Copper(I) Complexes of P-Stereogenic Josiphos and Related Ligands

Article abstract of DOI:10.1002/ejoc.202100146

Starting from (R)-Ugi's amine, diastereoselective lithiation followed by Ar'PCl₂ and then Ar’’MgBr led to the generation, as single diastereoisomers, of (R,S_p,S_phos) [Ar’=Ph, Ar’’=o-Tol] and (R,S_p,R_phos

Full text of DOI:10.1002/ejoc.202100146

Full Papers
doi.org/10.1002/ejoc.202100146
Copper(I) Complexes of P-Stereogenic Josiphos and Related
Ligands
Ross A. Arthurs,[a] Alice C. Dean,[a] David L. Hughes,[a] and Christopher J. Richards*[a]
Starting from (R)-Ugi’s amine, diastereoselective lithiation
followed by Ar’PCl2 and then Ar’’MgBr led to the generation, as
single diastereoisomers, of (R,Sp,Sphos) [Ar’=Ph, Ar’’=o-Tol] and
(R,Sp,Rphos) [Ar’=o-Tol, Ar’’=Ph] PPFA ligand derivatives. Amine
substitution of both with HPCy2 gave P-stereogenic Josiphos
ligands, and then addition of CuCl, the corresponding copper(I)
complexes. The latter were also generated by using borane P
and N protecting groups and in situ Cu(I) complexation,
avoiding the isolation of air-sensitive phosphine intermediates.
This protection methodology was also applied to the synthesis
of Josiphos/CuCl complexes derived from PCl3. In addition,
related bulky cobalt-sandwich complex-based derivatives were
also obtained. Preliminary investigation revealed isolated CuCl
complexes as competent catalyst precursors for enantioselec-
tive conjugate addition reactions.
Introduction
Although many ligands have been developed for application in
asymmetric synthesis, relatively few have found widespread
use. Of these so-called ‘privileged’ ligands,[1] one of the most
successful class are the ferrocene-based Josiphos ligands A.[2]
Numerous derivatives are known, obtained by varying the
identity of R’ and R’’, and many are commercially available. This
has aided the incorporation of A into many ligand screening
exercises, often leading to a successful outcome.[3] In particular,
in situ generated Josiphos-copper complexes have been applied
successfully to enantioselective conjugative CÀ C,[4] CÀ H,[5] and
CÀ B[6] bond-forming reactions of activated alkenes (Scheme 1).
In addition to e.e. optimisation by varying the steric and
electronic properties of the R’ and R’’ substituents, there is the
possibility of creating an additional phosphorus-based stereo-
genic centre, such that one of the two resulting diastereomeric
ligands B may lead to higher enantioslectivity.[7] We chose to
explore this aspect of ligand-optimisation by first synthesising a
series of readily isolated and air-stable copper(I) chloride
complexes C (Ar’¼Ar’’ and Ar’=Ar’’).
Scheme 1. Ligands and copper catalysts for asymmetric conjugate addition.
by Chen,[8,9] and applied by ourselves for the synthesis of P-
stereogenic Phosferrox derivatives.[10] As in our previous report,
our aim was to control the introduction of phenyl and o-tolyl
groups to create the phosphorus-based stereogenic centre.
These aryl groups are differentiated by a methyl substituent,
with this positioned ortho so that it influences the environment
about a coordinated metal, but ideally not to the extent that
this relatively small group changes the nature of ligand
coordination. Accordingly, (R)-N,N-dimethyl-1-ferrocenylethyl-
amine 1[11] (Ugi’s amine) was lithiated at 0 C with s-butyl lithium
followed by the addition of dichlorophenylphosphine. After
°
Results and Discussion
°
For the initial synthesis of P-stereogenic Josiphos derivatives we
adapted the diastereoslective methodology reported previously
warming to room temperature and re-cooling to À 78 C (o-
tolyl)magnesium bromide was added, leading after chromatog-
raphy and recrystallisation, to the isolation of (R,Sp,Sphos)-2 as a
single diastereoisomer (Scheme 2). Use of the same procedure,
but with dichloro(o-tolyl)phosphine and phenylmagnesium
bromide, gave (R,Sp,Rphos)-3, again as a single diastereoisomer.
The configuration of these new PPFA ligand analogues was
confirmed by X-ray crystallographic analysis (Figure 1 and
Figure 2).[12] Separate reaction of both diastereoisomers with
acetic anhydride at room temperature was followed by reaction
with dicyclohexylphosphine, again at room temperature (vide
infra), to give P-stereogenic Josiphos ligand derivatives
(R,Sp,Sphos)-4 and (R,Sp,Rphos)-5 as single diastereoisomers. These
[a] Dr. R. A. Arthurs, A. C. Dean, Dr. D. L. Hughes, Dr. C. J. Richards
School of Chemistry,
University of East Anglia,
Norwich Research Park, Norwich, NR4 7TJ, U.K.
