Olefin hydrogenation with rigid mono-P-stereogenic diphosphines: A flexible rhodium ring to rule them all?

Article abstract of DOI:10.1002/ejoc.201403570

The preparation of a rigid C₁-symmetric P-stereogenic diphosphine series possessing a P-o-diphenylphosphinophenyl moiety and its use in Rh^I-catalyzed hydrogenation of benchmark olefins, is presented. This simplified ligand design, wh

Full text of DOI:10.1002/ejoc.201403570

FULL PAPER
DOI: 10.1002/ejoc.201403570
Olefin Hydrogenation with Rigid Mono-P-stereogenic Diphosphines: A Flexible
Rhodium Ring to Rule Them All?
Slavko Rast,[a] Michel Stephan,*[a][‡] and Barbara Mohar*[a]
Keywords: Asymmetric synthesis / Hydrogenation / Ligand effects / Phosphane ligands / Rhodium
The preparation of a rigid C1-symmetric P-stereogenic di-
phosphine series possessing a P-o-diphenylphosphinophenyl
moiety and its use in RhI-catalyzed hydrogenation of bench-
mark olefins, is presented. This simplified ligand design,
wherein some specific features (backbone flexibility, iden-
tical nature of P,PЈ-substituents) of related C2-symmetric eth-
ylene-bridged diphosphines were knocked out, is a useful
study-tool with which to better understand the role of essen-
tial motifs that guide catalysis in the original model.
addition to the basic ethylene bridge spanning the two
phosphorus atoms, 1,1Ј-biarene-2,2Ј-diyl, o-phenylene, or
Introduction
cis-Chelating P-stereogenic diphosphine ligands of tran- ferrocene-1,1Ј-diyl scaffolds can be identified in several suc-
sition-metal catalysts have continued to be a research topic cessful ligand designs (Figure 1). Forming another category
of interest since Knowles’ pioneering work in RhI-catalyzed are nonsymmetric (C1-symmetric) chiral diphosphines,
asymmetric hydrogenation.[1,2] Although early advances in which are embodied by the methylene-, ethylene-, or o-
the field evolved toward C2-symmetric diphosphines with phenylene-bridged derivatives (Figure 2).[4] Among them,
backbone chirality, several P-stereogenic diphosphines or Hoge’s all-aliphatic trichickenfootphos (TCFP) and Imam-
diphosphines possessing chiral P-substituents proved later oto’s 3H-BenzP* and 3H-QuinoxP* displayed an excellent
to be superior in many cases under mild conditions.[3] In performance in RhI-catalyzed hydrogenation in particular.
Figure 1. Selection of C2-symmetric chiral diphosphines.
In some cases, C1-symmetric designs furnished even better
[a] National Institute of Chemistry,
Hajdrihova 19, 1000 Ljubljana, Slovenia
E-mail: barbara.mohar@ki.si
results than their C2-symmetric bis(chiral phosphorus unit)
counterparts.[5] In theory, efficient C1-symmetric diphos-
phines, if easily accessed, would offer an economical syn-
thetic advantage over the latter.
In our ongoing research program on metal-(chiral phos-
phine) complexes,[3a–3c,6] we present hereafter the synthesis
http://www.ki.si
[‡] Present address: PhosPhoenix SARL
115, rue de l’Abbé Groult, 75015 Paris, France
E-mail: mstephan@phosphoenix.com
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201403570.
2214
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
Figure 2. Selection of C1-symmetric PCP- or PCCP-type chiral diphosphines.
of an o-diphenylphosphinophenyl-based C1-symmetric di-
phosphine series L, wherein the chirality is solely borne by
a single phosphorus atom. Such combination has not been
developed or investigated in catalysis.[7,8] The diphosphine
ligands were screened in the RhI-catalyzed asymmetric
hydrogenation of benchmark activated olefins.
Results and Discussion
Synthesis of the Diphosphine Series L
The mono-P-stereogenic diphosphines L1–L4 were con-
veniently accessed in enantiopure form and with good over-
all yields (62, 39, 41, and 39% for L1, L2, L3, and L4,
respectively) by following an adapted Jugé–Stephan asym-
metric route to phosphines[9] (Scheme 1). This general strat-
egy relies upon the sequential displacement with organo-
lithium reagents of either (+)- or (–)-ephedrine auxiliary
from the derived enantiopure 1,3,2-oxazaphospholidine-2-
borane complex (oxazaPB), giving rise to either target
enantiomer. In the present study, such an approach allowed
advantageously simple tuning of the diphosphine structure
at a late stage.
Thus, the diastereomerically pure phosphino-amino-
phosphine-P-borane 1 was obtained in 92% isolated yield
by ring-opening of (2S,4R,5S)-(–)-oxazaPB [derived from
(1S,2R)-(+)-ephedrine] with o-diphenylphosphino-phen-
yllithium. X-ray crystal-structure analysis of 1 confirmed
its RP absolute configuration. Notably, no by-products were
formed, as is usually observed in this step during the intro-
duction of bulky ortho-substituted phenyllithium deriva-
tives.[10] The reactive phosphino-phosphinite-P-borane 2 in-
termediate was subsequently obtained as a viscous oil in
98%ee (by chiral HPLC) with 96% yield by H2SO4-cata-
lyzed methanolysis of 1. Its enantiopurity was enhanced to
Ͼ99.9%ee through recrystallization of its borane bis ad- derived enantiopure oxazaPB.
Scheme 1. Synthesis of the ligand series L1–L4 from ephedrine-
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2215

S. Rast, M. Stephan, B. Mohar
FULL PAPER
duct 2-PЈ(BH3). Most importantly, the use of CH2Cl2 in-
stead of tetrahydrofuran (THF) ensured preservation of the
otherwise labile PЈ-borane on Ph2P during workup and iso-
lation of 2-PЈ(BH3).
The ent-2-PЈ-oxide [derived from (1R,2S)-(–)-ephedrine]
was prepared after failed attempts to determine the absolute
configuration of 2 (Scheme 2).[11] Fortunately, its X-ray
crystal-structure analysis was successful and revealed an un-
expected (SP)-configuration; therefore, an (RP)-configura-
tion was assigned to 2 in Scheme 1. This retention of P-
configuration ensuing the P–N cleavage is in contrast to the
inversion of stereochemistry encountered in this step of the
Jugé–Stephan general route.[3a–3c,6,9] Given the high ee of
(RP)-2, this retention can be rationalized by a preferential
MeOH attack from any less-encumbered face of the P-
centered tetrahedron rather than from the hindered face by
an adjacent phenylene-H, and leading to a transient trigo-
nal bipyramidal (pentacoordinate) chiral phosphorus
(Scheme 3). Such conformation is probably imposed by the
six-membered ring formed by H+-bridging of the N and P
atoms of the ephedrino and Ph2P moieties, respectively. To
our knowledge, this is the first example of an SN-type attack
on a phosphorus atom with retention of P-configuration
encountered in the aminophosphine-P-borane series.
Scheme 3. Proposed mechanism of H2SO4-catalyzed methanolysis
of (RP)-1 leading to 2 with retention of P-configuration.[14] (N-
ephedrino = MeN-Eph).
ium (52–55%). Finally, the corresponding BH3-free diphos-
phines L1–L4 were readily obtained with 94–98% yield by
decomplexation in Et2NH (55–60 °C).[12]
Moreover, Scheme 2 depicts the preparation of (SP)-L3-
P(BH3),PЈ(O), which demonstrates the possibility of gain-
ing access to such a dissymmetrically-protected and con-
gested structure by following this strategy.[13]
The optical rotation sign of the reported diphosphine
L1[7] indicated an (SP)-configuration and the X-ray crystal-
structure analysis of L2 revealed an (RP)-configuration
(Figure 3). Consequently, an identical stereochemical out-
come (inversion then retention from 2) resulted, as expected
in both cases taking into account the priorities of the
groups according to CIP stereochemistry rules. By exten-
sion, an (RP)-configuration was assigned to L3 and L4.
Figure 3. ORTEP of diphosphine L2 drawn at the 50% probability
level.
Assessment of the Diphosphine Series L in RhI-Catalyzed
Hydrogenation
The mono-P-stereogenic diphosphine ligands L1–L4
were screened in the RhI-catalyzed hydrogenation of methyl
α-acetamidoacrylate (MAA), methyl (Z)-α-acetamidocinn-
amate (MAC), dimethyl itaconate (DMI), α-acetamidostyr-
ene (AS), and atropic acid (AA) (Table 1).
Scheme 2. Preparation of (SP)-2-PЈ(O) and X-ray determination of
its absolute configuration.
Unfortunately, the overall outcome of RhI-L catalysis in
terms of enantioselectivity and activity by current standards
Subsequently, the mono-P-borane-protected diphos- does not have a high synthetic value for the tested model
phines L1·BH3–L4·BH3 were prepared through displace- substrates. Most notably, the highest ee of 87% was ob-
ment of the P-OMe group of the enantioenriched (RP)-2 tained for MAC and AS by using the fully arylic bulky li-
(Ͼ99.9%ee) with methyllithium (86%) or o-RO-phenyllith- gand L4 (P-o-tBuOPh-substituted) under 1 bar H2, and this
2216
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
Table 1. [RhI-L]-Catalyzed hydrogenation of benchmark activated
The following comparative observations can be made:
(1) The RhI complex of the P-Me-substituted ligand L1 dis-
played a much higher activity than the P-o-ROPh-contain-
ing L2–L4 counterparts but afforded lower ee values.
(2) The enantioselectivity consistently increased for all sub-
strates with systematic increase in the RO bulk going from
MeO (L2) to iPrO (L3) to tBuO (L4). (3) Testing the RhI-
L1 catalyst on MAC showed a noticeable H2 pressure de-
pendence (but not on AS) as enantioselectivity decreased
with increase in pressure (shifting from 1 to 10 bar). Such
variation was not substantially noticeable with L2–L4 for
MAC or AS. (4) The enantioselectivities (under 1 bar H2)
for MAC and AS with L1 or L2 were similar per substrate
(69/67%ee for MAC and 19/25% for AS). (5) The enantio-
selectivities (under 1 bar H2) for MAC and AS with L3 or
L4 were similar per ligand (78%ee with L3 and 87% with
L4).
olefins.[a]
Notably, whereas the ee values and catalyst activity with
L2–L4 were inferior to those obtained by using the corre-
sponding C2-symmetric chiral ethylene-bridged diphos-
phines,[15] the ee value reached using L1 for MAC (69%) is
noticeably better than all the ee values obtained with the
corresponding C2-symmetric 1,2-bis[(methyl)(phenyl)phos-
phino]benzene,[16] or 1,2-bis[(aryl)(methyl)phosphino]eth-
anes (aryl = Ph,[17] Fc[18]), 1,2-bis[(tert-butyl)(phenyl)phos-
phino]ethane,[19] and the C1-symmetric “unsymmetrical (1-
Ad)-BisP*” ligands[4f,4g] (Figure 4).
