Chiral Cobalt(III) Tris(1,2-diamine) Catalysts That Incorporate Nitrogenous Base Containing Anions for the Bifunctional Activation of Nucleophiles and Electrophiles in Enantioselective Addition Reactions

Article abstract of DOI:10.1021/acscatal.1c01883

The lipophilic diastereomeric cobalt complexes Λ or Δ-[Co((S,S)-dpen)₃]³⁺2Cl^-BAr_f^-(Λ or Δ-(S,S)-2 ³⁺2Cl^-BAr_f^-; dpen/BAr_f^-= 1,2-dipheny

Full text of DOI:10.1021/acscatal.1c01883

pubs.acs.org/acscatalysis
Research Article
Chiral Cobalt(III) Tris(1,2-diamine) Catalysts That Incorporate
Nitrogenous Base Containing Anions for the Bifunctional Activation
of Nucleophiles and Electrophiles in Enantioselective Addition
Reactions
Connor Kabes, Reagan Lucas, Jack Gunn, and John Gladysz*
Cite This: ACS Catal. 2021, 11, 7762−7771
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The lipophilic diastereomeric cobalt complexes Λ
or Δ-[Co((S,S)-dpen)3]3+ 2Cl−BArf− (Λ or Δ-(S,S)-23+
2Cl−BArf−; dpen/BArf− = 1,2-diphenylethylenediamine/B(3,5-
−
C6H3(CF3)2)4 ) catalyze a number of enantioselective C−H
bond addition reactions in the presence of aliphatic tertiary amines,
but pyridine and N,N-dimethylaniline are much less effective.
However, when these bases are incorporated into counter anions
(4−) as in Λ or Δ-(S,S)-23+ 4−Cl−BArf−, highly enantioselective
bifunctional catalysts can be realized. Salts of nicotinates,
isonicotinates, related sulfonates, and N,N-dimethylaminobenzoate
are effective. The 6-chloronicotinate salt gives slower rates and
lower ee values, and the 6-aminonicotinate salt gives faster rates
and higher ee values. The 6-methyl, 2-methoxy, and unsubstituted
analogs afford intermediate results. The 6-aminonicotinate catalyst is applied to additions of dimethyl malonate to 14 aryl-
substituted nitroolefins and additions of 6 1,3-dicarbonyl compounds to di-t-butyl azodicarboxylate, with average yields/ee values of
90%/85% and 94%/77%, respectively. The authors are unaware of other ionic catalysts for which Brønsted bases have been
productively incorporated into the anions, which are seldom if ever purposefully functionalized in any manner.
KEYWORDS: cobalt, 1,2-diamines, smart anions, task specific anions, hydrogen bonding, bifunctional catalysts, enantioselective catalysis,
Michael additions
−
C6H3(CF3)2)4 ). The latter allows this chemistry to be carried
out in nonpolar solvents that lack functional groups that might
compete for the NH hydrogen bonding sites. However, the D3
symmetric parent trication, historically important as one of the
first inorganic species resolved into enantiomers,6 always gave
mediocre enantioselectivities in addition reactions of carbon−
hydrogen bonds commonly carried out with nitrogenous
Brønsted bases.8
INTRODUCTION
■
There are large numbers of reactions that require (1) a catalyst
and (2) a Brønsted base that is frequently nitrogenous.1,2 The
base is sometimes required in stoichiometric quantities,1
whereas other times catalytic loadings suffice.2 In certain
instances, it has proved possible to covalently link these entities
such that superior bifunctional catalysts are obtained.3−5 One
case in point would be Takemoto’s catalyst (Figure 1),3a in
which an achiral thiourea that is an effective hydrogen bond
donor is joined with a chiral dimethylaminocyclohexyl
fragment. However, applying this strategy to enantioselective
catalysis can pose complications. Introducing functional groups
will often require lengthier catalyst syntheses and potentially
generates additional stereocenters. This may lead to mixtures
of diastereomers, which are sometimes difficultly separable.
These issues arose in our research with hydrogen bond
donor catalysts based upon the substitution inert, chiral-at-
cobalt Werner salts Λ- or Δ-[Co(en)3]3+ 3X− (13+ 3X−; Figure
1).6−8 Here, Λ or Δ denotes the cobalt configurations, en
denotes ethylenediamine, and X− is most often the lipophilic
and poorly hydrogen bond accepting9 anion BArf− (B(3,5-
Happily, excellent enantioselectivities were obtained with
two types of modified catalysts (Figure 1, bottom). The first
featured three 1,2-diphenylethylenediamine (dpen) ligands.10
Both enantiomers of the non-meso diastereomer are
commercially available at surprisingly low prices.11 Further-
more, both the Λ and Δ diastereomers of various lipophilic
Received: April 25, 2021
Revised: June 1, 2021
Published: June 11, 2021
https://doi.org/10.1021/acscatal.1c01883
© 2021 American Chemical Society
ACS Catal. 2021, 11, 7762−7771
7762

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
hydrogen bond addition reactions and can be generalized by A
in Figure 2.10
The second featured two en ligands and a third in which one
CH hydrogen atom had been replaced by a tertiary-amine-
containing CH2CH2CH2NMe2 moiety, Λ-[Co(en)2((S)-
H2NCH((CH2)3NMe2)CH2NH2)]3+ 3BArf− (Λ-(S)-33+
3BArf−).13 This catalyst, which can be generalized by B in
Figure 2, provided even higher enantioselectivities and now in
the absence of an external base. However, the synthesis of the
enantiopure substituted ethylenediamine ligand was not
trivial,14 and the catalyst was obtained as a ca. 50:50 mixture
of diastereomers, one of which gave poor enantioselectivities
and required a chromatographic separation.15
For large numbers of metal-based catalysts, including the
cobalt complexes in Figure 1, the locus of reactivity is a cation.
Thus, we considered the possibility of incorporating nitro-
genous Brønsted bases into the accompanying counter anions,
as represented by C and I in Figure 2. It is well known that
counter anions can have significant influences upon rates,
product distributions, and enantioselectivities,16 but they are
seldom if ever purposefully functionalized.17 In earlier efforts,
we combined the trications Λ- and Δ-(S,S)-23+ with both
enantiomers of various chiral anions and observed ″matched″
and ″mismatched″ combinations that afforded enhanced or
diminished enantioselectivities, respectively.18 Some related
studies have been reported by others.19,20
Figure 1. Relevant previously studied catalysts.
mixed salts [Co((S,S)-dpen)3]3+ 2X−X′− ((S,S)-23+ 2X−X′−)
can be synthesized with high diasteroselectivities.10a,12 Due to
the six phenyl groups in the trication, one BArf− anion (X′)
suffices to provide good solubilities in nonpolar solvents. These
have proved, when used with Et3N or N-methylmorpholine, to
be highly enantioselective catalysts for a variety of carbon−
Figure 2. Top: catalyst design strategy (A → B → C). Bottom: left, embodiment of C in this study (I); right, dominant ion pairing motif for Λ-
(S,S)-23+ 2Cl−BArf− in solution (II; each chloride ion associated with one of the two C3 symmetric NH faces).