E-mail: Chris.Richards@uea.ac.uk
https://people.uea.ac.uk/chris_richards
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejoc.202100146
© 2021 The Authors. European Journal of Organic Chemistry published by
Wiley-VCH GmbH. This is an open access article under the terms of the
Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
Eur. J. Org. Chem. 2021, 2719–2725
2719
© 2021 The Authors. European Journal of Organic Chemistry
published by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
Figure 2. Representation of the X-ray crystal structure of (R,Sp,Rphos)-3 (hydro-
gen atoms omitted for clarity). Thermal ellipsoids are drawn at the 50%
probability level. Principal bond lengths [Å] include: C(2)À C(3) 1.507(5),
C(1)À P(1) 1.813(3), P(1)À C(4) 1.843(3), P(1)À C(6) 1.837(3). Principal torsion
Scheme 2. Synthesis of P-stereogenic PPFA ligands (R,Sp,Sphos)-2 and
(R,Sp,Rphos)-3 and Josiphos ligands (R,Sp,Sphos)-4 and (R,Sp,Rphos)-5.
°
angles [ ] include: C(1)À C(2)À C(3)À N(1) À 69.6(4), C(2)À C(1)À P(1)À C(4)
À 166.9(3), C(2)À C(1)À P(1)À C(6) 90.5(3), C(1)À P(1)À C(4)À C(5) 78.0(3), C(1)À P-
P(1)À C(6)À C(7) À 176.9(3). Absolute structure parameter=À 0.017(10).
stereogenic centre.[8] For the latter it was suggested that the
reaction proceeds via an NÀ P bond containing intermediate
such as (R,Sp,Sphos)-7 in which the phenyl group is oriented away
from ferrocene (Scheme 3). Subsequent addition of the
Grignard reagent (in this case o-TolMgBr) results in substitution
(SN2) to give (R,Sp,Sphos)-2. Whilst the origin of the stereo-
selection in this process could arise from the reaction of
lithiated (R)-1 with prochiral PhPCl2, an alternative explanation
is that this gives (R,Sp)-6 with little or no control at phosphorus,
followed by loss of the second chloride leaving group and
epimerisation of the resulting cyclic intermediate. This is
consistent with the complete control of this process by the
element of planar chirality, as observed also on the application
of this chemistry to lithiated (S,Sp) and (S,Rp) ferrocenyl-
oxazolines.[10]
Figure 1. Representation of the X-ray crystal structure of (R,Sp,Sphos)-2 (hydro-
gen atoms omitted for clarity). Thermal ellipsoids are drawn at the 50%
probability level. Principal bond lengths [Å] include: C(2)À C(3) 1.519(3),
C(1)À P(1) 1.822(2), P(1)À C(4) 1.839(2), P(1)À C(6) 1.838(2). Principal torsion
°
angles [ ] include: C(1)À C(2)À C(3)À N(1) À 98.4(3), C(2)À C(1)À P(1)À C(4)
130.27(19), C(2)À C(1)À P(1)À C(6) À 124.57(19), C(1)À P(1)À C(4)À C(5) 160.79(17),
C(1)À P(1)À C(6)À C(7) 106.00(17). Absolute structure parameter=À 0.013(6).
stereospecific substitution reactions proceed with retention of
configuration via an intermediate α-ferrocenyl carbenium ion.[13]
The configuration of the products (R,Sp,Sphos)-2 and
(R,Sp,Rphos)-3 are consistent with the known lithiation diastereo-
selectivity of amine (R)-1,[11] and subsequent addition of R’PCl2
followed by R’’MgX for the creation of the phosphorus-based
Scheme 3. Planar chiral stereocontrol in the synthesis of P-stereogenic
diastereoisomer (R,Sp,Sphos)-2.
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2720
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
Although successful, an issue with the procedure outlined
bromide and then borane-dimethylsulfide. This gave doubly
protected (R,Sp)-8, as revealed by the two (broad) 11B NMR
signals at À 12.2 and À 37.2 ppm (Scheme 4). Using the same
protocol, but with o-tolylmagnesium bromide, gave singly
protected (R,Sp)-9 for which, in addition to there being a single
11B NMR signal (À 10.4 ppm), the 31P NMR signal (18.2 ppm)
differs significantly from that observed for (R,Sp)-8 (9.9 ppm).
