15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023].

See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DOI: 10.1002/chem.202000081 Full Paper

& Asymmetric Catalysis

Catalytic Asymmetric Fluorination of Copper Carbene Complexes:

Preparative Advances and a Mechanistic Rationale
Michael Buchsteiner,[a] Luis Martinez-Rodriguez,[a] Paul Jerabek,[a, b] Iago Pozo,[a]
Michael Patzer,[a] Nils Nçthling,[a] Christian W. Lehmann,[a] and Alois Ferstner*[a]

Abstract: The Cu-catalyzed reaction of substituted a-diazo- carbene intermediate. This unusual step is followed by 1,2-
esters with fluoride gives a-fluoroesters with ee values of up fluoride shift; for this migratory insertion to occur, the car-
to 95 %, provided that chiral indane-derived bis(oxazoline) li- bene must rotate about the Cu@C bond to ensure orbital
gands are used that carry bulky benzyl substituents at the overlap. The directionality of this rotatory movement within
bridge and moderately bulky isopropyl groups on their core. the C2-symmetric binding site determines the sense of in-
The apparently homogeneous solution of CsF in C6F6/hexa- duction. This model is in excellent accord with the absolute
fluoroisopropanol (HFIP) is the best reaction medium, but configuration of the resulting product as determined by X-
CsF in the biphasic mixture CH2Cl2/HFIP also provides good ray diffraction using single crystals of this a priori wax-like
results. DFT studies suggest that fluoride initially attacks the material grown by capillary crystallization.
Cu- rather than the C-atom of the transient donor/acceptor

Introduction foray in this direction (Scheme 1).[9, 10] Specifically, it was shown
that a-diazoesters such as 1 a, on treatment with catalytic
The incorporation of fluorine into active pharmaceutical ingre- amounts of [Cu(MeCN)4]PF6 (or [Cu(OTf)]2·PhMe) and L1 as the
dients (API’s) or agrochemicals often entails substantial advan- preferred ligand, react with excess KF in a biphasic mixture
tages in chemical and/or biological terms.[1] (Chiral) a-fluoro
carbonyl compounds represent a privileged motif and are
therefore prominently featured in API’s;[1, 2] at the same time,
they constitute valuable building blocks for further elaboration.
Most methods for their synthesis invoke the carbonyl com-
pound as (masked) enolate or enamine; the price to pay is the
need for an electrophilic fluorine source as the reaction part-
ner.[3–5] The alternative approach of using an “umpoled” sub- Scheme 1. Lead finding of a copper catalyzed formation of a-fluoroesters.
strate in combination with an ordinary fluoride salt is currently
less developed, despite potential chemical virtues and practical
advantages.[6–8] comprised of hexafluoroisopropanol (HFIP) and 1,2-dichloro-
A recent report on the fluorination of highly electrophilic ethane (DCE) at 40 8C to give the corresponding a-fluoroester
copper carbenes generated in situ provides an interesting 2 a in good yield. Despite the excellent track record of chiral
bis(oxazolines) (BOX) and related ligands in asymmetric cataly-
sis,[11, 12] however, only poor enantioselectivity (, 31 % ee) was
[a] M. Buchsteiner, Dr. L. Martinez-Rodriguez, Dr. P. Jerabek, I. Pozo, M. Patzer,
N. Nçthling, Prof. C. W. Lehmann, Prof. A. Ferstner reached. Use of a monodentate chiral phosphoramidite in lieu
Max-Planck-Institut fer Kohlenforschung of L1 entailed higher optical purity (86 % ee) at the expense of
45470 Melheim/Ruhr (Germany) the chemical yield, which dropped to only 12 %.[9]
E-mail: [email protected]
[b] Dr. P. Jerabek
Present Address: Nanotechnology Department
Helmholtz-Zentrum Geesthacht Results and Discussion
21502 Geesthacht (Germany)
Reaction optimization and scope
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under: As part of our studies into structure, bonding and reactivity of
https://doi.org/10.1002/chem.202000081.
organotransition metal carbene complexes,[13–18] we sought to
T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
improve on these lead findings. We were fully apprehensive
This is an open access article under the terms of the Creative Commons At-
tribution License, which permits use, distribution and reproduction in any that the small size of the fluoride ion constitutes a formidable
medium, provided the original work is properly cited. and inherent challenge for asymmetric catalysis; moreover, any

