Desymmetric enantioselective reduction of cyclic 1,3-diketones catalyzed by a recyclable p-chiral phosphinamide organocatalyst

Article abstract of DOI:10.1021/jacs.1c00277

The P-stereogenic phosphinamides are a structurally novel skeletal class which has not been investigated as chiral organocatalysts. However, chiral cyclic 3-hydroxy ketones are widely used as building blocks in the synthesis of natural products and bioactive compounds. However, general and practical methods for the synthesis of such chiral compounds remain underdeveloped. Herein, we demonstrate that the P-stereogenic phosphinamides are powerful organocatalysts for the desymmetric enantioselective reduction of cyclic 1,3-diketones, providing a useful method for the synthesis of chiral cyclic 3-hydroxy ketones. The protocol displays a broad substrate scope that is amenable to a series of cyclic 2,2-disubstituted five- and six-membered 1,3-diketones. The chiral cyclic 3-hydroxy ketone products bearing an all-carbon chiral quaternary center could be obtained with high enantioselectivities (up to 98% ee) and diastereoselectivities (up to 99:1 dr). Most importantly, the reactions could be practically performed on the gram scale and the catalysts could be reused without compromising the catalytic efficiency. Mechanistic studies revealed that an intermediate formed from P-stereogenic phosphinamide and catecholborane is the real catalytically active species. The results disclosed herein bode well for designing and developing other reactions using P-stereogenic phosphinamides as new organocatalysts.

Full text of DOI:10.1021/jacs.1c00277

pubs.acs.org/JACS
Article
Desymmetric Enantioselective Reduction of Cyclic 1,3-Diketones
Catalyzed by a Recyclable P‑Chiral Phosphinamide Organocatalyst
Xu-Long Qin, Ang Li, and Fu-She Han*
Cite This: J. Am. Chem. Soc. 2021, 143, 2994−3002
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The P-stereogenic phosphinamides are a structurally
novel skeletal class which has not been investigated as chiral
organocatalysts. However, chiral cyclic 3-hydroxy ketones are widely
used as building blocks in the synthesis of natural products and bioactive
compounds. However, general and practical methods for the synthesis of
such chiral compounds remain underdeveloped. Herein, we demonstrate
that the P-stereogenic phosphinamides are powerful organocatalysts for
the desymmetric enantioselective reduction of cyclic 1,3-diketones,
providing a useful method for the synthesis of chiral cyclic 3-hydroxy
ketones. The protocol displays a broad substrate scope that is amenable to a series of cyclic 2,2-disubstituted five- and six-membered
1,3-diketones. The chiral cyclic 3-hydroxy ketone products bearing an all-carbon chiral quaternary center could be obtained with
high enantioselectivities (up to 98% ee) and diastereoselectivities (up to 99:1 dr). Most importantly, the reactions could be
practically performed on the gram scale and the catalysts could be reused without compromising the catalytic efficiency. Mechanistic
studies revealed that an intermediate formed from P-stereogenic phosphinamide and catecholborane is the real catalytically active
species. The results disclosed herein bode well for designing and developing other reactions using P-stereogenic phosphinamides as
new organocatalysts.
with different skeletal classes as represented by estrone methyl
ether (2),1,2 aplysiasecosterol A (3),3 cyathane terpenoids (4
and 5),4,5 hamigerans (6 and 7),6 crotogoudin (8),7 cortistatin
A (9),8,9 and paspaline (10)10,11 have been synthesized using
the five- or six-membered chiral cyclic 3-hydroxy ketones as
the key building blocks.
INTRODUCTION
■
The chiral cyclic 3-hydroxy ketones (1 in Figure 1) with an all-
carbon quaternary chiral center at 2-position are widely used as
versatile intermediates for the enantioselective synthesis of
bioactive compounds. Specifically, numerous natural products
Desymmetric enantioselective reduction of 2,2-disubstituted
1,3-diketones has been the most frequently investigated
strategy for the synthesis of chiral cyclic 3-hydroxy ketones.
Early methods include the oxazaborolidine-1,8,12 (CBS) and
enzyme-catalyzed reduction,7,10,11,13−15 and the transition-
metal-catalyzed hydrogenation.2,9,16−18 While high to excellent
enantioselectivity could be obtained by employing these
methods, they suffer at least one of the following drawbacks,
such as limited substrate scope, low to moderate diaster-
eoselectivity, and poor stability of the catalysts toward air and
moisture. Apart from these, the CBS reduction always requires
a slow addition of the substrates or reductants which is
problematic for reliable scale up; whereas the enzymatic
reduction suffers from troublesome isolation of the products
Received: January 9, 2021
Published: February 10, 2021
Figure 1. Selected natural products synthesized from chiral cyclic 3-
hydroxy ketones.
© 2021 The Authors. Published
byAmerican Chemical Society
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002
2994

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
due to the severe foaming and emulsification, and the need to
use large volumes. In fact, we did encounter these problems
during our syntheses of diterpenoid natural products.4−6
Conscious of these disadvantages, we turned toward
developing an alternative catalyst system aimed at overcoming
these issues.
RESULTS AND DISCUSSION
Optimization of the Reaction Conditions. To identify
the potential catalysts with an appropriate skeletal type, we
initially evaluated the catalytic efficiency of four types of
skeletally different P-stereogenic catalysts 11a−14a by using
1,3-cyclopentanedione 17a (Table 1) as a model substrate
■
Chiral phosphorus compounds have been extensively used in
catalytic asymmetric reactions as ligands19−23 or organo-
catalysts.24−28 Specifically, the P-stereogenic phosphorus
compounds have been compellingly demonstrated to exhibit
a prominent chiral inducing property stemming from chirality
proximate to the catalytic center.29−32 However, extensive
investigations of P-stereogenic compounds with structural
diversity as well as novelty in the area of asymmetric catalysis
remain restricted due to the paucity of efficient methods for
their synthesis. In our recent works, we have established an
efficient and versatile synthetic platform that could access on
the gram level the various types of skeletally novel P-
stereogenic phosphinamides33−35 such as acyclic (thio)-
phosphinamides (11 and 12) and cyclic (thio)phosphinamides
(13 and 14) (Figure 2A).
a
Table 1. Evaluation of the Category of the Catalyts
a
Conditions: 17a (0.2 mmol, 36.0 mg), CB (1.1 equiv), and cat. (10
mol %) in 1 mL of anhydrous toluene at room temperature for 24 h.