The absence of phosphorus protection was confirmed by X-ray
crystallographic analysis (Figure 3). This use of phosphorus
trichloride in three substitution reactions with organolithium/
Grignard reagents provides a simple alternative method to the
use of preformed Ar2PCl or ArPCl2 reagents (vide infra).
in Scheme 2 is the air sensitivity of the ferrocenylphosphine
species, such that oxidation to the corresponding phosphine
oxides occurs readily. We, therefore, investigated the use of a
borane protecting group to prevent this during ligand syn-
thesis, and in addition, aimed to avoid isolation of the
deprotected ligand by complexing in situ with copper(I)
chloride. The viability of this approach was investigated first by
adding phosphorus trichloride to lithiated Ugi’s amine, followed
by greater than two equivalents each of phenylmagnesium
Isolated copper complexes of Josiphos ligands are known,[15]
generated from the addition of the ligand to a copper(I) halide
salt dissolved in a polar solvent. On a mmol scale, and as an
alternative to weighing out small quantities of air-sensitive
copper(I) chloride, it was convenient to use commercially
available copper(I) chloride – bis(lithium chloride) complex in
THF, as illustrated by the complexation of the parent Josiphos
ligand (R,Sp)-10 to give air-stable and readily chromatographed
complex (R,Sp)-11 (Scheme 5). In the solid-state and as a
solution in dichloromethane this complex is dimeric, but it is
monomeric in strongly
a coordinating solvent such as
acetonitrile.[15b]
Putting together these components of P-stereogenic
phosphine generation, protection, and complexation, the direct
synthesis of new copper-Josiphos complexes was attempted
(Scheme 6). Addition of BH3.SMe2 after the initial sequence of
lithiation, phosphorus introduction, and Grignard substitution,
resulted in the isolation of (R,Sp,Sphos)-12 and (R,Sp,Rphos)-13 as
doubly protected P,N-ligands. Both were isolated readily as
single diastereoisomers, and the identity and configuration of
both were confirmed by X-ray crystallographic analysis (Figure 4
and Figure 5).[12] Subsequent treatment of (R,Sp,Sphos)-12 with
dicyclohexylphosphine in 5% acetic anhydride/acetic acid at
Scheme 4. Investigation into the use of borane protection in ligand syn-
thesis.
°
80 C followed, after workup, by copper(I) chloride – bis(lithium
chloride) addition, resulted in an air-stable complex isolated by
column chromatography as a single diastereoisomer. Signifi-
cantly, the same product was obtained from (R,Sp,Rphos)-13,
revealing epimerisation of the phosphorus-based stereogenic
centre for one of the diastereoisomers under the conditions
required for deprotection and nitrogen substitution. Deprotec-
tion of borane-protected P-stereogenic phosphines is known to
proceed with retention of configuration.[16] The identity of the
copper-complex formed from both sequences as (R,Sp,Rphos)-14
was revealed by the separate formation of this from complex-
ation of (R,Sp,Rphos)-5 with copper(I) chloride (Scheme 7). Com-
Figure 3. Representation of the X-ray crystal structure of (R,Sp)-9 (hydrogen
atoms omitted for clarity). Thermal ellipsoids are drawn at the 50%
probability level. Principal bond lengths [Å] include: C(2)À C(3) 1.510(4),
C(1)À P(1) 1.820(3), P(1)À C(4) 1.841(4), P(1)À C(6) 1.845(3). Principal torsion
°
angles [ ] include: C(1)À C(2)À C(3)À N(1) À 100.1(3), C(2)À C(1)À P(1)À C(4)
113.7(3), C(2)À C(1)À P(1)À C(6) À 141.2(3), C(1)À P(1)À C(4)À C(5) À 176.3(3),
C(1)À P(1)À C(6)À C(7) 110.1(3). Absolute structure parameter=À 0.014(6).
Scheme 5. Synthesis of copper chloride complex (R,Sp)-11.
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2721
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
Figure 5. Representation of the X-ray crystal structure of (R,Sp,Rphos)-13
(hydrogen atoms omitted for clarity). Thermal ellipsoids are drawn at the
50% probability level. Principal bond lengths [Å] include: C(2)À C(3) 1.520(4),
C(1)À P(1) 1.810(3), P(1)À C(4) 1.815(3), P(1)À C(6) 1.836(3). Principal torsion
°
angles [ ] include: C(1)À C(2)À C(3)À N(1) À 105.7(4), C(2)À C(1)À P(1)À C(4)
À 150.1(3), C(2)À C(1)À P(1)À C(6) 102.3(3), C(1)À P(1)À C(4)À C(5) 98.7(3), C(1)À
P(1)À C(6)À C(7) 178.7(3). Absolute structure parameter=0.020(9).
Scheme 6. Application of borane-protection and in-situ complexation for the
synthesis of P-stereogenic Josiphos-copper derivative (R,Sp,Rphos)-14.
Scheme 7. Synthesis by ligand complexation of Josiphos-copper derivatives
(R,Sp,Rphos)-14 and (R,Sp,Sphos)-15.
