Lipophilic chiral cobalt (III)complexes of hexaamine ligands: Efficacies as enantioselective hydrogen bond donor catalysts

Article abstract of DOI:10.1016/j.mcat.2019.03.018

Four known chiral enantiopure octahedral cobalt(III)trichloride salts with aliphatic hexaamine ligands are converted to new CH₂Cl₂ soluble tris(tetraarylborate)or 3BAr_f^? salts (BAr_f = B(3,5-C₆

Full text of DOI:10.1016/j.mcat.2019.03.018

Molecular Catalysis 473 (2019) 110360
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Lipophilic chiral cobalt (III) complexes of hexaamine ligands: Efficacies as
enantioselective hydrogen bond donor catalysts
⁎
William J. Maximuck, John A. Gladysz
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, TX 77842-3012, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
Four known chiral enantiopure octahedral cobalt(III) trichloride salts with aliphatic hexaamine ligands are
−
Sepulchrate complexes
Sarcophagine complexes
Cobalt
Hydrogen bonding
Enantioselective catalysis
Hexaamine ligands
Werner complexes
converted to new CH Cl soluble tris(tetraarylborate) or 3BArf salts (BAr = B(3,5-C H (CF ) ) in biphasic
2
2
f
6
3
3 2 4
reactions. These include sepulchrate and sarcophagine complexes in which the hexaamine ligands are bicyclic (Z
CH NHCH CH NHCH Z, Z=N, CX), as well as a truncated sarcophagine in which a missing carbon vertex
renders the ligand acyclic and tripodal, with three terminal NH groups that define a trigonal C symmetric face
CH C(CH NHCH CH NH ). In the presence of tertiary amine bases, these are effective catalysts for two
(
2
2
2
2 3
)
2
3
(
3
2
2
2
2 3
)
Michael type carbon-carbon bond forming reactions and a related carbon-nitrogen bond forming reaction.
However, only the last complex affords significant enantioselectivities (30–57% ee). Rationales, and directions
for future catalyst optimization, are proposed.
Introduction
Many of these complexes can be synthesized from [Co(en)
3
]3+ 3X−
using inexpensive reagents such as formaldehyde, ammonium hydro-
xide, and nitromethane [15,18,20,21]. Like their precursor, they pos-
sess metal centered chirality and idealized D symmetry, as represented
3
The last twenty years have seen the extensive development of chiral
hydrogen bond donor catalysts for enantioselective organic synthesis
[
1]. Nearly all of these were metal-free (organic) species prior to our
by the partial structures in Fig. 2. The Δ enantiomer can be viewed as a
right handed helix, as the chelates extend in a clockwise direction when
bridging from the front to the rear trigonal coordination planes; for the
Λ enantiomer, the chelates extend in a counter-clockwise direction
[22]. In most cases, the enantiomers can be resolved by fractional
crystallization of their diastereomeric tartrate salts [14,23–25]. Alter-
natively, enantiopure adducts can be accessed from enantiopure [Co
report of the use of a lipophilic salt of a classical enantiopure Werner
trication, Λ-[Co(en)
B(3,5-C (CF ) or the enantiomer [2]. This strong NH hydrogen
bond donor [3] is depicted in Fig. 1. Since then, we have continued to
develop a range of cobalt(III) tris(diprimary 1,2-diamine) catalysts
3
+
−
f
3
]
3BAr
f
(I; en = 1,2-ethylenediamine; BAr =
6
H
3
3 2 4
) )
3
+
−
−
f
[
4–7] such as Λ-[Co((S,S)-dpen)
3
]
2Cl BAr
(II; dpen = 1,2-di-
3+
−
phenylethylenediamine) [7] and the bifunctional species III [5,8], as
well as much different systems such as the ruthenium adducts IV, which
feature NH hydrogen bond donors remote from the metal [9,10]. Others
(en)
3
]
3X [15,20,21].
In this paper, we report that cobalt(III) complexes of the types VIa,b
can in fact serve as hydrogen bond donor catalysts for various carbon-
carbon and carbon-nitrogen bond forming reactions. The specific sys-
tems investigated are depicted in Fig. 3. However, the enantioselec-
tivities are much lower than those obtained with II and III. Nonetheless,
close relatives yield significant enantioselectivities, and the data pro-
vide insight that should aid future catalyst design and optimization.
[
11–13], especially the Meggers group [13], have explored com-
plementary motifs, usually with NH donor groups remote from the
metal as exemplified by V.
Since Werner's time, the coordination chemistry of cobalt(III) has
expanded beyond diprimary 1,2-diamines to include a number of
macrobicyclic hexadentate ligands featuring secondary amine moieties.
Prominent classes include the sepulchrates (sep) and sarcophagines
Results
(
sar), as illustrated by VIa and VIb in Fig. 1 [14]. These ligands, which
have been applied to a variety of other transition metal ions [14–18],
confer inherently greater thermodynamic stabilities as compared to
hexa(monoamine) or tris(1,2-diamine) analogs [14,19].
Catalyst synthesis
3 2
The enantiopure chloride salts Λ-[Co((NH )
sar)]5+ 5Cl [18], Λ-
−
⁎
Corresponding author.
E-mail address: gladysz@mail.chem.tamu.edu (J.A. Gladysz).
https://doi.org/10.1016/j.mcat.2019.03.018
Received 22 February 2019; Received in revised form 11 March 2019; Accepted 12 March 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
Fig. 1. Some types of chiral enantiopure metal-containing hydrogen bond donors.
Fig. 2. Λ/Δ configurations of metal stereocenters in octahedral complexes with three chelating ligands.
f
Fig. 3. Enantiopure catalysts investigated (X = BAr ) and their precursors (X = Cl).
2

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
Fig. 4. Syntheses of the cobalt(III) catalysts shown in Fig. 3. The photograph in the left panel depicts the reactants in the above equation (L = sep), that in the middle
the stirred reaction, and that in the right the products.