The yields were isolated yields for two steps, the ee%, and dr were
determined by chiral HPLC by converting the product 18a into the
corresponding ester 19a.
because the chiral product 18a was demonstrated to be a useful
intermediate for the synthesis various diterpenoid natural
products.4−6 A brief screening of the reaction parameters
revealed that among the four different types of catalysts, the
acyclic phosphinamide 11a provided the best results by using
catecholborane (CB) as reducing reagent in toluene solvent in
term of yield (82%), enantioselectivity (24% ee), and
diastereoselectivity (93:7 dr) as determined by converting
the product 18a into the corresponding p-nitrobenzoyl ester
19a. Notably, these initial results showed that 11a displayed a
much better diastereoselectivity than the conventional CBS
reduction. Accordingly, a further screening of solvents was
carried out by employing 11a as catalyst (Table S1 of the
Supporting Information, SI). Dichloromethane (DCM)
afforded a slightly improved outcome (75% yield, 28% ee,
and 95:5 dr of 19a). Next, the effect of N-containing
compounds was examined as additives because prior literature1
for CBS reduction revealed that the amine additives influence
profoundly the reaction efficiency. After a systematic
evaluation on the structures of an array of commercially
available amines and pyridine derivatives (Table S2), following
additionally an optimization of the loading amounts of the
optimal amine (Table S3) and the reaction temperature (Table
S4), we found that the addition of diisopropylethylamine
(DIPEA) could markedly improve the reaction efficiency. 19a
could be obtained in 72% yield with 75% ee and 97:3 dr in the
presence of 10 mol % 11a and 20 mol % of DIPEA at 23 °C in
dichloromethane (Table 2, entry 1).
Figure 2. Structural skeleton of P-stereogenic phosphinamides (A)
and the tentatively proposed working model for desymmetric
reduction of diketones (B).
In earlier reports, the carbon-centered chiral phosphina-
mides were investigated as catalysts for asymmetric reduction
of ketone compounds.36−49 However, the enantioselectivity of
such catalysts was less prominent, and moreover, their use in
desymmetric enantioselective reduction of diketones has
remained underexplored. Herein, taking the advantages of
the P-stereogenic feature and the skeletal diversity of our chiral
phosphinamides, we envisioned they could serve as a potential
organocatalyst platform toward the challenging desymmetric
enantioselective reduction of 1,3-cyclic diketones. As tenta-
tively illustrated in Figure 2B, the N−PO unit of a
phosphinamide 15, which provides both Lewis basic and
acidic sites via its dipolar form 16,50,51 may coordinate with
borane and diketone to form a well-positioned transition state
TS. As a result, such a working model would not only activate
both borane and diketone via bifunctional interactions but also
impose a beneficial impact on stereochemistry owing to the
close proximity of activating center to chiral inducing center.
The successful demonstration of P-stereogenic phosphina-
mides as efficient catalysts for the desymmetric enantioselec-
tive reduction of 1,3-cyclic diketones and the investigation of
the reaction mechanism will be presented herein.
Under these preliminary conditions, we then investigated
the effect of P-stereogenic phosphinamide catalysts of type 11
on the catalytic efficiency. Various catalysts bearing electron
donating and electron-withdrawing groups on the phenyl ring
were synthesized and evaluated (Table 2). It was found that
2995
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
(entries 11−15) to reaching 92.5% ee in the presence of 1.8
equiv of CB (entry 15). These results imply that the over
reduction of 18a and its enantiomer ent-18a should proceed
via a kinetic resolution manner, during which the ent-18a was
reduced faster than 18a. In addition, we also noted that the
reduction of 17a proceeded very fast virtually when the
amounts of CB was increased to higher than 1.3 equiv. The
conversion of 17a could be completed within ca. 5 min. As a
comparison, the substrate could not be entirely reacted within
24 h when 1.1 equiv of CB was used (entries 1−11). Our later
investigation revealed the incomplete conversion of the
substrate with 1.1 equiv of CB was indeed ascribed to the
insufficient amounts of CB in the reaction system because part
of which was consumed by the over reduction of 18a and by
the formation of catalytically active species with phosphina-
mide catalyst (vide infra). These observations provided useful
clue not only for further optimization of the reaction
conditions but also for an indepth understanding of the
reaction mechanism.
Table 2. Optimization of the Catalysts
To clarify the origin of the kinetic resolution reduction, a
series of control experiments were implemented. Accordingly,
the 11k-catalyzed reduction of racemic 18a (rac-18a) was
carried out by varying the molar equivalents of CB (Table S5).