To discuss enantioselectivity and hydrogenation rate for
MAC and AS versus backbone rigidity, and due to the vari-
ety of PCCP-type diphosphines considering the backbone
and P-substituents features (as in Figure 4), it seems neces-
sary to differentiate between the diphosphines bearing a
phospholano group, a (methyl)(phenyl)phosphino group,
an (RO-substituted Ph)(phenyl)phosphino group, or an
(RO-unsubstituted aryl)(phenyl)phosphino group (RO-un-
substituted aryl = ex. α-Nap).
In the case of phospholano-containing diphosphines and
as reported by the Pringle group for AS, RhI-catalyzed
hydrogenation by using Me-UCAP-Ph was completed much
more rapidly than with Me-DuPHOS, and 1,2-bis(diphenyl-
phosphino)benzene was in an intermediate position.[4n,22]
[a] The catalysts were preformed in situ from [Rh(nbd)2]BF4 and
the ligand L. Runs were carried out with 0.5 mmol substrate in
MeOH (7.5 mL) by using a substrate/catalyst molar ratio (S/C) of
100 at 25 °C for the time indicated (progress at 1 bar was monitored
for up to 3 h but reaction was continued for 16 h when conversion
was not complete). Conversion and ee were determined by chiral
GC analysis. [b] Et3N (1.1 equiv.) added.
was improved to 90%ee for AS operating at 10 bar. The This implies that the incorporation of a rigid aliphatic
MAC and AS hydrogenation data with L1–L4 and PCCP- phospholano group slows down hydrogenation compared
type diphosphines relevant to this study are listed in Fig- with incorporation of a flexible arylic diphenylphosphino
ure 4 (for additional listings, see the Supporting Infor- group, but the electronic dissymmetry of Me-UCAP-Ph
mation, Figure S1); because reported data related to AS could be responsible for its observed rate enhancement.
hydrogenation and to the approximate hydrogenation rate However, the impact on hydrogenation kinetics of Me-
for MAC is not complete, and because the adopted hyd- UCAP-Ph versus its ethylene-bridged analogue[4o] or of R-
rogenation conditions vary, it is difficult to comment on this DuPHOS versus its ethylene-bridged analogue R-BPE was,
aspect. Notably, L4 is enantioselectively competitive (within to our knowledge, never clearly exposed. Nevertheless in
7%) to DiPAMP for AS but the opposite is true for these cases, the enantioselectivity (for MAC and AS) was
MAC.[15] Furthermore, L4 and the phospholane-containing higher with ligands having a rigid o-phenylene back-
(stiffer) analogue Me-UCAP-Ph[4n,4o] results are compar- bone.[3n,4n,4o]
able. Concerning the sense of induction of L1–L4, the same
In the case of (methyl)(phenyl)phosphino-containing
trend is observed as when applying the corresponding C2- diphosphines, the results demonstrate that MAC and AS
symmetric chiral ethylene- or o-phenylene-bridged diphos- hydrogenation rates are lower with the rigid o-phenylene
phines but, curiously, not for AS with L1 [(R)-N-Ac-(1- backbone than with the flexible ethylene backbone {L1
phenylethyl)amine was obtained, albeit in very low 19%ee]. compared with 1,2-bis[(methyl)(phenyl)phosphino]eth-
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2217

S. Rast, M. Stephan, B. Mohar
FULL PAPER
Figure 4. Selection of P-stereogenic or phospholano-containing PCCP-type diphosphines with a summary of their RhI-catalyzed hyd-
rogenation results against MAC and AS (for additional results, see the Supporting Information, Figure S1).
ane[17a]} but the enantioselectivity for MAC is better with Background
the rigid o-phenylene backbone.
Most importantly, the hydrogenation rates for MAC and
Since the early advances in asymmetric hydrogenation,
AS were much higher with 1,2-bis[(aryl)(phenyl)phos- much effort has been directed towards elucidating the
phino]ethanes [wherein aryl = o-RO-Ph,[6a,15] 2,3-(MeO)2- mode of action and the origin of enantioselection of the
Ph, 2,3,4,5-(MeO)4-Ph[20]] than with L1–L4, demonstrating chiral RhI-(cis-chelating C2-symmetric diphosphine) cata-
a clear beneficial impact on the hydrogenation kinetics of lysts.[3e,23] In particular, the complex mechanistic aspects of
the flexible CH2CH2 backbone in such P,P,PЈ,PЈ-tetraaryl asymmetric hydrogenation of prochiral α-amido-olefins [es-
PCCP-type bridged structure. However, the P-α-Nap-sub- pecially (Z)-α-amidocinnamates and α-acetamidostyrene]
stituted analogue[21] displayed much lower hydrogenation catalyzed by RhI-(P-stereogenic PCCP-type diphosphine)
rates for MAC and AS compared with DiPAMP[15] or complexes are by now well-substantiated and diverse. It
1,2-bis[(methyl)(phenyl)phosphino]ethane[17a] (having com- seems that either a different or a combination of mecha-
parable rates), and 1,2-bis{[2,3-(MeO)2-Ph](phenyl)phos- nisms (dihydride vs. unsaturated) can occur depending on
phino}ethane,[20] but gave the same level of enantio- the conditions, catalytic system, and the steric and elec-
selectivities as with the latter, pointing here also to the unfa- tronic properties of the substrate.[23,24] The basicity and
vorable increase in the overall rigidity of the system.
CH2CH2 bridge flexibility of diphosphines may play a
It thus seems that the decreased steric demands of L1 major role in the rate-determining step by favoring an in-
compared with L2–L4 are responsible for its higher hyd- crease in the RhI complex affinity to H2, facilitating the
rogenation rates in the series. This is also seen from the low oxidative H2 addition, positioning of the two hydrides, and
hydrogenation rate of the unhindered MAA substrate.
influencing the H-transfer.
Finally, comparing MAC results using L2–L4 (67%ee
Some studies point out that the five-membered ring Rh-
with L2, 78%ee with L3, 87%ee with L4) against (o- (P,P,PЈ,PЈ-tetraaryl CH2CH2-backboned diphosphine)
An)(Ph)PCH2CH2PPh2 (80–85%ee),[4j] suggests that the chelates possess just the appropriate conformational mo-
negative impact on enantioselectivity due, in this case, to bility.[23c] Accommodating incoming molecules (olefin, H2,
the severely restricted flexibility of the o-phenylene back- coordinating solvent), the diphosphine is able to adapt to a
bone, can be offset by an increase in the P-o-ROPh bulk. certain extent by P–C bond free-rotation of the P-substitu-
Based on these results, it seems that the highest ee values ents, flexing the chelate ring with positional switching of
and hydrogenation rates for MAC and AS are obtained by these groups. Thus, substrate control as well as Rh-complex
using RhI complexes of flexible ethylene-bridge backboned control can occur.
diphosphines having an appropriate arrangement of the
(RO-substituted Ph)(phenyl)phosphino groups.
Interestingly, Rh-diphosphine quadrant diagrams allow
a simple illustration of the primary steric interaction be-
2218
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
tween the chiral Rh proximal environment and the incom- creased steric bulk from L2 to L3 to L4, which, in turn,
ing olefin substrate, and rationalize the sense of induc- increases the overall rigidity. This points to a steric factor
tion.[25,26]
in the control of enantioselectivity and substrate (or H2)
As ligand optimization and fine-tuning involve partial al- approach.
teration of the P-substituents, this can severely impact the
When (RP/SP)-L1 was resolved with {(S)-o-[1-(dimeth-
phosphine basicity and/or bulk, the catalyst overall confor- ylamino)ethyl]phenyl}palladium(II) chloride,[7] two dia-
mation, and can present difficulties in predicting catalyst stereomeric Pd complexes were obtained, wherein the ste-
performance. A seemingly slight modification of the origi- reogenic phosphorus atom and the Me2N group were in the
nal ligand structure can lead to a dramatic positive or nega- trans-position. However, when (o-An)(Ph)PCH2CH2PPh2
tive effect on stereoselectivity. As observed in several cases, was enantioenriched [70% ee (SP)] with {(R)-1-[1-(dimeth-
just an optimal design matches a given substrate [for exam- ylamino)ethyl]-2-naphthyl}palladium(II) chloride,[4i] all
ple in the case of R-BPE vs. R-DuPHOS against a series four possible mixed diastereomeric Pd complexes were ob-
of β-substituted or unsubstituted (Z)-dehydro-α-acetamido tained, with the major complex having the Ph2P group and
methyl esters[3n,3o]]. Therefore in addition to the hydro- the Me2N group in trans-position. Similar to the last case,
genation conditions, the substrate structure is critical when addition of MAC to RhI-[(RP)-(o-An)(Ph)PCH2CH2PPh2]
discussing the efficacy of a ligand. Unfortunately, mostly led to all four possible mixed diastereomeric Rh complexes
sketchy hydrogenation conditions and data for MAC are in slight favor (56:44) of the pair having the Ph2P group
found in the early reports and rarely for AS or other by- and the C=C in trans-position (the stereogenic P-atom and
now benchmark olefins, rendering comparison difficult. the amide oxygen atom are in trans-position).[4j,29] This in-
However, it is suggested that (Z)-dehydro-α-amido acids/ dicates that some coordination mode(s) of the nonsymmet-
esters are more strongly bound, are less bulky than α- rical bidentate ligand to metal complexes can be favored
amidostyrenes, and are generally hydrogenated faster.
based on a pronounced dissymmetry in their electronic
characters.
Given the low ee values of the present system, further
mechanistic studies were not undertaken.[30] From the re-
sults described above it seems likely that the RhI-L1 and
Stereochemical and Profile Considerations for RhI-L
In this investigation, our particular interest was to deter- RhI-L (L = L2–L4) catalysts operate differently in hyd-
mine the impact on hydrogenation of rigidifying the rogenation due to the unequal steric and electronic nature
CH2CH2 backbone of P-stereogenic P,P,PЈ,PЈ-tetraarylic of the p-alkyl versus p-aryl substituent. In the case of L1,
diphosphines. Screening on basic benchmark substrates re- this may privilege a reduced number of Rh intermediates
vealed that catalyst activity clearly decreased with the rigid and affect the H-positioning (electronic and steric factors
o-phenylene-backboned RhI catalysts, especially with L2– visible), but in the case of L2–L4 this seems to hinder the
L4 (P-o-ROPh-substituted) compared with L1 (P-Me-sub- approach of the incoming substrate or H2 (more steric fac-
stituted).
tors visible).
