A Heteroleptic Dirhodium Catalyst for Asymmetric Cyclopropanation with α-Stannyl α-Diazoacetate. “Stereoretentive” Stille Coupling with Formation of Chiral Quarternary Carbon Centers

Article abstract of DOI:10.1002/anie.202004377

The heteroleptic dirhodium paddlewheel catalyst 7 with a chiral carboxylate/acetamidate ligand sphere is uniquely effective in asymmetric [2+1] cycloadditions with α-diazo-α-trimethylstannyl (silyl, germyl) acetate. Originally discovered as a trace impurity in a sample of the homoleptic parent complex [Rh₂((R)-TPCP)₄] (5), it is shown that the protic acetamidate ligand is quintessential for rendering 7 highly enantioselective. The -NH group is thought to lock the ensuing metal carbene in place via interligand hydrogen bonding. The resulting stannylated cyclopropanes undergo “stereoretentive” cross coupling, which shows for the first time that even chiral quarternary carbon centers can be made by the Stille–Migita reaction.

Full text of DOI:10.1002/anie.202004377

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Heteroleptic Dirhodium Catalyst for Asymmetric
Cyclopropanation with α-Stannyl α-Diazoacetate.Stereoretentive
Stille Coupling with Formation of Chiral Quarternary Carbon
Centers
Authors: Alois Fürstner and Fabio P. Caló
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202004377
Link to VoR: https://doi.org/10.1002/anie.202004377

10.1002/anie.202004377
Angewandte Chemie International Edition
A Heteroleptic Dirhodium Catalyst for Asymmetric Cyclopropanation with -Stannyl -Diazoacetate.
‘Stereoretentive’ Stille Coupling with Formation of Chiral Quarternary Carbon Centers
Fabio P. Caló and Alois Fürstner*
Abstract: The heteroleptic dirhodium paddlewheel catalyst 7 with a chiral carboxylate/acetamidate
ligand sphere is uniquely effective in asymmetric [2+1] cycloadditions with -diazo--trimethylstannyl
(silyl, germyl) acetate. Originally discovered as a trace impurity in a sample of the homoleptic parent
complex [Rh2((R)-TPCP)4] (5), it is shown that the protic acetamidate ligand is quintessential for
rendering 7 highly enantioselective. The –NH group is thought to lock the ensuing metal carbene in
place via interligand hydrogen bonding. The resulting stannylated cyclopropanes undergo
“stereoretentive” cross coupling, which shows for the first time that even chiral quarternary carbon
centers can be made by the Stille-Migita reaction.
Introduction
During the course of our investigations into (chiral) metal carbene complexes,1,2,3,4,5,6 we became aware
that reactions of silylated, germylated or stannylated -diazoacetate derivatives 1 largely fail to meet
the standards of modern asymmetric catalysis. Substrates of this type are easy to make on multigram
scale and safe to handle;7,8 the derived transition metal carbenes are known to be well-behaved
intermediates in cyclopropanation and C-H insertion reactions, to mention but a few.9,10,11,12 Yet, highly
enantioselective versions are basically unknown,13 except for a single report of an intramolecular
case.14 This methodological gap is all the more regrettable as the resulting products featuring an ester
and a metalloid center next to each other provide ample opportunity for downstream manipulation.
In this context, stannylated (silylated) cyclopropanes 2 are deemed particularly relevant (Scheme 1),15
not least because of the rapidly increasing demand of contemporary medicinal as well as natural
*
F. P. Caló, Prof. A. Fürstner
Max-Planck-Institut für Kohlenforschung
45470 Mülheim/Ruhr (Germany)
Email: fuerstner@kofo.mpg.de
Supporting Information and the ORCID identification number(s) for the authors of this article can be found under
#################
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
product chemistry for small-ring systems.16 Provided that the tertiary alkylstannane moiety of 2a (E =
Sn) can be engaged in cross coupling  which in itself is a highly challenging transformation  such
building blocks should open access to products 3 and surrounding chemical space that can be difficult
to reach otherwise.
Scheme 1. Conceptual outline
Results and Discussion
Catalyst Development. In a first attempt to meet the challenge, a series of standard chiral dirhodium
tetracarboxylate catalysts was screened in reactions with the readily available stannylated ester 1a (for
17,18
details, see the SI).8 [Rh2((R)-TPCP)4] (5)
gave the only notable “hit”, but the outcome proved
extremely erratic when different batches of this catalyst were used. This puzzling situation suggested
that minor impurities might massively interfere with the results.
Therefore we embarked into a more systematic investigation and prepared samples of this catalyst by
two different routes (Scheme 2): Method A reacts [Rh2(OAc)4] with acid 4 in refluxing chlorobenzene.