Surprisingly, only racemic 18a was recovered for all cases
(Figure 3A). In contrast, when chiral borate 22 prepared from
CB and enantiomerically pure 21 was added to the reaction
system, the recovered 18a was optically active. Specifically, as
shown in Figure 3B (see also Table S6 for detailed data), with
the increase of the amounts of CB, the recovery yields of 18a
decreased gradually (blue curve), whereas the ee% values
entry catalyst (10 mol %) CB (equiv) yield (%) ee (%)
dr
1
2
11a
11b
11c
11d
11e
11f
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.5
1.6
1.7
1.8
68
68
65
68
64
72
62
68
72
70
70
68
62
50
39
75
76
97:3
97.5:2.5
98:2
97.5:2.5
96:4
96:4
97:3
93:7
97:3
96.5:3.5
97:3
97:3
97:3
97:3
3
4
77.5
77
5
76
6
7
8
9
58.5
43.5
38
11g
11h
11i
68
10
11
12
13
14
15
11j
64.5
78
83
86
88
11k
11k
11k
11k
11k
92.5
97:3
a
Conditions: 17a (0.2 mmol, 36.0 mg), CB (1.1 equiv), catalyst (10
mol %), and DIPEA (20 mol %) in 1 mL of anhydrous DCM at room
temperature for 24 h. The yields were isolated yields of 19a for two
steps, the ee% and dr were determined by chiral HPLC by converting
18a into the corresponding ester 19a.
catalysts decorated by electron-donating substituents such as
alkyl and alkoxy groups (entries 1−5 and 9−11) afforded 19a
in >75% ee and excellent diastereoselectivity of uniformly
>24:1 dr, and catalyst 11k (entry 11) bearing bulky isopropyl
groups provided the best overall results. In comparison,
incorporation of electron-deficient groups such as CO2Me
(entry 6), CF3 (entries 7 and 8), and naphthyl (entry 10) on
any phenyl ring resulted in an apparently diminished
enantioselectivity. Although we could not discover a
satisfactory catalyst from the survey of this round, we gained
an important clue that incorporation of functional groups with
appropriate electron-donating nature and steric hindrance
would be a feasible way to improve the efficiency of the
catalysts.
An interesting point we observed during this survey was that,
while the yield of 18a was decreased with the increase of the
loading amounts of CB reductant resulting from the over
reduction of 18a, the enantioselectivity was gradually increased
Figure 3. Control experiments: reduction of racemic 18 (A), profile
of the course of the reduction in the presence of 22 (B), and
desymmetric reduction of 17a catalyzed by 22 (C).
2996
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
increased gradually to reach a peak data of 18.6% ee and then
decreased (red curve). The results of these control experiments
unambiguously affirmed that the over reduction of 18a in the
real reaction system did proceed via a kinetic resolution
manner which was exclusively catalyzed by the borate that
formed in situ from chiral 18a and CB rather than the
phosphinamide 11k.
CB catalyzed by 11k. However, to compare with DIPEA, these
additives allowed the conversion to complete within 0.5−2 h,
which would facilitate the control of reaction process. A
parallel comparison showed that the electron-deficient amine
additives resulted in a significantly diminished enantioselectiv-
ity (A6 and A7) albeit the yields were improved. The effects of
electronic nature on the outcomes are congruous with those of
catalysts (Table 2). Considering the overall performances, we
Here, a question rising from the above results is whether the
desymmetric enatioselective reduction of diketone 17a is
governed by the P-stereogenic phosphinamide or by the chiral
borate formed from chiral 18a and CB. To clarify this issue, we
performed the desymmetric reduction of 17a with the presence
of chiral borate 22 without 11k (Figure 3C). The acquisition
of almost racemic 18a (ca. 5% ee) associated with a somewhat
lowered dr (91.5:8.5) clearly indicated that the desymmetric
enantioselective reduction of diketone 17a was overwhelm-
ingly governed by the phosphinamide catalyst but not the
borate 22. Thus, on the basis of the results of a series of
control experiments, we demonstrated that the overall
outcome including the yield and stereoselectivity for the
desymmetric enantioselective reduction of diketone 17a was
virtually controlled by two processes. The major one is the P-
stereogenic phosphinamide-catalyzed desymmetric enantiose-
lective reduction of 17a, generating chiral hydroxy ketone 18a.
The minor one is the partial over reduction of the product 18a
via a kinetic resolution mediated by the borate in situ formed
from CB and chiral 18a.
i
choose Pr2NPh (A1) as the most suitable additive. Next, a
brief examination of the amounts of CB showed that 1.5 equiv
was suitable, giving 19a in 65% yield with 83% ee and 95:5 dr
in the presence of 11k catalyst (Table 4).
a
Table 4. Re-optimization of the Catalysts
Thus, on the basis of the results from Table 2 and the
control experiments, we learned that the key toward further
improving the reaction efficiency should be concentrated on
finding an appropriate combination of catalysts, additives, and
the equivalents of CB to adjust the two processes as mentioned
above. At first, an array of amines were synthesized and re-
evaluated because, as mentioned above, DIPEA may lead the
reaction to proceed so fast that it would be troublesome for
large scale operation. Through an extensive survey (Tables 3
and S7), we identified five additional amines (A1−A5) that
could provide 19a with the results almost identical to those of
DIPEA (entry 15 in Table 2) in the presence of 1.8 equiv of
a
a
Conditions: 17a (0.2 mmol, 36.0 mg), CB (1.5 equiv), catalyst (10
Table 3. Re-optimization of the Amine Additives
mol %), and iPr2NPh (20 mol %) in 1 mL of anhydrous DCM at
room temperature for 2 h. The yields were isolated yields for two
steps, the ee% and dr were determined by chiral HPLC by converting
b
the product 18a into the corresponding ester 19a. Yield obtained
from 5 mol % catalyst loading.
Finally, we reoptimized the catalysts by precisely tuning the
steric hindrance and electron-donating nature (Table 4, 11l−
11r) based on the general rule drawn from the results in Table
2. An extensive survey along this line eventually led to the
t
discovery of a satisfactory catalyst 11r bearing bulky Bu and
electron rich OiPr groups, which afforded 19a in 67% yield
with 92% ee and 97:3 dr. Further inspection revealed that the
catalyst loading of 11r could be decreased to as low as 5 mol %
without eroding the outcomes of the reaction. Thus, through
an exhaustive and patient optimization, we could established a
set of satisfactory conditions for the desymmetric enantiose-
lective reduction of the 1,3-diketone 17a, that is, 5 mol % of P-
stereogenic catalyst 11r, 20 mol % of iPr2NPh additive, and 1.5
equiv of CB reductant in CH2Cl2 at room temperature for 2 h.