7763
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Accordingly, we report in this paper the combination of Λ-
and Δ-(S,S)-23+ with what can be termed ″smart″ or ″task
specific″ anions, namely, achiral carboxylate and related
oxyanions that contain nitrogenous Brønsted bases and are
furthermore readily available. These catalysts distinctly out-
perform benchmark systems that employ catalytic or
stoichiometric quantities of the oxyanion-free Brønsted bases.
This simple strategy, which represents a new paradigm for
bifunctional catalysts and has the potential for considerable
generality, has not to our knowledge been previously utilized to
enhance the performance of any type of catalyst, enantiose-
lective or otherwise.
RESULTS
■
Catalyst Design and Synthesis. The diastereomeric
mixed salts Λ or Δ-(S,S)-23+ 2Cl−BArf− were synthesized by
a two-step procedure from Co(OAc)2 and (S,S)-dpen as
previously described.10a,12 Both are isolated as hydrates, as are
nearly all new cobalt(III) salts below. For simplicity, the water
molecules are ignored in the main text and graphics but are
fully represented in the Experimental Section. They are
factored into all formula weights and yield calculations.
These salts can be chromatographed on silica gel (typical
Figure 3. Biphasic synthesis of a representative mixed salt, Λ-(S,S)-
23+ 4b−Cl−BArf− (4b− = nicotinate), and other Brønsted base
containing anions examined.
1
eluent: 98:2 v/v DCM/MeOH), and H NMR spectra always
show the same relative integration of CH protons in the cation
and anion (the NH protons can exchange in some deuterated
solvents12). Furthermore, the D3 symmetric trication features
two C3 symmetric faces with three nearly synperiplanar NH
bonds, each ideally positioned for hydrogen bonding to a
chloride anion as shown in II in Figure 2. Accordingly, six NH
protons exhibit 1H NMR chemical shifts considerably
downfield from the others. Thus, these systems can be
confidently represented as mixed 2Cl−BArf− salts, both in
solution and in the solid state. One with chiral sulfonate anions
in place of the chloride anions has been crystallographically
characterized.18
Given the effectiveness of the dimethylamino-containing
catalyst Λ-(S)-33+ 3BArf− in Figure 1, it would be logical to
investigate salts of Λ- and Δ-(S,S)-23+ with carboxylate or
other oxyanions that contain aliphatic tertiary amines. There
are a number of obvious candidates, many of which would be
available in enantiopure form. However, such salts gave
disappointing results in initial screens, some of which are
briefly described in the Discussion section.
The chemical shifts of six NH protons were considerably
downfield of the other six (3.85−2.25 ppm for the Λ
diastereomers). The microanalytical data also matched the
proposed formulations. However, in solution, it would be easy
to envision alternative ion pairings, as further treated in the
Discussion section.
Catalyst Screening: First Series. The initial catalyst
screening was carried out with salts of the anions 4a−e−,
representing the conjugate bases of isonicotinic, nicotinic,
picolinic, 2-pyridinesulfonic, and 3-(dimethylamino)benzoic
acid or, expressed differently, all three possible pyridine
carboxylates, one pyridine sulfonate, and N,N-dimethylamino-
benzoate. Two test transformations, shown in Charts 1 and 2,
were selected. The first, the reaction of dimethyl malonate (5a)
and trans-β-nitrostyrene (6a) to give the addition product 7a,
has been rather widely studied.10a,21 The other, the addition of
the 1,3-dicarbonyl compound 9a to di-t-butyl azodicarboxylate
(8) to give 10a, has been less investigated.10b,22 Both can be
effected with various chiral hydrogen bond donor catalysts and
stoichiometric or catalytic quantities of trialkylamine bases,
some of which are quite effective, as summarized in recent
reviews.21,22
Chart 1 shows the initial results for the former reaction,
carried out at 0 °C in acetone-d6 with 10 mol % catalyst
loadings and an NMR standard to determine absolute yields.
First, the parent catalyst Λ-(S,S)-23+ 2Cl−BArf−, which lacks
Brønsted basic nitrogen atoms, did not effect catalysis alone.
Further, as shown in entries 1a and 1b, yields of 7a increased
to only 3−7% when catalytic or stoichiometric amounts of
pyridine were added. N,N-Dimethylaniline, modeling the free
base in 3-dimethylaminobenzoate (4e−), gave even lower
yields (entries 2a and 2b). However, the five salts Λ-(S,S)-23+
4a−e−Cl−BArf− afforded 7a in 83−85% ee (as determined by
HPLC) and 60−94% yields after 24 h. Aside from 7a, only the
unreacted starting material was evident by NMR.
Thus, attention was turned to carboxylic or sulfonic acids
that contained more weakly basic pyridyl or N,N-dimethyla-
nilinyl moieties. As shown in Figure 3, DCM solutions of Λ-
(S,S)-23+ 2Cl−BArf− were treated with aqueous solutions of
Na2CO3 and the inexpensive commercially available acids H4
corresponding to the conjugate bases 4− in Figure 3 (3.0−3.3
equiv each). The organic phases were separated and taken to
dryness to give the catalysts Λ-(S,S)-23+ 4−Cl−BArf− as air-
stable hydrated orange solids in 88−99% yields. Even though
the loadings of the acids H4 were sufficient to displace both
chloride anions, only one was exchanged, giving rather rare
examples of trication salts that feature, at least in a formal
sense, three different monoanions. Selected salts of the
opposite diastereomer, Δ-(S,S)-23+ 4−Cl−BArf− (epimeric at
cobalt), were similarly prepared.
All catalysts were characterized by NMR (1H, 13C) and
microanalyses, as summarized in the Supporting Information.
The 1H NMR spectra exhibited the correct relative integrations
of CH protons in the trication and the 4− and BArf− anions.
As illustrated by entries 8 and 9, the opposite diastereomers
Δ-(S,S)-23+ 4a,b−Cl−BArf− afforded lower ee values, with the
opposite enantiomer of 7a predominating. This paralleled
earlier results with Δ-(S,S)-23+ 2Cl−BArf− and the external
7764
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
base Et3N and shows the cobalt configuration to be the
primary determinant of the product configuration. Analogous
behavior was found for all salts of Δ-(S,S)-23+ examined.
Chart 2 summarizes the data for the latter reaction, similarly
carried out but in CD3CN. Entries 1 and 2 show that the
combination of Δ-(S,S)-23+ 2Cl−BArf− and external pyridine
(10−100 mol %) affords the addition product 10a in moderate
yields (27−65%) and enantioselectivities (52−62% ee).
However, entries 3−7 establish that significantly better results
are achieved with the five bifunctional catalysts Λ-(S,S)-23+
4a−e−Cl−BArf− (83−99% yields, 65−77% ee). Aside from
10a, only the unreacted starting material was evident by NMR.
