 (d, J = 6.8, 1H),
7.67ꢀ7.62 (m, 2H), 7.48 (t, J = 7.6, 1H), 7.44ꢀ7.35 (m, 6H), 4.69 (s, br, 2H), 4.53 (s, br, 1H),
4.49 (s, br, 1H), 4.00 (s, br, 5H). 13C{1H} NMR (101 MHz): 156.6ꢀ111.8 (C, CH, Ar), 74.1 (d,
JCP = 13.5, CH), 73.0 (d, JCP = 7.8, CH), 71.9 (d, JCP = 7.7, CH), 71.6 (d, JCP = 8.5, CH), 69.7 (s,
21

Dalton Transactions
Page 22 of 39
View Article Online
DOI: 10.1039/C6DT00995F
5CH), 67.7 (d, JCP = 70.5, C). 31P{1H} NMR (162 MHz): +12.6 (d, br, J ≈ 43). HRMS: calcd.
for C28H21FeOP ([M] – BH3), 460.0679; found, 460.0663. [α]D = +65.4º.
4ꢀDBTꢀMe, (S)ꢀ(4ꢀdibenzothiophenyl)methylphenylphosphineꢀborane
The procedure was the same used to prepare 4ꢀDBFꢀMe. Starting from 3ꢀDBT (580 mg, 1.7
mmol) and 1.6 M MeLi solution (1.2 mL, 1.7 mmol) the desired phosphineꢀborane was obtained
as a colourless oil. Yield: 357 mg (70%).
1H NMR (400 MHz): 8.31 (dt, J = 8.0, 1.2, 1H), 8.16 (m, 1H), 8.08 (ddd, J = 12.8, 7.2, 1.2,
1H), 7.74 (m, 1H), 7.70ꢀ7.64 (m, 2H), 7.61 (td, J = 7.6, 1.6, 1H), 7.52ꢀ7.40 (m, 5H), 2.09 (d,
1
2JHP = 10.0, 3H). 13C{1H} NMR (101 MHz): 142.2ꢀ121.5 (C, CH, Ar), 10.0 (d, JCP = 40.2,
CH3). 31P{1H} NMR (121 MHz): +13.2 (d, br, J ≈ 77). HRMS: calcd. for C19H22BNPS ([M] +
NH4), 338.1304; found, 338.1293. [α]D = +41.5º.
4ꢀDBTꢀiPr, (S)ꢀ(4ꢀdibenzothiophenyl)isopropylphenylphosphineꢀborane
The procedure was the same used to prepare 4ꢀDBFꢀiPr. Starting from 3ꢀDBT (1.00 g, 3.0
mmol) and 0.7 M iꢀPrLi solution in pentane (6.4 mL, 4.5 mmol) the desired phosphineꢀborane
was obtained as a colourless oil. Yield: 985 mg (95%).
1H NMR (400 MHz): 8.10 (dt, J = 8.0, 1.6, 1H), 8.02 (dd, J = 7.6, 1.2, 1H), 7.96 (m, 1H), 7.71ꢀ
7.66 (m, 2H), 7.58 (m, 1H), 7.41 (td, J = 7.6, 2.0, 1H), 7.34ꢀ7.24 (m, 3H), 7.11 (d, J = 7.2, 1H),
7.04 (d, br, J = 8.4, 1H), 3.17 (m, 1H), 1.23 (dd, 3JHP, 3JHH = 16.0, 6.8, 3H), 1.01 (dd, 3JHP, 3JHH
= 16.8, 6.8, 3H). 13C{1H} NMR (101 MHz): 142.3ꢀ121.5 (C, CH, Ar), 21.3 (d, 1JCP = 36.0, CH),
17.3 (d, 2JCP = 1.6, CH3), 17.1 (d, 2JCP = 2.5, CH3). 31P{1H} NMR (121 MHz): +29.7 (d, br, J ≈
50). HRMS: calcd. for C21H26BNPS ([M] + NH4), 366.1617; found, 366.1622. [α]D = +40.4º.
4ꢀDBTꢀFc, (S)ꢀ(4ꢀdibenzothiophenyl)ferrocenylphenylphosphineꢀborane
The procedure was the same used to prepare 4ꢀDBTꢀFc. Starting from ferrocene (1.31 g, 7.0
mmol) and 3ꢀDBT (1.18 g, 3.5 mmol) the desired phosphineꢀborane was obtained as an orange
solid. Yield: 1.36 g (79%).