−
3+
−
3+
−
[
[
Co(sep)]3+ 3Cl , Δ-[Co(sep)]
3Cl [26], Λ-[Co(sar)]
3Cl
in the presence of a tertiary amine base. This reaction could be cata-
lyzed with high enantioselectivities by the tris(1,2-diphenylethylene-
3
+
−
15,27], and Λ-[Co(sen)]
depicted in Fig. 3, were prepared according to previously reported
3Cl [23,28], the cations of which are
−
−
diamine) complex II (Fig. 1) and related salts with 2BF
4
BAr
anion sets [4]. As summarized in entries 2–6 for additions in
in the presence of Et N, all of the cobalt(III) cage complexes
f
and
3
+
−
−
procedures. The hexadentate ligand in Λ-[Co(sen)]
Hence, the complex is of lower symmetry than the others (C
and possesses a trigonal coordination plane with three primary amine
donor groups (−NH ).
These complexes were dissolved in water, and in the case of the
3Cl is acyclic.
3BF
CD
4
3
vs. D ),
3
2
Cl
2
3
were effective catalysts. Over the course of 2 h at 0 °C, > 99% conver-
sions were realized, as assayed by NMR versus an internal standard.
Unfortunately, significant enantioselectivities were not achieved, as
assayed by HPLC (see supporting information). The top performer was
2
5
+
−
pentacation Λ-[Co((NH
3
)
2
sar)]
5Cl , NaOH was added to deproto-
Cl
(3.0 equiv) were added. The characteristic or-
ange color of the cobalt(III) cations transferred from the upper aqueous
to the lower CH Cl phases, as depicted in Fig. 4.
Workups of the CH Cl phases gave the corresponding tris(tetra-
salts in 89–98% yields as bright orange solids.
3+
−
nate the alkylammonium groups and generate a trication. Then CH
2
2
Λ-[Co(sen)] 3BAr
f
with only a 10% ee (entry 6). The ee values for
+
−
solutions of Na BAr
f
the hexadentate secondary amine complexes (entries 2–5) were also
3+
−
f
surpassed by that of Λ-[Co(en)
3
]
3BAr
(I; entry 1). The dominant
2
2
product configurations were determined by HPLC elution orders es-
tablished previously, always with the same type of chiral column
2
2
−
arylborate) or 3BAr
f
[4,30,31]. The catalyst in entry 2 features two NH
the experiment was repeated without added Et N, no reaction occurred.
No background reactions were detected in the absence of catalyst.
Analogous experiments were also conducted in CD CN using N-
methylmorpholine as the base. Reactions became slower, but the ee
2
groups, but when
1
These contained 4–15 water molecules per cobalt, as estimated from H
NMR and microanalytical data. The water molecules are accounted for
in all the molar quantities and yields reported herein, but for simplicity
are only represented in the experimental section. To our knowledge,
these represent the first lipophilic salts of the trications in Fig. 3.
3
3
3+
−
f
3+
−
values obtained with Λ-[Co(en)
3
]
3BAr
and Λ-[Co(sen)] 3BAr
f
1
13
1
The H and C{ H} NMR spectra of the new salts were recorded in
acetone-d , and data for all CH and carbon signals closely matched
those reported for the trichloride salts in D
were found for the NH and NH protons of Λ-[Co((NH
, and the NH and diastereotopic NHH' protons of Λ-[Co(sen)]
increased significantly (Fig. 5, entries 7 and 8). Those for the other
catalysts remained very low. Hence, the presence of a trigonal co-
ordination plane with three primary amine donor groups appears to
enhance enantioselectivities.
6
n
1
2
O. Separate H NMR signals
+
2
2 2
)
sar)]3
−
3+
3
3
BAr
BAr
f
f
Fig. 6 depicts a different type of base promoted malonate addition
−
3+
−
f
. However, these peaks were not observed with the trichloride
O, presumably due to H/D exchange. All of the new salts were
soluble in other moderately polar organic solvents, such as acetonitrile.
Analogous anion metatheses were attempted with the trichloride
that we studied earlier with Δ-[Co(en)
3
]
3BAr
(I), but could never
salts in D
2
achieve an ee value of over 33% [2]. Under the conditions of Fig. 6, the
ee with I was 20% (entry 1). All of the cobalt(III) cage complexes were
competent catalysts (entries 2–6), but reactions were slower than those
in Fig. 5. All of the hexadentate secondary amine complexes (entries
3
+
−
3+
−
salts Λ-[Co(Cl
[
2
sar)]
3Cl
[27] and Λ-[Co((NO
2
)
2
sar)]
3Cl
3
+
15,27], in which more electron withdrawing substituents have re-
2–5) again gave very low enantioselectivities. However, Λ-[Co(sen)]
3
+
−
−
placed the apical amino groups of Λ-[Co((NH
2
)
2
sar)]
3Cl . These
values of 8.85
, chromium(III) analog) [16]
have been reported. Interestingly, the initially formed bright orange
CH Cl solutions turned deep violet. The same color change occurs
f
3BAr , the best performing catalyst in Fig. 5, afforded a significant
groups render the secondary NH units more acidic; pK
a
enantioselectivity of 57% ee (entry 6).
(
NO ) [15], 10.36 (Cl) [29], and 12.0 (NH
2
2
Fig. 7 depicts a base promoted addition of a 1,3-dicarbonyl com-
pound to an azo linkage. This transformation is catalyzed by the tris
(1,2-diphenylethylenediamine) complex II, and an enantioselectivity of
76% ee was obtained using N-methylmorpholine as the base; ee values
of > 90% could be realized with other solvents, anion sets, or the Δ
diastereomer [6a]. All cobalt(III) cage complexes were again competent
catalysts, with rates generally similar to those in Fig. 5 (entries 2–6). As
2
2
when the analogous trichloride or tris(perchlorate) salts are mono-
deprotonated, a process noted to happen (at least to some extent)
spontaneously in the presence of organic solvents (or upon vacuum
dessication) [15].
3+
−
f
in Figs. 5 and 6, Λ-[Co(sen)] 3BAr
was again the clear winner, with
an ee of 40%.