As illustrated by entries 8 and 9, the opposite diastereomers
Δ-(S,S)-23+ 4a,b−Cl−BArf− give 10a with comparable yields
(99%) and enantioselectivities (67−83% ee) and opposite
absolute configurations. By a modest margin, the highest ee
value is now realized in the Δ diastereomer series (83% vs
77%), paralleling earlier results with Δ-(S,S)-23+ 2Cl−BArf−
and the external base N-methylmorpholine.10b Thus, the
diastereomer that gives the best enantioselectivities depends
upon the test reaction.23
Chart 1. Catalyst Screening: Additions of Dimethyl
Malonate (5a) to trans-β-Nitrostyrene (6a)
Catalyst Screening: Second Series. Overall, the top
performer among (S,S)-23+ 4a−e−Cl−BArf− was judged to be
nicotinic acid derived (S,S)-23+ 4b−Cl−BArf−, although other
salts were not without promise. In the interest of further
optimization, analogs with the substituted nicotinates 4f−i−
(Figure 3) were investigated. These feature both electron-
donating and -withdrawing groups. As summarized in Chart 3,
Λ-(S,S)-23+ 4f−i−Cl−BArf− were evaluated in the same
malonate/nitroolefin addition reaction employed in Chart 1
but now carried out for 48 h at room temperature unless
noted.
a
1
Assayed by H NMR relative to the internal standard Ph2SiMe2.
b
c
Assayed by chiral HPLC. 24 h reaction time.
Chart 2. Catalyst Screening: Additions of Methyl 2-
oxocyclopentane-1-carboxylate (9a) to Di-t-butyl
Azodicarboxylate (8)
Chart 3. Catalyst Optimization: Effect of Nicotinate Anion
Substituents upon Yields and Enantioselectivities for the
Addition in Chart 1
a
1
Assayed by H NMR relative to the internal standard Ph2SiMe2.
b
c
Assayed by chiral HPLC. This yield is after 22 h.
As shown in entry 1, the 6-chloronicotinate catalyst Λ-(S,S)-
23+ 4f−Cl−BArf− afforded the poorest results, with a distinctly
lower yield of 7a (33%) and a slightly lower enantioselectivity
(79% ee). The 2-methoxy and 6-methyl substituted catalysts
Λ-(S,S)-23+ 4g,h−Cl−BArf− (entries 2 and 4) performed
comparably to Λ-(S,S)-23+ 4b−Cl−BArf−, giving yields of 69−
58% and enantioselectivities of 84% ee. More importantly, 6-
amino substituted Λ-(S,S)-23+ 4i−Cl−BArf− (entry 5) afforded
a
Assayed after by 1H NMR relative to the internal standard Ph2SiMe2.
b
Assayed by chiral HPLC.
7765
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
a distinctly higher yield and rate (95% after 22 h as compared
to 68−33% after 48 h) and a slightly higher enantioselectivity
(87% ee). This corresponds to the least acidic nicotinic acid or
the most basic and presumably most hydrogen bond accepting
nicotinate.
Chart 4. Substrate Scope: Additions of Dimethyl Malonate
(5a) to Aryl-Substituted Nitroolefins (6a−n) with the
Optimized Catalyst Λ-(S,S)-23+ 4i−Cl−BArf−
To better define the trends in Chart 3, yields were
monitored as a function of time. The results, shown in Figure
4, dramatically illustrate the superior qualities of the 6-
Figure 4. Effect of nicotinate anion substituents upon the rates of
addition of dimethyl malonate (5a) to trans-β-nitrostyrene (6a) under
the conditions of Chart 3.
a
Yields assayed by 1H NMR relative to the internal standard
b
Ph2SiMe2; ee values assayed by chiral HPLC. This configuration is
inferred as described in the text. This yield is after 48 h; an additional
10 mol % of catalyst is added after 24 h.
c
aminonicotinate catalyst Λ-(S,S)-23+ 4i−Cl−BArf− and the
deleterious effect of an electron-withdrawing 6-chloro sub-
stituent. However, outside of the product 7a, in all cases, only
the starting material could be detected by NMR.
With the nitroolefin trans,trans-PhCHCHCHCHNO2,
in which the aryl groups of 6a−n have been replaced by a β-
styryl moiety, the yields and enantioselectivities dropped to
14% and 73% ee. Thus, extensions of these protocols to
nitroolefins that lack aryl substituents may be problematic.
When dimethyl malonate (5a) was replaced by diethyl
malonate (5a-Et) in the addition to 6a, the yield and
enantiomeric excess of the product (7a-Et) fell only slightly
(90% and 80% ee). However, the decreases were dramatic for
diisopropyl and di-t-butyl malonate.24
The scope of 1,3-dicarbonyl compound additions to di-t-
butyl azodicarboxylate (8) that can be catalyzed by the
opposite diastereomer Δ-(S,S)-23+ 4i−Cl−BArf− at 0 °C in
CD3CN was similarly explored. As sketched in Chart 5, six
different substrates (9a−f) afforded products (10a−f) in yields
ranging from 90 to 99% (average 94%) and enantioselectivities
ranging from 86 to 51% ee (average 77%). The dominant
absolute configurations were assigned as for Chart 4.
Catalyst Scope. The 6-aminonicotinate salt Λ-(S,S)-23+
4i−Cl−BArf− was then applied to the additions of dimethyl
malonate to 14 aryl substituted nitroolefins (6a−n) at 0 °C in
acetone-d6. As summarized in Chart 4, after 24 h, the products
7a−n were obtained in average/median yields and ee values of
82%/90% and 85%/87%, respectively. The dominant absolute
configurations of most adducts could be assigned by previously
established chiral HPLC relationships, as documented in the
Supporting Information. These always corresponded to the
same relative configuration. Thus, for the three cases for which
HPLC assignments have not yet been established (7e, l, and
m), identical relative configurations were presumed.
Under the default conditions for Chart 4, the two lowest
yields were encountered with 2-(benzyloxy)phenyl and 2-furyl
substituents (7k and n; 37% and 28%). However, NMR
analyses showed substantial amounts of starting materials.
Accordingly, after the standard 24 h reaction time, another 10
mol % charge of catalyst was added. After an additional 24 h,
the yields had markedly increased (95% and 87%), raising the
average/median yields to 90%/93%. To ensure that the NMR
yields in the charts translate into comparable preparative
yields, the reaction of 5a and 6f was repeated on a 0.1 g scale.
A chromatographic workup gave 7f in 88% yield and 96% ee.
DISCUSSION
■
The most significant and seemingly unprecedented aspect of
this report involves the extraordinary synergism that can be
realized when nitrogenous Brønsted bases are incorporated
into the counter anions of the tricationic cobalt(III) hydrogen
7766
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Mechanistic analysis is also impeded by the many NH donor
groups that might participate at any stage on the reaction
coordinate.28 It would be easy to envision simultaneous
interactions involving four or fivea much more complex
situation than with Takemoto’s catalyst (Figure 1), which can
only offer two NH groups. Thus, it becomes pure speculation
to propose a detailed mechanism. The nicotinate anion almost
certainly deprotonates the 1,3-dicarbonyl reactants in Charts
1−5, but there are numerous possible motifs for this step,
especially with respect to hydrogen bonding.29 However, such
issues sometimes prove computationally tractable, as for a
bifunctional cationic chiral ruthenium hydrogen bond donor
catalyst with a covalently linked nitrogenous Brønsted base.4
Despite these uncertainties, certain phenomenological
observations may offer insight. For example, the pKa(BH)
value for the nicotinate anion 4b− is 4.75 and the pKa(BH+)
value for nicotinic acid H4b is 2.07.30,31 Thus, both species are
weaker bases than Et3N (pKa(BH+) 10.7) and N-methyl-
morpholine (pKa(BH+) 7.4),30 the most effective external
bases for the reactions in Charts 1−5. Both are also less basic
than pyridine (pKa(BH+) 5.17) or N,N-dimethylaniline
(pKa(BH+) 5.15),30 the other external base that is ineffective
(Chart 1, entries 2a and 2b) until incorporated into an
oxyanion. However, any initial impression that a lower basicity
may be of importance is dashed by the much poorer results
with the 6-chloronicotinate anion 4f− (pKa(BH) 3.24)33 and
the superior results with the 6-aminonicotinate anion 4i−
(pKa(BH) 6.30).24,33
Another issue is whether the nicotinate anions function as
oxygen or nitrogen bases. The equilibrium between the N-
protonated zwitterionic and O-protonated non-zwitterionic
forms of nicotinic acid H4b is highly dependent upon the
solvent and temperature,32 with the latter generally favored in
organic solvents. Thus, the O-protonation of 4b− would be
thermodynamically preferred. However, with 6-aminonicotinic
acid H4i, the equilibrium should shift in the N-protonated
zwitterion direction, as supported by the much enhanced
basicity of 2-aminopyridine (pKa (BH+) 6.71)30 versus
pyridine. Since 6-aminonicotinate gives the most effective
catalyst, (S,S)-23+ 4i−Cl−BArf−, this argues for a key role of the
nitrogenous basic site.