1H NMR (400 MHz): 8.26 (d, J = 7.6, 1H), 8.14 (dd, J = 6.4, 2.4, 1H), 7.76ꢀ7.67 (m, 4H), 7.56ꢀ
7.39 (m, 6H), 4.75 (s, br, 1H), 4.56 (s, br, 2H), 4.43 (s, br, 1H), 4.08 (s, br, 5H). 13C{1H} NMR
(101 MHz): 139.9ꢀ121.5 (C, CH, Ar), 74.1 (d, JCP = 12.4, CH), 72.8 (d, JCP = 7.9, CH), 72.1 (d,
JCP = 7.3, CH), 71.9 (d, JCP = 8.4, CH), 69.9 (s, 5CH), 67.8 (d, JCP = 69.7, C). 31P{1H} NMR
(162 MHz): +18.9 (s, br). HRMS: calcd. for C28H22FePS ([M] + H – BH3), 477.0529; found,
477.0537. [α]D = –98.2º.
4ꢀTAꢀMe, (S)ꢀmethylphenyl(1ꢀthianthrenyl)phosphineꢀborane
22

Page 23 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
The procedure was the same used to prepare 4ꢀDBFꢀMe. Starting from 3ꢀTA (200 mg, 0.5
mmol) and 1.6 M MeLi solution (0.7 mL, 1.1 mmol) the desired phosphineꢀborane was obtained
as a white pasty solid. Yield: 166 mg (87%).
1H NMR (400 MHz): 7.94 (ddd, J = 12.8, 8.0, 1.6, 1H), 7.71 (dt, J = 7.6, 1.6, 1H), 7.57 (dt, J =
11.2, 1.6, 1H), 7.55 (dd, J = 11.2, 1.6, 4H), 7.44ꢀ7.36 (m, 2H), 7.23 (m, 1H), 7.16 (td, J = 7.6,
2
1.2, 1H), 7.06 (dd, J = 7.6, 1.6, 1H), 2.09 (d, JHP = 10.0, 3H). 13C{1H} NMR (101 MHz):
140.7ꢀ126.3 (C, CH, Ar), 11.8 (d, 1JCP = 40.5, CH3). 31P{1H} NMR (162 MHz): +15.9 (d, br, J ≈
51). HRMS: calcd. for C19H17BPS2 ([M] – H), 351.0602; found, 351.0602. [α]D = +115.8º.
L1 (5ꢀDBFꢀOMe), (R)ꢀ(4ꢀdibenzofuranyl)methoxyphenylphosphine
Phosphiniteꢀborane 3ꢀDBF (240 mg, 0.72 mmol) was dissolved in 5 mL of morpholine and the
solution was stirred at 40 ºC for 14 h. Morpholine was removed under vacuum and the gummy
residue was purified by column chromatography (Al2O3, toluene) to yield the title product as a
dense, colourless oil. Yield: 190 mg (81%).
1H NMR (400 MHz): 7.98ꢀ7.93 (m, 3H), 7.64 (td, J = 8.0, 2.0, 1H), 7.58 (d, J = 8.0, 1H), 7.53
3
(m, 1H), 7.49ꢀ7.42 (m, 2H), 7.40ꢀ7.31 (m, 4H), 3.80 (d, JHP = 14.0, 3H). 13C{1H} NMR (101
MHz): 156.2ꢀ111.0 (C, CH, Ar), 57.3 (d, 2JCP = 20.6, CH3). 31P{1H} NMR (121 MHz): +106.8
(s).
L2 (5ꢀDBTꢀOMe), (R)ꢀ(4ꢀdibenzothiophenyl)methoxyphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 3ꢀDBT (328 mg, 0.98
mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 220 mg (70%).
1H NMR (400 MHz): 8.19ꢀ8.13 (m, 3H), 7.88ꢀ7.83 (m, 2H), 7.65ꢀ7.57 (m, 2H), 7.53ꢀ7.43 (m,
3
3H), 7.37ꢀ7.34 (m, 2H), 3.78 (d, JHP = 14.0, 3H). 13C{1H} NMR (101 MHz): 139.4ꢀ121.5 (C,
CH, Ar), 57.2 (d, 2JCP = 20.0, CH3). 31P{1H} NMR (162 MHz): +114.2 (s).
L3 (5ꢀTAꢀOMe), (R)ꢀMethoxyphenyl(1ꢀthianthrenyl)phosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 3ꢀTA (630 mg, 1.71
mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 510 mg (84%).
1H NMR (400 MHz): 7.61ꢀ7.41 (m, 5H), 7.38ꢀ7.30 (m, 4H), 7.26ꢀ7.17 (m, 2H), 3.71 (d, 3JHP
14.0, 3H). 13C{1H} NMR (101 MHz): 142.3ꢀ127.4 (C, CH, Ar), 57.1 (d, JCP = 21.8, CH3).
31P{1H} NMR (162 MHz): +104.4 (s).
=
2
L4 (5ꢀDBTO2ꢀOMe), (R)ꢀ(4ꢀdibenzothiophenylꢀS,Sꢀdioxide)methoxyphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 3ꢀDBTO2 (200 mg,
0.54 mmol) the desired phosphineꢀborane was obtained as a white solid. Yield: 150 mg (78%).