Chem. Eur. J. 2020, 26, 2509 – 2515 2509 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

uncatalyzed and hence racemic background reaction had to

Table 1. Ligand screening and optimization of the reaction conditions
be prevented.[19] To this end, it was deemed imperative to find
milder reaction conditions and to carry out a broad ligand
screening. The first goal was readily achieved in that the use of
CsF instead of KF allowed the reaction to proceed at ambient
temperature in the biphasic mixture CH2Cl2/HFIP as a ligand-
accelerated process.[20] Of arguably higher relevance—even
beyond the present context—is the observation that mixtures Entry R Medium[a] Ligand ee [%][b] (yield)
of HFIP and hexafluorobenzene form an apparently homoge- 1 Me A L2 31
neous phase capable of dissolving CsF, in which the a-fluorina- 2 tBu A L2 35
tion proceeds particularly cleanly. 3 tBu A L3 54
4 tBu A L4 45
Not unexpectedly, the search for effective chiral ligands 5 tBu A L5 66
proved challenging, not least because clear-cut structure/selec- 6 tBu A L6 61
tivity relationships were difficult to deduce from the acquired 7 tBu A L7 59
data (for the full list, see the Supporting Information). As one 8 tBu A L8 75
9 tBu A L9 71
consistent trend, however, it was noticed that standard BOX li- 10 tBu A L10 16
gands led to clean reactions. The poor chiral induction not- 11 Me A L11 52
withstanding (cf. Scheme 1 and the Supporting Information),[9] 12 tBu A L11 73
the ligand optimization exercise remained largely focused on 13 tBu A L12 76
14 tBu A L13 66
this privileged scaffold.[11, 12] Our efforts were guided by the 15 tBu A L14 71
notion that efficient steering of the small incoming fluoride 16 tBu A L15 52
anion likely mandates a tight chiral binding site. We conjec- 17 tBu A L16 72
tured that this goal might be reached by adjustment of the fol- 18 tBu A L17 70
19 tBu A L18 81
lowing three parameters: (i) implementation of substituents at 20 tBu A L19 85 (72)[c]
the methylene bridge to exert remote control,[21, 22] (ii) exten- 21 tBu A L19 87 (65)[c,d]
sion of the lateral “walls” of the C2-symmetric ligand scaffold 22 tBu B L19 89 (87)[c]
beyond the tert-butyl substituents in L1, and (iii) increase of [a] A: HFIP (10 equiv), CH2Cl2 ; B: HFIP (10 equiv), C6F6. [b] ee of the isolat-
the size of the electrophilic partner by using bulkier a-diazo- ed pure compound; for details, see the Supporting Information. [c] Isolat-
esters. ed yield of analytically pure product. [d] Using [Cu(OtBu)]4 instead of
[Cu(OTf)]2·PhMe
Variation of any of these factors alone did not grant success;
gratifyingly though, they were found to synergize. While the
commercially available indane-based bis(oxazoline) L2 gave
methyl ester 2 a with only 31 % ee (Figure 1) and the corre- tors are likely relevant in the present context too, additional
sponding tert-butyl ester 2 b with equally disappointing 35 % aspects seem to play an important role. Specifically, introduc-
ee (Table 1, entries 1, 2), gem-disubstitution of the methylene tion of a cycloheptyl or cyclopropyl group improved the ee to
bridge entailed some improvement. This effect has ample 45 % and 54 % for product 2 b, respectively (entries 3, 4). Equal-
precedent in the literature and is usually ascribed to a but- ly good or even better results were obtained with ligands such
tressing effect, the fine-tuning of the ligand’s bite-angle, and as L5–L9 carrying different benzyl substituents at this position,
suppression of enolization at this site.[11, 12] Although these fac- which furnished 2 b with ee values of up to 75 % (entries 5–

Figure 1. Subset of the tested BOX-ligands; for the full list, see the Supporting Information.

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2510 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

9);[23] moreover, the size of the ester group in the diazo com- ent temperature (ca. + 3 to + 22 8C) and a remarkably strong
pound had a marked effect (cf. entries 11 and 12). hysteresis upon cooling (see the Supporting Information).
Next, we considered placement of substituents at strategic Therefore crystallization was expected to be non-trivial.
positions of the indane core or an extension of the scaffold by The challenge was met by capillary crystallization: the neat
annulation of additional aryl rings. Such modifications are “liquid” sample was filled into a capillary, which was tightly
rare,[24, 25] but it was hoped that they might entail a narrower fused before being chilled in a stream of cold air; the resulting
trajectory for the incoming fluoride anion. Screening of a polycrystalline material was then locally warmed to slightly
number of variants showed that bis(oxazoline) L19 carrying above melting temperature. This cooling/warming cycle led to
bulky p-tert-butyl substituted benzyl moieties at the bridge single crystals suitable for X-ray diffraction. The capillary was
and an only moderately big isopropyl substituent on the transferred quickly to the diffractometer; the collected data set
indane nucleus gave the best result (for the full data set, see afforded a statistically significant absolute structure parameter
the Supporting Information). This particular ligand furnished a- and allowed the absolute configuration to be determined;[27]
fluorophenylacetate 2 b in 87 % yield with 89 % ee when the re- to the best of our knowledge is this the first example of its
action was carried out in HFIP/C6F6 as the optimal medium kind[28] and bodes well for configurational assignments of
(entry 22). More electron-deficient a-diazocarbonyl derivatives other chiral liquid samples in the future. As shown in Figure 3,
performed even better, especially those carrying the electron- the major isomer of the a-fluoroester 2 b (½aA20 D = + 65.9)
withdrawing group at the para-position of the aryl ring formed in the copper catalyzed reaction with ligand L19 is (S)-
(Figure 2); the corresponding products were isolated in high configured. All other compounds shown in Figure 2 were as-
yields with optical purities of up to 95 % ee. Comparison of the signed by analogy, based on the fact that they are invariably
results obtained for products 4 and 5 illustrates the influence dextrorotatory.
of the substitution pattern on the outcome; in this context it is
noteworthy that attempted formation of the corresponding
ortho-chlorinated analogue failed due to decomposition of the
diazo ester precursor.