Chart 5. Substrate Scope: Additions of Dicarbonyl
Compounds (9a−f) to Di-t-butyl Azodicarboxylate (8) with
the Optimized Catalyst Δ-(S,S)-23+ 4i−Cl−BArf−
a
Yields assayed by 1H NMR relative to the internal standard
b
Ph2SiMe2; ee values assayed by chiral HPLC. This configuration is
inferred as described in the text.
bond donor catalyst (S,S)-23+. This synergism is enormous for
the test reaction in Chart 1 (″internal″ versus external pyridine
or N,N-dimethylaniline) and substantial for that in Chart 2.
The catalysts can be said to have ″smart″ or ″task specific″
counter anions that feature ″built in bases″, as generalized by C
in Figure 2. Counter anions have been modified to achieve
certain phase affinities,25 and chiral counter anions have been
used to optimize the performance of certain cationic chiral
metal catalysts.18−20 However, we remain unaware of any
purposeful attempts to introduce functional groups into
counter anions that would take part in bond breaking and
bond making.
In the neighboring field of ionic liquids, considerable
attention has been given to ″task specific″ cations and anions
for optimizing performances in diverse applications, including
catalysis.26 However, even in this voluminous literature, we are
unable to find prior efforts to achieve function by incorporating
sp2- or sp3-hybridized nitrogen donor atoms into the anions.27
Searches involving ″ionic liquids″ and ″nicotinate″ or related
combinations give no hits.
In Charts 1 and 2, the yield and ee data are benchmarked,
appropriately, to the external bases pyridine and N,N-
dimethylaniline, which are ineffective or only moderately
effective. However, for a full picture, it is also important to
compare the new data to optimized results with Λ-(S,S)-23+
2Cl−BArf− and the external bases Et3N and N-methylmorpho-
line.10a,b While the yields are similar, the average ee values are
slightly lower. In Chart 4, those for the eight substrates
common to both studies are 87% ee (this work) versus 89% ee
(Et3N external base).10a In Chart 5, those for the five
substrates common to both studies are 76% ee (this work)
versus 93% ee (N-methylmorpholine external base).10b
Charts 1−3 by no means exhaust the possible optimization
strategies for this class of catalysts. For example, the overall
efficacy of the isonicotinate catalyst Λ-(S,S)-23+ 4c−Cl−BArf− is
not so different from that of Λ-(S,S)-23+ 4b−Cl−BArf−. Indeed,
a variety of substituted isonicotinic acids are commercially
available, and certain adducts might outperform the bench-
mark 6-aminonicontinate catalyst Λ-(S,S)-23+ 4i−Cl−BArf−. In
the course of this work, a number of salts of Λ-(S,S)-23+ with
various trialkylamine-containing counter anions (e.g., Me2N-
The overall constitution of the mixed salt catalysts (S,S)-23+
4−Cl−BArf− is secure from the NMR and microanalytical data.
Nonetheless, it can be questioned whether they might be
better represented by other formulations, as in solution it
would be easy to envision ion pairing equilibria. We suggest
that the dominant species feature two of the hydrogen bond
accepting (or ″non-BArf−″) anions per cobalt tricationi.e.,
(S,S)-23+ 4−Cl−BArf−, (S,S)-23+ 24−BArf−, and (S,S)-23+
2Cl−BArf−so as to maximize the types of interactions in II
(Figure 2). However, at this time, one can only guess at the
most productive assembly for catalysis.
−
−
(CH2)nCO2 , n = 1−3; Me2N(CH2)2SO3 ) were prepared,
7767
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
partially characterized, and screened as catalysts. While some
gave moderate enantioselectivities, they did not attain the
levels in Chart 1.
(0.0060 mmol, 10 mol %) was added in one portion with
stirring. After 50 h at 0 °C, the solution was transferred to an
1
NMR tube and the yield of 7a was assayed by H NMR. The
Furthermore, there is seemingly an excellent opportunity to
apply multivariate analysis or machine learning methods to
guide future generations of catalyst optimization,34 especially
considering the ″drop in″ nature of the base-containing
counter anions and flexibility in the other counter anions. For
solvent was removed by rotary evaporation, and the residue
was chromatographed on silica (glass pipet, 25:75 v/v EtOAc/
hexanes). The solvent was removed from the product
containing fractions by rotary evaporation to give 7a as a
colorless oil. Enantiomeric excesses were assayed by HPLC
with a Chiralcel AD column (98:2 v/v hexane/isopropanol, 1
mL/min, λ = 220 nm); tR = 32.9 min (major) and 43.6 min
(minor).10a.
−
example, B(C6F5)4 salts are sometimes slightly superior to
BArf− salts.10b−d Finally, the many enantiopure carboxylate
anions that can be derived from chiral alkaloids, which often
feature pyridine or related heteroarene units, clearly merit
future investigation.
NMR (CDCl3, δ/ppm): 1H (400 MHz) 7.35−7.26 (m, 3H),
3
7.23−7.18 (m, 2H), 4.97−4.80 (m, 2H), 4.23 (td, JHH = 8.9,
5.3 Hz, 1H), 3.85 (d, 3JHH = 9.0 Hz, 1H), 3.75 (s, 3H), 3.55 (s,
3H); 13C{1H} (100 MHz): 168.0, 167.4, 136.3, 129.2, 128.6,
128.0, 77.5, 54.9, 53.2, 53.0, 43.0 (11 × s).
CONCLUSIONS
■
Bifunctional catalysis is an immense field with numerous
contributors and much current activity. Despite this breadth,
we remain unaware of any case in which bond breaking and
making for a reaction promoted by an ionic catalyst are
distributed between the cation and functional groups built into
the anion. This strategy, as implemented with the title catalysts
(S,S)-23+ 4−Cl−BArf−, is clearly competitive with covalently
tethered bifunctional catalysts (e.g., C versus B in Figure 2).