23

Dalton Transactions
Page 24 of 39
View Article Online
DOI: 10.1039/C6DT00995F
1H NMR (400 MHz): 7.82 (dq, J = 7.6, 0.4, 1H), 7.75ꢀ7.63 (m, 5H), 7.59 (tt, J = 8.0, 0.8, 1H),
3
7.56ꢀ7.47 (m, 2H), 7.41ꢀ7.30 (m, 3H), 3.78 (d, JHP = 14.4, 3H). 13C{1H} NMR (101 MHz):
141.1ꢀ121.4 (C, CH, Ar), 57.2 (d, 2JCP = 21.9, CH3). 31P{1H} NMR (162 MHz): +101.4 (s).
L5 (5ꢀDBFꢀMe), (S)ꢀ(4ꢀdibenzofuranyl)methylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBFꢀMe (500 mg,
1.56 mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 329 mg
(69%).
1H NMR (300 MHz): 8.00ꢀ7.92 (m, 2H), 7.62ꢀ7.54 (m, 3H), 7.51ꢀ7.43 (m, 2H), 7.40ꢀ7.30 (m,
5H), 1.86 (d, 2JHP = 3.9, 3H). 13C{1H} NMR (101 MHz): 158.0ꢀ111.6 (C, CH, Ar), 11.1 (d, 1JCP
= 12.8, CH3). 31P{1H} NMR (121 MHz): –37.0 (s)
L6 (5ꢀDBFꢀiPr), (S)ꢀ(4ꢀdibenzofuranyl)isopropylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBFꢀiPr (185 mg,
0.56 mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 120 mg
(68%).
1H NMR (300 MHz): 7.94 (d, J = 7.5, 2H), 7.67ꢀ7.52 (m, 4H), 7.45 (td, J = 7.2, 1.2, 1H), 7.37ꢀ
7.29 (m, 5H), 2.91 (dd, J = 7.2, 6.9, 1H), 1.16 (dd, J = 13.2, 6.9, 3H), 1.11 (dd, J = 13.2, 6.9,
3H). 13C{1H} NMR (101 MHz): 159.7ꢀ110.0 (C, CH, Ar), 22.4 (d, 1JCP = 6.7, CH), 19.9 (d, 2JCP
= 7.1, CH3), 19.7 (d, 2JCP = 8.8, CH3). 31P{1H} NMR (121 MHz): –10.7 (s).
L7 (5ꢀDBFꢀtBu), (S)ꢀ(tertꢀbutyl)(4ꢀdibenzofuranyl)phenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBFꢀtBu (600 mg,
1.73 mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 488 mg
(85%).
1H NMR (300 MHz): 7.97 (dd, J = 7.5, 1.2, 1H), 7.95 (ddd, J = 7.8, 1.5, 0.8, 1H), 7.68ꢀ7.57 (m,
4H), 7.44 (td, J = 7.2, 1.5, 1H), 7.38ꢀ7.30 (m, 5H), 1.28 (d, 3JHP = 13.2, 9H). 31P{1H} NMR (121
MHz): +0.6 (s).
L8 (5ꢀDBFꢀFc), (S)ꢀ(4ꢀdibenzofuranyl)ferrocenylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBFꢀFc (600 mg,
1.26 mmol) the desired phosphineꢀborane was obtained as an orange solid. Yield: 490 mg
(84%).
1H NMR (400 MHz): 7.95 (dq, J = 7.6, 0.8 1H), 7.94 (ddd, J = 7.6, 1.2, 0.4, 1H), 7.57 (d, J =
8.0, 1H), 7.48ꢀ7.41 (m, 3H), 7.35ꢀ7.29 (m, 4H), 7.27 (d, J = 7.2, 1H), 7.15 (m, 1H), 4.39 (m,
2H), 4.18 (m, 1H), 4.15 (m, 1H), 4.08 (m, 5H). 13C{1H} NMR (101 MHz): 156.0ꢀ111.6 (C, CH,
Ar), 73.1 (d, JCP = 15.0, CH), 73.0 (d, JCP = 15.4, CH), 70.9 (d, JCP = 4.0, CH), 70.7 (d, JCP
4.0, CH), 69.1 (s, 5CH), 67.9 (s, C). 31P{1H} NMR (162 MHz): –32.1 (s).
=
24

Page 25 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
L9 (5ꢀDBFꢀDBT), (S)ꢀ(4ꢀdibenzofuranyl)(4ꢀdibenzothiophenyl)phenylphosphine
Dibenzothiophene (210 mg, 1.1 mmol) was dissolved in 20 mL of THF in a Schlenk flask. The
solution was cooled to –78 ºC and then 1.6 M nꢀBuLi solution in hexanes (0.7 mL, 1.1 mmol)
was added by syringe. The resulting brown solution was removed from the cold bath, left
stirring at 0 ºC for 5 h and recooled to –78 ºC. At the same time phosphiniteꢀborane 5ꢀDBFꢀ
OMe (350 mg, 1.1 mmol) was dissolved in 20 mL of THF and the solution was cooled down to
–78 ºC. The content of the first flask was slowly transferred to the second Schlenk flask via
cannula and the resulting mixture was stirred for 14 h. Around 20 mL of deoxygenated water
were added to the brownꢀyellow solution and THF was evaporated. The white residue was
extracted with dichloromethane (3 x 30 mL) under nitrogen atmosphere and the combined
organic phases were washed with deoxygenated water and dried with anhydrous sodium sulfate.