Catalyst screening
Finally, some data on relative catalyst rates were sought. The mal-
onate addition in Fig. 5 was repeated, but with lower loadings of
Fig. 5 depicts the addition of a malonate ester to trans-ß-nitrostyrene
3

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
Fig. 5. Additions of dimethyl malonate to trans-ß-nitrostyrene catalyzed by the cobalt(III) complexes in Fig. 3a.
Fig. 6. Additions of diethyl malonate to 2-cyclopenten-1-one catalyzed by the cobalt(III) complexes in Fig. 3a.
4

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
a
Fig. 7. Additions of methyl 2-oxocyclopentanecarboxylate to di(t-butyl) azodicarboxylate catalyzed by the cobalt(III) complexes in Fig. 3 .
which also contain RNH
catalysts. However, a single run without the Et
2
and NR
3
moieties, are potentially bifunctional
N base (vide supra) gave
3
no reaction, and we were discouraged from attempting additional ex-
periments by the very low ee values.
There is perhaps still some basis for cautious optimism regarding the
3
D catalysts. For certain complexes, the modes of hydrogen bonding to
enantiomeric tartrate anions have been studied in detail, and it has
been possible to demonstrate that one diastereomer can be substantially
more favored than the other [24] – in other words, an appreciable level
of chiral recognition. When high diastereoselectivities can be estab-
lished for ground states, it becomes more plausible that significant se-
lectivities can be achieved for competing transition states.
Regardless, the most interesting lead that emerges from this work is
3
the less symmetric (C ) catalyst containing an acyclic hexaamine li-
3
+
−
gand, Λ-[Co(sen)] 3BAr
the corresponding D sarcophagine (VIb, X = CH
the two CH C(CH apices, thereby exposing a C
three NH moieties. This motif can also be found (twice) in the tris(1,2-
f
. This species can be formally derived from
) by removing one of
face that features
3
3
3
2
)
3
3
Fig. 8. Rate profiles for the addition in Fig. 5 with three catalysts under con-
2
3 2 2
ditions modified to give slower reactions (2 mol% cat, 0.36 equiv Et N, CD Cl ,
rt).
ethylenediamine) complex I and the tris(1,2-diphenylethylenediamine)
complex II, with the latter being an enantioselective catalyst of wide
applicability. For at least some reactions, there is good evidence that
catalyst and base so as to slow the reaction and allow more precise
the C face of II is the active site [4]. Also, note that the rate trends in
3
3
+
−
comparisons. As shown in Fig. 8, Λ-[Co(en)
3
]
3BAr
3BAr
f
(I) was the
and Λ-[Co
Fig. 8 correlate to the number of C faces in the catalysts. Thus, we view
3
3
+
−
f
3+
−
most active catalyst, followed by Λ-[Co(sen)]
Λ-[Co(sen)]
3BArf as deserving of follow-up studies.
3
+
−
(
sar)]
f
3BAr . Some possible rationales are suggested in the discus-
This recommendation is bolstered by unpublished results. For ex-
3+
−
sion section.
ample, for the addition in Fig. 6 (entry 6), Λ-[Co(sen)]
3BArf ac-
tually outperforms the tris(1,2-diphenylethylenediamine) complex Λ-
3+
−
[
Co((S,S)-dpen)
3
]
3BAr
f
[7], which gives a 47% ee under analogous
Discussion
conditions. Furthermore, other solvents, bases, and anion sets – at least
those investigated to date – afford lower ee values. This includes the
mixed salt catalyst II, which features two strongly hydrogen bonding
chloride anions and gives a very sluggish reaction.
One of the most obvious conclusions from Figs. 5–7 is that all of the
cobalt(III) catalysts derived from bicyclic hexaamine ligands (and
which have D symmetry) afford dismal enantioselectivities. Appar-
3
Finally, we note that some of the complexes in Fig. 3 have de-
monstrated proficiencies as oxidative catalysts [35], electron transfer
mediators [36], and photosensitizers [37]. Furthermore, analogous ar-
chitectures can be developed from other cobalt(III) building blocks,
ently the secondary NH moieties are effective enough hydrogen bond
donors to give active catalysts, but the diastereomeric transition states
leading to the different product enantiomers seem to be very close in
3
+
−
3+
−
energy. Both Λ-[Co((NH
2
)
2
sar)]
3BAr
f
and Λ-[Co(sep)]
f
3BAr ,
5

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
1
−
such as the tris(trans-1,2-cyclohexanediamine) complex [38,39]. With
this work, the large class of complexes represented by Fig. 3 [14], many
of which are readily available in enantiopure form, are shown to have
new relevancy as potential enantioselective catalysts. Although the ee
values in Figs. 5–7 are often modest, there are innumerable ways to
build upon or augment the hydrogen bonding interactions that underlie
the reactions. Efforts along these lines will be reported in due course.
NMR (acetone-d
6
, δ in ppm): H (500 MHz), BAr
f
at 7.79 (s, 24H,
o), 7.67 (s, 12H, p); sep at 6.91 (br s, 6H, NH), 4.81–4.69 (m, 6H,
NCHH′N), 4.14–4.07 (m, 6H, NCHH′N), 3.85–3.72 (m, 6H,
NCHH′CHH′N), 3.40–3.26 (m, 6H, NCHH′CHH′N), 3.01 (br s, 17H,
1
−
1
H
2
O). 13C{ H} (126 MHz), BAr
f
1
at 162.6 (q, JBC = 49.8 Hz, i), 135.6
CF = 272.0 Hz, CF ), 118.5 (m, p); sep
N), 54.9 (s, NCH CH N).
(s, o), 130.0 (m, m), 125.4 (q,
at 68.7 (s, NCH
J
3
2
2
2
Experimental section
Λ-[Co(sar)]3+ 3BAr −
f
3
+
−
General data
Λ-[Co(sar)] 3Cl ·1.5H
distilled H O (15 mL), CH
0.0521 mmol, 3.0 equiv) were combined in a procedure analogous to
2
O (0.0083 g, 0.017 mmol [15,27]; Fig. 3),
Cl
2
(15 mL), and Na+ BAr
f
−
(0.0462 g,
2
2
All reactions were conducted under aerobic conditions. Solvents for
NMR (Cambridge Isotopes), HPLC (hexanes, Fisher; isopropanol, JT
3+
−
3+
that for [Co(sep)]
3BAr
O as a bright orange solid (0.0489 g, 0.0163 mmol, 96%),
mp (open capillary) 120–131 °C (dec). Anal. Calcd. for
CoF72 ·4H O (3005.11): C 43.97, H 2.55, N 2.80; found C
44.24, H 2.62, N 2.82.