Figure 3. Structure of compound 2 b in the solid state which allowed the ab-
solute configuration of the a-fluorinated ester derivative to be determined;
all hydrogen atoms—except the one on the (S)-configured chiral center
C7—are omitted for clarity.

Mechanistic discussion
Despite numerous attempts, we failed to grow single crystals
Figure 2. Chiral a-fluoroester derivatives prepared by copper-catalyzed di- of the copper precatalyst carrying the optimal ligand L19,
azoester decomposition. Reaction conditions: [Cu(OTf)]2·PhMe (2.5 mol %), most likely because of the greasy lateral substituents. There-
L19 (5 mol %), CsF (3.5 equiv), RT, 12 h, HFIP (10 equiv)/CH2Cl2 (blue) or HFIP
fore, we resorted to solving the structure of the related com-
(10 equiv)/C6F6 (red).
plex derived from L3, which leads to the same sense of induc-
tion in the fluorination reaction but to a lower ee of 54 %
(Table 1, entry 3). In contrast to the copious information on
Configurational assignment via capillary crystallization
BOX complexes of CuII and other (Lewis-acidic) transition
Since none of the resulting a-fluorinated esters shown in metals,[29] surprisingly few crystal structures of (chiral) [BOX·-
Figure 2 had previously been prepared in the literature in opti- CuI]X complexes are known in the literature;[30–33] actually,
cally active form, we faced the need to rigorously establish the some of them are intricate dimeric or oligomeric arrays, which
absolute configuration prior to any mechanistic discussion.[26] likely have to disassemble before any catalytic reaction can
To this end, a sample of 2 b (89 % ee) as the parent compound take place.
of this series was subjected to preparative HPLC on a chiral In contrast, complex [L3·Cu(MeCN)]BF4 is a monomeric entity
stationary phase to remove the remaining minor enantiomer. (Figure 4); the coordination geometry about the Cu-center is
Low temperature differential scanning calorimetry showed that distorted trigonal. The N1-Cu1-N2 bite-angle of 95.9(1)8 is
2 b (ee + 99 %) features a broad melting range close to ambi- wider than that of the few comparable cases in the litera-

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2511 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

Figure 5. Graphical representation of relevant NOE contacts in [L11·Cu(-

MeCN)]BF4, which indicate that the substituted benzyl groups are, on aver-
age, “forward” oriented on the NMR time scale.

Figure 4. Structure of complex [L3·Cu(MeCN)]BF4 in the solid state; hydro-

gen atoms on the ligand backbone are omitted for clarity.