The suspension was filtered under nitrogen and the solvent was evaporated to dryness, leaving a
white pasty solid. 10 mL of morpholine were added and the solution was stirred at 40 ºC for 14
h. Morpholine was removed under vacuum and the gummy residue was purified by column
chromatography (Al2O3, toluene) to yield the title product as a pale brown solid. Yield: 307 mg
(61%).
1H NMR (400.1 MHz): 8.17 (m, 2H), 7.96 (m, 2H), 7.86 (m, 1H), 7.58 (d, J = 8.4, 1H), 7.46
(m, 7H), 7.35 (m, 3H), 7.27 (t, J = 7.6, 1H), 7.10 (m, 1H), 7.02 (m, 1H). 31P{1H} NMR (162
MHz): –23.1 (s). HRMS: calcd. for C30H20OPS ([M] + H), 459.0972; found, 459.0975.
L10 (5ꢀDBTꢀMe), (S)ꢀ(4ꢀdibenzothiophenyl)methylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBTꢀMe (450 mg,
1.41 mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 380 mg
(88%).
1H NMR (400 MHz): 8.17ꢀ8.13 (m, 2H), 7.84 (m, 1H), 7.51ꢀ7.42 (m, 6H), 7.35ꢀ7.31 (m, 3H),
2
1
1.78 (d, JHP = 3.2, 3H). 13C{1H} NMR (101 MHz): 145.0ꢀ121.5 (C, CH, Ar), 11.2 (d, JCP
=
13.1, CH3). 31P{1H} NMR (162 MHz): –30.6 (s).
L11 (5ꢀDBTꢀiPr), (S)ꢀ(4ꢀdibenzothiophenyl)isopropylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBTꢀiPr (490 mg,
1.41 mmol) the desired phosphineꢀborane was obtained as a colourless oil. Yield: 400 mg
(85%).
1H NMR (400 MHz): 8.19ꢀ8.10 (m, 3H), 7.85 (m, 1H), 7.61ꢀ7.55 (m, 2H), 7.52ꢀ7.41 (m, 4H),
7.33ꢀ7.29 (m, 2H), 2.70 (m, 1H), 1.18 (dd, 3JHP, 3JHH = 6.8, 2.4, 3H), 1.14 (dd, 3JHP, 3JHH = 6.8,
2.4, 3H). 13C{1H} NMR (101 MHz): 147.2ꢀ121.5 (C, CH, Ar), 25.0 (d, 1JCP = 7.7, CH), 19.8 (d,
2JCP = 6.7, CH3), 19.6 (d, 2JCP = 7.8, CH3). 31P{1H} NMR (162 MHz): –5.1 (s).
25

Dalton Transactions
Page 26 of 39
View Article Online
DOI: 10.1039/C6DT00995F
L12 (5ꢀDBTꢀFc), (S)ꢀ(4ꢀdibenzothiophenyl)ferrocenylphenylphosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀDBTꢀFc (350 mg,
0.71 mmol) the desired phosphineꢀborane was obtained as an orange solid. Yield: 300 mg
(89%).
1H NMR (400 MHz): 8.14 (d, J = 4.4, 1H), 8.11 (d, J = 7.2, 1H), 7.80 (t, J = 4.8, 1H), 7.54ꢀ7.48
(m, 2H), 7.45ꢀ7.38 (m, 3H), 7.35ꢀ7.30 (m, 3H), 7.22 (m, 1H), 4.45 (s, br, 1H), 4.41 (s, br, 1H),
4.38 (s, br, 1H), 4.09 (s, br, 5H), 4.06 (s, br, 1H). 13C{1H} NMR (101 MHz): 144.6ꢀ121.6 (C,
CH, Ar), 74.2 (d, JCP = 4.2, C), 73.9 (d, JCP = 23.2, CH), 72.4 (d, JCP = 6.8, CH), 71.2 (d, JCP
5.9, CH), 70.7 (d, JCP = 2.4, CH), 69.2 (s, 5CH). 31P{1H} NMR (162 MHz): –20.5 (s).
=
L13 (5ꢀTAꢀMe), (S)ꢀmethylphenyl(1ꢀthianthrenyl)phosphine
The procedure was the same used to prepare 5ꢀDBFꢀOMe. Starting from 4ꢀTAꢀMe (350 mg,
0.99 mmol) the desired phosphineꢀborane was obtained as colourless, dense oil. Yield: 280 mg
(83%).