NMR (acetone-d
f
. An identical workup gave Λ-[Co(sar)]
−
Baker), and reactions (CH
acetate (2 × ACS grade, ≥99.5%), hexanes (ACS reagent, ≥98.5%), 4
Aldrich) were used as received. Other chemicals were used as re+-
ceived from the following sources: NaOH (ACS grade), VWR; Na
2
Cl
2
(anhydrous, ≥99.8%), acetone and ethyl
f 2
3BAr ·4H
×
C
110
H
68
B
3
N
6
2
−
, δ in ppm): 1H (500 MHz), BAr −
at 7.79 (s, 24H,
BAr
trostyrene (98%), dimethyl malonate (98%), diethyl malonate (99%),
Et N (98%), and N-methylmorpholine (99%), 5 × Alfa Aesar; 2-cy-
clopenten-1-one ( > 98%), Acros; di-t-butyl azodicarboxylate (98%),
Aldrich; methyl 2-oxocyclopentanecarboxylate (> 97%), TCI; Ph SiMe
f
(97%), Ark Pharm (BAr
f
= B(3,5-C
6
3
H (CF
3
)
2
)
4
); trans-ß-ni-
6
f
o), 7.68 (s, 12H, p); sar at 6.70 (br s, 6H, NH), 3.79–3.72 (m, 6H,
NCHH′CHH′N), 3.69–3.62 (m, 6H, NCHH′CHH′N), 3.60 (br s, 15H,
H O), 3.25–3.15 (m, 6H, NCHH′CH), 3.09–3.02 (m, 6H, NCHH′CH),
3
2
13
1
−
2
2
2.83–2.79 (m, 2H, CH).
C{ H} (126 MHz), BAr
f
at 162.6 (q,
BC = 50.0 Hz, i), 135.6 (s, o), 130.0 (m, m), 125.4 (q, JCF = 271.8 Hz,
CF ), 118.5 (m, p); sar at 55.9 (s, NCH CH N), 51.0 (s, NCH CH), 40.8
(s, CH).
®
1
1
(
> 97%), TCI; silica gel (SiliaFlash F60), Silicycle.
NMR spectra were recorded on a 500 MHz spectrometer at ambient
J
3
2
2
2
1
probe temperatures and referenced (δ in ppm) to solvent signals ( H:
residual acetone-d
13
5
, 2.05, CHDCl
2
, 5.32, or CHD
2
CN, 1.94;
C:
acetone-d , 206.3). HPLC analyses were carried out with a Shimadzu
6
Λ-[Co(sen)]3+ 3BAr −
f
instrument package (pump/autosampler/detector LC-20AD/SIL-20A/
SPD-M20A). Melting points were determined using an Optimelt MPA
3
+
−
Λ-[Co(sen)] 3Cl ·3H
distilled H O (15 mL), CH
.2345 mmol, 3.0 equiv) were combined in a procedure analogous to
2
O (0.0364 g, 0.0782 mmol [23,28]; Fig. 3),
(15 mL), and Na+ BAr
−
(0.2078 g,
100 instrument. Microanalyses were conducted by Atlantic Microlab.
2
2
Cl
2
f
0
sar)]3+ 3BAr
−
that for [Co(sep)]
BAr ·15H O as a bright orange solid (0.2226 g, 0.07033 mmol, 90%),
f 2
3+
−
3+
Λ-[Co((NH
2
)
2
f
f
3BAr . An identical workup gave Λ-[Co(sen)]
−
3
A
beaker was charged with Λ-[Co((NH
0.0533 g, 0.0934 mmol [18]; Fig. 3) and distilled H
NaOH (aq, 2% w/w) was added until a pH of ca. 8 was attained, fol-
3
)
2
sar)]5+ 5Cl ·H
−
2
O
mp (open capillary) 143–147 °C (dec). Anal. Calcd. for
CoF72 ·15H O (3165.22): C 40.60, H 3.06, N 2.66; found C
40.15, H 2.78, N 2.65.
NMR (acetone-d
(
2
O (15 mL). Then
C
107
H
66
B
3
N
6
2
+
−
, δ in ppm): 1H (500 MHz), BAr −
at 7.79 (s, 24H,
lowed by CH
2
Cl
2
(15 mL) and Na BAr
f
(0.2480 g, 0.2798 mmol, 3.0
6
f
equiv). The biphasic mixture was vigorously stirred (10 min) and al-
lowed to stand. The lower bright orange phase was separated and the
solvent removed by passive evaporation (fume hood) to give Λ-[Co
o), 7.67 (s, 12H, p); sen at 6.90 (br s, 3H, NH), 5.46 (br s, 3H, NHH′),
5.37 (br s, 3H, NHH′), 3.70–3.62 (m, 3H), 3.51–3.43 (m, 6H),
3.28–3.19 (m, 6H, overlapping with H O), 3.19 (br s, 29H, H O),
2
2
3
+
−
13
1
−
f
(
0
C
4
(NH
2
)
2
sar)]
.0829 mmol, 89%), mp (open capillary) 70 °C (dec). Anal. Calcd. for
CoF72 ·7H O (3089.18): C 42.77, H 2.74, N 3.63; found C
2.69, H 2.50, N 3.39.