The ligand exchange is driven to completion by passing the high-boiling solvent through a Soxlet
extractor filled with K2CO3,19 which traps the released HOAc. All samples of 5 prepared in this manner
were essentially pure (NMR, HPLC) but invariably inactive in the model reaction (Table 1, entry 1). The
structure of 5 in the solid state shows quasi-C2 symmetric binding sites about the Rh atoms, which
might be too narrow to accommodate the stannylated diazoester (Figure 2).20 Method B reacts
Na4[Rh2(CO3)4]2.5(H2O) with acid 4 in boiling water according to the literature.17 In this case, the
catalyst samples were slightly less clean as evident from a representative HPLC trace which shows
several minor impurities in addition to a small amount of free ligand 4 (Figure 1). Such samples led to
highly variable but occasionally excellent ee’s; after HPLC separation, the pure sample of 5 (> 99%)
again failed to catalyze the test reaction (entries 2/3). Three additional fractions were collected,
delivering minuscule amounts of unknown rhodium-containing species: fraction 2 decomposed when
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
kept in CD3CN solution, but the two other samples could be tested despite the minute available
quantities. Fraction 4 gave only modest asymmetric induction, whereas the seemingly negligible
fraction 3 furnished the stannylated cyclopropanes cis-2aa and trans-2aa with  95% ee in what
appeared to be a fast and clean transformation (entries 4/6).
Scheme 2. Formation of the homoleptic complex 5 and by-products derived from impurities in
different samples of Na4Rh2(CO3)4, cf. Text; targeted syntheses of heteroleptic siblings
At this point, incomplete replacement of the carbonate ligands of Na4[Rh2(CO3)4] by the chiral acid 4
was deemed the most plausible explanation for the formation of these minor by-products. To the best
of our knowledge, only a single chiral heteroleptic paddlewheel complex comprising hydrogen
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
carbonate groups is known in the literature; its exact structure, however, is unclear and the catalytic
performance not fully convincing.21,22 This specific case notwithstanding, it seemed reasonable that a
heteroleptic dirhodium complex might outperform its homoleptic cousin 5 in certain applications; a
few such cases are known in the literature.23 For the lack of good and broadly applicable strategies for
the controlled introduction of two or more different (chiral) ligands about the Rh2-core, systematic
explorations of dirhodium complexes with mixed ligand spheres remain difficult. In line with this
notion, our attempts at partial substitution of the carbonate ligands of Na4[Rh2(CO3)4] by 4 basically
met with failure, despite considerable experimentation: 5 was the main product independent of the
chosen ligand/rhodium ratio and the experimental conditions. The reverse approach, that is partial
replacement of the chiral ligands in 5 on reaction with various carbonate sources, was equally
unrewarding.
Figure 1. Representative HPLC trace of a sample of 5 (= fraction 5) prepared by method B; fraction 1 is
unreacted acid 4; for the other fractions, see Text
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
Figure 2. Structure of [Rh2((R)-TPCP)4]2MeCN (52MeCN) in the solid state; coordinated and solute
MeCN is removed and H-atoms are omitted for clarity. The entire structure is shown in the SI, which
also contains a second crystal structure of the same complex in a different space group.
Table 1. Screening of different catalysts (for the full list, see the SI)[a]
Entry
Catalyst
ee (%)
cis-2aa
Yield (%)[b]
trans-2aa
1
2
3
4
5
6
7
8
5 (method A)
5 (crude, method B)
5 (> 99% pure, method B)
fraction 4
---
up to 92
---
54
53
96
95
30
---
up to 96
---
76
76
97
97
53
NR
up to quant.[c]
NR
n. d.
56
n.d.
76
n.d.
8
fraction 3
7
10
9
93
95
44
10
7
39
48
[c]
[a] in all entries, the cis:trans ratio was  1:1; [b] yield of isolated material, unless stated otherwise;
NMR yield; NR = no reaction; n. d. = not determined
Next, we turned out attention to common dirhodium tetracarboxylates as the point of departure. A
first important step was taken when we learnt that only three of the four acetate units of [Rh2(OAc)4]
are substituted by 4 when three equivalents of 4 are employed and the reaction is performed in boiling
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
toluene (Soxlet method). Unexpectedly, the heteroleptic complex 8 thus formed in 77% yield proved
identical with fraction 4, which is catalytically active but only modestly selective. This result implied
that the Na4[Rh2(CO3)4] sample used to make 5 must have contained some rhodium acetate impurity.24
Three-fold ligand exchange also worked well with [Rh2(tfa)4] (9, tfa = trifluororacetate), provided that
ethyl acetate25 or, preferentially, the higher boiling tert-butyl acetate was used as the solvent to furnish
10 in 65% yield.26 Complex 10 was then treated with [Et4N][HCO3] in MeCN: despite the presumably
better leaving group properties of trifluoroacetate, the reaction was again inefficient and furnished
two new products, which correspond to fraction 2 and the sought-after fraction 3. The instability of
the former in solution (see above) precluded full characterization; yet, a resonance in the 13C NMR
spectrum at C = 165.2 ppm, a cluster of indicative MS signals,27 and the fact that the material responds
to treatment with acid/base render the assignment as the heteroleptic mono-hydrogencarbonate
complex 6 highly likely.