Furthermore, the generality is strongly supported by two types
of test reactions involving carbon−carbon and carbon−
nitrogen bond formation (Charts 1−5), two types of otherwise
poorly effective ″built in″ bases (pyridine and N,N-
dimethylaniline), and two types of host oxyanions (carboxylate
and sulfonate). It cannot be dismissed as a one-off curiosity.
This work also adds to the growing body of catalytic
reactions that are promoted by noncovalent interactions
between ligands and substrates.35 Some of the more relevant
earlier efforts involve cationic bifunctional chiral-at-metal
containing hydrogen bond donor catalysts investigated by
Meggers and Gong36 and Belokon and Larionov.15,17,37,38
However, the data herein establish a mold-breaking new
paradigm for the design of bifunctional catalysts, which is
furthermore not limited to species with metal-containing
cations. Additional applications of ″smart″ or ″task specific″
functionality-containing anions in such systems will be
described in future reports.
Catalyst Screening: Azodicarboxylate Addition
(Chart 2). A vial was charged with methyl 2-oxocyclopen-
tane-1-carboxylate (9a; 0.0076 mL, 0.061 mmol, 1.0 equiv), di-
t-butyl azodicarboxylate (8; 0.014 g, 0.061 mmol, 1.0 equiv),
Ph2SiMe2 (0.0013 mL, internal standard), pyridine (1.0−0.0
equiv, delivered volumetrically), CD3CN (0.600 mL), and a
stir bar and cooled to 0 °C. The catalyst (0.0060 mmol, 10 mol
%) was added in one portion with stirring. After 50 h at 0 °C,
the solution was assayed and worked up per the preceding
procedure to give 10a as a colorless oil. Enantiomeric excesses
were determined by HPLC with a Chiralcel AD column (96:4
v/v hexane/isopropanol, 1 mL/min, λ = 210 nm); tR = 13.7
min (min) and 19.4 min (major).10b
1
NMR (CDCl3, δ/ppm): H (500 MHz) 6.70−6.03 (m,
1H), 3.76 (s, 3H), 2.97−2.03 (m, 5 H), 2.03−1.81 (s, 1H),
1.53−1.29 (m, 18H).
Catalyst Screening: Substituted Nicotinate Salts
(Chart 3 and Figure 3). A vial was charged with 6a (0.009
g, 0.060 mmol, 1.0 equiv), Ph2SiMe2 (0.0013 mL, internal
standard), catalyst (0.0060 mmol, 10 mol %), and acetone-d6
(0.600 mL). The solution was transferred to an NMR tube,
1
and an initial H NMR spectrum was recorded. Then, 5a
(0.0076 mL, 0.066 mmol, 1.1 equiv) was added and a second
1H NMR spectrum was immediately acquired. A stir bar was
added, and the solution was stirred (48 h, rt). The yield of 7a
1
was periodically assayed by H NMR (data: Figure 3). The
EXPERIMENTAL SECTION
solvent was removed by rotary evaporation, and the residue
was chromatographed on silica (glass pipet, 25:75 v/v EtOAc/
hexanes). The solvent was removed from the product
containing fractions by rotary evaporation to give 7a as a
colorless oil, which was analyzed as described for Chart 1.
Substrate Scope: Nitroolefin Additions (Chart 4). A
vial was charged with nitroolefin (6; 0.060 mmol, 1.0 equiv),
Ph2SiMe2 (0.0013 mL, internal standard), Λ-(S,S)-23+
4i−Cl−BArf−·2H2O (0.0107 g, 0.0060 mmol, 10 mol %), and
acetone-d6 (0.600 mL). The solution was transferred to an
■
General. All reactions and workups were conducted in air.
General procedures are provided below, and full details are
given in the Supporting Information. The instrumentation
utilized has been described in earlier papers in this series.10,18
Catalyst Synthesis: General Procedure. To a solution of
(S,S)-23+ 2Cl−BArf−·nH2O (Λ, n = 2 or Δ, n = 1; 0.050 g,
0.030 mmol, 1.0 equiv)12 in DCM (5 mL) was added a
solution of an aryl substituted carboxylic or sulfonic acid (4a−
i, Figure 3; 0.089 mmol, 3.0 equiv) and Na2CO3 (0.011 g,
0.099 mmol, 3.3 equiv) in water (5 mL). The biphasic mixture
was rapidly stirred for 30 min. The organic layer was separated,
washed with water (5 mL), dried (NaSO4), and taken to
dryness by rotary evaporation.
Catalyst Screening: Nitrostyrene Addition (Chart 1).
A vial was charged with trans-β-nitrostyrene (6a; 0.009 g, 0.060
mmol, 1.0 equiv), dimethyl malonate (5a; 0.0076 mL, 0.066
mmol, 1.1 equiv), Ph2SiMe2 (0.0013 mL, internal standard),
pyridine (1.0−0.0 equiv, delivered volumetrically), acetone-d6
(0.600 mL), and a stir bar and cooled to 0 °C. The catalyst
1
NMR tube, and an initial H NMR spectrum was recorded.
The sample was cooled to 0 °C. A stir bar and 5a (0.0076 mL,
0.066 mmol, 1.1 equiv) were added. The solution was stirred
(24 h, 0 °C). The yield of product 7 was assayed by 1H NMR,
the solvent was removed by rotary evaporation, and the residue
was chromatographed on silica (glass pipet, 25:75 v/v EtOAc/
hexanes). The solvent was removed from the product
containing fractions by rotary evaporation. Enantiomeric
excesses were determined by HPLC as described in the
Supporting Information.
7768
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Substrate Scope: Azodicarboxylate Additions (Chart
5). A vial was charged with a 1,3-dicarbonyl compound (9;
0.060 mmol, 1.0 equiv), Ph2SiMe2 (0.0013 mL, internal
standard), di-t-butyl azodicarboxylate (0.0153 g, 0.066 mmol,
1.1 equiv), and CD3CN (0.400 mL). The solution was
transferred to an NMR tube, and an initial 1H NMR spectrum
was recorded. The sample was cooled to 0 °C. A stir bar and a
solution of Λ-(S,S)-23+ 4i−Cl−BArf−·2H2O (0.0107 g, 0.0060
mmol, 10 mol %) in CD3CN (0.200 mL) were added. The
solution was stirred (24 h, 0 °C). The yield of product 10 was
Complete contact information is available at:
https://pubs.acs.org/10.1021/acscatal.1c01883
Notes
The authors declare the following competing financial
interest(s): The authors have filed a provisional patent related
to this work.
ACKNOWLEDGMENTS
■
The authors thank the Welch Foundation for financial support
(Grant A-1656) and Ms. Bailey Jameson and Mr. Aaron R.
Wegener for technical assistance and dedicate this paper to Dr.
Christian Bruneau (Université de Rennes 1) in recognition of
his many outstanding contributions to catalysis.
1
assayed by H NMR, the solvent was removed by rotary
evaporation, and the residue was chromatographed on silica
(glass pipet, 25:75 v/v EtOAc/hexanes). The solvent was
removed from the product containing fractions by rotary
evaporation. Enantiomeric excesses were determined by HPLC
as described in the Supporting Information.