NMR (acetone-d , δ in ppm): H (500 MHz), BAr
o), 7.68 (s, 12H, p); (NH sar at 6.85 (br s, 6H, NH), 3.80–3.71 (m, 6H,
NCCHH′N), 3.60–3.42 (m, 6H, NCCHH′N, overlapping with H O), 3.54
br s, 14H, H O), 3.25–3.17 (m, 6H, NCHH′CHH′N), 2.90–2.80 (m, 6H,
NCHH′CHH′N). 2.01–1.93 (m, 4H, NH
3BAr
f
·7H
2
O
as
a
bright orange solid (0.2562 g,
2.94–2.88 (m, 3H), 1.09 (s, 3H, CH
3
). C{ H} (126 MHz), BAr
BC = 49.8 Hz, i), 135.4 (s, o), 129.9 (m, m), 125.2 (q,
CF = 271.9 Hz, CF ), 118.3 (m, p); sen at 57.8 and 55.8 (2 s,
NHCH CH NH ), 46.2 (s, CCH ), 44.2 (s, NCH CCH ), 20.4 (s, CH ).
at
1
162.6 (q, J
1
110
H
70
B
3
N
8
2
J
3
2
2
2
3
2
3
3
1
−
6
f
at 7.82 (s, 24H,
2
)
2
Dimethyl 2-(2-nitro-1-phenylethyl)malonate; catalyst screening (Fig. 5)
2
(
2
An authentic sample of the racemate (colorless oil) was obtained by
adapting a literature procedure [4]. A 5 mm NMR tube was charged
with a solution of trans-ß-nitrostyrene (0.0054 g, 0.036 mmol, 1.0
equiv), catalyst (0.0036 mmol, 0.10 equiv), dimethyl malonate
(0.0045 mL, 0.0052 g, 0.039 mmol, 1.1 equiv), and a solution of
1
3
1
−
2
); C{ H} (126 MHz), BAr
BC = 49.9 Hz, i), 135.1 (s, o), 129.6 (m, m), 125.0 (q,
CF = 271.8 Hz, CF ), 118.0 (m, p); (NH sar at 57.9 (s, CNHC′), 56.8
s, CNHC'), 56.0 (s, CNH ).
f
at
1
1
62.2 (q, J
1
J
3
2 2
)
(
2
Ph
bar was added and the sample was cooled to 0 °C. Then Et
0.0050 mL, 0.0036 g, 0.036 mmol) or N-methylmorpholine
2 2 2 2 3
SiMe (0.0020 mL, standard) in CD Cl or CD CN (0.40 mL). A stir
Λ- and Δ-[Co(sep)]3+ 3BAr
−
f
3
N
(
3
+
−
Λ-[Co(sep)]
distilled H O (15 mL), CH
.1359 mmol, 3.0 equiv) were combined in a procedure analogous to
3Cl ·3H
2
O (0.0205 g, 0.0453 mmol [26]; Fig. 3),
(0.0040 mL, 0.0037 g, 0.036 mmol) was added and the mixture stirred.
+
−
After 2 h, the stir bar was removed and a 1H NMR spectrum recorded.
2
2
Cl
2
(15 mL), and Na BAr
f
(0.1204 g,
0
The product yield was assayed by integrating the CH(CO
2 2
Me) signal
3
+
−
that for Λ-[Co((NH
2
)
2
sar)] 3BAr
f
(but without NaOH). An identical
versus the CH signal of the standard. The solvent was removed under
3
3
+
−
f
workup gave Λ-[Co(sep)]
(
3BAr
·7H
2
O as a bright orange solid
reduced pressure to give an orange oil, which was added to a plug of
silica. The product was eluted with hexanes/EtOAc (1:1 v/v). The sol-
vent was removed from the product containing fraction under reduced
pressure. The enantiomeric excess was determined by HPLC with a
Chiralpak AD column (98:2 v/v hexanes/isopropanol, 1.0 mL/min,
0.1365 g, 0.0446 mmol, 98%), mp (open capillary) 122–133 °C (dec).
Anal. Calcd. for C108 CoF72 ·7H O (3061.13): C 42.38, H 2.63, N
.66; found C 42.30, H 2.68, N 3.61. The enantiomer was analogously
H
64
B
3
N
8
2
3
3
+
−
prepared from Δ-[Co(sep)]
2
3Cl ·3H O (0.0166 g, 0.0367 mmol)
[
26]; found C 42.27, H 2.61, N 3.44.
R
λ = 220 nm); for entry 6, t = 31.0 min (major), 41.0 min (minor)
6

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
[
4,30,31].
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.03.018.
3. -(Bis(ethoxycarbonyl)methyl)cyclopentanone; catalyst screening (Fig. 6)
An authentic sample of the racemate (colorless oil) was obtained by
References
adapting a literature procedure [2]. A 5 mm NMR tube was charged
with 2-cyclopenten-1-one (0.0030 mL, 0.0029 g, 0.036 mmol, 1.0
equiv), catalyst (0.0036 mmol, 0.10 equiv), diethyl malonate
[
1] (a) M.S. Taylor, E.N. Jacobsen, Asymmetric catalysis by chiral hydrogen-bond
donors, Angew. Chem. Int. Ed. 45 (2006) 1520–1543. Asymmetrische Katalyse
durch chirale Wasserstoffbrückendonoren, Angew. Chem. 118 (2006) 1550–1573;
(b) A.G. Doyle, E.N. Jacobsen, Small-molecule H-bond donors in asymmetric cat-
alysis, Chem. Rev. 107 (2007) 5713–5743;
(
0.0056 mL, 0.0059 g, 0.037 mmol, 1.0 equiv), and a solution of
Ph SiMe (0.0020 mL, standard) in CD Cl (0.40 mL). A stir bar was
added and the sample was cooled to −35 °C. Then Et N (0.0050 mL,
.0036 g, 0.036 mmol, 1.0 equiv) was added and the mixture stirred.
2
2
2
2
3
(c) X. Yu, W. Wang, Hydrogen-bond-mediated asymmetric catalysis, Chem. Asian
0
J. 3 (2008) 516–532;
1
After the specified time (Fig. 6), the stir bar was removed and a H NMR
(d) Y. Takemoto, Recognition and activation by ureas and thioureas: stereo-
selective reactions using ureas and thioureas as hydrogen-bonding donors, Org.