The successful synthesis of (R,Sp)-8 and (R,Sp)-9 from
phosphorus trichloride prompted the use of this reagent for the
synthesis of related P-stereogenic derivatives. Accordingly,
following the lithiation of (R)-1, sequential addition of
phosphorus trichloride, phenylmagnesium bromide, (o-tolyl)
magnesium bromide, and borane-dimethylsulfide resulted in
the formation of (R,Sp,Sphos)-12, but also in significant quantities
of the diphenyl and di(o-toly) derivatives (R,Sp)-8 and (R,Sp)-9, as
identified by NMR spectroscopy (Scheme 8). An additional
compound formed was identified by independent synthesis as
(R)-16, the borane adduct of the starting amine. Reversing the
sequence of Grignard reagent addition, i.e. adding (o-tolyl)
magnesium bromide and then phenylmagnesium bromide,
resulted in a 1:2.4:1.2:1.1 ratio of (R,Sp,Rphos)-13, (R,Sp)-8, (R,Sp)-9
and (R)-16. Although this method is not suitable for the clean
synthesis of the P-stereogenic derivatives, it is significant that
Figure 4. Representation of the X-ray crystal structure of (R,Sp,Sphos)-12
(hydrogen atoms omitted for clarity). Thermal ellipsoids are drawn at the
50% probability level. Principal bond lengths [Å] include: C(2)À C(3) 1.514(3),
C(1)À P(1) 1.806(2), P(1)À C(4) 1.819(2), P(1)À C(6) 1.828(2). Principal torsion
°
angles [ ] include: C(1)À C(2)À C(3)À N(1) À 101.4(3), C(2)À C(1)À P(1)À C(4)
96.9(2), C(2)À C(1)À P(1)À C(6) À 148.3(2), C(1)À P(1)À C(4)À C(5) À 169.19(17),
C(1)À P(1)À C(6)À C(7) À 75.7(2). Absolute structure parameter=À 0.006(6).
plexation of (R,Sp,Sphos)-4 gave (R,Sp,Sphos)-15, confirming that this
was not formed using the borane-based protocol.
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2722
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
Scheme 8. Investigation into the use of PCl3 for the synthesis of P-
stereogenic protected ligand-precursor (R,Sp,Sphos)-12.
only one isomer was obtained in each case. This supports the
earlier supposition that P-stereogenic stereocontrol does not
arise from the selective substitution of prochiral ArPCl2
reagents, and is instead a result of intermediate epimerisation
(Scheme 3).
Scheme 9. Synthesis of (R,Sp)-17 by use of PCl3 and from protected
intermediate (R,Sp)-9.
Epimerisation on heating P-stereogenic ferrocenylphos-
phines has been noted previously, with the diastereoisomer of
Sp*,Sphos* relative configuration being significantly less stable
than its Sp*,Rphos* epimer.[8,10] Further to the formation of only
(R,Sp,Rphos)-14 using the methodology outlined in Scheme 6,
diastereomerically pure samples of (R,Sp,Sphos)-2 and (R,Sp,Rphos)-3
°
were heated at 80 C in hexane for 24 h in the presence of
silica.[10] No change was observed for (R,Sp,Rphos)-3. In contrast,
the other sample gave a 3.8:1 ratio of (R,Sp,Rphos)-3: (R,Sp,Sphos)-2,
confirming the greater stability of the Sp*,Rphos* diastereoisomer.
This outcome is consistent with epimerisation to (R,Sp,Rphos)-5
under the conditions of its formation, prior to copper complex-
ation to give (R,Sp,Rphos)-14.
The utility of the in situ complexation methodology was
further illustrated with the synthesis of (R,Sp)-17 from (R,Sp)-9, a
reaction sequence requiring, in this instance, the removal of a
single borane protecting group (Scheme 9). Unprotected PÀ N
and PÀ P ligands, (R,Sp)-18 and (R,Sp)-19 respectively, were also
synthesised from PCl3, although phosphine oxidation was again
an issue resulting in a low yield of the latter. Subsequent
complexation with copper(I) chloride resulted in the quantita-
tive formation of (R,Sp)-17.
Scheme 10. Synthesis of bulky cobalt sandwich complex-based Rossiphos-
copper complexes (R,Sp)-21 and (R,Sp)-22.
Finally, in the context of ligand synthesis, the utility of
stereospecific phosphine introduction and in situ complexation
was further illustrated starting with bulky cobalt-sandwich
complex (R,Sp)-20 (PPCA).[17] This was transformed into new air-
stable Rossiphos[17] copper-complexes (R,Sp)-21 and (R,Sp)-22
that were readily purified by column chromatography
(Scheme 10).