Taking into account the incidence of the backbone
With their o-phenylene backbone, the RhI-L complexes
switch on P-configuration according to the CIP rules, the are more rigid than the RhI-DiPAMP system but they are
sense of stereochemical outcome was, as expected, from the more supple than the RhI-(Me-UCAP-Ph) complex. The
“same side” of the olefin (re-face for MAA, MAC, AS, and P,PЈ-phenyls and P-o-ROPh (or P-Me) substituents have
AA, and si-face for DMI), by analogy to the corresponding more rotational freedom around the corresponding P–C
C2-symmetric chiral ethylene- or o-phenylene-bridged coun- bonds. Although the L ligands provide only one half of the
terparts. However, L1 afforded the opposite configuration chiral array of the corresponding C2-symmetric chiral di-
for N-Ac-(1-phenylethyl)amine with a low but noticeable phosphines, the Ph2P group phenyls (both isoclinal) can
value (19%ee) pointing to a reversal in the major hydro- also participate to the induction process in the complex by
genation course seen for MAC.[27]
mutually affecting each other in response to the incoming
Although the use of RhI-L1 and RhI-L2 led to compar- molecules (olefin, H2) (Figure 5).[31,32] Thus, the reaction
able enantioselectivities for MAC (69/67%ee) and AS (19/ rate with L2–L4 should be comparable to that with 1,2-
25%ee) under 1 bar H2, operating at 10 bar resulted in a bis(diphenylphosphino)benzene, and the electronic dissym-
noticeable decrease in ee for MAC (from 69 to 53%ee) but metry of L1 could be responsible for the increased rate, as
not for AS (19/20%ee) by using the former,[28] whereas ee observed with Me-UCAP-Ph.[4n,22]
values were not affected (for both MAC and AS) by using
Finally, it is the geminal substituent of P-Me that induces
the latter (or the rest of catalysts). Furthermore, hydro- the steric hindrance with 1,2-bis[(Ar)(Me)phosphino]eth-
genation rates for MAC and AS with RhI-L1 (100% conv. anes or -benzenes, as in the case of R-BisP* (the P-Me
within 0.8 h) were much higher than with the rest of the mimics the bulkiest P-substituent of the corresponding
RhI-L catalysts (L = L2–L4), and, in particular, hydro- P,P,PЈ,PЈ-tetraarylic diphosphines). The geminal P-Ar sub-
genation rates with the latter for MAC (100% conv. within stituent is able to orient the sense of enantioselectivity by
16 h) were significantly lower than for AS (100% conv. more effectively blocking a quadrant. Thus, Imamoto and
within 2–3 h). It is clear that the ee values of MAC and AS Gridnev “reformulated quadrant rule”[26] accounts for the
hydrogenation products improved consistently with in- observed sense of induction with the P-stereogenic 1,2-bis-
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2219

S. Rast, M. Stephan, B. Mohar
FULL PAPER
phine[34] and (2S,4R,5S)-(–)-3,4-dimethyl-2,5-diphenyl-1,3,2-oxaza-
phospholidine-2-borane [(–)-oxazaPB; derived from (1S,2R)-
(+)-ephedrine] and (+)-oxazaPB [derived from (1R,2S)-(–)-
ephedrine];[9] noncommercial olefin: α-acetamidostyrene [N-(1-
phenylvinyl)acetamide, AS].[35] All reactions were conducted under
an inert atmosphere (nitrogen or argon) with anhydrous solvents.
Analytical thin-layer chromatography (TLC) was performed with
Silica Gel 60 F254 pre-coated plates (0.25 mm thickness). Rf values
are reported and visualization was accomplished by irradiation
with a UV lamp (254 nm) and/or staining with KMnO4 solution.
Flash column chromatography was performed with Silica Gel 60
(40–63 μm). 1H (299.9 MHz; internal Me4Si), 13C (75.4 MHz; in-
ternal CDCl3, δ = 77.00 ppm), and 31P (120 MHz, external 85%
H3PO4) NMR spectra were recorded with a Varian Unity plus 300
spectrometer for solutions in CDCl3. HRMS measurements were
obtained with a Waters Micromass Q-TOF Premier instrument
equipped with an orthogonal Z-spray ESI interface.
Figure 5. Quadrant diagram for [RhI-L] (RЈ = Me, o-RO-Ph)
[quadrants: unhindered (1), hindered (2), and quasi-hindered de-
pending on olefin accommodation (3, 4)].
[(Ar)(Me)phosphino]ethanes (wherein Ar = Ph, Fc), and
1,2-bis[(Ph)(Me)phosphino]benzene (Figure 4). Similarly
for L1, it is the P-Ph substituent geminal to P-Me that in-
duces the steric hindrance and the phenyls of the Ph2P
group (which are not influenced by the P-stereogenic center
as much as in CH2CH2-bridged diphosphines due to δ/λ-
conformations deletion) can adopt against each other either
a hindering or unhindering conformation to accommodate
molecules (Figure 5).
(RP)-(2-Diphenylphosphinophenyl)[(1S,2R)-(N-ephedrino)](phenyl)-
phosphine-P-borane [(RP)-1]: To a cold (0 °C) solution of (2-bromo-
phenyl)diphenylphosphine (6.10 g, 17.88 mmol, 1.32 equiv.) in
Et2O (100 mL) was added sBuLi (1.2 m in cyclohexane/hexane,
92:8; 14.6 mL, 1.3 equiv.) and the mixture was stirred for 1 h. To
this mixture at –15 °C, a solution of (–)-oxazaPB (3.86 g,
13.52 mmol) in THF (10 mL) was added. The resulting mixture
was warmed to 0 °C and stirred for 2 h, then the reaction was
quenched with H2O (10 mL). The residue was partitioned between
Et2O (150 mL) and H2O (50 mL), and the organic layer was dried
(Na2SO4) and concentrated. The residue was purified on silica gel
(toluene (Rf 0.2), then toluene/EtOAc, 9:1), and recrystallized (hex-
ane/CH2Cl2) to afford the product (6.82 g, 92%) as colorless crys-
tals; m.p. 130–132 °C; [α]2D5 = –14.8 (c = 1.0, CHCl3) (Ͼ99.9% de
Conclusions
A series of mono-P-stereogenic (P-o-diphenylphosphino-
phenyl)-based PCCP-type diphosphines L was prepared by
following the Jugé–Stephan asymmetric route, and studied
in RhI-catalyzed hydrogenation of benchmark activated ole-
fins. The observed retention of P-configuration during the
H2SO4-catalyzed methanolysis step of an aminophosphine-
P-borane means that extra caution must be taken when at-
tributing P-configuration to products derived from interme-
diates possessing “non-standard” substituents in the vicin-
ity of the reactive phosphorus center.
This “knockout diphosphine design” brought a further
mechanistic insight into the relationship between steric and
electronic factors affecting the RhI-catalyzed hydrogenation
of olefins. Consequently, this system can serve as a study-
tool for the translation of skeletal modifications of P-
stereogenic P,P,PЈ,PЈ-tetraarylic ethylene-bridged diphos-
phines to catalysis.
1
1
by H NMR). H NMR: δ = 0.88–1.98 (br. m, 3 H), 1.21 (d, J =
7 Hz, 3 H), 1.68 (br. s, 1 H), 2.63 (d, J = 7 Hz, 3 H), 4.35–4.57 (m,
1 H), 4.94–5.03 (m, 1 H), 7.06–7.45 (m, 21 H), 7.52–7.63 (m, 3
H) ppm. 13C NMR: δ = 11.5 (m), 31.9 (m), 58.4 (m), 78.9 (m),
125.9 (m), 127.2–133.4 (m), 137.1–139.0 (m), 141.4 (dd, J = 23,
13 Hz), 142.5 ppm. 31P NMR: δ = –16.5 (d, J = 23 Hz), 72.6 (br.
m) ppm. MS (ESI): m/z (%) = 548.2 (30) [M + H]+. HRMS (ESI):
m/z calcd. for C34H37BNOP2 [M + H]+ 548.244; found 548.243.
Absolute configuration determination: X-ray single-crystal-struc-
ture analysis revealed its (RP)-configuration.[36]
(SP)-(2-Diphenylphosphinophenyl)[(1R,2S)-(N-ephedrino)](phenyl)-
phosphine-P-borane [(SP)-1]: By following a similar procedure to
that described for (RP)-1, starting from (+)-oxazaPB, a crystalline
material (Ͼ99.9% de by 1H NMR) was obtained with identical
spectral characteristics to those described above; [α]2D5 = +14.9 (c =
1.0, CHCl3).
(SP)-(2-Diphenylphosphinoylphenyl)[(1R,2S)-(N-ephedrino)](phen-
yl)phosphine-P-borane [(SP)-1-PЈ(O)]: To a solution of (SP)-1
(84 mg, 0.153 mmol, Ͼ99.9% de) in acetone (10 mL) was added
50% aq. H2O2 (0.1 mL) at 0 °C and the mixture was stirred for 2 h.
The thoroughly studied asymmetric hydrogenation
mechanism can be sensitive to operating conditions, and
achieving an ideal match between a chiral catalyst system
and a given substrate towards highest enantioselective effi- The reaction was quenched with 10% aq. Na2SO3, concentrated,
and partitioned between CH2Cl2/H2O. The organic layer was fil-
tered through a pad of SiO2/Na2SO4 eluting with CH2Cl2 then with
EtOAc to afford the product (78 mg, 90 %) as a colorless oil.
[α]2D5 = –52.1 (c = 1.0, CHCl3) (Ͼ99.9% de by 1H NMR). 1H NMR:
δ = 0.21–1.50 (br. m, 3 H), 1.19 (d, J = 7 Hz, 3 H), 3.12 (d, J =
11 Hz, 3 H), 3.67–3.94 (m, 1 H), 4.57 (t, J = 4 Hz, 1 H), 4.68–4.77
(m, 1 H), 7.06–6.91 (m, 2 H), 7.67–7.07 (m, 21 H), 7.85–7.73 (m,
1 H) ppm. 13C NMR: δ = 12.1 (d, J = 3 Hz), 33.3 (d, J = 8 Hz),
58.6 (d, J = 3 Hz), 75.9, 126.6–132.2 (m), 133.5–136.0 (m), 137.1–
138.7 (m), 141.9 ppm. 31P NMR: δ = 36.2 (d, J = 4.4 Hz), 74.8 (br.
m) ppm.