ture,[31] which likely reflects the influence of the gem-disubsti-

tution at the bridge as well as the steric demand of the indane
core. Although the crystal structure provides only a static pic-
ture, it is noteworthy that the complex is not strictly C2-sym-
metric in the solid state; specifically, the N3-Cu1-N1 (121.8(1)8)
and the N3-Cu1-N2 (141.58) angles are notably uneven, as
are the distances Cu1@N1 (2.004(2) a) and Cu1@N2
(1.952(2) a).[31, 34]
NMR investigations provided some insights into the role of
the substituents at the bridge between the oxazoline rings.[34]
NOESY spectra of complex [L11·Cu(MeCN)]BF4 in CDCl3 show
numerous cross peaks between the protons of the substituted Figure 6. Computed ground state structure of the donor/acceptor copper
benzyl groups and those of the indane ring (see Figure 5). The carbene complex [L3·Cu = CPh(COOtBu)]BF4 formed from [L3·Cu(MeCN)]BF4
and the diazoester 1 b; illustration of the uneven barriers for clockwise and
benzyl substituents are hence “forward” oriented in at least
counterclockwise rotation about the Cu@C bond.
one averaged conformer on the NMR timescale; as such, they
(temporarily) reside in vicinity of the incipient carbene, above
and below the catalytically active Cu-center. The crystal struc- The plane defined by the trigonal donor/acceptor carbene
ture of a related BOX·Cu complex with pendant p-(tert-butyl)- center in [L3·Cu = CPh(COOtBu)]BF4 is orthogonal to the {N2Cu}
benzyl substituents corroborates the notion that these seem- plane of the chiral catalyst (Figure 6);[36] this arrangement is
ingly remote substituents actually shield the upper and lower geometrically favorable and, at the same time, arguably maxi-
face of the coordination plane.[31, 35] mizes back-bonding from the filled metal d-orbital that is most
If one takes the ligated acetonitrile as a dummy for the car- destabilized by the N,N-donor ligand into the empty carbene
bene to be formed upon reaction of the Cu-center with the a- p-orbital. As one might expect for a C2-symmetric ligand envi-
diazoester, attack onto the Re face of the electrophilic C-atom ronment,[34] the barriers for clockwise and counter-clockwise
seems most plausible at first sight, which would entail forma- rotation about the Cu@C bond are notably different. The ap-
tion of (R)-2 b (compare Figures 5 and 6). Since (S)-2 b, howev- preciable barrier heights of & 14 and & 25 kcal mol@1, respec-
er, has been shown to be the major product, a more involved tively, imply that the carbene moiety does not freely rotate at
scenario must be operative. Therefore, a computational study ambient temperature at which the reaction with fluoride does
was performed to gain a better understanding. Density func- occur, but rather oscillates about the conformation of lowest
tionals based on the generalized gradient approximation energy. Importantly, this swinging motion is directional and
(GGA) were chosen for the good balance between accuracy the amplitudes are uneven, in that clockwise movement will
and performance. The BP86 functional was found to reproduce prevail by large margins as it avoids a clash between the car-
the experimental structure of the copper complex [L3·Cu(- bene’s substituents and the rigid indane scaffold.[37]
MeCN)]BF4 particularly well and was therefore used throughout In exploring possible trajectories for the attack of fluoride
(for the computational methods, see the Supporting Informa- onto this complex, we were surprised to find that the Cu-
tion). center rather than the electrophilic carbene C-atom constitutes

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2512 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

Conclusions
The first highly enantioselective metal catalyzed addi-
tion of fluoride to a-diazoesters is described, featur-
ing ee values of up to 95 %. To properly assess this
result, one has to consider the inherent challenges
for asymmetric catalysis posed by this particular nu-
cleophile, which had basically thwarted an earlier at-
tempt at rendering this transformation asymmetric.[9]
Interestingly, the reaction most likely proceeds by ini-
tial attack of fluoride onto the Cu- rather than the C-
atom of the electrophilic metal carbene intermediate;
a subsequent 1,2-fluoride shift then leads to product
formation. For this migration to happen, the carbene
must rotate about the Cu@C bond to ensure the nec-
essary orbital overlap; the directionality of this rota-
tory motion, enacted by the C2-symmetric ligand en-
vironment, is the major enantiodetermining factor.
Further work on the structure and reactivity of transi-
tion metal carbene complexes and the stereoselec-
Figure 7. Reaction profile for nucleophilic attack of fluoride from the top-face of the Cu- tive preparation of organofluorine derivatives[44] is
carbene complex [L3·Cu = CPh(COOtBu)]BF4, followed by isomerization of F@Cutop into
(S)- or (R)-2 b; for the free energy profile of the bottom-face approach, see the Support- underway in our laboratory and will be reported in
ing Information. due course.