The HRMS data of the relevant complex contained in fraction 3 were suggestive: one of the recorded
signals at m/z = 1203.20886 matched the composition [C68H55O7NRh2] very well, which can be
interpreted as {[Rh2((R)-TPCP)3] + MeCN + OH]}. If traces of MeCN had been contained in the sample
of Na4[Rh2(CO3)4], it was almost certainly ligated to the axial sites at Rh. As this renders the nitrile group
susceptible to base, trace acetamide could have been generated in situ within the first coordination
sphere (A in Scheme 2), which might replace the trifluoroacetate group of 10 and give rise to the
enigmatic “fraction 3”.28,29,30 With this idea in mind, we pursued two targeted approaches to the
presumed heteroleptic acetamidate complex [Rh2((R)-TPCP)3(acam)] (7): to this end, [Et4N][HCO3] was
replaced by Bu4NOH in the reaction with 10 in MeCN, which indeed raised the yield of 7 to 22% after
ordinary flash chromatography. Alternatively, treatment of 10 with acetamide in refluxing
chlorobenzene (Soxhlet method)19 furnished 7 in 31% yield.31
Scope. Control experiments confirmed that complex 7 is indeed an active and highly enantioselective
catalyst for the model cyclopropanation reaction of 1a with 4-methoxystyrene; the missing
diastereoselectivity is somehow compensated by the ease of separation of cis-2aa (97% ee) and trans-
2aa (95% ee) by flash chromatography. The examples compiled in Figure 3 allow the scope of the
reaction to be assessed: styrene derivatives afforded the desired stannylated cyclopropane derivatives
in generally excellent optical purity, independent of whether they are electron-rich or -poor. Enamides,
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
enol esters and enol ethers are equally suitable substrates: for the three different functional groups,
the resulting products 2ad-2af are deemed particularly interesting building blocks. Even though this
study was mainly focused on the preparation of stannylated cyclopropanes as our premier candidates
for downstream functionalization, it was found that the corresponding silylated and germylated
products 2ba and 2ca are formed with similarly high ee’s. The standard donor/acceptor carbene
precursor p-MeOC6H4C(N2)COOEt, however, gave cyclopropane 13 with only 57% ee; gratifyingly, the
high reactivity of the new catalyst allowed the temperature to be lowered to 78°C and the outcome
to be improved to respectable 84% ee.32 The parent ethyl diazoacetate, in contrast, was found to react
well but furnished 14 with poor selectivity. These preliminary data suggest that a moderately bulky
substituent at the carbene site is mandatory in order to reach high levels of asymmetric induction in
reactions catalyzed by 7;33 this aspect is subject to further investigation in our laboratory.
Figure 3. Substrate scope (only the trans-isomer is drawn); unless stated otherwise, all reactions were
performed with catalyst 7 (1 mol%) in CH2Cl2 at ambient temperature; [a] at 78°C
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
The sense of induction was rigorously established by X-ray diffraction for two independent cases.
Statistically significant absolute structure parameters were obtained that allowed the configuration of
the stannylated cyclopropanes 2ac (see the SI) and 2ad (Figure 4) to be determined, which derive from
an electron-deficient and an electron-rich alkene, respectively. In both products the substituents on
the three-membered ring have the same orientation in space, even though the correct denomination
is different because of the formalism of the CIP-notation ((1R,2S)-2ac but (1R,2R)-2ad). All other
compounds were assigned by analogy.32
Figure 4. Structure of compound (1R,2R)-2ad in the solid state; hydrogen atoms omitted for clarity;
only one of four independent molecules in the unit cell is shown; for the whole structure, see the SI
Mechanistic Aspects. As yet another important prelude for a mechanistic discussion, the exact role of
the acetamidate ligand in 7 was examined. As already mentioned above, the heteroleptic complex 8
carrying an acetate reacts well but is much less enantioselective (Table 1, entry 5). The same disparate
behavior was observed for the pair 10 and 11 comprising a trifluoroacetate and a trifluoroacetamidate,
respectively: only the latter proved to be highly enantioselective (entries 8/9). Equally relevant is the
control experiment with complex 12, which differs from 7 in that its acetamide ligand is N-
methylated:34 the level of asymmetric induction is marginal (entry 10). Taken together, these results
suggest that the heteroleptic character accounts for the reactivity of the complexes, but the protic
ligand plays a quintessential role in the enantiodeterming step.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
This information has to guide the inspection of the structure of [Rh2((R)-TPCP)3(acam)] (7) in the solid
state. Crystals of good quality were obtained for an adduct carrying two molecules of DMF at the axial
sites (Figure 5). In comparison with the structure of the homoleptic parent complex [Rh2((R)-TPCP)4]
(5) (Figure 2), it is apparent that the incorporation of one small ligand leads to a significantly wider
binding site. Since the size of an oxygen atom and an –NH group are similar, the binding pockets of
complexes 8 and 10 are almost certainly akin. Therefore all heteroleptic complexes should be able to
accommodate fairly bulky incoming diazo derivatives,33 whereas the homoleptic complex 5 (Figure 2)
is not; this notion is in accord with the experimental reactivity data.