ciency remains largely an empirical process. The design and
screening of complementary new chiral ligands provides
probes with which to extend our knowledge. As Heller
rightfully put it “a chiral ligand is a necessary but not suf-
ficient condition for stereoselection”.[33]
Experimental Section
Materials and Methods: The following compounds were prepared
according to reported procedures: (2-bromophenyl)diphenylphos-
2220
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
Methyl (RP)-(2-Diphenylphosphinophenyl)(phenyl)phosphinite-P- Ͼ99.9%ee) in acetone (10 mL) was added 50% aq. H2O2 (0.5 mL)
borane [(RP)-2]: To a solution of (RP)-1 (6.00 g, 10.96 mmol) in at 0 °C and the mixture was stirred for 2 h. The reaction mixture
CH2Cl2 (30 mL) and absolute MeOH (120 mL), a solution of 96%
was quenched with 10 % aq. Na2SO3, and partitioned between
H2SO4 (1.11 g, 10.86 mmol) in absolute MeOH (30 mL) was added EtOAc and H2O. The organic layer was filtered through a pad of
at room temp. while stirring. The mixture was stirred overnight,
filtered through a bed of silica gel, and concentrated. The crude
product was purified on silica gel (toluene) to give the title com-
pound (4.32 g, 96%) as a colorless viscous oil. Rf = 0.7 (toluene);
[α]2D5 = +2.3 (c = 1.7, CHCl3); 98.2%ee by HPLC (Chiralpak AD-
H; hexane/2-PrOH, 98:2; 1.0 mL/min; λ 254 nm): tR = 7.6 (RP), 8.2
silica gel/Na2SO4 followed by recrystallization from MeOH to af-
ford the product (57 mg, 92 %) as white crystals; m.p. 100 °C;
1
[α]2D5 = +95.4 (c = 1.0, abs. MeOH). H NMR: δ = 0.3–1.7 (br. m,
3 H), 3.25 (d, J = 12 Hz, 3 H), 7.26–7.69 (m, 16 H), 7.78 (ddd, J
= 11, 8, 1 Hz, 2 H), 8.07 (ddd, J = 10, 8, 4 Hz, 1 H) ppm. 13C
NMR: δ = 53.5 (d, J = 3 Hz), 128.0–128.5 (m), 130.1–130.4 (m),
131.1–132.3 (m), 135.0–135.7 (m) ppm. 31P NMR: δ = 59.9 (d, J =
(SP) min [racemic mixture prepared by weighing equal amounts of
1
(RP)- and (SP)-2]. H NMR: δ = 0.59–1.74 (br. m, 3 H), 3.48 (d, J 5 Hz), 140.6 (br. m, J = 5 Hz) ppm. HRMS (ESI): m/z calcd. for
= 12 Hz, 3 H), 6.82–7.15 (m, 4 H), 7.17–7.53 (m, 12 H), 7.59–7.74
C25H23BO2P2 [M – H]+ 428.1375; found 428.1375. Absolute config-
(m, 2 H), 8.00–8.15 (m, 1 H) ppm. 13C NMR: δ = 53.6 (m), 128.0– uration determination: X-ray single-crystal-structure analysis re-
129.0 (m), 132.5–134.1 (m), 136.4–138.5 (m), 141.5 (dd, J = 23, vealed its (SP)-configuration.[36]
9 Hz) ppm. 31P NMR: δ = –14.5 (d, J = 23 Hz), +111.2 (br.
(SP)-(2-Diphenylphosphinophenyl)(methyl)(phenyl)phosphine-P-bor-
m) ppm. MS (ESI): m/z (%) = 415.1 (90) [M + H]+. HRMS (ESI):
ane [(SP)-L1·BH3]: To a cold (–20 °C) solution of (RP)-2 (2.00 g,
m/z calcd. for C25H26BOP2 [M + H]+ 415.155; found 415.156.
4.83 mmol, Ͼ99.9%ee) in THF (50 mL) was added dropwise MeLi
Methyl (SP)-(2-Diphenylphosphinophenyl)(phenyl)phosphinite-P-
borane [(SP)-2]: By following a similar procedure to that described
for (RP)-2, starting from (SP)-1, a viscous oil (98.4%ee by HPLC)
was obtained with identical spectral characteristics to those de-
scribed above.
(1.6 m in Et2O, 5.0 mL). The mixture was warmed to room temp.
overnight then the reaction was quenched with H2O (10 mL) and
the mixture was concentrated and extracted with Et2O. The organic
layer was filtered through a bed of silica gel/Na2SO4 and concen-
trated to afford the product (1.65 g, 86%) as a colorless syrup; Rf
0.3 (hexane/EtOAc, 9:1); [α]2D5 = –28.1 (c = 1.0, CHCl3). 1H NMR:
δ = 0.56–1.71 (br. m, 3 H), 2.16 (d, J = 10 Hz, 3 H), 6.78–6.90 (m,
2 H), 6.92–7.02 (m, 2 H), 7.09–7.30 (m, 9 H), 7.40–7.52 (m, 5 H),
8.12–8.27 (m, 1 H) ppm. 13C NMR: δ = 13.2 (m), 128.1–137.8 (m),
141.8 (dd, J = 20, 4 Hz) ppm. 31P NMR: δ = –15.7 (d, J = 23 Hz),
15.3 (br. m) ppm. MS (ESI): m/z (%) = 421.1 (15) [M + Na]+.
HRMS (ESI): m/z calcd. for C25H25BP2Na [M + Na]+ 421.142;
found 421.142. Absolute configuration of the title compound was
assigned as (SP) based on configuration of its BH3-free derivative
L1 (a reported compound,[7] the configuration of which has been
determined by X-ray analysis of a palladium complex), which was
determined by optical rotation measurement (see below), and as-
sumed retention of P-configuration during decomplexation as in
the case of related reported compounds such as (RP,RP)-
DioxyBenzP*·BH3 and P-stereogenic phosphine-P-boranes in ge-
neral.[3j,12]
Methyl (RP)-[2-(Diphenylphosphino-P-borane)phenyl](phenyl)phos-
phinite-P-borane [(RP)-2-PЈ(BH3)]: To a cold (0 °C) solution of
(RP)-2 (4.14 g, 10.0 mmol, 98.2%ee) in CH2Cl2 (30 mL) was added
Me2S·BH3 (5.0 mL, 50 mmol). After stirring for 1 h, the mixture
was filtered through a bed of silica gel and carefully concentrated
at room temp. to afford the product (4.36 g, 100% yield) as a white
powder; m.p. 141–143 °C; [α]2D5 = –178.6 (c = 1.0, CH2Cl2).
Recrystallization (hexane/CH2Cl2) afforded the product (3.77 g,
88 %) as fine white needles in enantiomerically pure form:
Ͼ99.9%ee [by HPLC analysis of regenerated (RP)-2]; [α]2D5 = –183.3
1
(c = 1.0, CH2Cl2). H NMR: δ = 0.07–2.61 (br. m, 6 H), 2.91 (d,
J = 12 Hz, 3 H), 7.07–7.22 (m, 1 H), 7.32 (ddd, J = 8, 3, 1 Hz, 1
H), 7.36–7.56 (m, 10 H), 7.57–7.74 (m, 5 H), 7.74–7.89 (m, 2
H) ppm. 13C NMR: δ = 52.2 (d, J = 3 Hz), 128.5–128.9 (m), 130.2–
131.1 (m), 131.8–132.6 (m), 133.9 (d, J = 9 Hz), 135.0 (dd, J = 23,
8 Hz), 136.5 (dd, J = 9, 7 Hz) ppm. 31P NMR: δ = +57.7 (br. m),
+143.4 (br. m) ppm.
(RP)-(2-Anisyl)(2-diphenylphosphinophenyl)(phenyl)phosphine-P-bor-
ane [(RP)-L2·BH3]: To a cold (0 °C) solution of 2-bromoanisole
(0.450 g, 2.40 mmol, 1.3 equiv.) in Et2O (10 mL) was added drop-
wise sBuLi (1.25 m in cyclohexane/hexaane, 92:8; 1.9 mL,
1.3 equiv.). After stirring at 0 °C for 45 min, a solution of (RP)-2
(0.760 g, 1.83 mmol, Ͼ99.9%ee) in Et2O (10 mL) was slowly added
at –78 °C. The reaction mixture was stirred at room temp. over-
night, quenched with H2O (1 mL), and concentrated. The residue
was partitioned between H2O (10 mL) and CH2Cl2 (10 mL), and
extracted with CH2Cl2 (2ϫ 10 mL). The organic layer was dried
(Na2SO4), concentrated, and the residue was recrystallized (EtOAc)
to afford the product (0.476 g, 54%) as a white powder; m.p. 166–
168 °C; Rf 0.2 (hexane/EtOAc, 9:1); Ͼ99.9 % ee by HPLC (Chi-
Methyl (SP)-[2-(Diphenylphosphino-P-borane)phenyl](phenyl)phos-
phinite-P-borane [(SP)-2-PЈ(BH3)]: By following a similar procedure
to that described for (RP)-2-PЈ(BH3), starting from (SP)-2
(98.4%ee), a white powder was obtained with identical spectral
characteristics to those described above with [α]2D5 = +183.4 (c =
1.0, CH2Cl2); Ͼ99.9%ee [by HPLC analysis of regenerated (SP)-
2].
Methyl (RP)-(2-Diphenylphosphinophenyl)(phenyl)phosphinite-P-
borane [(RP)-2; Ͼ99.9%ee]: A solution of (RP)-2-PЈ(BH3) (2.90 g,
6.77 mmol, Ͼ99.9 % ee) in CH2Cl2 (50 mL) and 96 % MeOH
(50 mL) was stirred at 40 °C overnight then concentrated. The resi-
due was filtered through a bed of silica gel (toluene) to afford the
title regenerated compound (2.75 g, 98%) as a colorless viscous oil;
[α]2D5 = +2.4 (c = 1.7, CHCl3); Ͼ99.9%ee by HPLC. 1H NMR data
were in accordance with those of 2 prepared as described above.
ralpak AD-H; hexane/2-PrOH, 95:5; 1.0 mL/min; λ 254 nm): tR
=
11.0 (SP), 12.9 (RP) min [a racemic mixture was prepared by weigh-
ing equal amounts of (RP)- and (SP)-L2·BH3]. 1H NMR: δ = 0.94–
2.09 (br. m, 3 H), 3.25 (s, 3 H), 6.42 (ddd, J = 8, 4, 1 Hz, 1 H),
6.91–7.05 (m, 5 H), 7.13–7.45 (m, 13 H), 7.46–7.62 (m, 1 H), 7.72–
7.87 (m, 2 H), 7.93 (ddd, J = 14, 8, 2 Hz, 1 H) ppm. 13C NMR: δ
= 54.6 (m), 110.1–138.4 (m), 141.1 (dd, J = 21, 10 Hz), 160.7 ppm.
31P NMR: δ = –16.6 (d, J = 30 Hz), 20.7 (br. m) ppm. MS (ESI):
m/z (%) = 489.2 (45) [M – H]+. HRMS (ESI): m/z calcd. for
C31H28BOP2 [M – H]+ 489.171; found 489.171. Absolute configu-
ration of the title compound was assigned as (RP) based on config-
Methyl (SP)-(2-Diphenylphosphinophenyl)(phenyl)phosphinite-P-
borane [(SP)-2; Ͼ99.9%ee]: By following a similar procedure to
that described for (RP)-2 (Ͼ99.9%ee), starting from (SP)-2-PЈ(BH3)
(Ͼ99.9 % ee), a viscous oil (Ͼ99.9 % ee by HPLC) was obtained
with identical spectral characteristics to those detailed above.