Experimental Section
the prime site of attack; no transition state was found for All experimental details can be found in the Supporting Informa-
direct C@F bond formation, despite considerable search ef- tion. The material includes compound characterization data, addi-
forts.[38] The incoming nucleophile approaches the metal tional crystallographic information, the full list of ligands investigat-
center from above or below the coordination plane to form ed, HPLC traces, supporting computational data, and copies of
spectra of new compounds.
the copper fluoride complexes F@Cutop and F@Cubot, respec-
tively. It is striking that top-side attack is barrierless with a sub-
merged transition state TSF!Cutop,[39] provided that a continuum Acknowledgements
solvent environment was used in the computations (Figure 7).
In stark contrast, attack from the bottom face shows an activa- Generous financial support by the Max-Planck-Society, the
tion barrier of no less than 12.3 kcal mol@1 (see the Supporting Alexander-von-Humboldt Foundation (return scholarship to
Information). Qualitatively, this differential is rather intuitive P.J.), and the Xunta de Galicia and European Union (European
and very well in line with the conclusions previously drawn Social Fund, predoctoral fellowship to I.P.) is gratefully ac-
from the solid-state structure of electrophilic donor/acceptor knowledged. We thank the analytical departments of our Insti-
dirhodium carbenes:[13] Any incoming nucleophile, especially tute for excellent support.
when charged, will avoid dipole repulsion with the lone-pairs
of the ester O-atoms[40] and hence preferentially follow a
Bergi–Dunitz trajectory alongside the aryl substituent of the Conflict of interest
donor/acceptor carbene.[13, 41]
Since the top–face approach wins out by far, only the fate of The authors declare no conflict of interest.
the resulting complex F@Cutop needs to be considered at this
point. Orbital alignment is necessary before the fluoride ligand Keywords: asymmetric catalysis · bisoxazoline ligands ·
can migrate from copper to carbon,[42, 43] which, in turn, man- carbene complexes · copper · organofluorine compounds
dates that the carbene unit rotates about the Cu@C bond. The
transition state for clockwise rotation is 1.9 kcal mol@1 lower in [1] a) S. Purser, P. R. Moore, S. Swallow, V. Gouverneur, Chem. Soc. Rev.
2008, 37, 320 – 330; b) J. Wang, M. S#nchez-Rosellj, J. L. Acena, C. del
energy (Figure 7); it leads to complex (R)-2 b@Cu, which ulti-
Pozo, A. E. Sorochinsky, S. Fustero, V. A. Soloshonok, H. Liu, Chem. Rev.
mately delivers (S)-2 b upon protonation (note that the “inver- 2014, 114, 2432 – 2506; c) E. P. Gillis, K. J. Eastman, M. D. Hill, D. J. Don-
sion” is due to the formalism of the CIP rules). Although this nelly, N. A. Meanwell, J. Med. Chem. 2015, 58, 8315 – 8359; d) Y. Zhou, J.
computational result overestimates the level of induction, it Wang, Z. Gu, S. Wang, W. Zhu, J. L. Acena, V. A. Soloshonok, K. Izawa, H.
Liu, Chem. Rev. 2016, 116, 422 – 518.
correctly predicts the sense of induction of the copper cata-
[2] For an instructive case, see: K. L. Lee, C. M. Ambler, D. R. Anderson, B. P.
lyzed a-fluorination reaction. Boscoe, A. G. Bree, J. I. Brodfuehrer, J. S. Chang, C. Choi, S. Chung, K. J.
Curran, J. E. Day, C. M. Dehnhardt, K. Dower, S. E. Drozda, R. K. Frisbie,

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2513 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