1H NMR (400 MHz): 7.50ꢀ7.41 (m, 4H), 7.36ꢀ7.33 (m, 3H), 7.26ꢀ7.18 (m, 4H), 7.15 (ddd, J =
7.6, 4.4, 1.2, 1H), 1.65 (d, 2JHP = 4.8, 3H). 13C{1H} NMR (101 MHz): 140.8ꢀ126.6 (C, CH, Ar),
12.3 (d, 1JCP = 14.6, CH3). 31P{1H} NMR (162 MHz): –32.1 (s).
Synthesis of the Ru complexes
Ru1, [RuCl2(η6ꢀpꢀcymene)(L1)]
Phosphinite L1 (214 mg, 0.70 mmol) was dissolved in 20 mL of dichloromethane, Ru dimer D1
(150 mg, 0.25 mmol) was added and the dark red solution was stirred for 1 h. The solvent was
removed under vacuum and the residue was recrystallised in dichloromethane/hexane, to furnish
the title product as a dark red solid. Yield: 246 mg (80%).
IR: 3051, 2958, 2869, 1580, 1469, 1450, 1400, 1185, 1109, 1032, 845, 804, 757, 696, 562. 1H
NMR (400 MHz): 8.37 (ddd, J = 11.2, 7.6, 1.2, 1H), 8.05ꢀ8.00 (m, 3H), 7.93 (dt, J = 7.6, 1.2,
1H), 7.43ꢀ7.32 (m, 7H), 5.41 (d, J = 6.6, 1H), 5.37 (d, J = 6.0, 1H), 5.33 (d, J = 6.0, 1H), 5.05
(d, J = 6.0, 1H), 3.63 (d, 3JHP = 12.0, 3H), 2.72 (setp, 3JHH = 6.8, 1H), 1.97 (s, 3H), 1.01 (d, 3JHH
= 6.8, 3H), 0.88 (d, 3JHH = 6.8, 3H). 13C{1H} NMR (101 MHz): 155.6ꢀ111.6 (C, CH, Ar), 110.9
2
(d, 2JCP = 1.7, C), 96.5 (s, C), 92.6 (d, JCP = 5.6, CH), 90.3 (d, 2JCP = 3.9, CH), 88.3 (d, 2JCP
=
7.2, CH), 86.5 (d, 2JCP = 5.6, CH), 55.1 (d, 2JCP = 5.1, CH3), 30.0 (s, CH), 21.8 (s, CH3), 21.0 (s,
CH3), 17.6 (s, CH3). 31P{1H} NMR (121 MHz): +112.7 (s). Anal.: calcd. for C29H29Cl2O2PRu,
C 56.87%, H 4.77%; found, C 57.29%, H 5.03%.
Ru2, [RuCl2(η6ꢀpꢀcymene)(L2)]
26

Page 27 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
The procedure was the same followed to prepare Ru1. Starting from L2 (220 mg, 0.68 mmol)
and Ru dimer D1 (149 mg, 0.24 mmol), the desired complex was obtained as a dark red solid.
Yield: 217 mg (72%).
IR: 3053, 2958, 2870, 1439, 1375, 1103, 1028, 756, 695, 554. 1H NMR (300 MHz): 8.34 (ddd,
J = 13.2, 7.5, 0.9, 1H), 8.23 (dt, J = 8.0, 1.5, 1H), 8.18ꢀ8.14 (m, 1H), 7.99ꢀ7.93 (m, 2H), 7.84ꢀ
7.81 (m, 1H), 7.53ꢀ7.45 (m, 3H), 7.39ꢀ7.32 (m, 3H), 5.44 (d, J = 6.0, 1H), 5.36 (d, J = 6.9, 1H),
5.33 (d, J = 7.8, 1H), 5.24 (d, J = 6.0, 1H), 3.70 (d, 3JHP = 11.7, 3H), 2.70 (setp, 3JHH = 7.2, 1H),
3
3
1.89 (s, 3H), 1.02 (d, JHH = 6.9, 3H), 1.00 (d, JHH = 6.9, 3H). 13C{1H} NMR (101 MHz):
140.0ꢀ121.4 (C, CH, Ar), 111.6 (d, 2JCP = 1.1, C), 96.9 (s, C), 91.7 (d, 2JCP = 4.1, CH), 91.4 (d,
2JCP = 4.6, CH), 87.5 (d, 2JCP = 6.6, CH), 87.2 (d, 2JCP = 5.9, CH), 54.6 (d, 2JCP = 3.6, CH3), 30.1
(s, CH), 21.7 (s, CH3), 21.5 (s, CH3), 17.5 (s, CH3). 31P{1H} NMR (121 MHz): +118.3 (s).
Anal.: calcd. for C29H29Cl2OPRuS, C 55.42%, H 4.65%, S 5.10%; found, C 55.97%, H 5.01%,
S 4.89%.