Methyl (SP)-(2-Diphenylphosphinoylphenyl)(phenyl)phosphinite-P-
borane [(SP)-2-PЈ(O)]: To a solution of (SP)-2 (60 mg, 0.145 mmol,
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2221

S. Rast, M. Stephan, B. Mohar
FULL PAPER
uration of its BH3-free derivative L2, which was determined by X-
ray crystal diffraction analysis (see below), and assumed retention
of P-configuration during decomplexation as in the case of related
reported compounds such as (RP,RP)-DioxyBenzP*·BH3 and P-ste-
reogenic phosphine-P-boranes in general.[3j,12]
3 H), 6.91–7.00 (m, 1 H), 7.17–7.34 (m, 18 H) ppm. 13C NMR: δ
= 12.4 (m), 127.7–129.1 (m), 131.7–134.1 (m), 137.1 (dd, J = 12,
6 Hz), 137.6 (dd, J = 12, 6 Hz), 140.2 (dd, J = 13, 5 Hz), 143.5 (dd,
J = 32, 11 Hz), 145.9 (dd, J = 31, 13 Hz) ppm. 31P NMR: δ = –36.2
(d, J = 158 Hz), –13.2 (d, J = 158 Hz) ppm. MS (ESI): m/z (%) =
385.1 (40) [M + H]+. HRMS (ESI): m/z calcd. for C25H23P2 [M
+ H]+ 385.128; found 385.127. Absolute configuration of the title
compound was determined as (SP) by comparing its optical rota-
tion with the reported compound [(S)-isomer: [α]589 = –51 (c =
2.38, acetone); (R)-isomer: [α]589 = +51 (c = 2.50, acetone)].[7]
(SP)-(2-Anisyl)(2-diphenylphosphinophenyl)(phenyl)phosphine-P-bor-
ane [(SP)-L2·BH3]: By following a similar procedure to that de-
scribed for (RP)-L2·BH3, starting from (SP)-2 (Ͼ99.9%ee), a white
powder was obtained with identical spectral characteristics to those
described above.
(RP)-(2-Anisyl)(2-diphenylphosphinophenyl)(phenyl)phosphine [(RP)-
L2]: Prepared by following a similar procedure to that described
for L1, from (RP)-L2·BH3 (0.40 g, 0.81 mmol, Ͼ99.9%ee), to af-
ford the product (0.373 g, 96 %) as colorless crystals; m.p. 158–
(RP)-(2-Diphenylphosphinophenyl)(2-isopropoxyphenyl)(phenyl)phos-
phine-P-borane [(RP)-L3·BH3]: By following a similar procedure to
that described for (RP)-L2·BH3 from o-bromo-isopropoxybenzene
(0.516 g, 2.40 mmol, 1.3 equiv.), the product (0.513 g, 55%) was ob-
tained as a white powder; m.p. 170–173 °C; Rf 0.3 (hexane/EtOAc,
1
160 °C; Rf 0.6 (toluene); [α]2D5 = –49.7 (c = 1.2, CHCl3). H NMR:
1
9:1); [α]2D5 = –15.5 (c = 0.8, CHCl3). H NMR: δ = 0.51–1.89 (br.
δ = 3.63 (s, 3 H), 6.63 (ddd, J = 7, 4, 2 Hz, 1 H), 6.72–6.85 (m, 2
H), 6.94–7.08 (m, 2 H), 7.11–7.29 (m, 18 H) ppm. 13C NMR: δ =
55.6, 110.2 (m), 120.8, 126.0 (dd, J = 14, 7 Hz), 128.1–130.0 (m),
133.7–134.3 (m), 136.5 (dd, J = 12, 5 Hz), 137.4 (dd, J = 12, 5 Hz),
137.5 (dd, J = 12, 4 Hz), 143.3 (dd, J = 32, 11 Hz), 143.7 (dd, J =
33, 10 Hz), 161.0 (d, J = 15 Hz) ppm. 31P NMR: δ = –22.9 (d, J =
166 Hz), –13.4 (d, J = 166 Hz) ppm. MS (ESI): m/z (%) = 477.2
(100) [M + H]+. HRMS (ESI): m/z calcd. for C31H27OP2 [M + H]+
477.154; found 477.155. Absolute configuration determination: X-
ray single-crystal-structure analysis revealed its (RP)-configura-
tion.[36]
m, 3 H), 0.70 (d, J = 6 Hz, 3 H), 0.96 (d, J = 6 Hz, 3 H), 4.06
(sept, J = 6 Hz, 1 H), 6.32–6.46 (m, 1 H), 6.91–7.05 (m, 5 H), 7.11–
7.46 (m, 13 H), 7.54–7.69 (m, 1 H), 7.75–7.91 (m, 2 H), 7.99 (ddd,
J = 14, 8, 2 Hz, 1 H) ppm. 13C NMR: δ = 21.1 (m), 69.1 (m), 111.1
(m), 117.5 (dd, J = 58, 4 Hz), 120.3 (m), 127.9–138.7 (m), 141.3
(dd, J = 22, 11 Hz), 158.9 ppm. 31P NMR: δ = –16.7 (d, J = 30 Hz),
20.7 (br. m) ppm. MS (ESI): m/z (%) = 517.2 (80) [M – H]+. HRMS
(ESI): m/z calcd. for C33H32BOP2 [M – H]+ 517.202; found 517.203.
Absolute configuration of the title compound is assumed to be (RP)
by extension from L2·BH3.
(SP)-(2-Diphenylphosphinophenyl)(2-isopropoxyphenyl)(phenyl)phos-
phine-P-borane [(SP)-L3·BH3]: By following a similar procedure to
that described for (RP)-L3·BH3, starting from (SP)-2 (Ͼ99.9%ee),
the product was obtained as a white powder with identical spectral
characteristics to those described above.
(SP)-(2-Anisyl)(2-diphenylphosphinophenyl)(phenyl)phosphine [(SP)-
L2]: By following a similar procedure to that described for (RP)-
L2, starting from (SP)-L2·BH3 (Ͼ99.9%ee), the product was ob-
tained as colorless crystals with identical spectral characteristics to
those described above.
(RP)-(2-tert-Butoxyphenyl)(2-diphenylphosphinophenyl)(phenyl)phos-
phine-P-borane [(RP)-L4·BH3]: To a solution of tert-butoxybenzene
(300 mg, 2.00 mmol, 1.4 equiv.) in cyclohexane (2 mL), tBuLi
(1.6 m in pentane, 1.25 mL, 1.4 equiv.) was added dropwise at room
temp. After stirring at 60 °C for 6 h, the solution was cooled to
–78 °C and a solution of (RP)-2 (637 mg, 1.54 mmol, Ͼ99.9%ee)
in Et2O (5 mL) was added slowly. The mixture was stirred at room
temp. overnight, then the reaction was quenched with H2O (1 mL),
and concentrated. The residue was partitioned between H2O
(10 mL) and CH2Cl2 (10 mL), and extracted with CH2Cl2 (2 ϫ
10 mL). The organic layer was dried (Na2SO4), concentrated, and
purified on silica gel (hexane then hexane/EtOAc, 95:5 to 90:10) to
give the product (0.426 g, 52 %) as a white powder; m.p. 132–
(RP)-(2-Diphenylphosphinophenyl)(2-isopropoxyphenyl)(phenyl)phos-
phine [(RP)-L3]: Prepared by following a similar procedure to that
described for L1, from (RP)-L3·BH3 (0.40 g, 0.77 mmol) to afford
the product (0.381 g, 98%) as colorless crystals; m.p. 130–132 °C;
Rf 0.6 (toluene); [α]2D5 = –50.9 (c = 1.5, CHCl3). 1H NMR: δ = 0.97
(d, J = 6 Hz, 3 H), 1.09 (d, J = 6 Hz, 3 H), 4.44 (sept, J = 6 Hz, 1
H), 6.61 (ddd, J = 8, 4, 2 Hz, 1 H), 6.67–6.83 (m, 2 H), 7.02–7.28
(m, 20 H) ppm. 13C NMR: δ = 21.7, 21.8, 70.1, 111.9 (m), 120.3,
127.4 (dd, J = 14, 8 Hz), 128.0–129.5 (m), 133.7–134.7 (m), 136.5
(dd, J = 11, 4 Hz), 137.4 (dd, J = 12, 5 Hz), 137.7 (dd, J = 13,
7 Hz), 143.4 (dd, J = 32, 10 Hz), 143.9 (dd, J = 32, 10 Hz), 159.0
(d, J = 14 Hz) ppm. 31P NMR: δ = –21.4 (d, J = 164 Hz), –14.0
(d, J = 164 Hz) ppm. MS (ESI): m/z (%) = 505.2 (100) [M + H]+.
HRMS (ESI): m/z calcd. for C33H31OP2 [M + H]+ 505.185; found
505.185.
134 °C; [α]2D5 = –13.7 (c = 1, CHCl3). H NMR: δ = 0.76–2.40 (br.
1
m, 3 H), 1.05 (s, 9 H), 6.53–6.62 (m, 1 H), 6.81–6.94 (m, 2 H),
6.94–7.04 (m, 2 H), 7.08–7.23 (m, 4 H), 7.23–7.45 (m, 9 H), 7.45–
7.57 (m, 1 H), 7.61–7.76 (m, 4 H) ppm. 13C NMR: δ = 28.4, 79.6,
115.2 (d, J = 5 Hz), 119.9 (d, J = 11 Hz), 127.9–128.1 (m), 128.8
(d, J = 11 Hz), 130.3–130.7 (m), 132.3 (m), 133.0–133.4 (m), 134.1–
134.2 (m), 134.7–135.0 (m), 135.7–135.9 (m), 137.6–137.8 (m),
158.4 ppm. 31P NMR: δ = –16.1 (d, J = 31 Hz), 22.4 (br. m) ppm.
HRMS (ESI): m/z calcd. for C34H34BOP2 [M – H]+ 531.2176;
found 531.2176. Absolute configuration of the title compound is
assumed to be (RP) by extension from L2·BH3.
(SP)-(2-Diphenylphosphinophenyl)(2-isopropoxyphenyl)(phenyl)phos-
phine [(SP)-L3]: By following a similar procedure to that described
for (RP)-L3, starting from (SP)-L3·BH3, colorless crystals were ob-
tained with identical spectral characteristics to those described
above.