Biomol. Chem. 3 (2005) 4299–4306;
spectrum recorded. The product yield was assayed by integrating the
CH(CO Et) signal versus the CH signal of the standard. The solvent
2 2 3
(e) F.E. Held, S.B. Tsogoeva, Asymmetric cycloaddition reactions catalyzed by bi-
was removed under reduced pressure to give an orange oil, which was
added to a plug of silica. The product was eluted with hexanes/acetone
functional thiourea and squaramide organocatalysts: recent advances, Catal. Sci.
Technol. 6 (2016) 645–667.
2] C. Ganzmann, J.A. Gladysz, Phase transfer of enantiopure werner cations into or-
ganic solvents: an overlooked family of chiral hydrogen bond donors for en-
antioselective catalysis, Chem. Eur. J. 14 (2008) 5397–5400.
[
[
(
9:1 v/v). The solvent was removed from the product containing frac-
tion under reduced pressure. The enantiomeric excess was determined
by HPLC with a Chiralpak AS-H column (97.5:2.5 v/v hexanes/iso-
3] S.K. Ghosh, A. Ehnbom, K.G. Lewis, J.A. Gladysz, Hydrogen bonding motifs in
structurally characterized salts of the tris(ethylenediamine) cobalt trication, [Co
propanol, 0.8 mL/min, λ = 220 nm); for entry 6,
t
R
= 45.1 min
3+
(en)
3
]
; an interpretive review, including implications for catalysis, Coord. Chem.
(
major), 51.3 min (minor) [32,33].
Rev. 350 (2017) 30–48.
[
[
4] K.G. Lewis, S.K. Ghosh, N. Bhuvanesh, J.A. Gladysz, Cobalt(III) Werner complexes
with 1,2-diphenylethylenediamine ligands: readily available, inexpensive, and
modular chiral hydrogen bond donor catalysts for enantioselective organic synth-
esis, ACS Cent. Sci. 1 (2015) 50–56.
5] (a) S.K. Ghosh, C. Ganzmann, N. Bhuvanesh, J.A. Gladysz, Werner complexes with
ω-dimethylaminoalkyl substituted ethylenediamine ligands: bifunctional hydrogen-
bond-donor catalysts for highly enantioselective Michael additions, Angew. Chem.
Int. Ed. 55 (2016) 4356–4360;
N,N'-Bis(t-butoxycarbonyl)-1-hydrazino-2-oxocyclopentanecarboxylic acid
methyl ester; catalyst screening (Fig. 7)
An authentic sample of the racemate (colorless oil) was obtained by
adapting a literature procedure [6a]. A 5 mm NMR tube was charged
with di-t-butyl azodicarboxylate (0.0083 g, 0.036 mmol, 1.0 equiv),
catalyst (0.0036 mmol, 0.10 equiv), methyl 2-oxocyclopentanecarbox-
ylate (0.0046 mL, 0.0053 g, 0.037 mmol, 1.0 equiv), and a solution of
(b) Werner-Komplexe mit ω-Dimethylaminoalkyl-substitutierten
Ethylendiaminliganden: bifunktionale H-Brückendonor-Katalysatoren für hoch en-
antioselektive Michael-Additionen, Angew. Chem. 128 (2016) 4429–4433.
6] (a) A. Kumar, S.K. Ghosh, J.A. Gladysz, Tris(1,2-diphenylethylenediamine)cobalt
[
(III) complexes: Chiral hydrogen bond donor catalysts for enantioselective α-ami-
Ph
added and the sample was cooled to 0 °C. Then Et
.0036 g, 0.036 mmol, 1.0 equiv) was added and the mixture stirred.
2
SiMe
2
(0.0020 mL, standard) in CD
2
Cl
2
(0.40 mL). A stir bar was
nations of 1,3-dicarbonyl compounds, Org. Lett. 18 (2016) 760–763;
(b) H. Joshi, S.K. Ghosh, J.A. Gladysz, Enantioselective additions of stabilized
carbanions to imines generated from α-amido sulfones by using lipophilic salts of
chiral tris(1,2-diphenylethylenediamine) cobalt(III) trications as hydrogen bond
donor catalysts, Synthesis 49 (2017) 3905–3915.
3
N (0.0050 mL,
0
1
After 2 h, the stir bar was removed and a H NMR spectrum recorded.
The product yield was assayed by integrating the C(CH and C′
C′H signals versus the CH signal of the standard. The solvent was
3 3
)
[
[
[
7] S.K. Ghosh, K.G. Lewis, A. Kumar, J.A. Gladysz, Syntheses of families of enantiopure
(
3
)
3
3
and diastereopure cobalt catalysts derived from trications of the formula [Co
3
+
(NH
2
CHArCHArNH
2
)
3
]
, Inorg. Chem. 56 (2017) 2304–2320.
removed under reduced pressure to give an orange oil, which was
added to a plug of silica. The product was eluted with hexanes/EtOAc
8] S.K. Ghosh, C. Ganzmann, J.A. Gladysz, Synthesis of a series of ω-dimethylami-
noalkyl substituted ethylenediamine ligands for use in enantioselective catalysis,
Tetrahedron Asymmetry 26 (2015) 1273–1280.
9] T. Mukherjee, C. Ganzmann, N. Bhuvanesh, J.A. Gladysz, Syntheses of enantiopure
bifunctional 2-guanidinobenzimidazole cyclopentadienyl ruthenium complexes:
highly enantioselective organometallic hydrogen bond donor catalysts for carbon-
carbon bond forming reactions, Organometallics 33 (2014) 6723–6737.
(
7:3 v/v). The solvent was removed from the product containing frac-
tion under reduced pressure. The enantiomeric excess was determined
by HPLC with a Chiralpak AD column (96:4 v/v hexanes/isopropanol,
1
.0 mL/min, λ = 210 nm); for entry 6, t
R
= 13.6 min (minor), 20.3 min
[
10] C. Thomas, J.A. Gladysz, Highly active families of catalysts for the ring opening
polymerization of lactide: metal templated organic hydrogen bond donors derived
from 2-guanidinobenzimidazole, ACS Catal. 4 (2014) 1134–1138 See also.