REFERENCES
■
Typical Preparative Reaction. A vial was charged with 6f
(see Chart 4; 0.100 g, 0.52 mmol, 1.0 eq), Λ-(S,S)-23+
4i−Cl−BArf−·2H2O (0.092 g, 0.052 mmol, 10 mol %), and
acetone (6 mL) and cooled to 0 °C. Then, 5a (0.082 g, 0.620
mmol, 1.2 eq) was added with stirring. After 48 h, the solvent
was removed by rotary evaporation. The residue was
chromatographed on silica (glass column, 20:80 v/v EtOAc/
hexanes). The solvent was removed from the product
containing fractions by rotary evaporation to give 7f as a
white solid (0.148 g, 0.46 mmol, 88%, 96% ee), mp 86.0−88.6
(1) Familiar examples that employ transition metal catalysts would
include the Heck and Sonogashira reactions: (a) Jagtap, S. Heck
Reaction − State of the Art. Catalysts 2017, 7, 267. (article number)
(b) Chinchilla, R.; Nájera, C. Recent advances in Sonogashira
reactions. Chem. Soc. Rev. 2011, 40, 5084−5121.
(2) For one major reaction class requiring a catalytic amount of base,
see (a) Reyes, E.; Uria, U.; Vicaro, J. L.; Carrillo, L. The catalytic,
enantioselective Michael reaction. Org. React. 2016, 90, 1−898.
(b) Soloshonok, V. A.; Kacharov, A. D.; Avilov, D. V.; Ishikawa, K.;
Nagashima, N.; Hayashi, T. Transition Metal/Base-Catalyzed Aldol
Reactions of Isocyanoacetic Acid Derivatives with Prochiral Ketones,
a Straightforward Approach to Stereochemically Defined ß,ß-
Disubstituted-ß-hydroxy-α-amino Acids. Scope and Limitations. J.
Org. Chem. 1997, 62, 3470−3479.
(3) (a) Takemoto, Y. Development of Chiral Thiourea Catalysts and
Its Application to Asymmetric Catalytic Reactions. Chem. Pharm. Bull.
2010, 58, 593−601. (b) Held, F. E.; Tsogoeva, S. B. Asymmetric
cycloaddition reactions catalyzed by bifunctional thiourea and
squaramide organocatalysts: recent advances. Catal. Sci. Technol.
2016, 6, 645−667.
(4) (a) Mukherjee, T.; Ganzmann, C.; Bhuvanesh, N.; Gladysz, J. A.
Syntheses of Enantiopure Bifunctional 2-Guanidinobenzimidazole
Cyclopentadienyl Ruthenium Complexes: Highly Enantioselective
Organometallic Hydrogen Bond Donor Catalysts for Carbon-Carbon
Bond Forming Reactions. Organometallics 2014, 33, 6723−6737.
(b) Wititsuwannakul, T.; Mukherjee, T.; Hall, M. B.; Gladysz, J. A.
Computational Investigations of Enantioselection in Carbon-Carbon
Bond Forming Reactions of Ruthenium Guanidinobenzimidazole
Second Coordination Sphere Hydrogen Bond Donor Catalysts.
Organometallics 2020, 39, 1149−1162.
(5) For a general treatment of bifunctional catalysts and potential
advantages, see Catalysis from A to Z. A Concise Encyclopedia; Cornils,
B.; Herrmann, W. A.; Wong, C.-H.; Zanthoff, H. Eds., 4th Edition.
Wiley-VCH: Weinheim, 2013, pp. 244−246, DOI: 10.1002/
9783527671380.
°C (lit: 86.0−87.5 °C).39 NMR (CDCl3, δ/ppm, 500 MHz):
2
1H (500 MHz) 6.70 (m, 3H), 5.94 (s, 2H), 4.87 (dd, JHH
=
13.1 Hz, 3JHH = 4.9, 1H), 4.80 (dd, 2JHH = 13.1, 3JHH = 9.3 Hz,
1H), 4.15 (td, 3JHH = 9.2, 4.9 Hz, 1H), 3.80 (d, 3JHH = 9.1 Hz,
1H), 3.76 (s, 3H), 3.70 (s, 3H); 13C{1H} (125 MHz) 167.9,
167.3, 148.2, 147.7, 129.7, 121.5, 108.8, 108.2, 101.4, 77.7,
55.0, 53.2, 53.0, 42.8, (14 × s). HPLC data are given in the
Supporting Information.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acscatal.1c01883.
■
sı
General information; syntheses of nitroolefin substrates
and catalysts; nitroolefin addition products accessed by
the general procedure; di-t-butyl azodicarboxylate
addition products accessed by the general procedure;
NMR spectra of catalysts; and HPLC traces (PDF)
AUTHOR INFORMATION
Corresponding Author
■
John Gladysz − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0002-7012-4872;
Email: gladysz@mail.chem.tamu.edu
(6) Werner, A. Zur Kenntnis des asymmetrischen Kobaltatoms. V.
Ber. Dtsch. Chem. Ges. 1912, 45, 121−130.
(7) Wegener, A. R.; Kabes, C. Q.; Gladysz, J. A. Launching Werner
Complexes into the Modern Era of Catalytic Enantioselective Organic
Synthesis. Acc. Chem. Res. 2020, 53, 2299−2313.
(8) (a) Ganzmann, C.; Gladysz, J. A. Phase Transfer of Enantiopure
Werner Cations into Organic Solvents: An Overlooked Family of
Chiral Hydrogen Bond Donors for Enantioselective Catalysis. Chem.
− Eur. J. 2008, 14, 5397−5400. (b) Maximuck, W. J.; Ganzmann, C.;
Alvi, S.; Hooda, K. R.; Gladysz, J. A. Rendering classical hydrophilic
enantiopure Werner salts [M(en)3]n+ nX− lipophilic (M/n = Cr/3,
Co/3, Rh/3, Ir/3, Pt/4); new chiral hydrogen bond donor catalysts
and enantioselectivities as a function of metal and charge. Dalton
Trans. 2020, 49, 3680−3691.
Authors
Connor Kabes − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0001-8383-5713
Reagan Lucas − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States
Jack Gunn − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States
(9) Wititsuwannakul, T.; Hall, M. B.; Gladysz, J. A. A Computational
Study of Hydrogen Bonding Motifs in Halide, Tetrafluoroborate,
7769
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Hexafluorophosphate, and Tetraarylborate Salts of Chiral Cationic
Ruthenium and Cobalt Guanidinobenzimidazole Hydrogen Bond
Donor Catalysts; Acceptor Properties of the ’BArf’ Anion. Polyhedron
2020, 187, 114618.
thylenediamine) Cobalt(III) Hydrogen Bond Donor Catalysts with
Defined Carbon/Metal Configurations; Matched/Mismatched Effects
upon Enantioselectivities with Enantiomeric Chiral Counter Anions.
ACS Catal. 2020, 10, 3249−3263.
(10) (a) Lewis, K. G.; Ghosh, S. K.; Bhuvanesh, N.; Gladysz, J. A.
Cobalt(III) Werner Complexes with 1,2-Diphenylethylenediamine
Ligands: Readily Available, Inexpensive, and Modular Chiral Hydro-
gen Bond Donor Catalysts for Enantioselective Organic Synthesis.
ACS Cent. Sci. 2015, 1, 50−56. (b) Kumar, A.; Ghosh, S. K.; Gladysz,
J. A. Tris(1,2-diphenylethylenediamine)cobalt(III) Complexes: Chiral
Hydrogen Bond Donor Catalysts for Enantioselective α-Aminations
of 1,3-Dicarbonyl Compounds. Org. Lett. 2016, 18, 760−763.