(RP)-(2-tert-Butoxyphenyl)(2-diphenylphosphinophenyl)(phenyl)phos-
phine [(RP)-L4]: Prepared by following a similar procedure to that
described for L1, from (RP)-L4·BH3 (0.35 g, 0.657 mmol), to afford
the product (0.174 g, 98%) as colorless crystals; m.p. 128–130 °C;
(SP)-(2-Diphenylphosphinophenyl)(methyl)(phenyl)phosphine [(SP)-
L1]: (SP)-L1·BH3 (1.10 g, 2.76 mmol) in Et2NH (30 mL) was Rf 0.7 (toluene); [α]2D5 = –19.2 (c = 1.0, CHCl3). 1H NMR: δ = 1.28
heated to reflux for 2–3 h under argon then concentrated. The resi-
due was filtered through a pad of silica gel with toluene under an
inert atmosphere to afford the product (0.998 g, 94%) as a colorless
syrup; Rf 0.5 (hexane/EtOAc, 9:1); [α]2D5 = –30.1 (c = 1.0, CHCl3),
(s, 9 H), 6.50–6.70 (m, 1 H), 6.74–6.78 (m, 1 H), 6.93–7.42 (m, 21
H) ppm. 13C NMR: δ = 29.0 (d, J = 1 Hz), 79.5, 118.4 (m), 121.5,
128.0–129.0 (m), 131.1 (dd, J = 12, 8 Hz), 133.7–134.7 (m), 136.8–
137.9 (m), 143.2 (dd, J = 32, 11 Hz), 144.1 (dd, J = 32, 11 Hz),
158.3 (d, J = 16 Hz) ppm. 31P NMR: δ = –14.1 (d, J = 166 Hz),
[α]2D3 = –74.1 (c = 2.5, acetone). H NMR: δ = 1.47 (d, J = 4 Hz,
1
2222
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
–20.4 (d, J = 166 Hz) ppm. HRMS (ESI): m/z calcd. for C34H33OP2 esterification with TMSCHN2 (CP-Chiralsil-DEX CB; 25m ϫ
[M + H]+ 519.2007; found 519.2011.
0.25 mm; 90 °C): tR = 19.7 (S), 20.5 (R) min.
(SP)-(2-Diphenylphosphinoylphenyl)(2-isopropoxyphenyl)(phenyl)-
phosphine-P-borane [(SP)-L3-P(BH3),PЈ(O)]: To a solution of (SP)-
L3·BH3 (50 mg, 0.097 mmol) in CH2Cl2 (10 mL) was added 50%
aq. H2O2 (0.1 mL) at –10 °C and the mixture was stirred for 2 h.
The reaction was quenched with 10% aq. Na2SO3, and the mixture
was partitioned between CH2Cl2 and H2O, and the organic layer
was filtered through a pad of MgSO4/Na2SO4 to afford the product
(48 mg, 95%) as a colorless oil. [α]2D5 = +101.9 (c = 1.0, CHCl3).
1H NMR: δ = 0.61–2.10 (br. m, 3 H), 0.78 (d, J = 6.1 Hz, 3 H),
1.01 (d, J = 6.0 Hz, 3 H), 3.97 (sept, J = 6.0 Hz, 1 H), 6.05–6.15
(m, 1 H), 6.79–6.89 (m, 1 H), 7.02–7.54 (m, 16 H), 7.55–7.70 (m,
2 H), 7.73–7.84 (m, 2 H), 8.49–8.63 (m, 1 H) ppm. 13C NMR: δ =
20.3, 21.6, 68.6, 110.6 (d, J = 6 Hz), 119.1 (d, J = 10 Hz), 127.3–
128.3 (m), 129.3–132.2 (m), 134.1 (d, J = 9 Hz), 135.0 (dd, J = 14,
7 Hz), 135.5–135.6 (m), 138.4 (dd, J = 19, 10 Hz), 158.2 (d, J =
4 Hz) ppm. 31P NMR: δ = 26.0 (m), 29.7 (d, J = 5.4 Hz) ppm.
Acknowledgments
This work was supported by the Ministry of Higher Education,
Science, and Technology of the Republic of Slovenia (grant number
P1-242). Barbara Modec, University of Ljubljana, is thanked for
X-ray crystal structures determination on the Supernova dif-
fractometer of EN-FIST Centre of Excellence, Slovenia. Borut
Zupancˇicˇ is thanked for assistance.
[1] a) W. S. Knowles, Acc. Chem. Res. 1983, 16, 106–112; b) W. S.
Knowles, Adv. Synth. Catal. 2003, 345, 3–13; c) B. D. Vineyard,
W. S. Knowles, M. J. Sabacky, G. L. Bachman, D. J. Weinkauff,
J. Am. Chem. Soc. 1977, 99, 5946–5952.
[2] a) A. Grabulosa, P-Stereogenic Ligands in Enantioselective Ca-
talysis, 1st ed., RSC Publishing, Cambridge, UK, 2011; b)
P. C. J. Kamer, P. W. N. M. van Leeuwen, Phosphorous(III) Li-
gands in Homogenous Catalysis: Design and Synthesis vol. 44,
John Wiley & Sons, West Sussex, UK, 2012; c) A. Börner,
Phosphorus Ligands in Asymmetric Catalysis, Wiley-VCH,
Weinheim, Germany, 2008.
Preparation of [Rh(L)(MeOH)2]BF4 Solvate: To a solution of
bis(2,5-norbornadiene)rhodium tetrafluoroborate {[Rh(nbd)2]BF4}
(9.3 mg, 0.025 mmol) in MeOH (0.5 mL), a solution of ligand L
(0.025 mmol, 1 equiv. to Rh atom) in MeOH (0.5 mL) was added
dropwise at room temp. The mixture was hydrogenated at 1 atm H2
for 30 min, and this amount was equally divided for five hydrogen-
ation tests. The same procedure was followed for the preparation
of RhI complexes of (SP,SP)-1,2-bis[(methyl)(phenyl)phosphino]
ethane and (SP,SP)-1,2-bis[(1-naphthyl)(phenyl)phosphino]ethane.
[3] The following are representative works. For R-SMS-Phos, see:
a) B. Mohar, M. Stephan, Adv. Synth. Catal. 2013, 355, 594–
600; b) B. Zupancˇicˇ, B. Mohar, M. Stephan, Org. Lett. 2010,
ˇ
12, 1296–1299; c) M. Stephan, D. Sterk, B. Mohar, Adv. Synth.
Catal. 2009, 351, 2779–2786; for (S,S)-tBu-BisP*, see: d) I. D.
Gridnev, Y. Yamanoi, N. Higashi, H. Tsuruta, M. Yasutake, T.
Imamoto, Adv. Synth. Catal. 2001, 343, 118–136; e) I. D. Grid-
nev, M. Yasutake, N. Higashi, T. Imamoto, J. Am. Chem. Soc.
2001, 123, 5268–5276; for (1S,1ЈS,2R,2ЈR)-TangPhos, Du-
anPhos, and ZhangPhos, see: f) X. W. Zhang, K. X. Huang,
G. H. Hou, B. N. Cao, X. M. Zhang, Angew. Chem. Int. Ed.
2010, 49, 6421–6424; Angew. Chem. 2010, 122, 6565; g) D. Liu,
X. Zhang, Eur. J. Org. Chem. 2005, 646–649; h) W. Tang, X.
Zhang, Angew. Chem. Int. Ed. 2002, 41, 1612–1614; Angew.
Chem. 2002, 114, 1682; for BIBOPs, see: i) W. J. Tang, B. Qu,
A. G. Capacci, S. Rodriguez, X. D. Wei, N. Haddad, B. Naray-
anan, S. L. Ma, N. Grinberg, N. K. Yee, D. Krishnamurthy,
C. H. Senanayake, Org. Lett. 2010, 12, 176–179; for BenzP*,
QuinoxP*, and similar ligands, see: j) T. Imamoto, K. Tamura,
Z. Zhang, Y. Horiuchi, M. Sugiya, K. Yoshida, A. Yanagisawa,
I. D. Gridnev, J. Am. Chem. Soc. 2012, 134, 1754–1769; k) K.
Tamura, M. Sugiya, K. Yoshida, A. Yanagisawa, T. Imamoto,
Org. Lett. 2010, 12, 4400–4403; l) T. Miura, T. Imamoto, Tetra-
hedron Lett. 1999, 40, 4833–4836; for R-DuPhos and R-BPE,
see: m) M. J. Burk, Y. M. Wang, J. R. Lee, J. Am. Chem. Soc.
1996, 118, 5142–5143; n) M. J. Burk, J. E. Feaster, W. A. Nug-
ent, R. L. Harlow, J. Am. Chem. Soc. 1993, 115, 10125–10138;
o) M. J. Burk, J. Am. Chem. Soc. 1991, 113, 8518–8519; for R-
FerroTANE, see: p) U. Berens, M. J. Burk, A. Gerlach, W.
Hems, Angew. Chem. Int. Ed. 2000, 39, 1981–1984; Angew.
Chem. 2000, 112, 2057.
Procedure for Hydrogenation Tests (S/C 100): To a solution of the
substrate (0.5 mmol) in MeOH (7.5 mL), three freeze-pump-thaw
cycles were applied and the system was filled with argon. To the
substrate solution was added under argon a solution of the above
preformed [Rh(L)(MeOH)2]BF4 solvate (0.2 mL, 0.005 mmol). A
vacuum was applied to this system then it was backfilled with H2.
The mixture was stirred at room temp. and H2 pressure as indi-
cated. Progress of the hydrogenation was monitored by the dimin-
ution of the volume of the closed reaction system at 1 bar (until
H2 uptake ceased and the color of the solution changed); for reac-
tions at 10 bar, analyses were carried out after 16 h. The reaction
mixture was analyzed by chiral GC and the absolute configuration
was assigned by comparison with the reported data of tR.[6] The
following hydrogenation products with the mentioned absolute
configurations were obtained by using (RP)-L3 or L4 (Table 1).
(S)-N-Acetyl-alanine Methyl Ester: 1H NMR: δ = 1.41 (d, J = 7 Hz,
3 H), 2.02 (s, 3 H), 3.76 (s, 3 H), 4.60 (m, 1 H), 6.26 (br. s, 1
H) ppm. GC (Lipodex-E; 25m ϫ 0.25 mm; 120 °C): tR = 8.2 (S),
9.2 (R) min.
1
(S)-N-Acetyl-phenylalanine Methyl Ester: H NMR: δ = 1.99 (s, 3
H), 3.13 (m, 2 H), 3.73 (s, 3 H), 4.90 (dt, J = 6, 8 Hz, 1 H), 5.90
(br. d, J = 6 Hz, 1 H), 7.09 (m, 2 H), 7.28 (m, 3 H) ppm. GC
(Chiralsil-l-Val; 25m ϫ 0.25 mm; 160 °C): tR = 4.7 (R), 4.9 (S) min.
[4] For trichickenfootphos (TCFP), see: a) G. Hoge, H. P. Wu,
W. S. Kissel, D. A. Pflum, D. J. Greene, J. Bao, J. Am. Chem.
Soc. 2004, 126, 5966–5967; b) H. P. Wu, G. Hoge, Org. Lett.
2004, 6, 3645–3647; for MeO-POP, see: c) W. J. Tang, A. G.
Capacci, A. White, S. L. Ma, S. Rodriguez, B. Qu, J. Savoie,
N. D. Patel, X. D. Wei, N. Haddad, N. Grinberg, N. K. Yee, D.
Krishnamurthy, C. H. Senanayake, Org. Lett. 2010, 12, 1104–
1107; for other C1-symmetrical PCP-type ligands, see: d) Q.