Figure 5. Structure of [Rh2((R)-TPCP)3(acam)]2DMF (72DMF) in the solid state in two different
orientations; the axial DMF ligands were removed for a better view onto the binding site about Rh1,
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
to which the N-atom of the acetamidate ligand is coordinated; hydrogen atoms are omitted for clarity.
The full structure is contained in the SI
The –NH group constitutes the critical determinant for high selectivity. The effect that it imparts,
however, cannot be steric in origin: Figure 5 shows that Rh1 and Rh2 of 7 are both well accessible. A
purely electronic argument is equally unlikely: in consideration of the well-founded trend that amidate
ligands tend to render dirhodium catalysts less reactive (but often very selective),29 one might assume
that carbene formation occurs preferentially or exclusively at Rh2 surrounded by the four O-atoms. If
this were the case, however, complexes 7 and 12 differing only in the substituent on the acetimidate
N-atom (NH versus NMe) should lead to similar levels of asymmetric induction; experimentally, the
outcome is dramatically different (Table 1, entries 7/10). The fact that 12 with the more bulky N-
substituent is also chemically somewhat less effective also speaks for diazo-decomposion occuring at
the N-containing binding site. Moreover, if the reaction takes place at an all-oxygen coordinated Rh-
center, the (trifluoro)acetate-containing complexes 8 and 10 comprising two essentially equivalent
such binding sites should be highly selective too, which is clearly not the case.
Figure 6. Possible rationale: interligand hydrogen bonding
These facts and arguments suggest that Rh1 is the relevant reaction center that reigns the asymmetric
process. We assume that the NH-group plays an active role that outweighs any electronic handicap:
the protic ligand might engage the diazocarbonyl derivative 1 in intermolecular hydrogen bonding and
hence recruite the substrate to this site. Once it is bound and nitrogen extruded, the then
intramolecular hydrogen bonding array locks the resulting carbene in place within the chiral binding
pocket, such that it eclipses the ORhN axis (Figure 6).35,36 Under this proviso, however, the fairly
bulky Me3E- group (E = Si, Ge, Sn) might force the top phenyl ring, which protrudes over the binding
10
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
site of 7 (Figure 5), to relocate and change the chiral microenvironment. Therefore it seems prudent
at this point not to over-interpret the structure of the precatalyst 7 in the solid state: in any case, it is
non-obvious from this X-ray structure which enantiotopic face of the carbene is exposed to the
reaction partner and which one is shielded. Moreover, it is unclear in this particular case whether the
alkene approaches the electrophilic carbene alongside the R3E-substitutent or the ester;37 these and
related aspects are subject to ongoing investigations. The largely missing diastereoselectivity,
however, means that 7 fails to determine the orientation of the incoming olefin, which is plausible for
a catalyst with a fairly wide binding site.
Scheme 3. a) Pd(dba)2 (10 mol%), JackiePhos (18) (20 mol%, CuCl, KF, THF, 60°C
Cross Coupling. Formation of Chiral Quarternay Carbon Centers. Bifunctional cyclopropanes of type
2 bearing a metalloid center adjacent to an ester open many possibilities for downstream
functionalization. Even though Stille coupling of tert-alkylstannanes with formation of stereogenic
quarternary carbon had been unknown at the outset of our investigation,15,38 it seemed promising to
pursue this tantalizing prospect in view of the special bonding situation in cyclopropanes (Walsh
orbitals); in case of 2 one can also think of this transformation as an -arylation process.39 However,
the generation of tin enolates by facile CO migration of the Me3Sn-group with concomitant
planarization of the chiral center must be strictly avoided; premature protodestannation is yet another
serious threat. The prototype examples shown in Scheme 3 illustrate that these challenges can indeed
be met using conditions previously developed for the cross coupling of secondary azastannatranes:40
cis-2aa and trans-2aa were coupled with iodobenzene in appreciable yield and perfect integrity of the
stereocenter as manifest in a dr > 20:1 (NMR, HPLC) in both cases.41 While product 15 could certainly
11
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
be made directly by asymmetric cyclopropanation via a convenitonal donor/acceptor carbene,1 the
new cross-coupling approach provides additional opportunities as illustrated by the formation of 16
and 17 comprising a terminal alkene and an aldehyde, respectively: either functionality is incompatible
with a transient carbene intermediate. As many more such examples reaching beyond the traditional
scope can be envisaged, these promising results mark just the starting point of a more comprehensive
study in our laboratory.
Conclusion
The present report rigorously exemplifies that the switch from a homoleptic to a heteroleptic ligand
sphere about a dirhodium core can unlock entirely new reactivity and selectivity in carbene chemistry;
the effect per se is known, but equally striking cases are exceedingly rare.Fehler! Textmarke nicht definiert. For
more systematic forays into this promising area, however, innovative new concepts and techniques
are deemed vital that allow heteroleptic complexes to be crafted in a (more) rational and productive
manner. At the same time, catalyst 7 is thought to showcase the power of interligand hydrogen
bonding in catalysis.36 Finally, we note that the conclusions of this detective story may arguably be of
conceptual relevance in that the success hinges on a ligand that plays an active role rather than being
solely a passive divider of (chiral) space.42 This notion guides our future investigations in the field.
Acknowledgements
Generous financial support by the Fonds der Chemischen Industrie (Kekulé stipend to F. P. C.) and the
Max-Planck-Society is gratefully acknowledged. We thank Mr. Nils Nöthling, Mr. J. Rust, Dr. R. Goddard
and Prof. C. W. Lehmannn for solving the X-ray structures and all analytical departments of our
Institute for excellent support.
12
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
For the Table of Contents
Less is more: Take one chiral ligand from a standard paddlewheel dirhodium tetracarboxylate complex
out and replace it by acetamidate and you switch on reactivity and harness outstanding
enantioselectivity in cyclopropanations with -stannylated--diazoesters. Even though the products
contain a tert-alkylstannane moiety, “stereoretentive” Stille cross coupling is shown to be possible.
Keywords: asymmetric catalysis  cyclopropanation  heteroleptic complexes  quarternary chiral
centers  rhodium carbenes  Stille coupling
13
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
1
General reviews: a) M. P. Doyle, D. C. Forbes, Chem. Rev. 1998, 98, 911-935; b) H. M. L. Davies,
R. E. J. Beckwith, Chem. Rev. 2003, 103, 2861-2904; c) H. Lebel, J.-F. Marcoux, C. Molinaro, A. B.
Charette, Chem. Rev. 2003, 103, 977-1050; d) Z. Zhang, J. Wang, Tetrahedron 2008, 64, 6577-
6605; e) A. Ford, H. Miel, A. Ring, C. N. Slattery, A. R. Maguire, M. A. McKervey, Chem. Rev. 2015,
115, 9981-10080; f) A. DeAngelis, R. Panish, J. M. Fox, Acc. Chem. Res. 2016, 49, 115-127.
a) C. Werlé, R. Goddard, P. Philipps, C. Farès, A. Fürstner, Angew. Chem. Int. Ed. 2016, 55, 10760-
10765; b) C. Werlé, R. Goddard, P. Philipps, C. Farès, A. Fürstner, J. Am. Chem. Soc. 2016, 138,
3797-3805; c) C. Werlé, R. Goddard, A. Fürstner, Angew. Chem. Int. Ed. 2015, 54, 15452-15456;
d) D. J. Tindall, C. Werlé, R. Goddard, P. Philipps, C. Farès, A. Fürstner, J. Am. Chem. Soc. 2018,
140, 1884-1893.
2
3
4
5
a) L. R. Collins, S. Auris, R. Goddard, A. Fürstner, Angew. Chem. Int. Ed. 2019, 58, 3557-3561; b)
L. R. Collins, M. van Gastel, F. Neese, A. Fürstner, J. Am. Chem. Soc. 2018, 140, 13042-13055.
M. Buchsteiner, L.- Martinez-Rodriquez, P. Jerabek, I. Pozo, M. Patzer, N. Nöthling, C. W.
Lehmann, A. Fürstner, Chem. Eur. J. 2020, 26, 2509-2525.
a) G. Seidel, A. Fürstner, Angew. Chem. Int. Ed. 2014, 53, 4807-4811; b) G. Seidel, B. Gabor, R.
Goddard, B. Heggen, W. Thiel, A. Fürstner, Angew. Chem. Int. Ed. 2014, 53, 879-882; c) A. G.
Tskhovrebov, J. Lingnau, A. Fürstner, Angew. Chem. Int. Ed. 2019, 58, 8834-8838; d) G. Seidel, R.
Mynott, A. Fürstner, Angew. Chem. Int. Ed. 2009, 48, 2510-2513; f) W. Debrouwer, A. Fürstner,
Chem. Eur. J. 2017, 23, 4271-4275; g) A. G. Tskhovrebov, R. Goddard, A. Fürstner, Angew. Chem.
Int. Ed. 2018, 57, 8089-8094.
6
a) A. Fürstner, J. Am. Chem. Soc. 2019, 141, 11-24; b) A. Guthertz, M. Leutzsch, L. M. Wolf, P.
Gupta, S. M. Rummelt, R. Goddard, C. Farès, W. Thiel, A. Fürstner, J. Am. Chem. Soc. 2018, 140,
3156-3168; c) T. Biberger, C. P. Gordon, M. Leutzsch, S. Peil, A. Guthertz, C. Copéret, A. Fürstner,
Angew. Chem. Int. Ed. 2019, 58, 8845-8850; d) S. Peil, A. Guthertz, T. Biberger, A. Fürstner,
Angew. Chem. Int. Ed. 2019, 58, 8851-8856; e) S. Peil, A. Fürstner, Angew. Chem. Int. Ed. 2019,
58, 18476-18481.
14
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
7
a) T. Allspach, H. Gümbel, M. Regitz, J. Organomet. Chem. 1985, 290, 33-39; b) U. Schöllkopf, B.
Banhidai, H.-U. Scholz, Liebigs Ann. Chem. 1972, 761, 137-149.
8
a) J. Lorberth, J. Organomet. Chem. 1968, 15, 251-253; b) J. Lorberth, J. Organomet. Chem. 1971,
27, 303-325; c) J. Lorberth, S.-H. Shin, H. Donath, S. Wocadlo, W. Massa, J. Organomet. Chem.
1991, 407, 167-171.
9
Y. Hari, T. Aoyama, T. Shioiri, Science of Synthesis Knowledge Updates 2010, 4, 60-68.
R. E. Gawley, S. Narayan, Chem. Commun. 2005, 5109-5111.
10
11
a) V. Gettwert, F. Krebs, G. Maas, Eur. J. Org. Chem. 1999, 1213-1221; b) G. Maas, S. Bender,
Synthesis 1999, 1175-1180; c) G. Maas, M. Gimmy, M. Alt, Organometallics 1992, 11, 3813-3820.
a) S. P. Marsden, W.-K. Pang, Tetrahedron Lett. 1998, 39, 6077-6080; b) S. N. Kablean, S. P.
Marsden, A. M. Craig, Tetrahedron Lett. 1998, 39, 5109-5112.
12
13
14
15
16
A. Ghanem, F. Lacrampe, H. Y. Aboul-Enein, V. Schurig, Monatsh. Chem. 2005, 136, 1205-1219.
S. Inoue, K. Nagatani, H. Tezuka, Y. Hoshino, M. Nakada, Synlett 2017, 28, 1065-1070.
Review on cyclopropylstannanes: M. Rubina, V. Gevorgyan, Tetrahedron 2004, 60, 3129-3159.
a) T. T. Talele, J. Med. Chem. 2016, 59, 8712-8756; b) C. Ebner, E. M. Carreira, Chem. Rev. 2017,
117, 11651-11679.
17
For this catalyst class, see the following for leading references: a) C. Qin, V. Boyarskikh, J. H.
Hansen, K. I. Hardcastle, D. G. Musaev, H. M. L Davies, J. Am. Chem. Soc. 2011, 133, 19198-19204;
b) K. Liao, S. Negretti, D. G. Musaev, J. Bacsa, H. M. L. Davies, Nature 2016, 533, 230-234.
The catalyst with this exact ligand set was used in: a) C. Qin, H. M. L. Davies, J. Am. Chem. Soc.
2013, 135, 14516-14519; b) Z. Yu, A. Mendoza, ACS Catal. 2019, 9, 7870-7875.
M. P. Doyle, W. R. Winchester, M. N. Protopopova, A. P. Kazala, L. J. Westrum, Org. Synth. 1998,
73, 13.
18
19
20
21
5 can crystallize in two different space groups; both structures are contained in the SI.
It is unknown if the phos ligands in [Rh2((+)-phos)2(HCO3)2]5H2O ((+)-PhosH = (S)-(+)-1,1’-
binaphthyl-2,2’-diyl hydrogen phosphate) are cis or trans to each other; applications to different
asymmetric diazocarbonyl reactions led to ee’s of 9-60%, see: N. McCarthy, M. A. McKervey, T.
Ye, M. McCann, E. Murphy, M. P. Doyle, Tetrahedron Lett. 1992, 33, 5983-5986.
15
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
22
An achiral heteroleptic formiate/carbonate complex is mentioned without any data in: R. N.
Shchelokov, A. G. Maiorova, O. M. Evstaf’eva, G. N. Emel’yanova, Zhurnal Neorganicheskoi
Khimii 1977, 22, 1414-1416 (engl. translation: Russ. J. Inorg. Chem. 1977, 22, 770-771).
For leading studies illustrating the synthetic challenges and potential promise of heteroleptic
complexes, see: a) F. Estevan, P. Lahuerta, J. Pérez-Prieto, S.-E. Stiriba, M. A. Ubeda, Synlett
1995, 1121-1122; b ) M. P. Doyle, M. Yang, H.-M. Gau, E. C. Blossey, Org. Lett. 2003, 5, 561-
563;c) Y. Lou, M. Horikawa, R. A. Kloster, N. A. Hawryluk, E. J. Corey, J. Am. Chem. Soc. 2004,
126, 8916-8918; d) Y. Lou, T. P. Remarchuk, E. J. Corey, J. Am. Chem. Soc. 2005, 127, 14223-
14230; e) B. H. Brodsky, J. Du Bois, Chem. Commun. 2006, 4715-4717; f) K. Takeda, T. Oohara,
M. Anada, H. Nambu, S. Hashimoto, Angew. Chem. Int. Ed. 2010, 49, 6979-6983; g) R.
Sambasivan, Z. T. Ball, J. Am. Chem. Soc. 2010, 132, 9289-9291; h) D. T. Boruta, O. Dmitrenko, G.
P. A. Yap, J. M. Fox, Chem. Sci. 2012, 3, 1589-1593; i) V. N. G. Lindsey, A. B. Charette, ACS Catal.
2012, 2, 1221-1225; j) R. A. Panish, S. R. Chintala, J. M. Fox, Angew. Chem. Int. Ed. 2016, 55,
4983-4987; k) C.-J. Yoo, D. Rackl, W. Liu, C. B. Hoyt, B. Pimentel, R. P. Lively, H. M. L. Davies, C.
W. Jones, Angew. Chem. Int. Ed. 2018, 57, 10923-10927.
23
24
25
26
Na4[Rh2(CO3)4] is best made from [Rh2(OAc)4]: a) G. H. P. Roos, M. A. McKervey, Synth. Commun.
1992, 22, 1751-1756; see also: b) C. R. Wilson, H. Taube, Inorg. Chem. 1975, 14, 405-409.
J. Liu, Y. Xu, P. B. Groszewicz, M. Brodrecht, C. Fasel, K. Hofmann, X. Tan, T. Gutmann, G.
Buntkowsky, Catal. Sci. Technol. 2018, 8, 5190-5200.
The rest of the material (33%) was a mixture of cis- and trans-[Rh2(TFA)2((R)-TPCP)2], from which
a sample of the pure trans-isomer was obtained by crystallization; the X-ray structure of this
complex is contained in the SI. Reaction monitoring by HPLC showed that trans-[Rh2(TFA)2((R)-
TPCP)2] is more rapidly formed from 9 and 4 than the cis-isomer and also converts somewhat
more rapidly into 13 (see the SI). This observation might be useful for future studies into
heteroleptic dirhodium complexes; compare: C. J. Welch, Q. Tu, T. Wang, C. Raab, P. Wang, X.
Jia, X… Bu, D. Bykowski, B. Hohenstaufen, M. P. Doyle, Adv. Synth. Catal. 2006, 348, 821-824.
m/z = 1186 [Rh2((R)-TPCP)3]+/MeCN]; 1265 [Rh2((R)-TPCP)3(HCO3)]+/MeCN/H2O]; 1268 [Rh2((R)-
TPCP)3]+/3 MeCN]; 1306 [Rh2((R)-TPCP)3(HCO3)]+/2 MeCN/H2O]
27
16
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
28
For a study on the stepwise ligand exchanges upon reaction of Rh2(OAc)4 with molten acetamide
and an insightful discussion of the effects that amidate ligands entail, see: M. Q. Ahsan, I. Bernal,
J. L. Bear, Inorg. Chem. 1986, 25, 260-265.
29
30
For a review on dirhodium carboxamidates, see: M. P. Doyle, J. Org. Chem. 2006, 71, 9253-9260.