For application in conjugate addition reactions, most
diphosphine/copper complexes are generated in situ from a
variety of Cu(I) and Cu(II) salts.[4–6,18] The validity of generating
and isolating the copper(I) chloride complexes described above
is contingent upon these being suitable as catalyst precursors
for such reactions. In an initial investigation of this issue two
reactions were examined; the reduction of unsaturated nitrile
23 to give 24,[18a] and the β-borylation of unsaturated ketone 25
to give, after oxidation, β-hydroxyketone 26 (Scheme 11).[18b]
Using the parent Josiphos ligand (R,Sp)-10, an e.e. value of ca.
65% has been reported for both reactions, giving scope for a
significant improvement in enantioselectivity.
Repetition of the conditions reported for the reduction of
23 using Cu(OAc)2 resulted in an improvement of the literature
e.e. value of 65% to 91%, albeit with poor conversion (Table 1,
entries 1 and 2). Use of CuCl for in situ catalyst generation
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2723
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
Conclusion
Methodology for the synthesis of P-stereogenic ferrocene
ligands was applied successfully to PPFA (PÀ N ligands) deriva-
tives containing phenyl and o-tolyl phosphorus substituents.
Subsequent transformation into the corresponding Josiphos
derivatives (PÀ P ligands) was also successful. Modification of
this methodology by use of a borane protecting group avoided
complications arising from phosphine oxidation, and this was
further assisted, following deprotection, by direct coordination
to copper, and isolation/purification of the resulting air-stable
complexes. Further adaptation of this methodology gave a new
Josiphos derivative from the use of lithiated Ugi’s amine, PCl3,
and two equivalents of aryl Grignard, although this approach
was only partially successful for the synthesis of P-stereogenic
ligands by sequential use of two different aryl Grignard
compounds. The P-stereogenic ligands/complexes are config-
urationally stable at room temperature, but epimerisation can
occur on heating (Sp*,S*phos to Sp*,R*phos). The copper chloride
complex of the parent Josiphos ligand is suitable as a catalyst
precursor, as illustrated by the comparable e.e. values to those
obtained from in situ generated complexes (up to 95% e.e.) for
representative conjugate reduction and β-borylation reactions.
For these, lower enantioselectivities were obtained with the P-
stereogenic derivatives, but marked increases in catalyst activity
were observed. Further application of these copper-chloride
complexes in asymmetric catalysis is under investigation.
Scheme 11. Representative copper-catalysed asymmetric conjugate-addition
reactions.
Table 1. Copper-catalysed asymmetric conjugate-addition reactions of 23
and 25.[a]
Entry
Substrate
Cu source[b]
Time
[h]
Conversion
[%]
e.e.
[%]
1[c]
2
3
4
5
23
23
23
23
23
23
25
25
25
25
25
Cu(OAc)2
Cu(OAc)2
CuCl
8.5
24
120
24
72
1
24
18
96
1.5
144
96
65
25[d]
100[d]
97[d]
19[d]
100[d]
100
91[e]
95[e]
95[e]
27[e]
57[e]
68
(R,Sp)-11
(R,Sp,Rphos)-14
(R,Sp,Sphos)-15[f]
CuCl
CuCl
(R,Sp)-11
6
7[g]
8
87[d]
94[d]
98[d]
80[d]
70[h]
71[h]
�0[h]
�0[h]
9
10
11
(R,Sp,Rphos)-14
(R,Sp,Sphos)-15[f]
[a] Using conditions outlined in Scheme 11. [b] With (R,Sp)-10 [R,Sp)-
Josiphos] for in situ generated catalysts unless otherwise stated. [c] Lit.
value.[18a] [d] Determined by 1H NMR spectroscopy. [e] Of (R)-24 as
determined by HPLC.[f] Generated in situ from CuCl and (R,Sp,Sphos)-4. [g]
Lit. value.[18b] [h] Of (S)-26 as determined by HPLC.
Experimental Section
Preparation of (R,Sp,Rphos)-13. (R)-1 (0.200 g, 0.78 mmol) added to a
flame dried Schlenk tube and dissolved in diethyl ether (6 mL). The
°
solution was cooled to 0 C and sec-butyllithium (1.4 M in Hexanes)
(0.61 ml, 0.86 mmol) added slowly. The reaction was allowed to
warm to room temperature and stirred for 2 h. Upon cooling to
further improved the e.e. to 95% (entry 3), and this was
maintained, together with a significantly shorter reaction time,
on employing preformed (R,Sp)-11 (entry 4). Use of the P-
stereogenic Josiphos ligands as either a preformed complex
(entry 5), or generated in situ (entry 6), resulted in a large
reduction in e.e. For β-borylation of 25 followed by oxidation,
repetition of the conditions reported with CuCl resulted in a
similar selectivity (ca. 70% e.e, entries 7 and 8). This was also
the case with (R,Sp)-11, although in this instance a longer
reaction time was required to achieve high conversion (entry 9).