L. K. Gavrin, J. A. Goldberg, S. Han, M. Hegen, D. Hepworth, H.-R. Hope, Chem. 2014, 126, 898 – 901; c) G. Seidel, R. Mynott, A. Ferstner, Angew.
S. Kamtekar, I. C. Kilty, A. Lee, L.-L. Lin, F. E. Lovering, M. D. Lowe, J. P. Chem. Int. Ed. 2009, 48, 2510 – 2513; Angew. Chem. 2009, 121, 2548 –
Mathias, H. M. Morgan, E. A. Murphy, N. Papaioannou, A. Patny, B. S. 2551; d) A. G. Tskhovrebov, J. B. Lingnau, A. Ferstner, Angew. Chem. Int.
Pierce, V. R. Rao, E. Saiah, I. J. Samardjiev, B. M. Samas, M. W. H. Shen, Ed. 2019, 58, 8834 – 8838; Angew. Chem. 2019, 131, 8926 – 8930.
J. H. Shin, H. H. Soutter, J. W. Strohbach, P. T. Symanowicz, J. R. Thoma- [19] For metal-free reactions of a-diazocarbonyl compounds with various
son, J. D. Trzupek, R. Vargas, F. Vincent, J. Yan, C. W. Zapf, S. W. Wright, J. fluoride sources see: a) G. A. Olah, J. Welch, Synthesis 1974, 896 – 898;
Med. Chem. 2017, 60, 5521 – 5542. b) E. L. Setti, O. A. Mascaretti, J. Chem. Soc. Perkin Trans. 1 1988, 2059 –
[3] T. Liang, C. N. Neumann, T. Ritter, Angew. Chem. Int. Ed. 2013, 52, 8214 – 2060; c) M. Ohno, M. Itoh, T. Ohashi, S. Eguchi, Synthesis 1993, 793 –
8264; Angew. Chem. 2013, 125, 8372 – 8423. 796; d) R. Pasceri, H. E. Bartrum, C. J. Hayes, C. J. Moody, Chem.
[4] For leading references on stoichiometric methods, see the following Commun. 2012, 48, 12077 – 12079.
and literature cited therein: a) F. A. Davis, P. V. N. Kasu, Tetrahedron Lett. [20] In the absence of ligand, reaction of 1 b with CsF (3 equiv) and
1998, 39, 6135 – 6138; b) J. Alvarado, A. T. Herrmann, A. Zakarian, J. Org. [Cu(OTf)]2·PhMe (5 mol %) in CH2Cl2/HFIP led to , 1 % conversion after
Chem. 2014, 79, 6206 – 6220. 4 h reaction time at ambient temperature.
[5] For leading references on catalytic procedures, see the following and [21] a) D. A. Evans, K. A. Woerpel, M. M. Hinman, M. M. Faul, J. Am. Chem.
literature cited therein: a) T. D. Beeson, D. W. C. MacMillan, J. Am. Chem. Soc. 1991, 113, 726 – 728; b) E. J. Corey, N. Imai, H.-Y. Zhang, J. Am.
Soc. 2005, 127, 8826 – 8828; b) T. Suzuki, Y. Hamashima, M. Sodeoka, Chem. Soc. 1991, 113, 728 – 729.
Angew. Chem. Int. Ed. 2007, 46, 5435 – 5439; Angew. Chem. 2007, 119, [22] For an excellent review, see: S. Liao, X.-L. Sun, Y. Tang, Acc. Chem. Res.
5531 – 5535; c) D. H. Paull, M. T. Scerba, E. Alden-Danforth, L.-R. Widger, 2014, 47, 2260 – 2272.
T. Lectka, J. Am. Chem. Soc. 2008, 130, 17260 – 17261; d) P. Wheeler, [23] However, the use of ligands with p-extended benzylic groups was
H. U. Vora, T. Rovis, Chem. Sci. 2013, 4, 1674 – 1679; e) J. Peng, D. M. Du, counterproductive, see the Supporting Information.
RSC Adv. 2014, 4, 2061 – 2067; f) F. Li, Z. Wu, J. Wang, Angew. Chem. Int. [24] C. Liu, J.-C. Yi, Z.-B. Zheng, Y. Tang, L.-X. Dai, S.-L. You, Angew. Chem. Int.
Ed. 2015, 54, 656 – 659; Angew. Chem. 2015, 127, 666 – 669; g) X. Dong, Ed. 2016, 55, 751 – 754; Angew. Chem. 2016, 128, 761 – 764.
W. Yang, W. Hu, J. Sun, Angew. Chem. Int. Ed. 2015, 54, 660 – 663; [25] V. L. Rendina, S. A. Goetz, A. E. Neitzel, H. Kaplan, J. S. Kingsbury, Tetrahe-
Angew. Chem. 2015, 127, 670 – 673; h) L. Wang, X. Jiang, J. Chen, Y. dron Lett. 2012, 53, 15 – 18.
Huang, Angew. Chem. Int. Ed. 2019, 58, 7410 – 7414; Angew. Chem. [26] Parenthetically, we note that the configurational assignment of a-
2019, 131, 7488 – 7492. fluoro-a-phenylacetic acid methyl ester, which is often used as a refer-
[6] a) C. Hollingworth, V. Gouverneur, Chem. Commun. 2012, 48, 2929 – ence point, was based on either plausibility arguments or on three-step
2942; b) J. Wu, Tetrahedron Lett. 2014, 55, 4289 – 4294. derivatization of an only modestly enriched sample, see ref. [8b] and
[7] P. Adler, C. J. Teskey, D. Kaiser, M. Holy, H. H. Sitte, N. Maulide, Nat. the following: K. Miyamoto, S. Tsuchiya, H. Ohta, J. Fluorine Chem. 1992,
Chem. 2019, 11, 329 – 334. 59, 225 – 232.
[8] Formation of scalemic a-aryl-a-fluoro esters from mandelate or arylgly- [27] H. D. Flack, Acta Crystallogr. A 1983, 39, 876 – 881.
cine derivatives is challenging and/or requires special reagents, see: [28] For a review on low-temperature crystallization, see: a) M. Veith, W.
a) S. Bresciani, D. O’Hagan, Tetrahedron Lett. 2010, 51, 5795 – 5797; b) S. Frank, Chem. Rev. 1988, 88, 81 – 92; for a recent example, see: b) R. W.
Hamman, N. Barrelle, F. Tetaz, C. G. Beguin, J. Fluorine Chem. 1987, 37, Seidel, R. Goddard, N. Nçthling, C. W. Lehmann, CrystEngComm 2019,
85 – 94. 21, 3295 – 3303.