Chiral and achiral homoleptic dirhodium-tetra-carboxamidate complexes are widely used, with
[Rh2(MEPY)4] and dirhodium caprolactamate as the most prominent examples, see ref. 19 and
the following for leading references: a) M. P. Doyle, W. R. Winchester, J. A. A. Hoorn, V. Lynch,
S. H. Simonsen, R. Ghosh, J. Am. Chem. Soc. 1993, 115, 9968-9978; b) A.J. Catino, R. E. Forslund,
M. P. Doyle, J. Am. Chem. Soc. 2004, 126, 13622-13623.
31
[Rh2((R)-TCPC)2(acam)2] is also formed, which was removed by HPLC, see the SI; immobilization
might be a way to avoid such by-products and hence improve the yield of the desired
heteroleptic complex.
32
33
34
Compound 13 was assigned by comparison with authentic material, see ref. 3a
Me3E- is a good compromise, since larger R3E-groups entail low reactivity, see the SI
The preparation of 12 was very low yielding but not optimized as this complex proved
incompetent; the difficulties in making dirhodium complexes with N-substituted acyclic
carboxamidate ligands are discussed in ref. 29
35
H-bonding was shown to determine binding of different dirhodiumcarboxamidate complexes to
nucleobases and DNA, see: a) S. U. Dunham, T. S. Remaley, B. S. Moore, D. L. Evans, S. U.
Dunham, Inorg. Chem. 2011, 50, 3458-3463; b) K. Aoki, M. A. Salam, Inorg. Chim. Acta 2002, 339,
427-437; H-bonding was also proposed to play a role in carbene CN insertions, see: c) S. Harada,
M. Kono, T. Nozaki, Y. Menjo, T. Nemoto, Y. Hamada, J. Org. Chem. 2015, 80, 10317-10333.
For other cases in which interligand hydrogen bonding was recently shown to be selectivity-
determining, see ref. 6 and the following: a) S. M. Rummelt, K. Radkowski, D.-A. Roşca, A.
Fürstner, J. Am. Chem. Soc. 2015, 137, 5506-5519; b) D.-A. Roşca, K. Radkowski, L. M. Wolf, M.
Wagh, R. Goddard, W. Thiel, A. Fürstner, J. Am. Chem. Soc. 2017, 139, 2443-2455; c) S. M.
Rummelt, G. Cheng, P. Gupta, W. Thiel, A. Fürstner, Angew. Chem. Int. Ed. 2017, 56, 3599-3604;
corrigendum: Angew. Chem. Int. Ed. 2017, 56, 5652; d) X. Mo, A. Letort, D.-A. Rosca, K. Higashida,
36
A. Fürstner, Chem. Eur. J. 2018, 24, 9667-9674.
17
This article is protected by copyright. All rights reserved.

10.1002/anie.202004377
Angewandte Chemie International Edition
37
The crystal structures of prototypical donor/acceptor dirhodium carbene complexes showed
that the ester is orthogonal to the carbene to minimize orbital overlap. This orientation
translates into stereoelectronic control over the trajectory of the incoming styrene, which is
forced to approach alongside the donor substituent (typically an arene); /-interactions help
to position the alkene substrate, see ref. 1 and literature cited therein. If a H-bonding array as
shown in Figure 6 is operative in the present case, the ester is held coplanar with rather than
orthogonal to the carbene center; therefore the whole scenario might change. Moreover, an
R3E-group as the donor-substituent is more space-filling than a flat arene and cannot entertain
/-interactions either.
38
39
For pertinent reviews on stereoselective (Stille) cross coupling, see: a) R. Jana, T. P. Pathak, M.
S. Sigman, Chem. Rev. 2011, 111, 1417-1492; b) E. C. Swift, E. R. Jarvo, Tetrahedron 2013, 69,
5799-5817; c) C.-Y. Wang, J.Derose, M. R. Biscoe, Chem. Sci. 2015, 6, 5105-5113.
For racemic -arylations of cyclopropyl nitriles, see: B. A. Wright, M. J. Ardolino, J. Org. Chem.
2019, 84, 4670-4679.
40
41
L. Li, C.-Y. Wang, R. Huang, M. R. Biscoe, Nature Chem. 2013, 5, 607-612.
The observed “retention“ refers to the orientation of the pertinent groups in space: note,
however, that cis-2aa formally leads to trans-15, cf. Scheme 3
42
A related concept was pursued by Fox and cowokers, who used hetereoleptic complexes
incorporating one ligand with a large aromatic surface to harness attractive non-covalent /
and/or CH/-interactions between ligand and substrate, cf. ref. 1f, Fehler! Textmarke nicht
definiert.h
18
This article is protected by copyright. All rights reserved.