Ru3, [RuCl2(η6ꢀpꢀcymene)(L3)]
The procedure was the same followed to prepare Ru1. Starting from L3 (270 mg, 0.76 mmol)
and Ru dimer D1 (186 mg, 0.30 mmol), the desired complex was obtained as a dark red solid.
Yield: 217 mg (55%).
1
IR: 3052, 2959, 2869, 1470, 1448, 1435, 1378, 1109, 1029, 752, 694, 550. H NMR (400
MHz): 8.30 (dd, J = 11.6, 7.6, 1H), 7.84 (t, J = 9.2, 2H), 7.64 (d, J = 7.6, 1H), 7.46ꢀ7.34 (m,
5H), 7.21 (td, J = 7.2, 1.2, 1H), 7.11 (td, J = 7.6, 1.2, 1H), 7.00 (dd, J = 7.6, 1.2, 1H), 5.41 (d, J
= 6.4, 1H), 5.31 (d, J = 6.0, 1H), 5.27 (d, J = 6.0, 1H), 5.18 (d, J = 5.6, 1H), 3.61 (d, 3JHP = 11.6,
3H), 2.63 (sept, 3JHH = 6.8, 1H), 1.93 (s, 3H), 0.92 (d, 3JHH = 7.2, 3H), 0.86 (d, 3JHH = 7.2, 3H).
31P{1H} NMR (162 MHz): +110.2 (s). HRMS: calcd. for C29H29ClOPRuS2 ([M] – Cl),
625.0123; found, 625.0126.
Ru4, [RuCl2(η6ꢀpꢀcymene)(L4)]
The procedure was the same followed to prepare Ru1. Starting from L4 (70 mg, 0.20 mmol)
and Ru dimer D1 (48 mg, 0.08 mmol), the desired complex was obtained as a dark red solid.
Yield: 70 mg (68%).
IR: 3060, 2959, 2869, 1446, 1436, 1387, 1308, 1154, 1095, 1045, 815, 764, 721, 701, 584, 568,
468. 1H NMR (400 MHz): 8.13 (tt, J = 8.4, 1.6, 2H), 7.98 (ddd, J = 13.2, 7.6, 0.8, 1H), 7.91 (d,
J = 6.8, 1H), 7.86 (t, J = 7.6, 2H), 7.71 (td, J = 7.6, 1.2, 1H), 7.64 (td, J = 7.6, 0.8, 1H), 7.52 (td,
J = 7.6, 2.0, 1H), 7.37ꢀ7.27 (m, 3H), 5.68 (d, J = 6.0, 1H), 5.62 (dd, J = 6.4, 1.2, 1H), 5.50 (d, J
= 6.0, 1H), 5.31 (d, J = 6.0, 1H), 3.72 (d, 3JHP = 11.6, 3H), 2.72 (sept, 3JHH = 6.8, 1H), 1.87 (s,
3H), 1.17 (d, 3JHH = 6.8, 3H), 0.97 (d, 3JHH = 6.8, 3H). 13C{1H} NMR (101 MHz): 141.9ꢀ121.2
(C, CH, Ar), 112.1 (s, C), 97.4 (s, C), 93.8 (d, 2JCP = 5.4, CH), 90.9 (d, 2JCP = 3.8, CH), 87.6 (d,
2JCP = 6.3, CH), 86.7 (d, 2JCP = 5.9, CH), 54.2 (d, 2JCP = 4.0, CH3), 30.1 (s, CH), 22.3 (s, CH3),
21.1 (s, CH3), 17.5 (s, CH3). 31P{1H} NMR (162 MHz): +117.0 (s). Anal.: calcd. for
C29H29Cl2O3PRuS, C 52.73%, H 4.42%, S 4.85%; found, C 51.15%, H 4.51%, S 4.42%.
27

Dalton Transactions
Page 28 of 39
View Article Online
DOI: 10.1039/C6DT00995F
Ru5, [RuCl2(η6ꢀpꢀcymene)(L5)]
The procedure was the same followed to prepare Ru1. Starting from L5 (185 mg, 0.64 mmol)
and Ru dimer D1 (162 mg, 0.26 mmol), the desired complex was obtained as a dark red solid.
Yield: 201 mg (65%).
IR: 3049, 2958, 2919, 2868, 1583, 1469, 1449, 1399, 1185, 1109, 1057, 898, 843, 802, 755,
1
725, 696, 556, 424. H NMR (300 MHz): 8.11 (dt, J = 7.8, 1.2, 1H), 8.04ꢀ8.00 (m, 1H), 7.96
(ddd, J = 7.8, 1.5, 0.9, 1H), 7.89 (ddd, J = 11.1, 7.8, 1.2, 1H), 7.78ꢀ7.71 (m, 2H), 7.59ꢀ7.56 (m,
1H), 7.52ꢀ7.31 (m, 5H), 5.59 (d, J = 6.3, 1H), 5.49 (d, J = 6.3, 1H), 5.47 (m, 1H), 4.76 (d, J =
5.7, 1H), 2.54 (sept, 3JHH = 6.9, 1H), 2.06 (s, 3H), 2.04 (d, 2JHP = 11.4, 3H), 0.90 (d, 3JHH = 6.9,
3H), 0.38 (d, 3JHH = 6.9, 3H). 13C{1H} NMR (101 MHz): 156.1ꢀ111.6 (C, CH, Ar), 107.8 (s, C),
94.2 (s, C), 93.8 (d, 2JCP = 6.7, CH), 89.6 (d, 2JCP = 8.8, CH), 86.4 (d, 2JCP = 2.6, CH), 81.4 (d,
1
2JCP = 3.2, CH), 29.8 (s, CH), 22.8 (s, CH3), 19.1 (s, CH3), 17.5 (s, CH3), 12.1 (d, JCP = 37.4,
CH3). 31P{1H} NMR (121 MHz): +15.2 (s). Anal.: calcd. for C29H29Cl2OPRu, C 58.39%, H
4.90%; found, C 60.59%, H 5.04%.
Ru6, [RuCl2(η6ꢀpꢀcymene)(L6)]
The procedure was the same followed to prepare Ru1. Starting from L6 (166 mg, 0.52 mmol)
and Ru dimer D1 (114 mg, 0.19 mmol), the desired complex was obtained as a dark red solid.
Yield: 145 mg (62%).
IR: 3054, 2958, 2925, 2867, 1581, 1469, 1449, 1434, 1398, 1264, 1182, 1109, 1039, 844, 802,
759, 699, 533, 516. 1H NMR (400 MHz): 8.10 (dt, J = 7.6, 1.2, 1H), 8.04 (d, J = 8.0, 1H), 8.01ꢀ
7.94 (m, 3H), 7.51ꢀ7.38 (m, 7H), 5.23 (d, J = 6.0, 2H), 5.11 (d, J = 6.0, 1H), 4.51 (d, J = 5.6,
3
1H), 3.71 (m, 1H), 2.66 (sept, JHH = 7.2, 1H), 1.89 (s, 3H), 1.10 (dd, J = 18.0, 7.2, 3H), 0.97
3
3
(dd, J = 14.0, 6.8, 3H), 0.92 (d, JHH = 6.8, 3H), 0.68 (d, JHH = 6.8, 3H). 13C{1H} NMR (101
MHz): 155.3ꢀ111.7 (C, CH, Ar), 109.2 (s, C), 93.9 (d, 2JCP = 4.5, C), 93.6 (s, CH), 88.9 (d, 2JCP
= 2.4, CH), 86.1 (d, 2JCP = 7.4, CH), 83.4 (d, 2JCP = 4.6, CH), 29.9 (s, CH), 26.2 (d, 1JCP = 23.5,
CH), 22.3 (s, CH3), 20.3 (s, CH3), 19.1 (d, 2JCP = 6.2, CH3), 18.7 (s, CH3), 17.5 (s, CH3). 31P{1H}
NMR (121 MHz): +21.5 (s). Anal.: calcd. for C31H33Cl2OPRu, C 59.62%, H 5.33%; found, C
59.08%, H 5.64%.
Ru8, [RuCl2(η6ꢀpꢀcymene)(L8)]
The procedure was the same followed to prepare Ru1. Starting from L8 (180 mg, 0.39 mmol)
and Ru dimer D1 (96 mg, 0.16 mmol), the desired complex was obtained as a dark red solid.
Yield: 180 mg (75%).
IR: 3051, 2957, 2924, 2868, 1624, 1579, 1469, 1449, 1435, 1398, 1306, 1263, 1183, 1158,
1108, 1058, 1028, 1002, 844, 821, 801, 755, 699, 560, 458. 1H NMR (400 MHz): 8.03ꢀ8.00 (m,
2H), 7.96 (d, J = 8.0, 1H), 7.93 (d, J = 7.2, 1H), 7.74 (dd, J = 10.8, 7.6, 1H), 7.53ꢀ7.47 (m, 1H),
7.46ꢀ7.41 (m, 3H), 7.40ꢀ7.29 (m, 3H), 5.56 (d, J = 6.4, 1H), 5.38 (d, J = 6.8, 1H), 5.37ꢀ5.35 (m,
1H), 4.97 (m, 1H), 4.61 (d, J = 5.6, 1H), 4.38 (m, 1H), 4.34 (m, 1H), 4.26 (m, 1H), 3.66 (s, 5H),
28

Page 29 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00995F
3
3
3
2.55 (sept, JHH = 6.8, 1H), 2.03 (s, 3H), 0.89 (d, JHH = 7.2, 3H), 0.33 (d, JHH = 6.8, 3H).