(c) Joshi, H.; Ghosh, S. K.; Gladysz, J. A. Enantioselective Additions
of Stabilized Carbanions to Imines Generated from α-Amido Sulfones
By Using Lipophilic Salts of Chiral Tris(1,2-diphenylethylenedi-
amine) Cobalt(III) Trications as Hydrogen Bond Donor Catalysts.
Synthesis 2017, 49, 3905−3915. (d) Luu, Q. H.; Gladysz, J. A. An Air
and Water Stable Hydrogen Bond Donor Catalyst for the
Enantioselective Generation of Quaternary Carbon Stereocenters by
Additions of Substituted Cyanoacetate Esters to Acetylenic Esters.
Chem. − Eur. J. 2020, 26, 10230−10239.
(19) For cases where catalysts consisting of chiral metal cations and
chiral anions have been accessed in both diastereomeric forms and
evaluated, see (a) Llewellyn, D. B.; Adamson, D.; Arndtsen, B. A. A
Novel Example of Chiral Counteranion Induced Enantioselective
Metal Catalysis: The Importance of Ion-Pairing in Copper-Catalyzed
Olefin Aziridination and Cyclopropanation. Org. Lett. 2000, 2, 4165−
4168. (b) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. A
Powerful Chiral Counterion Strategy for Asymmetric Transition
Metal Catalysis. Science 2007, 317, 496−499. (c) Li, C.; Wang, C.;
Villa-Marcos, B.; Xiao, J. Chiral Counteranion-Aided Asymmetric
Hydrogenation of Acyclic Imines. J. Am. Chem. Soc. 2008, 130,
14450−14451. (d) Aikawa, K.; Kojima, M.; Mikami, K. Synergistic
Effect: Hydroalkoxylation of Allenes through Combination of
Enantiopure BIPHEP-Gold Complexes and Chiral Anions. Adv.
Synth. Catal. 2010, 352, 3131−3135.
(20) (a) 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, 9047−9153. (b) 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,
603−614.
(21) Alonso, D. A.; Baeza, A.; Chinchilla, R.; Gómez, C.; Guillena,
G.; Pastor, I. M.; Ramón, D. J. Recent Advances in Asymmetric
Organocatalyzed Conjugate Additions to Nitroalkenes. Molecules
2017, 22, 895. (article number)
(22) Usman, M.; Zhang, X.-W.; Wu, D.; Guan, Z.-H.; Liu, W.-B.
Application of dialkyl azodicarboxylate frameworks featuring multi-
functional properties. Org. Chem. Front. 2019, 6, 1905−1928. (see
section 6)
(23) A reviewer has requested a comment on this point. While it will
take much additional effort to develop detailed models for the
enantioselective reactions in question, such dichotomies likely arise
from the ″mechanistic adaptability″ of this catalyst family. The six
NH2 groups can provide platforms for a wide range of transition
states, with the end result that either diastereomer may prove more
effective for a given reaction.
(24) A reviewer has asked about the effect of the pKa value of the
1,3-dicarbonyl compound upon the results. While this has not been
specifically interrogated, it is clear that this family of catalysts can be
successfully applied to a broad spectrum of ″active methylene″
substrates,7 ranging from cyanoacetate esters10d through ß-keto
esters10b to dialkyl malonates (pKa 9-14).10a
(11) As of the submission date of this manuscript, the best prices are
(R,R)-dpen, $291.55/100 g and (S,S)-dpen, $295.80/100 g (https://
www.ambeed.com, accessed 25 April 2021).
(12) Ghosh, S. K.; Lewis, K. G.; Kumar, A.; Gladysz, J. A. Syntheses
of Families of Enantiopure and Diastereopure Cobalt Catalysts
D e r i v e d f r o m T r i c a t i o n s o f t h e F o r m u l a [ C o -
(NH2CHArCHArNH2)3]3+. Inorg. Chem. 2017, 56, 2304−2320.
(13) (a) Ghosh, S. K.; Ganzmann, C.; Bhuvanesh, N.; Gladysz, J. A.
Werner Complexes with ω-Dimethylaminoalkyl Substituted Ethyl-
enediamine Ligands: Bifunctional Hydrogen-Bond-Donor Catalysts
for Highly Enantioselective Michael Additions. Angew. Chem., Int. Ed.
2016, 55, 4356−4360. (b) Ghosh, S. K.; Ganzmann, C.; Bhuvanesh,
N.; Gladysz, J. A. Werner-Komplexe mit ω-Dimethylaminoalkyl-
substitutierten Ethylendiaminliganden: bifunktionale H-Bru
nor-Katalysatoren fur hoch enantioselektive Michael-Additionen.
Angew. Chem. 2016, 128, 4429−4433.
̈
ckendo-
̈
(14) Ghosh, S. K.; Ganzmann, C.; Gladysz, J. A. Synthesis of a Series
of ω-Dimethylaminoalkyl Substituted Ethylenediamine Ligands for
use in Enantioselective Catalysis. Tetrahedron: Asymmetry 2015, 26,
1273−1280.
(15) For another chiral cobalt(III) hydrogen bond donor catalyst
incorporating a tertiary amine base, see Savel’yeva, T. F.; Khromova,
O. V.; Larionov, V. A.; Smol’yakov, A. F.; Fedyanin, I. V.; Belokon, Y.
N.; Maleev, V. I. Expanding the Family of Octahedral Chiral-at-Metal
Cobalt(III) Catalysts by Introducing Tertiary Amine Moiety into the
Ligand. Catalysts 2021, 11, 152. (article number)
(16) (a) Smidt, S. P.; Zimmermann, N.; Studer, M.; Pfaltz, A.
Enantioselective Hydrogenation of Alkenes with Iridium-PHOX
Catalysts: A Kinetic Study of Anion Effects. Chem. − Eur. J. 2004,
10, 4685−4693. (b) Appelhans, L. N.; Zuccaccia, D.; Kovacevic, A.;
Chianese, A. R.; Miecznikowski, J. R.; Macchioni, A.; Clot, E.;
Eisenstein, O.; Crabtree, R. H. An Anion-Dependent Switch in
Selectivity Results from a Change of C-H Activation Mechanism in
the Reaction of an Imidazolium Salt with IrH5(PPh3)2. J. Am. Chem.
Soc. 2005, 127, 16299−16311.
(25) Ghosh, S. K.; Ojeda, A. S.; Guerrero-Leal, J.; Bhuvanesh, N.;
Gladysz, J. A. New Media for Classical Coordination Chemistry:
Phase Transfer of Werner and Related Polycations into Highly
Nonpolar Fluorous Solvents. Inorg. Chem. 2013, 52, 9369−9378.
(26) Giernoth, R. Task-Specific Ionic Liquids. Angew. Chem., Int. Ed.
2010, 49, 2834−2839. Ionische Flussigkeiten fur Spezialaufgaben −
̈ ̈
von der Katalyse bis zur Analytik. Angew. Chem. 2010, 122, 2896−
2901, DOI: 10.1002/ange.200905981.