[9] E. E. Gray, M. K. Nielsen, K. A. Choquette, J. A. Kalow, T. J. A. Graham, [29] R. Rasappan, D. Laventine, O. Reiser, Coord. Chem. Rev. 2008, 252, 702 –
A. G. Doyle, J. Am. Chem. Soc. 2016, 138, 10802 – 10805. 714.
[10] For racemic “vinylogous” fluorination of diazo derivatives, see: a) C. Qin, [30] a) D. A. Evans, K. A. Woerpel, M. J. Scott, Angew. Chem. Int. Ed. Engl.
H. M. L. Davies, Org. Lett. 2013, 15, 6152 – 6154; b) S. Thompson, S. J. 1992, 31, 430 – 432; Angew. Chem. 1992, 104, 439 – 441; see also: b) S.
Lee, I. M. Jackson, N. Ichiishi, A. F. Brooks, M. S. Sanford, P. J. H. Scott, Bellemin-Laponnaz, L. H. Gade, Chem. Commun. 2002, 1286 – 1287.
Synthesis 2019, 51, 4401 – 4407. [31] C. Deng, L.-J. Wang, J. Zhu, Y. Tang, Angew. Chem. Int. Ed. 2012, 51,
[11] A. Pfaltz, Acc. Chem. Res. 1993, 26, 339 – 345. 11620 – 11623; Angew. Chem. 2012, 124, 11788 – 11791.
[12] a) J. S. Johnson, D. A. Evans, Acc. Chem. Res. 2000, 33, 325 – 335; b) H. A. [32] a) A. Walli, S. Dechert, M. Bauer, S. Demeshko, F. Meyer, Eur. J. Inorg.
McManus, P. J. Guiry, Chem. Rev. 2004, 104, 4151 – 4202; c) G. Desimoni, Chem. 2014, 4660 – 4676; b) V. E. Goswami, A. Walli, M. Forster, S. De-
G. Faita, K. A. Jorgensen, Chem. Rev. 2006, 106, 3561 – 3651; d) G. Desi- chert, S. Demeshko, M. C. Holthausen, F. Meyer, Chem. Sci. 2017, 8,
moni, G. Faita, K. A. Jorgensen, Chem. Rev. 2011, 111, PR284 – PR437. 3031 – 3037.
[13] a) C. Werl8, R. Goddard, A. Ferstner, Angew. Chem. Int. Ed. 2015, 54, [33] H. D. Srinivas, P. Maity, G. P. A. Yap, M. P. Watson, J. Org. Chem. 2015, 80,
15452 – 15456; Angew. Chem. 2015, 127, 15672 – 15676; b) C. Werl8, R. 4003 – 4016.
Goddard, P. Philipps, C. FarHs, A. Ferstner, J. Am. Chem. Soc. 2016, 138, [34] The spectra of [L11·Cu(MeCN)]BF4 in CDCl3 indicate C2 symmetry on the
3797 – 3805; c) C. Werl8, R. Goddard, P. Philipps, C. FarHs, A. Ferstner, NMR timescale.
Angew. Chem. Int. Ed. 2016, 55, 10760 – 10765; Angew. Chem. 2016, 128, [35] For a similar situation in [BOX·CuII] complexes, see: a) H. Xiong, H. Xu, S.
10918 – 10923. Liao, Z. Xie, Y. Tang, J. Am. Chem. Soc. 2013, 135, 7851 – 7854; b) Y. Li, K.
[14] a) L. R. Collins, M. van Gastel, F. Neese, A. Ferstner, J. Am. Chem. Soc. Zhou, Z. Wen, S. Cao, X. Shen, M. Lei, L. Gong, J. Am. Chem. Soc. 2018,
2018, 140, 13042 – 13055; b) L. R. Collins, S. Auris, R. Goddard, A. Ferst- 140, 15850 – 15858.
ner, Angew. Chem. Int. Ed. 2019, 58, 3557 – 3561; Angew. Chem. 2019, [36] This geometrical feature is well in line with the X-ray structures of ter-
131, 3595 – 3599. minal copper carbenes, see: a) X. Dai, T. H. Warren, J. Am. Chem. Soc.
[15] D. J. Tindall, C. Werl8, R. Goddard, P. Philipps, C. FarHs, A. Ferstner, J. 2004, 126, 10085 – 10094; b) B. F. Straub, P. Hofmann, Angew. Chem. Int.
Am. Chem. Soc. 2018, 140, 1884 – 1893. Ed. 2001, 40, 1288 – 1290; Angew. Chem. 2001, 113, 1328 – 1330; c) P.
[16] a) A. Ferstner, J. Am. Chem. Soc. 2019, 141, 11 – 24; b) A. Guthertz, M. Hofmann, I. V. Shishkov, F. Rominger, Inorg. Chem. 2008, 47, 11755 –
Leutzsch, L. M. Wolf, P. Gupta, S. M. Rummelt, R. Goddard, C. FarHs, W. 11762; d) I. V. Shishkov, F. Rominger, P. Hofmann, Organometallics 2009,
Thiel, A. Ferstner, J. Am. Chem. Soc. 2018, 140, 3156 – 3169; c) T. Biberg- 28, 1049 – 1059.
er, C. P. Gordon, M. Leutzsch, S. Peil, A. Guthertz, C. Cop8ret, A. Ferstner, [37] a) The directionality of the rotatory motion is also the essence of the
Angew. Chem. Int. Ed. 2019, 58, 8845 – 8850; Angew. Chem. 2019, 131, Pfaltz model explaining the course of Cu-catalyzed cyclopropanations;
8937 – 8942; d) S. Peil, A. Guthertz, T. Biberger, A. Ferstner, Angew. however, the incoming olefin directly interacts with the carbene-C
Chem. Int. Ed. 2019, 58, 8851 – 8856; Angew. Chem. 2019, 131, 8851 – atom rather than the Cu center, see: H. Fritschi, U. Leutenegger, A.
8856. Pfaltz, Helv. Chim. Acta 1988, 71, 1553 – 1565; b) for a later computation-
[17] A. G. Tskhovrebov, R. Goddard, A. Ferstner, Angew. Chem. Int. Ed. 2018, al study, see: T. Rasmussen, J. F. Jensen, N. Ostergaard, D. Tanner, T. Zie-
57, 8089 – 8094; Angew. Chem. 2018, 130, 8221 – 8226. gler, P.-O. Norrby, Chem. Eur. J. 2002, 8, 177 – 184.
[18] a) G. Seidel, A. Ferstner, Angew. Chem. Int. Ed. 2014, 53, 4807 – 4811; [38] For a cationic Fischer Ir – carbene complex, which is attacked by fluoride
Angew. Chem. 2014, 126, 4907 – 4911; b) G. Seidel, B. Gabor, R. Goddard, at the metal rather than carbon atom, see: J. M. O’Connor, L. Pu, S.
B. Heggen, W. Thiel, Angew. Chem. Int. Ed. 2014, 53, 879 – 882; Angew. Woolard, R. K. Chadha, J. Am. Chem. Soc. 1990, 112, 6731 – 6732.