11] (a) V.I. Maleev, M. North, V.A. Larionov, I.V. Fedyanin, T.F. Savel´yeva,
M.A. Moscalenko, A.F. Smolyakov, Y.N. Belokon, Chiral octahedral complexes of
cobalt(III) as “organic catalysts in disguise” for the asymmetric addition of a glycine
Schiff base ester to activated olefins, Adv. Synth. Catal. 356 (2014) 1803–1810;
(
major) [6a],34].
[
Rate profiles (Fig. 8)
A 5 mm NMR tube was charged with a solution of trans-ß-ni-
trostyrene (0.0075 g, 0.050 mmol, 1.0 equiv), catalyst (0.001 mmol,
(
b) P. Pardo, D. Carmona, P. Lamata, R. Rodríguez, F.J. Lahoz, P. García-Orduña,
6
L.A. Oro, Reactivity of the chiral metallic Bronsted acid [(η -p-MeC
6
H
4
iPr)Ru
0
.02 equiv), dimethyl malonate (0.0063 mL, 0.0072 g, 0.055 mmol, 1.1
equiv), and Ph SiMe (0.0020 mL, standard) in CD Cl (0.50 mL). Then
Et N (0.0025 mL, 0.0018 g, 0.018 mmol, 0.36 equiv) was added. The
3
'
(
κ P,O,O-POH)][SbF
aldimines, Organometallics 33 (2014) 6927–6936;
c) K.L. Skubi, J.B. Kidd, H. Jung, I.A. Guzei, M.-H. Baik, T.P. Yoon,
6 2 2 2
] (POH = SC1,RC2)-Ph PC(Ph)HC(OH)HCH OMe) toward
2
2
2
2
(
3
Enantioselective excited-state photoreactions controlled by a chiral hydrogen-
bonding iridium sensitizer, J. Am. Chem. Soc. 139 (2017) 17186–17192.
tube was inverted once (mixing) and placed in a NMR probe. The
1
product yield was assayed by H NMR as for the corresponding reaction
[12] (a) D.M. Nickerson, A.E. Mattson, Transition metal and hydrogen bond donor
hybrids: catalysts for the activation of alkylidene malonates, Chem. Eur. J. 18
in Fig. 5.
(2012) 8310–8314;
(b) D.M. Nickerson, V.V. Angeles, T.J. Auvil, S.S. So, A.E. Mattson, Internal Lewis
acid assisted ureas: tunable hydrogen bond donor catalysts, Chem. Commun. 49
2013) 4289–4291.
Conflicts of interest
(
[
13] (a) H. Huo, C. Fu, C. Wang, K. Harms, E. Meggers, Metal-templated en-
antioselective Enamine/H-bonding dual activation catalysis, Chem. Commun. 50
The authors have no conflicts of interest.
(2014) 10409–10411;
(b) Y. Hu, Z. Zhou, L. Gong, E. Meggers, Asymmetric aza-Henry reaction to provide
oxindoles with quaternary carbon stereocenter catalyzed by a metal-templated
chiral Bronsted base, Org. Chem. Front. 2 (2015) 968–972;
Acknowledgements
(
c) X. Ding, H. Lin, L. Gong, E. Meggers, Enantioselective Sulfa-Michael addition to
The authors thank the Welch Foundation (Grant A-1656) for sup-
port and Dr. Jack M. Harrowfield for helpful discussions.
α,β-unsaturated γ-oxoesters catalyzed by a metal-templated chiral Bronsted base,
Asian J. Org. Chem. 4 (2015) 434–437;
7

W.J. Maximuck and J.A. Gladysz
Molecular Catalysis 473 (2019) 110360
twist on an old theme in the inorganic laboratory, J. Chem. Educ. 66 (1989)
445–446.
(d) W. Xu, M. Arieno, H. Löw, K. Huang, X. Xie, T. Cruchter, Q. Ma, J. Xi, B. Huang,
O. Wiest, L. Gong, E. Meggers, Metal-templated design: enantioselective hydrogen-
bond-driven catalysis requiring only parts-per-million catalyst loading, J. Am.
Chem. Soc. 138 (2016) 8774–8780;
[27] G.A. Bottomley, I.J. Clark, I.I. Creaser, L.M. Engelhardt, R.J. Geue, K.S. Hagen,
J.M. Harrowfield, G.A. Lawrance, P.A. Lay, A.M. Sargeson, A.J. See, B.W. Skelton,
A.H. White, F.R. Wilner, The synthesis and structure of encapsulating ligands:
properties of bicyclic hexamines, Aust. J. Chem. 47 (1994) 143–179.
[28] R.J. Geue, G.H. Searle, Cation-exchange chromatography and selective complexa-
tion in the isolation of branched acyclic polyamines: syntheses of ethylidynetris
(methanamine) [tame], 2,2-bis(aminomethyl)propan-1-ol [hmmp], 4,4′,4″-
Ethylidynetris(3-azabutan-1-amine) [sen] and 5,5′,5″-ethylidynetris(4-azapentan-1-
amine) [stn], Aust. J. Chem. 36 (1983) 927–935.
[29] P. Comba, A.M. Sargeson, Is the E2 mechanism required to explain base hydrolysis
of cobalt(III) amine complexes in the limit of general base catalysis? J. Chem. Soc.
Chem. Commun. (1985) 51–52.
[30] J.M. Andrés, R. Manzano, R. Pedrosa, Novel bifunctional chiral urea and thiourea
derivatives as organocatalysts: enantioselective nitro-Michael reaction of malonates
and diketones, Chem. Eur. J. 114 (2008) 5116–5119.
[31] D. Almaşi, D.A. Alonso, E. Gómez-Bengoa, C. Nájera, Chiral 2-aminobenzimidazoles
as recoverable organocatalysts for the addition of 1,3-dicarbonyl compounds to
nitroalkenes, J. Org. Chem. 74 (2009) 6163–6168.
[32] P. Li, S. Wen, F. Yu, Q. Liu, W. Li, Y. Wang, X. Liang, J. Ye, Enantioselective or-
ganocatalytic Michael addition of malonates to α,β-unsaturated ketones, Org. Lett.
11 (2009) 753–756.