13C{1H} NMR (101 MHz): 155.0ꢀ111.4 (C, CH, Ar), 109.1 (s, C), 94.6 (d, 2JCP = 4.4, C), 93.8
2
(s, CH), 88.9 (d, JCP = 9.1, CH), 87.8 (s, CH), 81.5 (s, CH), 78.0 (d, JCP = 12.1, CH), 75.0 (d,
1JCP = 54.0, C), 74.3 (d, 2JCP = 8.2, CH), 70.0 (s, ov, 6CH), 69.6 (d, JCP = 8.1, CH), 29.6 (s, CH),
22.6 (s, CH3), 19.0 (s, CH3), 17.1 (s, CH3). 31P{1H} NMR (162 MHz): +15.8 (s). Anal.: calcd.
for C38H35Cl2FeOPRu, C 59.55%, H 4.60%; found, C 59.59%, H 5.00%.
Ru9, [RuCl2(η6ꢀpꢀcymene)(L9)]
The procedure was the same followed to prepare Ru1. Starting from L9 (140 mg, 0.31 mmol)
and Ru dimer D1 (74 mg, 0.12 mmol), the desired complex was obtained as a dark red solid.
Yield: 101 mg (55%).
1H NMR (400 MHz): 8.43 (dd, J = 13.2, 7.6 1H), 8.30ꢀ8.22 (m, 3H), 8.16ꢀ8.07 (m, 3H), 7.94
(d, J = 6.8, 1H), 7.59 (d, J = 6.4, 1H), 7.55 (d, J = 7.6, 1H), 7.43ꢀ7.20 (m, 9H), 5.27 (d, J = 6.0,
3
2H), 5.20 (d, J = 6.0, 1H), 5.04 (d, J = 6.0, 1H), 2.80 (sept, JHH = 6.8, 1H), 1.90 (s, 3H), 0.94
3
3
(d, JHH = 6.8, 3H), 0.86 (d, JHH = 6.8, 3H). 13C{1H} NMR (101 MHz): 155.9ꢀ120.7 (C, CH,
2
2
2
Ar), 111.5 (s, C), 95.7 (s, C), 91.1 (d, JCP = 2.4, CH), 89.7 (d, JCP = 3.4, CH), 86.2 (d, JCP
=
2
6.5, CH), 85.8 (d, JCP = 5.7, CH), 30.0 (s, CH), 21.6 (s, CH3), 21.3 (s, CH3), 17.6 (s, CH3).
31P{1H} NMR (162 MHz): +19.8 (s). HRMS: calcd. for C40H33ClOPRuS ([M] – Cl), 729.0716;
found, 729.0745.
Ru10, [RuCl2(η6ꢀpꢀcymene)(L10)]
The procedure was the same followed to prepare Ru1. Starting from L10 (190 mg, 0.62 mmol)
and Ru dimer D1 (140 mg, 0.23 mmol), the desired complex was obtained as a dark red solid.
Yield: 251 mg (89%).
IR: 3051, 2958, 2868, 2838, 1438, 1374, 1103, 1027, 817, 755, 694, 554. 1H NMR (400 MHz):
8.28 (dt, J = 7.6, 1.6, 1H), 8.19 (m, 1H), 8.15 (ddd, J = 12.4, 7.6, 1.2, 1H), 7.79ꢀ7.74 (m, 2H),
7.70 (dd, J = 7.2, 2.0, 1H), 7.61 (td, J = 7.6, 1.6, 1H), 7.54ꢀ7.42 (m, 5H), 5.64 (d, J = 5.6, 1H),
3
5.57 (d, J = 6.0, 1H), 5.34 (d, J = 6.0, 1H), 5.17 (d, J = 5.6, 1H), 2.55 (sept, JHH = 7.2, 1H),
2.08 (d, 2JHP = 9.6, 3H), 2.07 (s, 3H), 0.83 (d, 3JHH = 7.2, 3H), 0.57 (d, 3JHH = 7.2, 3H). 13C{1H}
2
NMR (101 MHz): 140.7ꢀ121.7 (C, CH, Ar), 107.2 (s, C), 94.5 (s, C), 91.7 (d, JCP = 5.8, CH),
88.7 (d, 2JCP = 6.9, CH), 88.1 (d, 2JCP = 3.7, CH), 83.1 (d, 2JCP = 4.6, CH), 29.8 (s, CH), 22.0 (s,
CH3), 20.0 (s, CH3), 17.5 (s, CH3), 11.2 (d, 1JCP = 37.1, CH3). 31P{1H} NMR (162 MHz): +22.6
(s). Anal.: calcd. for C29H29Cl2PRuS, C 56.86%, H 4.77%, S 5.24%; found, C 56.69%, H
5.07%, S 5.26%.
Ru11, [RuCl2(η6ꢀpꢀcymene)(L11)]
The procedure was the same followed to prepare Ru1. Starting from L11 (400 mg, 1.20 mmol)
and Ru dimer D1 (244 mg, 0.40 mmol), the desired complex was obtained as a dark red solid.
Yield: 405 mg (79%).
29