Essentially racemic product was obtained on application of the
P-stereogenic Josiphos ligands (entries 10 and 11). The reaction
time required for β-borylation of 25 was significantly reduced
with the catalyst derived from (R,Sp,Rphos)-5. In contrast, the
reaction time for reduction of 23 was significantly reduced with
the catalyst derived from (R,Sp,Sphos)-4. Thus for these two
reactions, the difference between the ligand diastereoisomers is
more apparent as catalyst activity rather than catalyst enantio-
selectivity. This may be the result of a different, perhaps
monodentate, coordination mode in the catalyst for one or
both P-stereogenic ligand diastereoisomers containing the
additional ortho-methyl substituent.[19]
°
À 78 C, dichloro ortho-tolylphosphine (0.165 g, 0.86 mmol) was
added and the reaction allowed to stir at room temperature for 1 h.
°
The reaction was re-cooled to À 78 C and phenylmagnesium
bromide (1.0 M in THF) (0.86 mL, 0.86 mmol) was added and the
reaction allowed to warm to room temperature and stirred for an
°
additional hour. The reaction was cooled to 0 C and borane
dimethyl sulfide complex (2.0 M in THF) (0.86 mL, 1.71 mmol) was
added and the reaction allowed to stir overnight at room temper-
ature. The reaction was quenched with saturated sodium carbonate
solution, extracted with dichloromethane, dried (MgSO4) and the
solvent removed in vacuo. Purification by column chromatography
(SiO2, 50/50 CH2Cl2/hexane) yielded an orange solid (0.18 g, 47%): Rf
°
0.30 (50/50 CH2Cl2/hexane); Mp 209–211 C; [α]D C =À 184 (c=
26
°
0.24, chloroform); 1H NMR (500 MHz, CDCl3) δ 7.83–7.76 (2H, m,
PhH), 7.56–7.51 (1H, m, PhH), 7.51–7.46 (2H, m, PhH), 7.34–7.27 (2H,
m, ArH), 7.16–7.12 (1H, m, ArH), 7.10 (2H, t, J=7.6 Hz, ArH), 5.12 (1H,
dd, J=3.3, 2.5 Hz, CpH), 4.82 (1H, q, J=6.6 Hz, CH), 4.63 (1H, t, J=
2.6 Hz, CpH), 4.24–4.21 (1H, m, CpH), 4.01 (5H, s, CpH), 2.38 (3H, s,
NCH3), 2.22 (3H, s, ArCH3), 2.10 (3H, s, NCH3), 1.94 (3H, d, J=6.8 Hz,
CHCH3); 13C NMR (125 MHz, CDCl3) δ 142.8 (d, J=12.8 Hz, ArC),
134.0 (d, J=5.7 Hz, ArC), 133.5 (d, J=9.4 Hz, PhC), 132.2 (d, J=
9.2 Hz, ArC), 131.4 (d, J=2.0 Hz, ArC), 131.3 (d, J=2.3 Hz, PhC),
130.7 (d, J=59.9 Hz, PhC), 129.1 (d, J=52.9 Hz, ArC), 128.5 (d, J=
10.1 Hz, PhC), 125.0 (d, J=8.4 Hz, ArC), 93.6 (d, J=16.6 Hz, CpC),
73.6 (d, J=8.0 Hz, CpC), 73.2 (d, J=2.5 Hz, CpC), 71.4 (C5H5), 71.3 (d,
’
J=5.9 Hz, CpC), 69.5 (d, J=61.9 Hz, CpC), 62.8 (CHCH3), 53.5 (NCH3 ),
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2724
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100146
50.3 (NCH3), 22.6 (d, J=4.4 Hz, ArCH3), 20.8 (CHCH3); 31P NMR
(202 MHz, CDCl3) δ 14.53 (PArPh); 11B NMR (160 MHz, CDCl3) δ
[1] T. P. Yoon, E. N. Jacobsen, Science 2003, 229, 1691–1693.
[2] a) T. Hayashi, T. Mise, M. Fukushima, M. Kagotani, N. Nagashima, Y.
Hamada, A. Matsumoto, S. Kawakami, M. Konishi, K. Yamamoto, M.
Kumada, Bull. Chem. Soc. Jpn. 1980, 53, 1138–1151; b) A. Togni, C.
Breutel, A. Schnyder, F. Spindler, H. Landert, A. Tijani, J. Am. Chem. Soc.
1994, 116, 4062–4066.
[3] Examples with metals other than copper include: a) B. Song, C.-B. Yu, Y.
Ji, M.-W. Chen, Y.-G. Zhou, Chem. Commun. 2017, 53, 1704–1707; b) W.
Baratta, G. Chelucci, S. Magnolia, K. Siega, P. Rigo, Chem. Eur. J. 2009, 15,
726–732; c) A. Yen, K.-L. Choo, S. K. Yazdi, P. T. Franke, R. Webster, I.
Franzoni, C. C. J. Loh, A. I. Poblador-Bahamonde, M. Lautens, Angew.
Chem. Int. Ed. 2017, 56, 6307–6311; Angew. Chem. 2017, 129, 6404–
6408; d) T. Nagano, A. Iimuro, R. Schwenk, T. Ohshima, Y. Kita, A. Togni,
K. Mashima, Chem. Eur. J. 2012, 18, 11578–11592; e) T. Sawano, K. Ou, T.
Nishimura, T. Hayashi, J. Org. Chem. 2013, 78, 8986–8993; f) T. Hibino, K.
Makino, T. Sugiyama, Y. Hamada, ChemCatChem 2009, 1, 237–240.
[4] B. L. Feringa, R. Badorrey, D. Peña, S. Harutyunyan, A. J. Minnaard, Proc.
Natl. Acad. Sci. USA 2004, 101, 5834–5838.
~
À 12.26 (BH3À NMe2), À 35.09 (BH3À PArPh). IR (film) v=3058, 3007,
2956, 2345 (BÀ H), 2278, 1471, 738; MS (EI) m/z calcd for
C27H36B2FeNP [M]+ 483.2; found 483.2. [HRMS of the borane
complex was not obtained due to the loss of BH3, HRMS (ASAP) [M-
2BH3 +H]+ C37H46FeP2 +H+, Calc. 609.2503, Obs. 609.2503].
Preparation of (R,Sp,Rphos)-14from (R,Sp,Rphos)-13. A flame dried
Schlenk tube (fitted with a glass stopper) containing glacial acetic
acid (3 mL) and acetic anhydride (0.15 mL) was degassed by the
freeze-pump-thaw method (x3) and the atmosphere replaced with
argon. Then (R,Sp,Rphos)-13 (0.040 g, 0.08 mmol) was added and the
sample degassed once more. After freezing for a 5th and final time
the stopper was replaced by a septum and the atmosphere flushed
with argon. Dicyclohexylphosphine (0.08 mL, 0.41 mmol) was then
added via syringe. The septum was again replaced by a glass
stopper and the vessel evacuated once more and allowed to thaw
under an atmosphere of argon. After stirring at room temperature
for 5 mins, the Schlenk tube was introduced to an oil bath pre-
[5] B. H. Lipshutz, J. M. Servesko, Angew. Chem. Int. Ed. 2003, 42, 4789–
4792; Angew. Chem. 2003, 115, 4937–4940.
[6] J.-E. Lee, J. Yun, Angew. Chem. Int. Ed. 2007, 47, 145–147.
[7] This statement is based on a study of literature examples where formal
addition of an element of chirality to an existing chiral non-racemic
ferrocene-based ligand gave rise to both resulting diastereoisomers. On
subsequent screening in a metal-based asymmetric transformation, the
e.e. values for the ‘offspring’ diastereoisomers were higher and lower
than that obtained for the ‘parent’ ligand in 20 out of 23 cases. See: C. J.
Richards, R. A. Arthurs, Chem. Eur. J. 2017, 23, 11460–11478.
[8] W. Chen, W. Mbafor, S. M. Roberts, J. Whittall, J. Am. Chem. Soc. 2006,
128, 3922–3923.
°
heated to 80 C and stirred for 3 hours. The reaction was allowed to
°
cool to room temperature and then cooled to 0 C before adding
copper(I) chloride bis(lithium chloride) complex (1 M in THF)
(0.62 mL, 0.62 mmol) and stirred at room temperature for 1 hour.
°
The reaction was re-cooled to 0 C, quenched with saturated
sodium hydrogen carbonate solution (30 mL), and the resulting
yellow precipitate collected by filtration. The solids were washed
with water followed by hexane and then dissolved in dichloro-
methane. This was washed with water, dried (MgSO4) and the
solvent removed in vacuo. Purification by column chromatography
[9] P-Stereogenic Josiphos derivatives have also been synthesised by the
separation of diastereoisomers: R. Schwenk, A. Togni, Dalton Trans.
2015, 44, 19566–19575.
[10] R. A. Arthurs, D. L. Hughes, C. J. Richards, J. Org. Chem. 2020, 85, 4838–
(SiO2, 5% Et2O/CH2Cl2) yielded an orange solid° (0.046 g, 79%): Rf
4847.
26
0.11 (10% EtOAc/hexane); Mp 148–149 C; [α]D C =À 186 (c=0.64,
°
[11] D. Marquarding, H. Klusacek, G. Gokel, P. Hoffmann, I. Ugi, J. Am. Chem.
Soc. 1970, 92, 5389–5393.
1
CHCl3); H NMR (500 MHz, CHCl3) δ 7.93–7.86 (4H, m, PhH), 7.47–
7.41 (6H, m, PhH), 7.23–7.17 (2H, m, ArH), 7.11–7.06 (6H, m, ArH),
4.52 (2H, brs, CpH), 4.45 (2H, brs, CpH), 4.13 (2H, brs, CpH), 3.90
(10H, s, CpH), 3.50–3.42 (2H, m, CH), 2.25 (6H, s, ArCH3), 2.04–0.64
(44H, m, PCy2), 1.45 (6H, t, J=7.7 Hz, CHCH3); 13C NMR (125 MHz,
CHCl3) δ 142.0 (d, J=18.1 Hz, ArC), 134.6 (d, J=17.1 Hz, PhC), 134.2
(dd, J=25.5, 11.7 Hz, PhC), 133.6 (d, J=26.4 Hz, ArC), 132.0 (d, J=
4.1 Hz, ArC), 131.2 (d, J=6.6 Hz, ArC), 130.3 (PhC), 129.3 (ArC), 128.7
(d, J=10.3 Hz, PhC), 125.3 (d, J=4.8 Hz, ArC), 95.4 (d, J=21.2 Hz,
CpC), 73.8 (CpC), 72.5–72.3 (m, CpC), 71.2 (d, J=32.8 Hz, CpC), 70.5
(C5H5), 70.1 (d, J=3.2 Hz, CpC), 34.2–25.6 (m, CyC), 30.7 (d, J=
10.4 Hz, CHCH3), 22.5 (d, J=15.5 Hz, ArCH3), 17.7 (CHCH3); 31P NMR
(202 MHz, CHCl3) δ 6.28 (d, J=203.5 Hz, PCy2), À 27.43 (d, J=
[12] Deposition Numbers 1956990 (2), 1956991 (3), 2033033 (9), 2033035
(12), and 2033034 (13) contain the supplementary crystallographic
data for this paper. These data are provided free of charge by the joint
Cambridge Crystallographic Data Centre and Fachinformationszentrum
Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures.
[13] G. Gokel, P. Hoffmann, H. Klusacek, D. Marquarding, E. Ruch, I. Ugi,
Angew. Chem. Int. Ed. Engl. 1970, 9, 64–65; Angew. Chem. 1970, 82, 77–
78.
[14] Borane protection has been use with secondary phosphines derived
from Ugi’s amine: S. Pandey, M. B. Sárosi, P. Lönnecke, E. Hey-Hawkins,
Eur. J. Inorg. Chem. 2017, 256–262.
[15] a) B. H. Lipshutz, B. Frieman, H. Birkedal, Org. Lett. 2004, 6, 2305–2308;
̇
b) S. R. Harutyunyan, F. López, W. R. Browne, A. Correa, D. Pena, R.
Badorrey, A. Meetsma, A. J. Minnaard, B. L. Feringa, J. Am. Chem. Soc.
2006, 128, 9103–9118; c) F. Caprioli, M. Lutz, A. Meetsma, A. J. Minnaard,
S. R. Harutyunyan, Synlett 2013, 24, 2419–2422.
~
203.6 Hz, PArPh). IR (film) v=3055, 2926, 2849, 2222, 1449, 918,
731; HRMS (EI) m/z calcd for C37H46ClCuFeP2 704.1450 [M]+; found
704.1452.
[16] T. Imamoto, T. Oshiki, T. Onozawa, T. Kusumoto, K. Sato, J. Am. Chem.
Soc. 1990, 112, 5244–5252.
[17] R. A. Arthurs, P. N. Horton, S. J. Coles, C. J. Richards, Chem. Eur. J. 2018,
24, 4310–4319.
[18] a) H.-S. Sim, X. Feng, J. Yun, Chem. Eur. J. 2009, 15, 1939–1943; b) D. Lee,
D. Kim, J. Yun, Angew. Chem. Int. Ed. 2006, 45, 2785–2787; Angew. Chem.
2006, 118, 2851–2853.
Acknowledgements
°
[19] Compared to Ph, o-Tol adds ca. 15 to the cone angle of a ligand
The EPSRC (EP/N019393/1) (R.A.A.) is thanked for financial support.
We also thank the EPSRC national Mass Spectrometry Centre
(University of Wales, Swansea).
containing these groups: T. E. Pickett, F. X. Roca, C. J. Richards, J. Org.
Chem. 2003, 68, 2592–2599.
Conflict of Interest
The authors declare no conflict of interest.
Manuscript received: February 9, 2021
Revised manuscript received: April 12, 2021
Accepted manuscript online: April 14, 2021
Keywords: Asymmetric Catalysis · Copper · Ligand design ·
Metallocene · Protecting groups
Eur. J. Org. Chem. 2021, 2719–2725
www.eurjoc.org
2725
© 2021 The Authors. European Journal of Organic Chemistry published
by Wiley-VCH GmbH