Dai, W. Li, X. M. Zhang, Tetrahedron 2008, 64, 6943–6948; e)
K. X. Huang, X. W. Zhang, T. J. Emge, G. H. Hou, B. N. Cao,
X. M. Zhang, Chem. Commun. 2010, 46, 8555–8557; for un-
symmetrical BisP*, see: f) A. Ohashi, T. Imamoto, Tetrahedron
Lett. 2001, 42, 1099–1101; g) A. Ohashi, S. Kikuchi, M. Yasu-
take, T. Imamoto, Eur. J. Org. Chem. 2002, 2535–2546; for
1
Dimethyl (R)-2-Methylsuccinate: H NMR: δ = 1.23 (d, J = 7 Hz,
3 H), 2.42 (dd, J = 6, 16 Hz, 1 H), 2.75 (dd, J = 8, 16 Hz, 1 H),
2.93 (m, 1 H), 3.68 (s, 3 H), 3.70 (s, 3 H) ppm. GC (Lipodex-E;
25m ϫ 0.25 mm; 80 °C): tR = 8.6 (S), 9.0 (R) min.
(S)-N-Acetyl-(1-phenylethyl)amine: 1H NMR: δ = 1.48 (d, J =
6.9 Hz, 3 H), 1.97 (s, 3 H), 5.12 (m, 1 H), 5.84 (br. s, 1 H), 7.22–
7.39 (m, 5 H) ppm. GC (CP-Chiralsil-DEX CB; 25m ϫ 0.25 mm;
140 °C): tR = 10.6 (S), 11.7 (R) min.
1
(S)-2-Phenylpropionic Acid: H NMR: δ = 1.52 (d, J = 7 Hz, 3 H),
3.74 (q, J = 7 Hz, 1 H), 7.31 (m, 5 H) ppm. GC analysis after
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2223

S. Rast, M. Stephan, B. Mohar
FULL PAPER
(RP)- and (SP)-(Nmen)(Ph)PCH2CH2PPh2, see: h) R. B. King, [17]
J. Bakos, C. D. Hoff, L. Marko, J. Org. Chem. 1979, 44, 1729–
1731; for (RP)- and (SP)-(o-An)(Ph)PCH2CH2PPh2, see: i) J. A.
Ramsden, J. M. Brown, M. B. Hursthouse, A. I. Karalulov,
Tetrahedron: Asymmetry 1994, 5, 2033–2044; j) J. A. Ramsden,
T. D. W. Claridge, J. M. Brown, J. Chem. Soc., Chem. Commun.
1995, 2469–2471; for (R)-Prophos, see: k) M. D. Fryzuk, B.
Bosnich, J. Am. Chem. Soc. 1978, 100, 5491–5494; for (R)-
Cycphos, see: l) J. D. Oliver, D. P. Riley, Organometallics 1983,
2, 1032–1038; for Me-UCAP-R (or SemiPHOS with R = Ph)
and close analogues, see: m) Q. Lin, PhD Thesis 1999, The
University of New South Wales Sydney (AU); n) S. Basra, J. G.
de Vries, D. J. Hyett, G. Harrison, K. M. Heslop, A. G. Orpen,
P. G. Pringle, K. von der Luehe, Dalton Trans. 2004, 1901–
1905; o) A. Baber, J. G. de Vries, A. G. Orpen, P. G. Pringle, K.
von der Luehe, Dalton Trans. 2006, 4821–4828; p) K. Matsu-
a) For this work, we have prepared (SP,SP)-1,2-bis[(methyl)-
(phenyl)phosphino]ethane by following ref.[17b] and tested it in
RhI-catalyzed hydrogenation (1 bar H2, MeOH, room temp.,
S/C = 100) of MAC (15 min, 100% conv., 17%ee (R)), AS
(20 min, 100% conv., 61%ee (R)), and DMI (20 min, 100%
conv., 23%ee (S)). (SP,SP)-1,2-Bis[(methyl)(phenyl)phosphino]-
ethane was prepared but not evaluated in catalysis; for this, see:
b) F. Chaux, S. Frynas, H. Laureano, C. Salomon, G. Morata,
M. L. Auclair, M. Stephan, R. Merdes, P. Richard, M. J. On-
del-Eymin, J. C. Henry, J. Bayardon, C. Darcel, S. Jugé, C. R.
Chim. 2010, 13, 1213–1226; c) A. R. Muci, K. R. Campos,
D. A. Evans, J. Am. Chem. Soc. 1995, 117, 9075–9076; therein
it is claimed that (SP,SP)-1,2-bis[(aryl)(methyl)phosphino]eth-
anes wherein aryl = Ph, o-An, o-Tol, or α-Nap, were found to
be configurationally unstable in CDCl3, which was not the case
in our hands for aryl = Ph.
mura, H. Shimizu, T. Saito, H. Kumobayashi, Adv. Synth. Ca-
tal. 2003, 345, 180–184; q) H. Shimizu, T. Saito, H. Kumobaya-
shi, Adv. Synth. Catal. 2003, 345, 185–189; for 3H-BenzP* and
3H-QuinoxP*, see: r) Z. Zhang, K. Tamura, D. Mayama, M.
Sugiya, T. Imamoto, J. Org. Chem. 2012, 77, 4184–4188.
[18]
[19]
[20]
N. Oohara, K. Katagiri, T. Imamoto, Tetrahedron: Asymmetry
2003, 14, 2171–2175.
F. Maienza, F. Spindler, M. Thommen, B. Pugin, C. Malan, A.
Mezzetti, J. Org. Chem. 2002, 67, 5239–5249.
a) B. Zupancˇicˇ, B. Mohar, M. Stephan, Adv. Synth. Catal.
2008, 350, 2024–2032; therein, MAC was hydrogenated (1 bar
H2, MeOH, room temp., S/C = 100) within 3–4 min, leading
to 97/99%ee (S) using (RP,RP)-1,2-bis[(aryl)(phenyl)phosphi-
no]ethanes (ex. aryl = 2,3-(MeO)2-Ph and 2,3,4,5-(MeO)4-Ph);
b) B. Zupancˇicˇ, B. Mohar, M. Stephan, Tetrahedron Lett. 2009,
50, 7382–7384; therein, AS was hydrogenated (1 bar H2,
MeOH, room temp., S/C = 100) within 4 min leading to 71/
87%ee (S) using (RP,RP)-1,2-bis[(aryl)(phenyl)phosphino]eth-
anes (ex. aryl = 2,3-(MeO)2-Ph and 2,3,4,5-(MeO)4-Ph); c)
Using the P-(o-unsubstituted MeO-aryl)-containing (RP,RP)-
1,2-bis[(MeO-aryl)(phenyl)phosphino]ethanes, wherein MeO-
aryl is 3-MeO-Ph, 4-MeO-Ph, 3,4-(MeO)2-Ph, and 3,5-
(MeO)2-Ph, RhI-catalyzed hydrogenation (1 bar H2, MeOH,
room temp., S/C = 100) led respectively to 6, 17, 25, and 12%ee
(S) for MAC (100% conv. within Ͻ10 min), and to 1, 15, 27,
and 5%ee (S) for AS[20b] (100% conv. within Ͻ10 min). For
the former set, see: B. Zupancˇicˇ, PhD Thesis 2011, University
of Ljubljana, Slovenia.
a) For this work, we have tested (SP,SP)-1,2-bis[(1-naphth-
yl)(phenyl)phosphino]ethane ([α]2D2 = –19 (c = 1, CHCl3),
Ͼ99%ee, 95% purity, purchased from Sigma–Aldrich Co.) in
RhI-catalyzed hydrogenation (1 bar H2, MeOH, room temp.,
S/C = 100) of MAC [1.5 h, 100% conv., Ͼ99%ee ((R)-configu-
ration with measured [α]2D2 = –15.6 (c = 1.0, abs. MeOH))], AS
(2.4 h, 100% conv., 68%ee (R)), DMI (2.3 h, 100% conv.,
40%ee (S)), and AA (10 bar H2 in the presence of 1.1 equiv.
Cy2NH; 16 h, 100% conv., 20%ee (R)). Thus, in light of these
results, it is clear that the reported[19] MAC hydrogenation
using RhI-{(SP,SP)-1,2-bis[(1-naphthyl)(phenyl)phosphino]eth-
ane} (1.1 bar H2, MeOH, 25 °C) in 98.6%ee (S), and the au-
thors advising to apply the “quadrant rule” with caution, no
longer hold for this ligand. Moreover, and in line with our
findings, MAC was hydrogenated (3 atm H2, iPrOH, 50 °C) in
92 to Ͼ99%ee (R) by using the closely related RhI-{(SP,SP)-
1,2-bis[(aryl)(phenyl)phosphino]ethane} catalysts, wherein aryl
= o-Tol, o-Et-Ph, o-iPr-Ph, or 5,6,7,8-tetrahydronaphth-1-yl;
for this, see: b) Y. Wada, T. Imamoto, H. Tsuruta, K. Yamagu-
chi, I. D. Gridnev, Adv. Synth. Catal. 2004, 346, 777–788; RhI-
catalyzed hydrogenation (3 atm H2, iPrOH, 50 °C) of α-acet-
amidocinnamic acid (ACA) using (RP,RP)-1,2-bis[(o-An)(Cy)-
phosphino]ethane (DiCAMP) led to 64%ee (S), (RP,RP)-1,2-
bis[(o-F-Ph)(Ph)phosphino]ethane to 60%ee (S), and (RP,RP)-
1,2-bis[(o-An)(Et)phosphino]ethane to 60%ee (S). For this,
see: c) W. S. Knowles, W. C. Christopfel, K. E. Koenig, C. F.
For example, see MeO-POP vs. MeO-BIBOP in ref.[4c] and
[5]
[6]
ref.[3i]
ˇ
a) M. Stephan, D. Sterk, B. Zupancˇicˇ, B. Mohar, Org. Biomol.
Chem. 2011, 9, 5266–5271; b) B. Zupancˇicˇ, B. Mohar, M. Ste-
phan, Org. Lett. 2010, 12, 3022–3025.
Resolution of (RP/SP)-(o-diphenylphosphinophenyl)(methyl)-
phenylphosphine ((RP/SP)-L1) via PdII complexes was de-
scribed but no related use in hydrogenation was reported. For
this, see: N. Gabbitas, G. Salem, M. Sterns, A. C. Willis, J.
Chem. Soc., Dalton Trans. 1993, 3271–3276.
[7]
[8]
[9]
For catalysis study of (R,R)-1-(2,5-diMe-phospholano)-2-di-
phenylphosphino-benzene ((R,R)-Me-UCAP-Ph), see ref.[4n,4o]
a) M. Stephan, B. Modec, B. Mohar, Tetrahedron Lett. 2011,
52, 1086–1089; b) S. Jugé, M. Stephan, J. A. Laffitte, J. P.
Genêt, Tetrahedron Lett. 1990, 31, 6357–6360.
ˇ
[10]
[11]
M. Stephan, D. Sterk, B. Modec, B. Mohar, J. Org. Chem.
2007, 72, 8010–8018.
Obtaining 2-PЈ(BH3) as fine needles was favorable for its en-
antioenrichment, but the crystals were unfortunately unsuitable
for X-ray diffraction analysis.
We assume full retention of P-configuration in the decomplex-
ation step with Et2NH by analogy to reported related
compounds such as the mono-BH3-protected (RP,RP)-
DioxyBenzP*·BH3.[3j] Furthermore, decomplexation of P-ste-
reogenic phosphine-P-boranes occurs with full retention of P-
configuration, see: T. Imamoto, T. Kusumoto, N. Suzuki, K.
Sato, J. Am. Chem. Soc. 1985, 107, 5301–5303.
[21]
[12]
[13]
[14]
For the synthesis of Me-DuPHOS-P(BH3),PЈ(O), see: A. A.
Boezio, J. Pytkowicz, A. Côté, A. B. Charette, J. Am. Chem.
Soc. 2003, 125, 14260–14261.
Methanolysis attempt with (SP)-1-PЈ(O) (prepared as shown in
Scheme 2) led to acidolysis compounds [o-(diphenylphos-
phoryl)phenyl](hydroxy)(phenyl)phosphine-P-borane and [o-
(diphenylphosphoryl)phenyl](phenyl)phosphine P-oxide (prob-
ably resulting from slow BH3 loss of the latter) according to
31P NMR analysis of the crude reaction mixture before workup
(products not isolated).
I
[15]
Rh [(RP,RP)-diphosphine] hydrogenation (1 bar H2, MeOH,
room temp., S/C = 100) of MAC and AS afforded (S)-products.
With DiPAMP: 95%ee within 18 min and 84%ee within
11 min, respectively; with iPr-SMS-Phos: 99.7%ee within
4 min and 97.8%ee within 3 min, respectively; and with tBu-
SMS-Phos: 99.8%ee within 2 min and 99.4%ee within 2 min,
respectively.[3b,3c,6a]
MAC was hydrogenated (1 atm H2, EtOH, 20 °C, S/C = 100) in
51%ee (S) with RhI{(SP,SP)-1,2-bis[(methyl)phenylphosphino]-
benzene}. For this, see: D. G. Allen, S. B. Wild, D. L. Wood,
Organometallics 1986, 5, 1009–1015.
Hobbs, Adv. Chem., Ser. 1982, 325–336.
[4n]
[16]
[22]
In ref.
the catalysts were formed in situ from [Rh(cod or
nbd)(diphosphine)]BF4 in the presence of AS during hydrogen-
ation (5 bar H2, 3 h), hence the induction period toward the
formation of the active species was not eliminated. We learned
2224
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2214–2225

Olefin Hydrogenation with Rigid Diphosphines
from the corresponding author that the H2 uptake plots vs.
time actually correspond to different reaction scales and that
the H2 uptake values derive from pressure variation taking into
account H2 solubility. The slopes of the plots indicate a 10:2:1
hydrogenation rate ratio for the RhI-catalysts of Me-UCAP-
Ph, 1,2-bis(diphenylphosphino)benzene, and Me-DuPHOS,
respectively.
PHOS). For this, see also: D. Gridnev, N. Higashi, K. Asakura,
T. Imamoto, J. Am. Chem. Soc. 2000, 122, 7183–7194.
I
[27]
Interestingly, a reverse in induction in Rh -[(SP)-(1-Ad)(Me)-
PCH2CH2PR2] MAC and o-MeO-AS hydrogenations occurred
upon swapping R = Cy (or Me) with R = Ph (Figure 4).[4g]
This demonstrates that Ph2P and (alkyl)2P have different effects
on the mechanism.
[23] a) C. R. Landis, J. Halpern, J. Am. Chem. Soc. 1987, 109, 1746–
1754; b) B. Mcculloch, J. Halpern, M. R. Thompson, C. R.
Landis, Organometallics 1990, 9, 1392–1395; c) J. S. Giovan-
netti, C. M. Kelly, C. R. Landis, J. Am. Chem. Soc. 1993, 115,
4040–4057; d) S. Feldgus, C. R. Landis, Organometallics 2001,
20, 2374–2386; e) A. Ohashi, Chirality 2002, 14, 573–577; f) T.
Schmidt, W. Baumann, H. J. Drexler, A. Arrieta, D. Heller, H.
Buschmann, Organometallics 2005, 24, 3842–3848; g) P. J. Don-
oghue, P. Helquist, O. Wiest, J. Org. Chem. 2007, 72, 839–847;
h) T. Schmidt, Z. Dai, H. J. Drexler, M. Hapke, A. Preetz, D.
Heller, Chem. Asian J. 2008, 3, 1170–1180; i) I. D. Gridnev, T.
Imamoto, G. Hoge, M. Kouchi, H. Takahashi, J. Am. Chem.
Soc. 2008, 130, 2560–2572; j) I. D. Gridnev, T. Imamoto, Chem.
Commun. 2009, 7447–7464; k) P. J. Donoghue, P. Helquist, P. O.
Norrby, O. Wiest, J. Am. Chem. Soc. 2009, 131, 410–411. For
studies related to electronic effects, see: l) D. Z. Wang, Tetrahe-
dron 2005, 61, 7125–7133; m) H. C. Wu, S. A. Hamid, J. Q. Yu,
J. B. Spencer, J. Am. Chem. Soc. 2009, 131, 9604–9605; n) S.
Giri, D. Z. Wang, P. K. Chattaraj, Tetrahedron 2010, 66, 4560–
4563. For a study of enantioselectivity and hydrogenation rate
vs. conformation of five-membered chelate rings of [Rh(PP)-
(diene)]+-type, wherein PP is a stereogenic DPPE-derived li-
gand with chirality on the ethylene bridge, see ref.[4l]
[28]
[29]
The more rigid iPr-SMS-Phos was shown to be less sensitive
to H2 pressure than DiPAMP.[3c]
Some supporting information (spectra and other proofs) re-
lated to the study in ref.[4j] is unfortunately lacking and no ar-
chives were kept as we learned from the corresponding author.
31P NMR analysis at room temperature of {Rh[(RP)-
L2](MAC)}BF4 showed four diastereomeric adducts in ca.
50:25:20:5 ratio (see the Supporting Information, Figure S2).
[30]
[31]
X-ray spectroscopic analysis of {Pt[(R,R)-Me-UCAP-Ph]-
Cl2}[4n] revealed two molecules in the unit cell, wherein in one
the two phenyls have an edge-face disposition[32] and a face-
edge disposition in the second, however X-ray analysis of
{Pt[(R,R)-1-(2,5-diMe-phospholano)-2-diphenylphosphino-eth-
ane]Cl2}[4o] revealed the two phenyls to be in an edge-face dis-
position (starting from the Me-blocked quadrant). Further-
more, X-ray analysis of cis-{Pd[(SP)-L1][(S)-o-(1-(dimeth-
[7]
ylamino)ethyl)phenyl]}PF6 revealed the three phenyls in an
edge-edge-face arrangement (from below the P-Me quadrant).
Unfortunately, the X-ray analysis of cis-{Pd[(SP)-(o-An)(Ph)-
CH2CH2PPh2][(R)-1-(1-(dimethylamino)ethyl)-2-naphthyl]}-
[4i]
BF4 is poorly characterized and its cif is lacking. Indepen-
dently, a C2-symmetric diarsine complex {Pd[(RAs,RAs)-1,2-bis-
[(methyl)(phenyl)arsino]benzene][(S)-o-(1-(dimethylamino)eth-
yl)phenyl]}PF6 revealed the two phenyls to have edge disposi-
tions. For this, see: B. W. Skelton, A. H. White, J. Chem. Soc.,
Dalton Trans. 1980, 1556–1566.
[24] For example, the extent and sense of enantioselection in α-sub-
stituted acetamidoethylenes hydrogenation has been shown to
be quite sensitive to the nature of the α-substituent influencing
the substrate orientation.[3e,23d,23g,23j]
[25] Knowles early empirical “quadrant rule”[1] predicts, in many
cases, and for α-amido acids/esters in particular, the sense of
the product enantioselectivity in relation to the C2-symmetric
P-stereogenic diphosphine chirality. It states that (S)-α-amino
acid derivatives are obtained by using (RP,RP)-diphosphines
and vice versa (note that the α-substituent of the resulting
amido acid/ester should have the 3rd CIP priority number).
This is valid for P-stereogenic 1,2-bis[(Ar)(Ph)phosphino]eth-
anes such as DiPAMP, R-SMS-Phos, and 1,2-bis[(o-alkyl-
Ph)(Ph)phosphino]ethanes.
[32] For P-aryl edge/face-exposed determination in metal com-
plexes, see: a) H. Brunner, A. Winter, J. Breu, J. Organomet.
Chem. 1998, 553, 285–306; b) J. M. Brown, P. L. Evans, Tetra-
hedron 1988, 44, 4905–4916.
[33] H. J. Drexler, S. L. Zhang, A. L. Sun, A. Spannenberg, A. Arri-
eta, A. Preetz, D. Heller, Tetrahedron: Asymmetry 2004, 15,
2139–2150.
[34] D. Quintard, M. Keller, B. Breit, Synthesis 2004, 905–908.
[35] a) J. T. Reeves, Z. L. Tan, Z. X. S. Han, G. S. Li, Y. D. Zhang,
Y. B. Xu, D. C. Reeves, N. C. Gonnella, S. L. Ma, H. Lee, B. Z.
Lu, C. H. Senanayake, Angew. Chem. Int. Ed. 2012, 51, 1400–
1404; b) M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem.
1998, 63, 6084–6085.
[36] CCDC-1034830 [for (RP)-1], -1034831 [for (RP)-L2], and
-1034832 [for (SP)-2-PЈ(O)] contain the supplementary crystal-
lographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
[26] Imamoto and Gridnev[21b] reformulated Knowles “quadrant
rule” in order to fit the selectivity observed with the new intro-
duced P-stereogenic ligands: (1) the bulky P,PЈ-substituents in
the top-left and bottom-right quadrants give (R)-hydrogen-
ation products and the opposite orientation gives (S); (2) if the
P,PЈ-substituents are the same or very similar in size, more ste-
ric hindrance is given by the quasi-axial substituents: viz. (R)-α-
amino acid derivatives are obtained using (SP,SP)-1,2-bis[(bulky
alkyl)(Me)phosphino]ethanes (e.g., tBu-BisP*) and (RP,RP)-
1,2-bis[(bulky alkyl)(Me)phosphino]methanes (e.g., tBu-Mini-
Received: December 3, 2014
Published Online: February 17, 2015
Eur. J. Org. Chem. 2015, 2214–2225
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2225