Under these conditions, the corresponding p-nitrobenzoyl
a
Conditions: 17a (0.2 mmol, 36.0 mg), CB (1.8 equiv), catalyst (10
mol %), and additive (20 mol %) in 1 mL of anhydrous DCM at room
temperature for 2 h. The yields were isolated yields of 19a for two
steps, the ee% and dr were determined by chiral HPLC by converting
18a into the corresponding ester 19a.
2997
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
ester 19a was isolated in 66% yield over two steps with 92% ee
and 97:3 dr.
Substrate Scope, Application for Gram Scale Syn-
thesis, and Recyclability. To examine the compatibility of
the protocol, an array of 2,2-disubstituted 1,3-cyclopentane-
dione substrates were investigated (Table 5). The substrates
catalyzed2 and Zhou’s Pd-catalyzed17 hydrogenation, which
afforded R,R and S,R stereoisomers, respectively.
To further demonstrate the broad applicability of the
method, we extended the reaction to the six-membered 2,2-
disubstituted 1,3-diketones. Prior literature2 showed that,
compared with the five-membered 1,3-cyclodiketones, the
desymmetric enantioselective reduction of the six-membered
1,3-cyclodiketones was much more challenging and has rarely
been investigated. To the best of our knowledge, only very few
methods using transition-metal-catalyzed hydrogenative de-
symmetrization have been reported.18 Herein, by employing
our standard conditions, we examined a broad variety of 2,2-
disubstituted and 2,2,5,5-tetrasubstituted 1,3-cyclohexanedione
derivatives, as well as the 4,4-disubstituted six-membered 3,5-
cyclodiketones containing heteroatoms (23) (Table 6). The
method exhibited broad compatibility with various functional
groups, such as allyl (24a, 24b, and 24n), benzyl (24c−24k,
and 24o), alkynyl (24l), and cyano (24m) groups. High to
excellent enantioselectivities as well as diastereoselectivities
were observed for a wide range of the products, except for 24b
a
Table 5. Substrate Scope for 1,3-Cyclopentanediones
a
Table 6. Substrate Scope for 1,3-Cyclohexanediones
a
Conditions: 17 (0.5 mmol), CB (1.5 equiv), catalyst (5 mol %), and
iPr2NPh (20 mol %) in 1 mL of anhydrous DCM at room
temperature for 2 h. The yields were isolated yields, the ee% and dr
were determined by chiral HPLC (the ee% and dr for 18a, 18b, 18l,
18m, and 18n were determined by converting the products into the
corresponding p-nitrobenzoyl ester).
substituted with allyl groups were viable substrates, affording
the products in excellent enantioselectivities as well as
diastereoselectivities (18a and 18b). In addition, various
substrates with aryl substituents were also well tolerated.
Satisfactory yields as well as high to excellent enantio- and
diastereoselectivities were obtained no matter whether the aryl
group was undecorated or decorated by electron-donating as
well as electron-withdrawing groups at different position on the
benzene ring (18c−18k). Importantly, the alkynyl and cyano
functionalities were also compatible (18l−18n). Finally, the
method could also be extended to the disubstituted
indandiones, giving the desired products in high yields and
excellent enantio- and diastereoselectivities (18o and 18p).
Thus, we have successfully developed an efficient and generally
applicable organocatalyst for the desymmetric enantioselective
reduction of 2,2-disubstituted 1,3-cyclopentanediones. The
absolute configuration of the products was S,S as determined
from the known compounds 18a and 18b. Moreover, this
protocol also provides a complementary way for the efficient
and versatile synthesis of the diastereoisomers to Zhang’s Ir-
a
Conditions: 23 (0.5 mmol), CB (1.5 equiv), catalyst (5 mol %), and
iPr2NPh (20 mol %) in 1 mL of anhydrous DCM at room
temperature for 2 h. The yields were isolated yields, and the ee% and
dr were determined by chiral HPLC (the ee% and dr for 24a, 24b,
24l, 24m, and 24n were determined by converting the products into
the corresponding p-nitrobenzoyl ester).
2998
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
which was obtained with a slightly diminished ee (81%) and
moderate dr (87:13). It is worth mentioning that, as an
individual example reported in prior literature,2 compound 24c
was also synthesized through Ir-catalyzed desymmetric
enantioselective hydrogenation with 82% ee and 6:1 dr. In
comparison, a much better outcome with up to 93% ee and
24:1 dr was afforded by employing our P-stereogenic
organocatalyst. Also noteworthy, substrates containing heter-
oatoms were also viable, giving the desired products in
satisfactory yields and high stereoselectivity (24p and 24q).
These overall results demonstrated that the protocol could also
be extensively used for the desymmetric enantioselective
reduction of six-membered cyclodiketones. The absolute
configuration of the products was S,S as assigned based on
the X-ray single crystal diffraction of 24i (CCDC 2044061).
Having demonstrated the substrate scope of the method, we
then investigated the practicality by conducting gram scale
syntheses and examining the recyclability of recovered catalyst,
respectively (Figure 4). Happily, the reactions could be reliably
using 11r as catalyst, iPr2NPh (A1) as additive, catecholborane
(CB) as reductant, and 2-methyl-2-(4-methylbenzyl)-1,3-
cyclopentanedione 17d as substrate in CD2Cl2 solvent at
ambient temperature. For a 1:1 mixture of catalyst 11r and CB,
1
the H (Figure S1) and 31P NMR (Figure 5) spectra showed
Figure 5. Mechanistic investigation: The 31P NMR spectra in CD2Cl2
(upper) and the structures of possible intermediates (lower).
that resonance signals of the proton of NH and phosphorus for
catalyst 11r shifted apparently toward the downfield region
from 4.63 to 5.16 ppm (Figure S1, A vs B) and 26.4 to 32.3
ppm (Figure 5, A vs B), respectively, to compare with those of
the free catalyst 11r. Associated with these changes, the peak of
boron for CB in the 11B NMR spectrum became broad and
shifted upfield from ca. 29.0 to 15.0 ppm (Figure S2, A vs B).
These observations imply that a complex 25 should be formed
from the Lewis basic 11r and acidic CB. Control experiments
revealed that this complex should not be the catalytically active
species since the reduction reaction proceeded very slowly
under this system without amine additive.
Figure 4. Gram scale reaction (A) and recyclability (B).
performed on the gram scale and exhibited excellent
reproducibility as seen from the two representative reactions
(18a and 18d) (Figure 4A). Notably, unlike the CBS reduction
which usually requires slow addition of the reductants or
substrates, the reaction catalyzed by P-stereogenic phosphina-
mide was carried out by adding all the substances in one
portion, allowing for a much more convenient operation. Most
importantly, it was demonstrated that the catalyst could be
readily recovered in >90% yield by simple column chromatog-
raphy and reused with entirely retained catalytic efficiency to
compare with the freshly prepared catalyst (Figure 4B).
Mechanistic Study. To understand the reaction mecha-
nism of the protocol, we sought to identify the possible
reaction intermediates by monitoring the detailed reaction
In stark contrast, remarkable changes were observed when
i
11r, CB, and Pr2NPh (A1) were mixed in a 1:1:1 ratio.
Specifically, the proton of NH for 11r almost disappeared and
the protons of CH on both the isopropyl groups for 11r and A1
1
were split into two sets of peaks as seen from the H NMR
spectrum (Figure S1, A−C vs D and E); in the 31P NMR
spectrum, two peaks were observed, one of which shifted
upfield to 27.3 ppm and the other moved largely downfield to
1
1
course using H, 31P, and 11B NMR spectroscopies. More to
50.3 ppm (Figure 5, C vs B); in line with the changes of H
the point, this study would also set an important basis for
designing other asymmetric reactions catalyzed by P-stereo-
genic catalysts. The NMR monitoring studies were carried out
and 31P NMR spectra, two peaks at 14.2 and 8.1 ppm appeared
in the 11B NMR spectrum (Figure S2, B vs C). The
spectroscopic data suggest that a couple of species correspond-
2999
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
ing to boraphosphinimidate 26 and boraphosphonium 27
should be formed from 11r and CB by extrusion of a molecule
of H2 under the assistance of A1 base, which may then interact
with A1 through Lewis base-acid coordination. More
interestingly, upon adding 1 equiv of 1,3-cyclopentanedione
17d to the 1:1:1 mixture composed of 11r/CB/A1, forming a
1:1:1:1 mixture of 11r/CB/A1/17d, the peak at 50.3 ppm
corresponding to 27 decreased gradually with the prolongation
of standing time and eventually disappeared (Figure 5, D−F).
Correlated with this variation, the intensity of the peak at 27.3
ppm increased gradually and shifted very slightly upfield to
26.8 ppm. These spectroscopic variations should result
rationally from the coordination effect between 26 and
diketone 17d, which in turn may act as the driving force to
promote the interconversion of boraphosphinimidate 27 into
26.
The above assumption was strongly supported by the results
obtained from the reduction of diketone 17d with CB.
Namely, upon adding the 2nd equiv of CB to the above 1:1:1:1
mixture composed of 11r/CB/A1/17d, providing a mixture of
11r/CB/A1/17d in 1:2:1:1 ratio, the peak at 50.33 ppm
corresponding to 27 reappeared rapidly within 30 min (Figure
5, G), indicating the reducing reaction proceeded quickly;
meanwhile, in the 11B NMR spectrum (Figure S2, E), a new
peak at 23.1 ppm was observed which could be assigned to the
borate 28 produced from the product 18d and CB. The
spectroscopic data clearly indicated that accompanied by the
consumption of diketone 17d, boraphosphinimidate 26 was
regenerated which then interconverted into a pair of
equilibrium intermediates 26 and 27, and ultimately,
completing the first catalytic cycle. The NMR monitoring
experiments revealed that 26 and 27 were catalytically active
and promoted the desymmetric reduction of diketone 17d.
Coincidently with the profile of the NMR changes as seen in
the first catalytic cycle, subsequent addition of the second
equivalent of diketone 17d to the above reaction mixture (the
final ratio of 11r/CB/A1/17d = 1:2:1:2) again resulted in a
gradual decrease of the 31P peak at 50.3 ppm corresponding to
27 until it almost disappeared; synchronously, a gradual
increase of the intensity associated with a slight upfield shift of
the peak corresponding to 26 at 27.3 ppm was observed
(Figure 5, H−J). These observations suggested that the
reaction entered into the next catalytic cycle upon the addition
of the 2nd equiv of diketone 17d. In the end, product 18d was
obtained from the NMR monitoring experiments and
displayed high enantioselectivity with 84% ee, indicating
apparently that the reaction proceeded in a highly stereo-
selective manner during the course of NMR monitoring.
Thus, on the basis of the extensive NMR tracking
experiments, we could propose a plausible reaction mechanism
(Figure 6). The P-stereogenic phosphinamide catalyst forms
complex A with CB which is then transformed to
boraphosphinimidate B by extrusion of a molecule of H2
under the assistance of the iPr2NPh base. Tautomerization of B
produces boraphosphonium C which exists with B as a couple
of equilibrium intermediates. Acting as a bifunctional catalyti-
cally active species, the intermediate B then coordinates with
diketone 17 or 23 to preferentially form the sterically favored
D over the disfavored E. Successively, the Lewis basic site of D
bonds with Lewis acidic CB to afford the complex F. As a
result, both diketone and CB are activated through
coordination interactions with B. Finally, face-selective hydride
transfer from CB to diketone regenerates the boraphosphini-
Figure 6. Proposed reaction mechanism.
midate B and delivers borate G. Finally, B brings the
desymmetric enantioselective reduction into the next cycle.
As has been demonstrated by the control experiments and
NMR monitoring experiments, the stereochemistry of the
reaction is predominantly controlled by this catalytic cycle.
Alternatively, a small part of borate G is over-reduced by a self-
mediated kinetic resolution reduction, which makes a minor,
yet non-negligible contribution to improve the stereoselectivity
of the final products 18 and 24.
CONCLUSIONS
■
In summary, through a patient and thorough investigation, we
have developed an efficient and practical protocol for the
desymmetric enantioselective reduction of 2,2-disubstituted
1,3-cyclodiketones using the P-stereogenic phoshphinamides
as a type of novel organocatalyst. This protocol is highlighted
by a number of indisputable advantages compared with the
reported methods as listed below: (i) it displays a wide
substrate scope that works well with both five- and six-
membered substrates. High to excellent enantio- and
diastereoselectivity are obtained for a series of substrates of
both types; (ii) the reaction could be reliably performed on the
gram scale and easily operated by adding all substances in one
portion; and (iii) most importantly, the catalysts could be
readily recovered and reused without losing the catalytic
efficiency. These advantages render the current protocol to
have great potential for practical application. Also significantly,
the identification of the intermediates B and C derived from P-
stereogenic phoshpinamide and catecholborane (CB) as the
bifunctional catalytic active intermediates should have further
implications beyond this work for the de novo design of other
relevant asymmetric reactions. These works are currently
underway in our laboratory.
3000
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(3) Lu, Z.; Zhang, X.; Guo, Z.; Chen, Y.; Mu, T.; Li, A. Total
Synthesis of Aplysiasecosterol A. J. Am. Chem. Soc. 2018, 140, 9211.
(4) Wu, G.-J.; Zhang, Y.-H.; Tan, D.-X.; Han, F.-S. Total Synthesis
of Cyrneines A-B and Glaucopine C. Nat. Commun. 2018, 9, 2148.
(5) Wu, G.-J.; Zhang, Y.-H.; Tan, D.-X.; He, L.; Cao, B.-C.; He, Y.-
P.; Han, F.-S. Synthetic Studies on Enantioselective Total Synthesis of
Cyathane Diterpenoids: Cyrneines A and B, Glaucopine C and
(+)-Allocyathin B2. J. Org. Chem. 2019, 84, 3223.
(6) Cao, B.-C.; Wu, G.-J.; Yu, F.; He, Y.-P.; Han, F.-S. A Total
Synthesis of (−)-Hamigeran B and (−)-4-Bromohamigeran B. Org.
Lett. 2018, 20, 3687.
(7) Breitler, S.; Carreira, E. M. Total Synthesis of (+)-Crotogoudin.
Angew. Chem., Int. Ed. 2013, 52, 11168.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c00277.
■
sı
Detailed procedures for the optimization of reaction
conditions; the preparation of amine additives, P-
stereogenic catalysts, and 1,3-diketone substrates; the
desymmetric enantioselective reduction of 1,3-dike-
tones; chemical compound information including
NMR and HRMS data and copies of NMR spectra for
all compounds; 1H and 11B NMR spectra for
mechanistic study; single X-ray crystal data of 24i
(CIF); and HPLC charts of chiral catalysts and cyclic 3-
hydroxy ketone products (PDF)
(8) Liu, L. L.; Chiu, P. An Expeditious Asymmetric Synthesis of the
Pentacyclic Core of the Cortistatins by an Intramolecular (4 + 3)
Cycloaddition. Chem. Commun. 2011, 47, 3416.
(9) Kuang, L.; Liu, L. L.; Chiu, P. Formal Total Synthesis of
(+)-Crotistatins A and J. Chem. - Eur. J. 2015, 21, 14287.
(10) Sharpe, R. J.; Johnson, J. S. A Global and Local
Desymmetrization Approach to the Synthesis of Steroidal Alkaloids:
Stereocontrolled Total Synthesis of Paspaline. J. Am. Chem. Soc. 2015,
137, 4968.
(11) Sharpe, R. J.; Johnson, J. S. Asymmetric Total Synthesis of the
Indole Diterpene Alkaloid Paspaline. J. Org. Chem. 2015, 80, 9740.
(12) Shimizu, M.; Yamada, S.; Fujita, Y.; Kobayashi, F. Highly
Stereocontrolled Reduction of 1,3-Cyclopentanediones Using Ox-
azaborolidine-BH3. Tetrahedron: Asymmetry 2000, 11, 3883.
(13) Csuk, R.; Glanzer, B. I. Baker’s Yeast Mediated Trans-
formations in Organic Chemistry. Chem. Rev. 1991, 91, 49.
(14) Brooks, D. W.; Mazdiyasni, H.; Grothaus, P. G. Asymmetric
Microbial Reduction of Prochiral 2,2-Disubstituted Cycloalkane-
diones. J. Org. Chem. 1987, 52, 3223.
Accession Codes
CCDC 2044061 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, U.K.; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
■
Fu-She Han − CAS Key Lab of High-Performance Synthetic
Rubber and its Composite Materials, Changchun Institute of
Applied Chemistry, Chinese Academy of Sciences, Changchun,
Jilin 130022, P. R. China; University of Science and
Technology of China, Hefei, Anhui 230026, P. R. China;
orcid.org/0000-0003-0456-7461; Email: fshan@
ciac.ac.cn
(15) Baumgartner, C.; Ma, S.; Liu, Q.; Stoltz, B. M. Efforts toward
Rapid Construction of the Cortistatin A Carbocyclic Core via Enyne-
ene Metathesis. Org. Biomol. Chem. 2010, 8, 2915.
(16) Shi, B.; Merten, S.; Wong, D. K. Y.; Chu, J. C. K.; Liu, L. L.;
Authors
̈
Lam, S. K.; Jager, A.; Wong, W.-T.; Chiu, P.; Metz, P. The Rhodium-
Xu-Long Qin − CAS Key Lab of High-Performance Synthetic
Rubber and its Composite Materials, Changchun Institute of
Applied Chemistry, Chinese Academy of Sciences, Changchun,
Jilin 130022, P. R. China; University of Science and
Technology of China, Hefei, Anhui 230026, P. R. China
Ang Li − CAS Key Lab of High-Performance Synthetic Rubber
and its Composite Materials, Changchun Institute of Applied
Chemistry, Chinese Academy of Sciences, Changchun, Jilin
130022, P. R. China; University of Science and Technology of
China, Hefei, Anhui 230026, P. R. China
Catalyzed Carbene Cyclization Cycloaddition Cascade Reaction of
Vinylsulfonates. Adv. Synth. Catal. 2009, 351, 3128.
(17) Yu, C.-B.; Wang, H.-D.; Song, B.; Shen, H.-Q.; Fan, H.-J.;
Zhou, Y.-G. Reversal of Diastereoselectivity in Palladium-Arene
Interaction Directed Hydrogenative Desymmetrization of 1,3-
Diketones. Sci. China: Chem. 2020, 63, 215.
(18) Yu, C.-B.; Song, B.; Chen, M.-W.; Shen, H.-Q.; Zhou, Y.-G.
Construction of Multiple-Substituted Chiral Cyclohexanes through
Hydrogenative Desymmetrization of 2,2,5-Trisubstituted 1,3-Cyclo-
hexanediones. Org. Lett. 2019, 21, 9401.
(19) Tang, W.; Zhang, X. New Chiral Phosphorus Ligands for
Enantioselective Hydrogenation. Chem. Rev. 2003, 103, 3029.
(20) Wei, Y.; Shi, M. Multifunctional Chiral Phosphine Organo-
catalysts in Catalytic Asymmetric Morita-Baylis-Hillman and Related
Reactions. Acc. Chem. Res. 2010, 43, 1005.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c00277
Notes
(21) van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Claver, C.; Pamies,
O.; Diéguez, M. Phosphite-Containing Ligands for Asymmetric
Catalysis. Chem. Rev. 2011, 111, 2077.
(22) Xie, J.-H.; Zhu, S.-F.; Zhou, Q.-L. Recent Advances in
Transition Metal-Catalyzed Enantioselective Hydrogenation of
Unprotected Enamines. Chem. Soc. Rev. 2012, 41, 4126.
(23) Xie, J.-H.; Zhou, Q.-L. Magical Chiral Spiro Ligands. Huaxue
Xuebao 2014, 72, 778.
(24) You, S.-L.; Cai, Q.; Zeng, M. Chiral Brønsted Acid Catalyzed
Friedel-Crafts Alkylation Reactions. Chem. Soc. Rev. 2009, 38, 2190.
(25) Mahlau, M.; List, B. Asymmetric Counteranion-Directed
Catalysis: Concept, Definition, and Applications. Angew. Chem., Int.
Ed. 2013, 52, 518.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from National Natural Science Foundation
of China (22071235) is acknowledged.
REFERENCES
■
(1) Yeung, Y.-Y.; Chein, R.-J.; Corey, E. J. Conversion of Torgov’s
Synthesis of Estrone into a Highly Enantioselective and Efficient
Process. J. Am. Chem. Soc. 2007, 129, 10346.
(2) Gong, Q.; Wen, J.; Zhang, X. Desymmetrization of Cyclic 1,3-
Diketones via Ir-Catalyzed Hydrogenation: an Efficient Approach to
Cyclic Hydroxy Ketones with a Chiral Quaternary Carbon. Chem. Sci.
2019, 10, 6350.
(26) Brak, K.; Jacobsen, E. N. Asymmetric Ion-Pairing Catalysis.
Angew. Chem., Int. Ed. 2013, 52, 534.
3001
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(27) Phipps, R. J.; Hamilton, G. L.; Toste, F. D. The Progression of
Chiral Anions from Concepts to Applications in Asymmetric
Catalysis. Nat. Chem. 2012, 4, 603.
(28) Parmar, D.; Sugiono, E.; Raja, S.; Rueping, M. Complete Field
Guide to Asymmetric BINOL-Phosphate Derived Brønsted Acid and
Metal Catalysis: History and Classification by Mode of Activation;
Brønsted Acidity, Hydrogen Bonding, Ion Pairing, and Metal
Phosphates. Chem. Rev. 2014, 114, 9047.
(46) Li, K.-Y.; Zhou, Z.-H.; Wang, L.-X.; Zhao, G.-F.; Choi, M. C.
K.; Zhou, Q.-L.; Tang, C.-C. Asymmetric Borane Reduction of
Prochiral Ketones Catalyzed by Phosphinamides Prepared from L-
Serine. Heteroat. Chem. 2003, 14, 288.
(47) Basavaiah, D.; Reddy, G. J.; Rao, K. V. Toward Effective Chiral
Catalysts Containing the N-P=O Structural Framework for the
Borane-Mediated Asymmetric Reduction of Prochiral Ketones.
Tetrahedron: Asymmetry 2004, 15, 1881.
(48) Basavaiah, D.; Reddy, G. J.; Chandrashekar, V. A New Chiral
Catalytic Source with an N-P=O Structural Framework Containing a
Proximal Hydroxyl Group for the Borane-Mediated Asymmetric
Reduction of Prochiral Ketones. Tetrahedron: Asymmetry 2004, 15,
47.
(49) Du, D.-M.; Fang, T.; Xu, J.; Zhang, S.-W. Structurally Well-
Defined, Recoverable C3-Symmetric Tris(β-hydroxy phosphoramide)-
Catalyzed Enantioselective Borane Reduction of Ketones. Org. Lett.
2006, 8, 1327.
(50) Denmark, S. E.; Dorow, R. L. Solution and Solid-State
Structure of a Phosphorus-Stabilized Carbanion. J. Am. Chem. Soc.
1990, 112, 864.
(51) Denmark, S. E.; Miller, P. C.; Wilson, S. R. Configuration,
Conformation, and Colligative Properties of a Phosphorus-Stabilized
Anion. J. Am. Chem. Soc. 1991, 113, 1468.
(29) Crépy, K. V. L.; Imamoto, T. New P-Chirogenic Phosphine
Ligands and Their Use in Catalytic Asymmetric Reactions. Top. Curr.
Chem. 2003, 229, 1.
(30) Imamoto, T.; Sugita, K.; Yoshida, K. An Air-Stable P-Chiral
Phosphine Ligand for Highly Enantioselective Transition-Metal-
Catalyzed Reactions. J. Am. Chem. Soc. 2005, 127, 11934.
(31) Fu, W.; Tang, W. Chiral Monophosphorus Ligands for
Asymmetric Catalytic Reactions. ACS Catal. 2016, 6, 4814.
(32) Xu, G.; Senanayake, C. H.; Tang, W. P-Chiral Phosphorus
Ligands Based on a 2,3-Dihydrobenzo[d][1,3]oxaphosphole Motif for
Asymmetric Catalysis. Acc. Chem. Res. 2019, 52, 1101.
(33) Du, Z.-J.; Guan, J.; Wu, G.-J.; Xu, P.; Gao, L.-X.; Han, F.-S.
Pd(II)-Catalyzed Enantioselective Synthesis of P-Stereogenic Phos-
phinamides via Desymmetric C-H Arylation. J. Am. Chem. Soc. 2015,
137, 632.
(34) Chen, Y.-H.; Qin, X.-L.; Han, F.-S. Efficient Synthesis of Cyclic
P-Stereogenic Phosphinamides from Acyclic Chiral Precursors via
Radical Oxidative Intramolecular Aryl C-H Phosphinamidation.
Chem. Commun. 2017, 53, 5826.
(35) Chen, Y.-H.; Qin, X.-L.; Guan, J.; Du, Z.-J.; Han, F.-S. Pd-
Catalyzed Enantioselective C-H Arylation of Phosphinamides with
Boronic Acids for the Synthesis of P-Stereogenic Compounds.
Tetrahedron: Asymmetry 2017, 28, 522.
(36) Burns, B.; Studley, J. R.; Wills, M. New Catalysts Containing an
N-P=O Structural Unit for the Asymmetric Reduction of Ketones.
Tetrahedron Lett. 1993, 34, 7105.
(37) Burns, B.; King, N. P.; Studley, J. R.; Tye, H.; Wills, M.
Stereoelectronic Requirements for a New Class of Asymmetric
Ketone Reduction Catalysts Containing an N-P=O Structural Unit.
Tetrahedron: Asymmetry 1994, 5, 801.
(38) Gamble, M. P.; Studley, J. R.; Wills, M. New Chiral
Phosphinamide Catalysts for Highly Enantioselective Reduction of
Ketones. Tetrahedron Lett. 1996, 37, 2853.
(39) Gamble, M. P.; Studley, J. R.; Wills, M. Design, Synthesis and
Applications of a Ketone Reduction Catalyst Containing a
Phosphinamide Combined with a Dioxaborolidine Unit. Tetrahedron:
Asymmetry 1996, 7, 3071.
(40) Hulst, R.; Heres, H.; Peper, N. C.M.W.; Kellogg, R. M.
Synthesis and Application of New Chiral Ligands for the Asymmetric
Borane Reduction of Prochiral Ketones. Tetrahedron: Asymmetry
1996, 7, 1373.
(41) Burns, B.; Gamble, M. P.; Simm, A. R. C.; Studley, J. R.; Alcock,
N. W.; Wills, M. Phosphinamides Catalysts Containing a Stereogenic
Phosphorus Atom for the Asymmetric Reduction of Ketones by
Borane. Tetrahedron: Asymmetry 1997, 8, 73.
(42) Burns, B.; King, N. P.; Tye, H.; Studley, J. R.; Gamble, M.;
Wills, M. Chiral Phosphinamides: New Catalysts for the Asymmetric
Reduction of Ketones by Borane. J. Chem. Soc., Perkin Trans. 1 1998,
1027.
(43) Gamble, M. P.; Smith, A. R. C.; Wills, M. A Novel
Phosphinamide Catalyst for the Asymmetric Reduction of Ketones
by Borane. J. Org. Chem. 1998, 63, 6068.
(44) Basavaiah, D.; Reddy, G. J.; Chandrashekar, V. A Novel and
Effective Chiral Phosphoramide Catalyst for the Borane-Mediated
Asymmetric Reduction of Prochiral α-Halo Ketones. Tetrahedron:
Asymmetry 2001, 12, 685.
(45) Li, K.; Zhou, Z.; Wang, L.; Chen, Q.; Zhao, G.; Zhou, Q.;
Tang, C. Asymmetric Carbonyl Reduction with Borane Catalyzed by
Chiral Phosphinamides Derived from L-Amino Acid. Tetrahedron:
Asymmetry 2003, 14, 95.
3002
https://dx.doi.org/10.1021/jacs.1c00277
J. Am. Chem. Soc. 2021, 143, 2994−3002