(17) There are many catalysts in which both cations and anions play
roles, particularly in the case of halide anions. Alternatively, simple
anions (such as acetate) may function as Brønsted bases. However,
the title claims of this paper involve counter anions to which
functionality has been added. For examples of halide anion
participation in the context of cationic cobalt(III) catalysts, see
Rulev, Y. A.; Larionov, V. A.; Lokutova, A. V.; Moskalenko, M. A.;
Lependina, O. L.; Maleev, V. I.; North, M.; Belokon, Y. N. Chiral
Cobalt(III) Complexes as Bifunctional Brønsted Acid-Lewis Base
Catalysts for the Preparation of Cyclic Organic Carbonates.
ChemSusChem 2016, 9, 216−222. and earlier papers cited therein
(18) Kabes, C. Q.; Maximuck, W. J.; Ghosh, S. K.; Kumar, A.;
Bhuvanesh, N.; Gladysz, J. A. Chiral Tricationic Tris(1,2-diphenyle-
(27) Some examples with pyridine-derived nitrogen donor atoms in
the cation can be found: Clarke, C. J.; Bui-Le, L.; Hallett, J. P.;
Licence, P. Thermally-Stable Imidazolium Dicationic Ionic Liquids
with Pyridine Functional Groups. ACS Sustainable Chem. Eng. 2020,
8, 8762−8772.
(28) When either dimethyl malonate (5a) or trans-ß-nitrostyrene
(6a) is titrated into these families of cobalt(III) catalysts, the NH 1H
NMR signals show regular downfield shifts. See Figure s3 of reference
10a and Figure 5 of Luu, Q. H.; Lewis, K. G.; Banerjee, A.;
Bhuvanesh, N.; Gladysz, J. A. The Robust, Readily Available
Cobalt(III) Trication [Co(NH2CHPhCHPhNH2)3]3+ is a Progenitor
of Broadly Applicable Chirality and Prochirality Sensing Agents.
Chem. Sci. 2018, 9, 5087−5099.
7770
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
(29) Crystal structures are available in which the nicotinate anion
4b− exhibits hydrogen bonding. Although the motif could very much
depend upon the hydrogen bond donor and also reflect crystal
packing forces, all structures we have examined exhibit hydrogen
bonding to the carboxyl group as opposed to the nitrogen atom. See
(a) Sridhar, B.; Ravikumar, K. Supramolecular hydrogen-bonded
networks in cytosinium nicotinate monohydrate and cytosinium
isonicotinate cytosine dihydrate. Acta Cryst. 2010, C66, o414−o417.
(b) Cole, L. B.; Holt, E. M. Calcium and Magnesium Complexation
with Nicotinic Acid Anion from Aqueous Media. [Ca-
(nicotinate)2(H2O)2](H2O)3 and [Mg(H2O)6](nicotinate)2(H2O)4;
C12H18CaN2O9 and C12H28MgN2O14. Inorg. Chim. Acta 1985, 108,
159−164. (c) Chen, C.; Chan, Z.-K.; Yeh, C.-W.; Chen, J.-D.
Synthesis, structures, and thermal properties of Cu(II) and Ni(II)
complexes containing diethylenetriamine and nicotinate ligands.
Struct. Chem. 2008, 19, 87−94. In this paper, refer to the nicotinate
counter anions in the Cu(II) complex 2
(30) Lange’s Handbook of Chemistry; 17th ed, Speight, J. G., Editor.
McGraw Hill: New York, 2017, Table 2.59.
(31) Evans, R. F.; Herington, E. F. G.; Kynaston, W. Determination
of Dissociation Constants of the Pyridine-Monocarboxylic Acids by
Ultra-Violet Photoelectric Spectrophotometry. Trans. Faraday Soc.
1953, 49, 1284−1292.
(32) Nagy, P. I.; Takács-Novák, K. Theoretical and Experimental
Studies of the Zwitterion ⇆ Neutral Form Equilibrium of Ampholytes
in Pure Solvents and Mixtures. J. Am. Chem. Soc. 1997, 119, 4999−
5006.
(33) This value is taken from SciFinder®, which gives predicted
acidities for a number of compounds.
(34) (a) Santiago, C. B.; Guo, J.-Y.; Sigman, M. S. Predictive and
mechanistic multivariate linear regression models for reaction
development. Chem. Sci. 2018, 9, 2398−2412. (b) Reid, J. P.;
Sigman, M. S. Holistic prediction of enantioselectivity in asymmetric
catalysis. Nature 2019, 571, 343−348.
(35) Fanourakis, A.; Docherty, P. J.; Chuentragool, P.; Phipps, R. J.
Recent Developments in Enantioselective Transition Metal Catalysis
Featuring Attractive Noncovalent Interactions Between Ligand and
Substrate. ACS Catal. 2020, 10, 10672−10714.
(36) (a) Ma, J.; Ding, X.; Hu, Y.; Huang, Y.; Gong, L.; Meggers, E.
Metal-templated chiral Brønsted base organocatalysis. Nature.
Commun. 2014, 5, 4531. (article number) (b) Huo, H.; Fu, C.;
Wang, C.; Harms, K.; Meggers, E. Chem. Commun. 2014, 50, 10409−
10411. (c) Hu, Y.; Zhou, Z.; Gong, L.; Meggers, E. Asymmetric aza-
Henry reaction to provide oxindoles with quaternary carbon
stereocenter catalyzed by a metal-templated chiral Brønsted base.
Org. Chem. Front. 2015, 2, 968−972. (d) Ding, X.; Lin, H.; Gong, L.;
Meggers, E. Enantioselective Sulfa-Michael Addition to α,ß-
Unsaturated γ-Oxoesters Catalyzed by a Metal-Templated Chiral
Brønsted Base. Asian J. Org. Chem. 2015, 4, 434−437. (e) Ding, X.;
Tian, C.; Hu, Y.; Gong, L.; Meggers, E. Tuning the Basicity of a
Metal-Templated Brønsted Base to Facilitate the Enantioselective
Sulfa-Michael Addition of Aliphatic Thiols to α,ß-Unsaturated N-
Acylpyrazoles. Eur. J. Org. Chem. 2016, 887−890.
(37) Larionov, V. A.; Yashkina, L. V.; Medvedev, M. G.; Smol’yakov,
A. F.; Peregudov, A. S.; Pavlov, A. A.; Eremin, D. B.; Savel’yeva, T. F.;
Maleev, V. I.; Belokon, Y. N. Henry Reaction Revisited. Crucial Role
of Water in an Asymmetric Henry Reaction Catalyzed by Chiral
NNO-Type Copper(II) Complexes. Inorg. Chem. 2019, 58, 11051−
11065.
(38) Review covering contributions by several groups: Cruchter, T.;
Larionov, V. A. Asymmetric catalysis with octahedral stereogenic-at-
metal complexes featuring chiral ligands. Coord. Chem. Rev. 2018, 376,
95−113.
(39) Watanabe, M.; Ikagawa, A.; Wang, H.; Murata, K.; Ikariya, T.
Catalytic Enantioselective Michael Addition of 1,3-Dicarbonyl
Compounds to Nitroalkenes Catalyzed by Well-Defined Chiral Ru
Amido Complexes. J. Am. Chem. Soc. 2004, 126, 11148−11149.
7771
https://doi.org/10.1021/acscatal.1c01883
ACS Catal. 2021, 11, 7762−7771