[33] V. Wascholowski, K.R. Knudsen, C.E.T. Mitchell, S.V. Ley, A general organocatalytic
enantioselective malonate addition to α,β-unsaturated enones, Chem. Eur. J. 114
(2008) 6155–6165.
[34] S. Saaby, M. Bella, K.A. Jørgensen, Asymmetric construction of quaternary stereo-
centers by direct organocatalytic amination of α-cyanoacetates and β-dicarbonyl
compounds, J. Am. Chem. Soc. 126 (2004) 8120–8121. The chiral HPLC column
employed with the addition product featured the same stationary phase but a dif-
ferent particle size from that used earlier by these researchers. It is presumed that
the well separated enantiomers exhibit identical orders of elution. Configurations
were assigned accordingly.
(e) X. Ding, C. Tian, Y. Hu, L. Gong, E. Meggers, Tuning the basicity of a metal-
templated Bronsted base to facilitate the enantioselective Sulfa-Michael addition of
aliphatic thiols to α,β-unsaturated N-acylpyrazoles, Eur. J. Org. Chem. 2016 (2016)
8
87–890.
[14] L.R. Gahan, J.M. Harrowfield, Sepulchrate: four decades on, Polyhedron 94 (2015)
1
–51.
[
15] R.J. Geue, T.W. Hambley, J.M. Harrowfield, A.M. Sargeson, M.R. Snow, Metal ion
encapsulation: cobalt cages derived from polyamines, formaldehyde, and ni-
tromethane, J. Am. Chem. Soc. 106 (1984) 5478–5488.
[
[
[
[
[
[
[
16] P. Comba, I.I. Creaser, L.R. Gahan, J.M. Harrowfield, G.A. Lawrance, L.L. Martin,
A.W.H. Mau, A.M. Sargeson, W.H.F. Sasse, M.R. Snow, Macrobicyclic chromium(III)
hexaamine complexes, Inorg. Chem. 25 (1986) 384–389.
17] H.A. Boucher, G.A. Lawrance, P.A. Lay, A.M. Sargeson, A.M. Bond, D.F. Sangster,
J.C. Sullivan, Macrobicyclic (hexaamine)platinum(IV) complexes: synthesis, char-
acterization, and electrochemistry, J. Am. Chem. Soc. 105 (1983) 4652–4661.
18] H. Cai, J. Fissekis, P.S. Conti, Synthesis of a novel bifunctional chelator AmBaSar
based on sarcophagine for peptide conjugation and Cu radiolabelling, Dalton
Trans. 27 (2009) 5395–5400.
19] A.E. Martell, R.D. Hancock, R.J. Motekaitis, Factors affecting stabilities of chelate,
macrocyclic, and macrobicyclic complexes in solution, Coord. Chem. Rev. 133
6
4
(1994) 39–65.
20] I.I. Creaser, J.M. Harrowfield, A.J. Herlt, A.M. Sargeson, J. Springborg, R.J. Geue,
M.R. Snow, Sepulchrate: a macrobicyclic nitrogen cage for metal ions, J. Am. Chem.
Soc. 99 (1977) 3181–3182.
21] I.I. Creaser, R.J. Geue, J.M. Harrowfield, A.J. Herlt, A.M. Sargeson, M.R. Snow,
J. Springborg, Synthesis and reactivity of aza-capped encapsulated Co(III) ions, J.
Am. Chem. Soc. 104 (1982) 6016–6025.
22] A. Ehnbom, S.K. Ghosh, K.G. Lewis, J.A. Gladysz, Octahedral Werner complexes
with substituted ethylenediamine ligands: a stereochemical primer for a historic
series of compounds now emerging as a modern family of catalysts, Chem. Soc. Rev.
[35] S. Ribeiro, L. Cunha-Silva, S.S. Balula, S. Gago, Cobalt(III) sepulchrate complexes:
application as sustainable oxidative catalysts, New J. Chem. 38 (2014) 2500–2507.
[36] F.M.C. He, P.V. Bernhardt, Cobalt cage complexes as mediators of protein electron
transfer, J. Biol. Inorg. Chem. 22 (2017) 775–788.
4
5 (2016) 6799–6811 Review of stereoisomerism in salts of the trication [Co
3+
(en)
3
]
, and homologs with substituted 1,2-diamine ligands:.
[
[
[
[
23] J.E. Sarneski, F.L. Urbach, A model for the circular dichroism perturbations which
arise upon ion pairing between tris(diamine)-metal chelates and certain poly-
oxyanions, J. Am. Chem. Soc. 93 (1971) 884–888.
24] P.V. Bernhardt, T.E. Dyahningtyas, J.M. Harrowfield, J.-Y. Kim, Y. Kim, E. Ruk-
mini, Chiral resolution of hexaamine cobalt(III) cages: substituent effects on chiral
discrimination, Aust. J. Chem. 56 (2003) 1187–1191.
25] G.W. Walker, R.J. Geue, A.M. Sargeson, C.A. Behm, Surface-active cobalt cage
complexes: synthesis, surface chemistry, biological activity, and redox properties,
Dalton Trans. 15 (2003) 2992–3001.
[37] F. Pina, Q.G. Mulazzani, M. Venturi, M. Ciano, V. Balzani, Photochemistry of Co
3
+
(sep) -oxalate ion pairs: a novel system for dihydrogen evolution from aqueous
solutions, Inorg. Chem. 24 (1985) 848–851.
[38] R.J. Geue, M.G. McCarthy, A.M. Sargeson, Synthesis, chiroptical properties, and
electron self-exchange reactivity of a rigid pentacyclic metal ion cage system with
3
D symmetry, J. Am. Chem. Soc. 106 (1984) 8282–8291.
[39] G.W. Walker, R.J. Geue, K.J. Haller, A.D. Rae, A.M. Sargeson, New synthetic routes
to hexa-aza cages using cobalt(III) tris(1,2-diamine) templates, Dalton Trans.
(2003) 279–281.
26] L.R. Gahan, P.C. Healy, G.J. Patch, Synthesis of cobalt (III) “Cage” complexes: a
8