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2514 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2020, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202000081 by Nat Prov Indonesia, Wiley Online Library on [14/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

[39] This phenomenon is particularly well precedented for reactions in 2011, 1, 153 – 165; see also: e) A. Cairncross, W. A. Sheppard, J. Am.
which one partner is small and highly reactive; it is thought to indicate Chem. Soc. 1968, 90, 2186 – 2187.
stabilizing effects near the transition state, resulting in a strong non-Ar- [43] For Cu-mediated trifluoromethylation and trifluoromethylthiolation of
rhenius behavior, cf.: N. M. Donahue, Chem. Rev. 2003, 103, 4593 – 4604. diazo esters, which seem to proceed by a similar mechanism, see: a) M.
[40] In the ground state, the ester is orthogonal to the carbene to minimize Hu, C. Ni, J. Hu, J. Am. Chem. Soc. 2012, 134, 15257 – 15260; b) X. Wang,
orbital overlap and avoid further withdrawal of electron density from Y. Zhou, G. Ji, G. Wu, M. Li, Y. Zhang, J. Wang, Eur. J. Org. Chem. 2014,
this highly electrophilic site; the barrier for rotation about the C1@C3 3093 – 3096; c) Q. Lefebvre, E. Fava, P. Nikolaienko, M. Rueping, Chem.
bond is & 8 kcal mol@1, see the Supporting Information. Commun. 2014, 50, 6617 – 6619.
[41] For crystallographic evidence for the validity of the Bergi – Dunitz trajec- [44] a) H. Sommer, A. Ferstner, Chem. Eur. J. 2017, 23, 558 – 562; b) X. Mo, A.
tory in metal carbene chemistry, see: S. Peil, A. Ferstner, Angew. Chem. Letort, D.-A. Rosca, K. Higashida, A. Ferstner, Chem. Eur. J. 2018, 24,
Int. Ed. 2019, 58, 18476 – 18481; Angew. Chem. 2019, 131, 18647 – 18652. 9667 – 9674; c) Z. Meng, L. Souillart, B. Monks, N. Huwyler, J. Herrmann,
[42] Related migratory insertion reactions were proposed for different R. Meller, A. Ferstner, J. Org. Chem. 2018, 83, 6977 – 6994.
metals, including copper, cf.: a) Q. Xiao, Y. Zhang, J. Wang, Acc. Chem.
Res. 2013, 46, 236 – 247; b) X. Zhao, Y. Zhang, J. Wang, Chem. Commun.
2012, 48, 10162 – 10173; c) J. Barluenga, C. Vald8s, Angew. Chem. Int. Ed. Manuscript received: January 7, 2020
2011, 50, 7486 – 7500; Angew. Chem. 2011, 123, 7626 – 7640; d) N. M. G. Accepted manuscript online: January 9, 2020
Franssen, A. J. C. Walters, J. N. H. Reek, B. de Bruin, Catal. Sci. Technol. Version of record online: February 18, 2020

Chem. Eur. J. 2020, 26, 2509 – 2515 www.chemeurj.org 2515 T 2020 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim