pubs.acs.

org/acscatalysis Research Article

Halogen Bond Activation in Gold Catalysis

Helgi Freyr Jónsson, Daniel Sethio, Julian Wolf, Stefan M. Huber, Anne Fiksdahl,* and Mate Erdelyi*
Cite This: ACS Catal. 2022, 12, 7210−7220 Read Online

ACCESS Metrics & More Article Recommendations *

sı Supporting Information

ABSTRACT: Gold catalysis has, over the past decades, provided

innovative organic transformations under mild conditions with
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

high chemoselectivities. It receives steadily growing attention

thanks to its wide synthetic applicability. The catalytically active
form, [Ln-Au]+, of ligated gold complexes, [Ln-Au-Cl], is formed
via halide abstraction. This is typically achieved by anion exchange
upon the addition of an appropriate silver salt to the reaction
Downloaded via 185.183.146.239 on June 1, 2022 at 21:57:17 (UTC).

mixture. Herein, an alternative silver-free route for gold activation

is presented, making use of halogen bonding to promote halide
abstraction. We demonstrate that a catalytic amount of a strong
halogen bond donor efficiently activates both gold(I) and gold(III)
catalysts. Following the reaction, both the catalyst and the activator
are easily recovered. Importantly, this not only reduces the metal
waste in a gold-catalyzed process but also enables its upscaling, possibly opening new avenues for its use in industrial settings. Gold is
an expensive and limited resource, and its recyclability is of supreme importance. Based on systematic reaction kinetics, NMR
spectroscopic, and computational investigations, we describe the mechanism of halogen bond-activated gold(I/III) catalysis using
cyclopropanation as a model reaction. Our discovery paves the way for the development of gold-mediated transformations that allow
recycling of the precious gold catalyst and that may thereby become useful also for the large-scale generation of complex molecules.
KEYWORDS: gold, catalysis, halogen bond, bond activation, halide abstraction, recyclability

■ INTRODUCTION
Homogeneous gold catalysis has become a powerful tool for
Several approaches for silver-free gold catalysis have been
developed.24 These include chloride abstraction by use of
the synthesis of complex organic molecules.1−13 Offering ambiphilic ligands containing a hydrogen bond donor25 and an
internal Lewis acid26 as well as halide abstraction by alternative
outstanding chemoselectivities under mild conditions, it has
metal salts.27
found a wide range of applications in a variety of research
Halogen bond28-mediated catalysis is a novel approach in
fields.14−18 In most transformations, gold’s Lewis acidity is
synthesis.29−31 Halogen bond donors have recently been used
used for the activation of carbon−carbon multiple bonds to
to activate carbonyl groups to promote Diels−Alder reactions
promote the formation of a new carbon−carbon bond.
and Michael additions, for instance, and to assist the
Initially, gold chloride salts were used as catalysts, whereas
dissociation of halides to facilitate SN1-type processes.
later protocols apply ligands to modulate gold’s reactivity and
Activation of a metal−halogen bond by a halogen bond
stability.19 A common feature of the majority of gold-catalyzed
donor to enable catalytic reactions has recently been reported,
processes, ligated or non-ligated, is that the precatalyst
without a detailed mechanistic discussion however, by Huber
possesses a gold−chloride bond. Usually, this is activated
and co-workers, using the example of the gold(I)-catalyzed
through chloride abstraction by anion exchange using a silver cyclization of a propargylic amide and the cycloisomerization
salt with a weakly coordinating anion, generating the active of 1,6-enyne.32
catalyst, [Ln-Au]+, along with silver chloride. The potential The interaction of halogen bond donors with gold(III)
influences of silver20,21 and its counterion22 on gold catalysis complexes has not yet been assessed and that with gold(I) has
have been a matter of debate over the past decades. Silver is of only been scarcely explored.32 Halogen bond catalysis was
concern in ecosystems as it is one of the most toxic metals to
aquatic organisms.23 Recent efforts in synthesis target not only
the development of new processes but also the improvement of Received: April 16, 2022
the sustainability of existing transformations. Besides the Revised: May 14, 2022
decrease in catalyst loading and recycling, the substitution of
silver salts with alternatives capable of activating gold
precatalysts is therefore of vast interest, both from a
mechanistic perspective and from a sustainability perspective.
© XXXX The Authors. Published by
American Chemical Society https://doi.org/10.1021/acscatal.2c01864
7210 ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis
proposed based on intuition but without the support of
comprehensive experimental and computational proof. We,
therefore, set the aim to evaluate whether the halogen bond
activation of gold(I) and gold(III) species may be a generally
applicable alternative to activation with the traditionally used
silver salts (Figure 1). In addition to disclosing a sustainable
Figure 1. Activation (a) with a silver salt is the common route in gold
catalysis (X− is a weakly coordinating anion). Herein, (b) gold
activation using a halogen bond donor (XBD) is explored as an
alternative sustainable route.
alternative synthetic route for gold(I/III) activation, we
provide a detailed mechanistic understanding of the halogen
bond-activated gold-catalyzed process.
■
RESULTS AND DISCUSSION
Design. To demonstrate the real-life applicability of a
halogen bond activation strategy, we explored the most
commonly used gold(I) complexes, namely, the gold(I)
JohnPhos complex 1, the chiral dinuclear BINAP gold(I)
complex 2, the gold(I) N-heterocyclic carbene complexes
possessing IMes (3) and IPr (5) ligands, and the respective
gold(III) complexes 4 and 6 (Figure 2). Compounds 7−11
Figure 2. Common gold complexes 1−6 were explored in the
investigation of halogen bond activation in gold catalysis.
(Figure 3) were chosen as halogen bond donors to be explored
for their halide abstraction ability. Due to their monodentate
nature, compounds 7 and 8 are weak halogen bond donors.
The cationic iodolium complex 9 and the bidentate 10 and 11
are strong halogen bond donors. We chose to use one of the
most well-studied gold-mediated synthetic procedures, the
cyclopropanation of propargyl acetate 12 with the vinyl
derivative styrene 13 to cyclopropane 14 (Scheme 1) as a
model reaction,33 which has previously been used for the
evaluation of the catalytic ability of novel gold(I) and gold(III)
catalysts.34−37
Reaction Rates. Reactions were initially performed by
mixing 1 equiv of propargyl acetate 12 and 2 equiv of styrene
13 with 20 mol % each of gold complexes 1−6 and halogen
7211
Research Article
Figure 3. Halogen bond donors 7−11 were explored for their ability
to activate gold(I) and gold(III) catalysts.
bond donor 7 or 8 (Scheme 1). For the dinuclear complex 2,
10 mol % was used as this equals 20 mol % of gold(I) as
compared to the substrate. The conversion to product 14 was
monitored by 1H NMR. Most importantly, the reaction
progressed when a halogen bond donor was introduced to
the mixture. The fastest reaction was observed using the
combination of IPrAuCl3 6 and halogen bond donor 7,
reaching completion in 18 h, whereas most other halogen
bond-activated reactions proceeded but did not give full
conversions within 60 h (Table 1, Figures 4 and 5). The
observed relative reaction rates suggest that the identities of
both the gold complex and the halogen bond donor play a role,
and simple conclusions, such as the stronger the halogen bond
donor, the quicker the reaction rate, cannot be drawn, but
factors other than being purely electronic might be at play as
well. Hence, halogen bond donor 8, which has a strong
electron-withdrawing CF3 group, is not always a stronger
activator than the somewhat less electron-poor 7 and gave
greater reaction rates only in combination with gold catalysts 2,
4, and 5 (Table 1). The high reactivity of dimesityl-substituted
iodoimidazolium salts may be due to the aryl groups’
preference to adopt conformations that minimize steric
repulsions, which breaks the conjugation with the imidazolium
ring.38 Accordingly, the positive charge is localized close to the
halogen bond donor iodine in these complexes.
We performed several control experiments to confirm that
the activation truly originates from halogen bonding. Using 20
mol % gold complex 3 without any activator, we observed no
conversion after 24 h. This confirms that the gold catalyst does
not show any background reactivity. No conversion was seen
using 20 mol % of halogen bond donor 7 without a gold
catalyst. Hidden acid catalysis was ruled out by detection of no
conversion upon the addition of small amounts of HCl along
with 20 mol % of gold complex 3. There was no conversion
either using 1,3-dimesitylimidazolium chloride, the nonhalogen
bond donor analogue of 7, in combination with catalyst 3,
which corroborates our hypothesis that the halogen bond
donor iodine of 7 is crucial for the activation of the gold
catalyst.
Subsequently, we turned our attention to 9, which,
possessing a charged iodolium ion, is expected to be a stronger
halogen bond donor than 7 and 8. Indeed, using as little as 0.5
mol % 9 and 0.5 mol % gold catalyst 3, full conversion was
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis Research Article

Scheme 1. Gold-Catalyzed (1−6) Cyclopropanation of Propargyl Acetate 12 Using the XBDs 7−11 as Activatorsa

a
In the reactions using 7 and 8 as activators, 20 mol % gold catalyst (1−6) and 20 mol % 7 or 8 were applied. In the reactions using 9−11 as the
activators, 0.5 mol % gold catalyst and 0.5 mol % chosen activator were used.

Table 1. Reactivity of Gold Complexes (Au) 1−6 upon

Activation with Halogen Bond Donors (Activator) 7 and 8
in Cyclopropanationa
activator Au conversionb krelc
7 1 31 3.2
7 2 38 3.9
7 3 60 5
7 4 25 2.6
7 5 10 1
7 6 84 8.8
8 1 16 1.7 Figure 4. Conversions for the cyclopropanation of propargyl acetate
8 2 70 7.3 12 to 14 (Scheme 1 and Table 1) as a function of reaction time using
8 3 43 4.5 7 (triangles) or 8 (circles) to activate gold(I) catalysts 1 (green), 2
8 4 29 3.0 (red), 3 (turquoise), and 5 (blue). Reactions were run at 25 °C,
8 5 63 6.6 followed by 1H NMR. The runs with the same gold catalyst are shown
8 6 45 4.7 with the same color. The control experiments, using 20 mol % gold
none 3 0d catalyst 3 without a halogen bond donor, using 20 mol % halogen
bond donor 7 without a gold catalyst, using 20 mol % each 1,3-
7 none 0d
dimesitylimidazolium chloride and gold catalyst 3, and finally, using
IMes-Cl 3 0d
20 mol % gold catalyst 3 with a small amount of HCl, are all shown as
HCl 3 0d black squares. None of the controls gave conversion.
a
Reactions were performed with 20 mol % of activators 7 and 8 and
gold complexes 1−6 in CD2Cl2 at 25 °C, [12] = 0.075 M, [Au] =
[XBD] = 0.015 M. Both cis and trans isomers of 14 are formed, and
the reported conversions are given as the sum of those of the isomers.
The isomerization is a slow background reaction that takes place in
the presence of gold complexes.34 bConversion. cRelative reaction
rates measured by 1H NMR at 12 h reaction time (before
isomerization). dControl experiment. Under “classical” activation
conditions, using 5 mol % AgSbF6 to activate the 5 mol % gold
complex, the reactions reach full conversion within 5 min, regardless
of the type of gold catalyst used.

obtained within 2.5 h (see Table 2 and Figure 6). Iodolium ion
9 was significantly better at activating gold(I) as compared to
gold(III) catalysts. The relative rate of cyclopropanation using
activator 9 in combination with catalyst 6 is higher than that
using 4 as a catalyst (Table 2). Thus, the former only reached
60% conversion while the latter 89% within 24 h. This may
indicate the decomposition of 6 during the reaction.
Next, gold activation using the bidentate halogen bond
donors 10 and 11 was explored. As the iodine of these can
simultaneously form two halogen bonds with the same
chlorine, they are highly efficient in facilitating the cleavage
of a Au−Cl bond and thereby activating the catalysts.
Accordingly, full conversion in the cyclopropanation shown
in Scheme 1 was reached within 10 min when using gold(I)
catalysts. To obtain reliable reaction rates, we ran these
reactions at −20 °C (Table 3). Similar to the reactions using
activator 9, those using 10 and 11 were significantly slower
when a gold(III) catalyst was applied and were, therefore, run
at 25 °C. The relative rates shown in Table 3 and Figure 7 for
gold(I)- and Figure 8 for gold(III)-catalyzed reactions are
therefore not directly comparable.
7212
Figure 5. Conversions observed for the cyclopropanation of propargyl
acetate 12 to 14 (Scheme 1 and Table 1) as a function of reaction
time using 7 (triangles) or 8 (circles) to activate gold(III) catalysts 4
(bright green) and 6 (dark red). Reactions were run at 25 °C, and
conversion was followed by 1H NMR.
Similar to the reactions using activator 9, the gold(III)-
mediated reactions ran with halogen bond donors 10 and 11
were faster when using catalyst 4 (IMesAuCl3) as compared to
those using 6 (IPrAuCl3). It should be underlined that
activators 10 and 11 provided comparable reaction rates to
that using NaBArF as the activator. Halogen bond donor 11
gave faster initial reaction rates than NaBArF for both gold
complexes studied. The fact that the NaBArF-activated
reactions reached completion earlier might indicate an
induction period for these reactions.
Overall, the halogen bond donor strength positively
correlates with the reaction rate. Accordingly, the monodentate
imidazolium-type 7 and 8 are weak gold(I/III) activators, even
at catalyst loadings as high as 20 mol %. The stronger
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis Research Article

Table 2. Conversion of Alkyne 12 and Styrene 13 to Table 3. Conversion of Alkyne 12 and Styrene 13 to
Cyclopropane 14 (Scheme 1) Using Gold Complexes 1−6 Cyclopropane 14 (Scheme 1) Using Gold Catalysts 1−6
(Au) in Combination with Activator 9 and the (Au) in Combination with Halogen Bond Donor Activators
Corresponding Relative Reaction Rates (krel)a 10 and 11, and the Corresponding Relative Reaction Rates
(krel)a
Au conversionb krelc
1 80 10 activator Au conversionb krelc
2 48 6 With Gold(I) Catalysts, Reactions Were Run at −20 °C
3 88 11 10 1 20 2.2
4 8 1 10 2 46 4.9
5 51 6.4 10 3 41 4.4
6 18 2.3 10 5 9 1
a 11 1 37 4.0
Reactions were performed with 0.5 mol % halogen bond donor 9
and gold complexes 1−6 (0.0015 M each), 0.30 M 12, and 0.60 M 13 11 2 61 6.4
in CD2Cl2 at 25 °C, monitoring the conversions by 1H NMR. 11 3 80 8.5
b
Conversions. cRelative reaction rates (krel) following 60 min reaction 11 5 24 2.6
time are shown. As a consequence of the short reaction times, no cis- NaBArF 1 77 7.4
to-trans isomerization was observed. NaBArF 2 89 9.5
NaBArF 3 91 9.7
NaBArF 5 45 4.8
With Gold(III) Catalysts, Reactions Were Run at 25 °C
10 4 25 2.5
10 6 10 1
11 4 86 8.6
11 6 24 2.4
NaBArF 4 78 7.8
NaBArF 6 20 2
a
Reactions were performed with 0.5 mol % halogen bond donor 10 or
11 and gold complex 1−6 (0.0015 M each), 0.30 M 12 and 0.60 M
13 in CD 2 Cl 2 , monitoring the conversions by 1 H NMR.
b
Conversions. cRelative reaction rates (krel) following 30 min reaction
Figure 6. Conversions observed for the cyclopropanation of propargyl time are shown. Due to the short reaction times, no cis-to-trans
acetate 12 (Table 2) to 14 as a function of reaction time using isomerization was observed.
halogen bond donor 9 for activation of gold(I) catalysts 1 (green), 2
(red), 3 (turquoise), and 5 (blue) and gold(III) complexes 4 (bright
green) and 6 (dark red).

iodolium-type 9 and the bidentate 10 and 11 halogen bond

donors provide quicker conversions at very low catalyst
loadings of 0.5 mol %. Gold(III)-catalyzed reactions are
generally slower than the gold(I)-activated ones, even when
activated by a reagent other than a halogen bond donor
(Figure 9).
Catalyst and Activator Recycling. As reagent recycling is
an important aspect of sustainability, we evaluated the
recyclability of the gold catalyst and the activator. NMR
monitoring indicated that the gold precatalysts remained
unchanged throughout all the above reactions (Tables 1−3
and Figures 4−7), and hence, the intact precatalysts were
observed in all crude product mixtures. To evaluate
recyclability, we have used the cyclopropanation reaction
mediated by gold precatalyst 1 and activator 9 as a model
system and performed the reaction on a ∼100 mg scale of
propargyl acetate 12 (93% yield). Following the reaction, one
single flash column enabled the isolation of the target
cyclopropane 14 (93%), gold precatalyst 1 (88%), and
activator 9 (75%). The possibility to easily recover both the
precatalyst and the activator is highly significant as the gold
precatalyst is not recoverable from reactions using anion
exchange when using a silver(I) salt for activation. Under such
classical conditions, the active gold catalyst is quantitatively
prepared from a precatalyst, either in advance or in situ, with a
silver salt that is converted into a AgCl precipitate. The latter
cannot be directly reused. Throughout the reaction, gold
7213
Figure 7. Conversions observed for the cyclopropanation of propargyl
acetate 12 (Table 2) to 14 as a function of reaction time using 10
(triangles), 11 (circles), or NaBArF (squares) for activation of gold(I)
catalysts 1 (green), 2 (red), 3 (turquoise), or 5 (blue). Reactions were
run at −20 °C, and conversions were followed by 1H NMR.
catalyst decomposition is common, but even if it would not be
the case, the gold catalyst still gets decomposed in the
subsequent quenching process. As gold is an expensive and
rare element, the decomposition of its catalytic complexes
poses a serious limitation, particularly for large-scale
applications. Accordingly, an alternative route allowing smooth
recovery of the gold precatalyst, and of the activator, is a great
leap toward sustainability, especially for large-scale synthetic
applications.
The above recyclability implies a dynamic catalyst activation,
where the cationic active gold species is in equilibrium with the
chloride-ligated inactive complex throughout the reaction. This
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis Research Article

Figure 10. The formation of a halogen bond between the Lewis basic
chlorine (halogen bond acceptor, blue) of the gold catalyst and the
Lewis acidic iodine (halogen bond donor, purple) was detected on the
NMR-active nuclei that are closest to the interacting atoms (in bold,
marked with a star) using 13C and 31P NMR spectroscopy. The
coordination shifts observed upon complex formation are given in
Figure 8. Conversions observed for the cyclopropanation of propargyl Tables 4 and 5.
acetate 12 (Table 2) to 14 as a function of reaction time using
halogen bond donors 10 (triangles) or 11 (circles) or NaBArF Table 4. 31P and 13C NMR Coordination Shifts (Δδ, ppm)
(square) for the activation of gold(III) catalysts 4 (green) and 6 for Gold Complexes 1−6 with Activators 7−11a
(red). Reactions were run at 25 °C, and the conversions were
followed by 1H NMR. catalyst 31
P NMR 13
C NMR
activator 1 2 3 4 5 6
7 0.045 0.45 −0.18 −0.36 −0.41 −0.46
8 0.060 0.51 −0.31 −0.39 −0.47 −0.51
9 0.96 7.15 −2.79 −2.56 −3.21 −3.65
10 0.74 5.27 −1.71 −2.10 −2.43 −2.75
11 0.96 6.26 −2.63 −2.85 −3.15 −3.63
a
NMR spectra were acquired in CD2Cl2 at 25 °C with a 0.03 M
concentration of the gold complex and the activator. For the accurate
determination of 31P NMR chemical shifts, PPh3 was used as the
internal standard.

bound to halogen bond donor iodine (Table 5). This

observation is in line with the formation of a halogen bond,

Table 5. 13C NMR Coordination Shifts (Δδ, ppm) of C−I

for XBDs 7−11 with Gold Complexes 1−6a
activator
catalyst 7 8 9 10 11 6
1 1.65 1.33 0.39 3.56 4.93 1.65
2 0.38 0.57 0.41 1.58 2.72 0.38
Figure 9. Normalized relative reaction rates observed for the
cyclopropanation of propargyl acetate 12 (Table 2) to 14 using 3 0.31 0.61 0.32 2.33 3.95 0.31
halogen bond donors 7−11 for activation of gold(I) and gold(III) 4 0.57 0.42 0.38 1.08 2.52 0.57
catalysts 1−6. For comparison, reactions ran at different conditions 5 0.98 1.07 0.24 2.79 3.94 0.98
were normalized to 0.5% catalyst concentration and 25 °C. For 6 0.33 0.48 0.39 1.35 2.94 0.33
details, see Table S1 in the Supporting Information. a
NMR spectra were acquired in CD2Cl2 at 25 °C with a 0.03 M
concentration of the gold complex and the activator.
is different from silver(I)-activated reactions that generate the
cationic gold species irreversibly. The dynamic gold activation corroborating our hypothesis. Following literature conven-
of the halogen bond-activated route is expected to be beneficial tions,42 the coordination shift is defined as the chemical shift
for the stability of the gold catalyst since catalyst decom- difference between the halogen bonded and the free ligand,
position often takes place during the off-cycle.39 Similarly, the that is, δ31Pcoord = δ31Pcomplex − δ31Pligand, when expressed for
31
slow release of cationic, catalytically active gold has been P NMR chemical shifts. Upon halogen bonding, the 31P
reported by using Cu(OTf)2 as the activating agent40 or by chemical shifts of 1 and 2 increased (Table 4). While the
applying thiolate-bridged, dinuclear NHC−Au(I) complexes.41 chemical shift change is small upon interacting with weak
NMR Chemical Shift Perturbation. To gain under- halogen bond donors 7 and 8, it dramatically increases when
standing of the mechanism of gold activation by halogen bond the strong halogen bond donors 9−11 are involved. The
donors, we measured the interaction-induced 13C and 31P magnitude of the chemical shift change shows a clear
NMR chemical shift changes. These were largest for the atoms dependence on the type of gold catalyst (Table 4). Similar
closest to the presumptive interaction sites of the gold to that observed for gold complexes 1 and 2, the coordination
complexes 1−6 and activators 7−11 (Figure 10). Thus, shifts induced by the strong halogen bond donors 9−11 on 3−
upon mixing the gold complexes with the activators (1:1), the 6 were a magnitude larger than those induced by the weak
largest chemical shift changes were observed for carbon or halogen bond donors 7 and 8. The 31P NMR chemical shifts
phosphorous bound to gold (Table 4) and carbon directly show an inverse trend as compared to that observed by 13C
7214 https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis
NMR for 3−6, in line with the previous literature.43 Besides a
few exceptions, most literature studies report the inverse
correlation of 31P NMR chemical shifts to the electron density.
This has been rationalized by electron-withdrawing substitu-
ents (here, charge transfer via halogen bonding) decreasing the
electron density around phosphorus and thereby contracting
its d-orbitals; this is suggested to lead to an increased π back-
bonding with neighboring atoms that, in turn, results in
increased shielding. We further observed that the signal of the
carbenic carbon of gold catalyst 5 split into two upon its
interaction with 9 (Figure S93, Supporting Information),
whereas a single signal was observed for its complexes with 7−
8 and 10−11. The cause of this signal doubling is not entirely
clear; however, 7−8 and 10−11 are structurally closely related
2-iodoimidazolium salts, whereas 9 is an iodolium ion. The
latter is the only halogen bond donor among 7−11 capable of
forming both mono- and dicoordinated complexes with
halogen bond acceptors, which may explain the above
observation.
Diffusion NMR. The coordination of gold catalysts with
halogen bond donor activators was further corroborated by the
observation of the consistent decrease in translational diffusion
coefficient (D) of 1−6 and 7−11 upon their mixing (Figures
11 and 12; for tabulated values, see Supporting Information,
Figure 11. Alteration of the translational diffusion coefficients of gold
complexes 1−6 upon addition of an equivalent of activators 7 (blue),
8 (red), 9 (green), 10 (purple), and 11 (cyan). NMR spectra were
acquired in CD2Cl2 at 25 °C with [Au] = [activator] = 0.03 M.
Figure 12. Alteration of the translational diffusion coefficients of
halogen bond donors 7−11 upon addition of an equivalent of gold
complexes 1 (blue), 2 (red), 3 (green), 4 (purple), 5 (cyan), and 6
(orange). NMR spectra were acquired in CD2Cl2 at 25 °C with [Au]
= [XBD] = 0.03 M.
Tables S4 and S5). This indicates an increase in their Stokes
radii, which is expected upon complexation. The diffusion
coefficient of the component of noncovalent complexes is the
population-weighted average of that of the free and of the
bound forms. Hence, the magnitude of the observed diffusion
7215
Research Article
coefficients is not just modulated by the molecular volume but
also by the interactionhere, the halogen bond strength.
Presuming a comparable volume for each catalyst and
activator, which is simplistic yet not unrealistic, a greater
decrease in diffusion coefficient is indicative of a larger
population of the halogen-bonded complex and hence a
stronger halogen bond. In line with this presumption, we
detected a greater decrease in diffusion coefficient for all gold
complexes when interacting with the stronger halogen bond
donors 9−11 as compared to the weak donors 7 and 8 (Figure
11). Activators 10 and 11 are, however, larger than 9, which in
turn is also smaller than 7 and 8 (Scheme 1). Taking this into
consideration, the decrease in diffusion coefficient suggests 9
to be as strong halogen bond donor as 10−11, which agrees
with the observed chemical shift perturbations shown in Table
4. Accordingly, the largest decrease in the diffusion coefficient
of 7−11 upon complexation with 1−6 was observed for
iodolium complex 9 (Figure 12). The relative alteration in
between the coefficients of 7−11 is, however, difficult to
interpret.
NMR Titration. NMR titration experiments were per-
formed to provide further proof that halogen bonding is the
dominant interaction between the gold catalysts and halogen
bond donors. Halogen bond donor 7 was titrated with 1−10
equiv of gold catalyst 1 while keeping its concentration
constant and monitoring the chemical shift change of all nuclei
of the donor (Figure 13). Upon increasing the concentration
Figure 13. 13C NMR chemical shift change of nuclei C-1 (orange), C-
2 (blue), and C-3 (pink) of halogen bond donor 7 when titrated with
gold catalyst 1 in 2.5 equiv increments. NMR spectra were acquired in
CD2Cl2 at 25 °C with concentrations [7] = 0.03 M and [1] = 0.03−
0.30 M.
of the gold catalyst, the fraction of complexed halogen bond
donor in solution increases, and the chemical shifts are
expected to approach those of the halogen-bonded complex.
The greatest chemical shift increase of the activator throughout
the titration was observed on carbon neighboring iodine, C-1
(1.67 to 5.61 ppm), while other nuclei showed an order of
magnitude smaller shift alterations. This is in line with the
formation of the expected C−I···Cl−Au halogen bond. While
binding curves were obtained in the NMR titrations, the data
were judged to not be complete enough to allow reliable
determination of binding constants but only to estimate that
the binding affinity of the halogen bond donor to the gold
catalyst is expected to be in the low millimolar range.
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis Research Article

Scheme 2. Free Energy Profile Computed for the Cyclopropanation of Propargyl Acetate 12 with Styrene 13 Using a Halogen
Bond Donor to Activate the Gold(III) (Blue) or Gold(I) (Green) Catalysta

a
Simplified models of the gold(I/III) catalysts 3/4 were used due to the time limitations of DFT calculations. Following Int4, the reaction route
leading to the trans product is shown in red whereas that to the cis product in blue. Relative stabilities are given in kcal/mol, with respect to the
energy of the isolated fragments of the reactants (Tables S8 and S9). The full free energy diagram for the gold(I)-mediated process is given in
Figure S109 in the Supporting Information to avoid an overcrowded graph. Computations were done at the B3LYP-D3BJ/Def2-TZVP; for further
details, see the Supporting Information.

Computations. The above observations give a strong
experimental indication that gold catalysts are activated by
halogen bond donors. The latter is expected to lower the
energy barrier for the Au−Cl bond cleavage and thereby
promote the formation of the active gold catalyst. To gain an
insight into the mechanism of halogen bond activation, we
carried out density functional theory (DFT) calculations for
the envisioned reaction intermediates and related transition
states of the gold-catalyzed cyclopropanation (Scheme 1). To
avoid computational limitations posed by the size of the
studied system, a truncated model (Scheme 1) was used, and
the computational method was carefully optimized (see Table
S7 in the Supporting Information). Geometries were optimized
using B3LYP44-D3BJ45,46/Def2-SVP47 /CPCM (CH2Cl2),
hence employing Grimme’s correction for dispersion effects.
Gibbs free energies were corrected using the quasi-harmonic
approximation48 and standard state correction.49
The mechanism of gold activation (Scheme 2) is expected to
begin with the iodine of the activator forming a halogen bond
with chlorine of the gold(I) or gold(III) catalyst. This leads to
the formation of intermediate Int1, whereas the formation of
binary adducts of the two reactants is higher in energy (see
Tables S8 and S9, Supporting Information). According to
DFT, a Au−Cl···I → Au···Cl−I reorganization takes place
simultaneously to coordinate propargyl 12 with gold(I) or
gold(III), leading to the transition state TS1. Here, the
association of alkyne is facilitated by the concerted weakening
of the Au−Cl covalent bond upon strengthening the Cl···I
halogen bond. Since the free energies of TS1 are predicted to
be 26.12 and 20.76 kcal/mol compared to those of the
reactants for the gold(III)- and gold(I)-mediated processes,
respectively, the formation of TS1 is the rate-limiting step.
This energy is in good agreement with the experimentally
observed reaction rates for the gold(I)-mediated process,
leading to a full conversion in ca. 60 h reaction time. The
alternative reaction route analogous to an SN1-type halide
abstraction followed by substrate coordination would have a
>50 kcal/mol energy barrier and is hence not realistic (see
7216
Supporting Information, Figure S112). This observation is in
good agreement with the literature studies.21,50−57 In agree-
ment with the experimentally observed trend of reaction rates,
TS1 of the gold(I)-mediated process has a somewhat lower
energy than that of the analogous gold(III)-mediated reaction
route (Scheme 2). The fully computed free energy profile for
the gold(III)-catalyzed process is shown in Scheme 2, whereas
the corresponding free energy diagram for the gold(I)-
catalyzed reaction is shown in Figure S109 of the Supporting
Information. The gold complex TS1 is converted to Int2 upon
completion of the chloride abstraction by the activator,
adjoined with a simultaneous strengthening of the π-
coordination of propargylacetate to gold. This leads to
activation of the unsaturated bond of the substrate for
cyclization. The chloride complex of the halogen bond donor
leaving the gold(I/III) complex is slightly endothermic and
leads to the formation of Int3. At this point, the halogen bond
donor has completed its role as the activator. The subsequent
1,2-acyl shift takes place by rearrangement of Int3 via TS2,
which involves the carboxylate group and leads to the
formation of the five-membered ring of Int4. The subsequent
ring opening to give the 2-acyl intermediate may proceed via
two possible paths through a higher energy cis (TS3c) (blue)
or a lower energy trans (TS3t) (red) configured transition
state, yielding Int5c and Int5t, respectively. Formation of a
styrene (13) π···π interaction stabilizes the Int6c complex,
whereas no significant energy is gained for the trans-configured
intermediate Int6t. Formation of the cyclopropane ring from
these intermediates passes intermediates of significantly
different activation energies, TS4t and TS4c, and leads to the
cis- and trans-configured products Pc and Pt, respectively.
Overall, both transformations leading to the cis- and trans-
configured products are feasible, which is in agreement with
the experimental observations. Following dissociation of the
active cationic gold catalyst, [Ln-Au]+, from the product, the
neutral precatalyst, [Ln-Au−Cl], is regenerated by chloride
transfer from the halogen bond donor−chloride complex. The
proposed mechanism is energetically feasible.
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis
We further evaluated the influence of the halogen bond
strength on the reaction rate. The electron densities at the
Au−Cl and I···Cl bond critical points ρAu−Cl and ρI···Cl,
respectively, were estimated as these can be taken as indicators
for the interaction strengths. Computations (DFT; for details,
see the Supporting Information) confirm that the rate of the
gold(I)-mediated reaction increases as the electron density at
the I···Cl bond critical point increases, and the electron density
at the Au−Cl bond critical point simultaneously decreases, that
is, as a halogen bond is formed. As the halogen bond strength
of the activator increases, the I···Cl interaction strengthens, and
the Au−Cl bond weakens, which is accordingly manifested in
the alteration of these bond lengths. The noteworthy
correlation (R2 0.962, Figure 14) of the relative reaction rate
Figure 14. Experimental reaction rates of gold(I) complex 3 as a
function of the 13C NMR coordination shifts (ppm) using activators
7−11. Apart from one outlier, all data points suggest the correlation
of the chemical shifts of carbon directly bound to gold(I) and the
reaction rates (see Table S20 in the Supporting Information).
with the 13C NMR coordination shift of carbon directly bound
to gold(I) corroborates the formation of a Cl···I halogen bond
(Figure 14). Unexpectedly, the correlation of the reaction rate
to the electron density (Figures S114 and S115, Supporting
Information) and to the 13C NMR coordination shift of the
halogen bond donor (Figure S116, Supporting Information)
for gold(III) complexes is poor. Topological analysis of the
electron density suggests that the Au(III)−Cl bond is
predominantly covalent, while the I···Cl interaction is
dominantly electrostatic.
The gold catalyst−activator complexes that form the
strongest halogen bonds are expected to provide the highest
reaction rates because these promote the most efficient
chloride abstraction. To confirm this hypothesis, we calculated
(at the M06-2X/Def2-SVP level of theory, Figure S104 in the
Supporting Information) the molecular electrostatic potential
of halogen bond donors 7−11. In line with our expectations,
the halogen bond donors that provided the highest
experimental reaction rates also showed the most positive
computed molecular electrostatic potentials, that is, 11 > 10 >
9 ≫ 8 > 7 (Table S13, Supporting Information).
The correlation of the experimental reaction rates and the
sum of the computed molecular electrostatic potentials of the
activators and the catalysts (Figures S105 and S106) confirms
that the halogen bond strength modulates the rate of
transformation of the studied cyclopropanation reaction.
7217
Research Article
Hence, the stronger halogen bond the activator forms with
the catalyst, the faster the reaction.
Principal component analysis (PCA) of the experimental
observables and computational parameters for gold(I) complex
3 and gold(III) complex 4 indicates that the observed reaction
rate correlates with the experimentally observed chemical shift
of carbon C1 that is covalently bound to gold(I) in 3 (Figure
15) and gold(III) in 4 (Figure 16). The rates of the gold(I)-
Figure 15. PCA of the data of IMesAuCl (3). The rate is proportional
to the chemical shift of C1 (carbene) of the gold catalyst and is
inversely proportional to the nitrogen chemical shift. Here, ρ
represents the electron density at the bond critical point; C and N
are the chemical shifts of C and N atoms with the index indicating the
position of the atoms, respectively; r is the bond length; Hc/ρc is the
bond energy expressed as the energy density divided by the electron
density at the bond critical point; and log10 rate is the natural
logarithm of the reaction rate.
and gold(III)-catalyzed reactions are inversely proportional to
the chemical shift of nitrogen two bonds away from gold. The
inverse correlation is expected and is explained by the 15N
NMR chemical shifts of aromatic heterocycles, in contrast to
the 1H and 13C NMR chemical shifts, being dominantly
influenced by the paramagnetic term instead of the electron
density-related diamagnetic contribution to shielding.58,59 Not
unexpectedly, the reaction rate also correlates to the computed
Au−Cl bond length for both the gold(I)- and gold(III)-
mediated reactions. Furthermore, a somewhat weaker
correlation is seen for the halogen bond energy, expressed as
the energy density divided by the electron density at the bond
critical point (Hc/ρc). These observations are in good
agreement with the fact that halogen bonding plays an
activating role in the gold-mediated process.
■
CONCLUSIONS
Recyclable, homogeneous gold catalyst systems are so far
scarce, and their design is typically based on the use of water-
soluble gold complexes and the reuse of the aqueous
solutions.60−63 Ionic liquids have also been used in a similar
manner.64,65 These techniques do not allow for the isolation of
the catalyst and the activator as pure chemicals and catalysis in
the commonly applied organic solvents. Halogen bond
activation introduced herein as a resource-friendly alternative
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis
Figure 16. PCA of the data of IMesAuCl3 (4). The rate is
proportional to the C1 chemical shift of the gold complex and to
the Au−Cl bond length and is inversely proportional to the electron
density at the Au−Cl bond critical point. Here, ρ represents the
electron density at the bond critical point; C and N are the chemical
shifts of C and N atoms with the index indicating the position of the
atoms, respectively; r is the bond length; Hc/ρc is the bond energy
expressed as the energy density divided by the electron density at the
bond critical point; and log10 rate is the natural logarithm of the
reaction rate.
to the activation of gold(I) and gold(III) chloride precatalysts
by silver salts compensates for the above weaknesses. The
preparation of halogen bond activators 10 and 11 requires
significant synthetic effort. The synthesis of the strong activator
9 is, however, straightforward, and it may therefore be a
promising candidate for future applications. It should be
emphasized that activators 7−11 were used here as model
systems to prove a principle and are not proposed to be the
most optimal alternatives but rather starting points for future
optimization.
The formation of a halogen bond between the activator and
the gold catalyst is indicated by chemical shift perturbation,
diffusion NMR, and chemical shift titration data along with
PCA. The activation is shown to take place via an SN2-type
mechanism, and thus, there is a concerted transfer of chlorine
from gold and the coordination of the unsaturated bond of the
substrate to gold. The original gold catalyst remains intact
during the reaction. No reaction takes place without an
activator. Both the gold catalyst and the halogen bond donor
activator can be efficiently and easily regenerated on a
preparative scale. This is an important advantage, in addition
to avoiding the use of silver for activation, paving the way for
upscaling and for the development of gold-catalyzed ecological
industrial processes. The variety of gold catalysts and halogen
bond donors tested in this work suggests that this concept may
be applicable to a number of additional gold-catalyzed
processes.
The rate of the gold-mediated cyclopropanation is
determined by the halogen bond donor strength of the
activator, as demonstrated by the correlation of the relative
reaction rates and the coordination shifts of carbon directly
bound to gold and that to iodine in the catalyst and the
activator, respectively. An iodolium ion was observed to be a
strong activator as the bidentate bis(imidazolium)- and
7218
Research Article
bis(benzimidazolium)-type halogen bond donors, whereas
the weaker halogen bond monodentate imidazolium-type
donors are less strong activators of gold. DFT computations
estimate a ∼21 kcal/mol energy barrier for gold activation
through halogen bonding for a simplified model system, which
is in good agreement with the experimental reaction times.
In our experiments, activation by NaBArF along with
activator 11 provided the greatest catalytic activities. The
recycling of the activator in the halogen bond-activated process
also allows for the reuse of the expensive BArF− counterion.
Halogen bond activation is herein demonstrated as a resource-
friendly strategy for both gold(I)- and gold(III)-mediated
reactions. The first fragments of understanding of the
mechanism of halogen bond activation are given based on
NMR spectroscopic and computational data. Our results
provide the basis for the development of new, silver-free, and
ecological gold-catalyzed processes. The possibility of simple
and effective regeneration of both the gold catalyst and the
halogen bond donor activator may allow even future large-scale
applications.
■
ASSOCIATED CONTENT
*
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acscatal.2c01864.
Details on the synthesis and spectroscopic data for
compound identification and details on the NMR,
computational, and kinetic investigations (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
Anne Fiksdahl − Department of Chemistry, Norwegian
University of Science and Technology, Trondheim 7491,
Norway; orcid.org/0000-0003-2577-2421;
Email: anne.fiksdahl@ntnu.no
Mate Erdelyi − Department of ChemistryBMC, Uppsala
University, Uppsala SE-751 23, Sweden; orcid.org/0000-
0003-0359-5970; Email: mate.erdelyi@kemi.uu.se
Authors
Helgi Freyr Jónsson − Department of Chemistry, Norwegian
University of Science and Technology, Trondheim 7491,
Norway; orcid.org/0000-0002-1630-598X
Daniel Sethio − Department of ChemistryBMC, Uppsala
University, Uppsala SE-751 23, Sweden; orcid.org/0000-
0002-8075-1482
Julian Wolf − Faculty of Chemistry and Biochemistry, Organic
Chemistry I, Ruhr-Universität Bochum, Bochum 44801,
Germany
Stefan M. Huber − Faculty of Chemistry and Biochemistry,
Organic Chemistry I, Ruhr-Universität Bochum, Bochum
44801, Germany; orcid.org/0000-0002-4125-159X
Complete contact information is available at:
https://pubs.acs.org/10.1021/acscatal.2c01864
Funding
Research Council of Norway (226244/F50), Swedish Research
Council (2020-03431).
Notes
The authors declare no competing financial interest.
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis
■
pubs.acs.org/acscatalysis
ACKNOWLEDGMENTS
We gratefully acknowledge the Norwegian University of
Science and Technology for a Ph.D. studentship for Helgi
Freyr Jónsson. This work was partly supported by the Research
Council of Norway through the Norwegian NMR Platform,
NNP (226244/F50). A/S.M.H. Lundgreens Enkes Fond is
thanked for their generous support in covering NMR
instrument costs. We thank Stefano Battaglia (Uppsala
University) and Oriana Brea Noriega (KTH Royal Institute
of Technology) for helpful discussions. This project made use
of the NMR Uppsala infrastructure, which is funded by the
Department of Chemistry, BMC and the Disciplinary Domain
of Medicine and Pharmacy. We also thank Gaston Courtade
and Nebojsa Simic (Norwegian University of Science and
Technology) for their assistance with diffusion NMR. The
computations were enabled by resources provided by the
Swedish National Infrastructure for Computing (SNIC) at
Tetralith and partially funded by the Swedish Research
Council through grant agreement no. 2018-05973, under
project numbers 2020/5-435, 2021/5-359, and 2021/22-350.
J.W. and S.M.H. acknowledge financial support by the
Deutsche Forschungsgemeinschaft (DFG; grant number HU
1782/6-1).
■
REFERENCES
(1) Hashmi, A. S. K.; Hutchings, G. J. Gold catalysis. Angew. Chem.,
Int. Ed. Engl. 2006, 45, 7896−7936.
(2) Gorin, D. J.; Toste, F. D. Relativistic effects in homogeneous
gold catalysis. Nature 2007, 446, 395−403.
(3) Fürstner, A. Gold and platinum catalysisa convenient tool for
generating molecular complexity. Chem. Soc. Rev. 2009, 38, 3208−
3221.
(4) Krause, N.; Winter, C. Gold-Catalyzed Nucleophilic Cyclization
of Functionalized Allenes: A Powerful Access to Carbo- and
Heterocycles. Chem. Rev. 2011, 111, 1994−2009.
(5) Bandini, M. Gold-catalyzed decorations of arenes and
heteroarenes with C-C multiple bonds. Chem. Soc. Rev. 2011, 40,
1358−1367.
(6) Dorel, R.; Echavarren, A. M. Gold(I)-Catalyzed Activation of
Alkynes for the Construction of Molecular Complexity. Chem. Rev.
2015, 115, 9028−9072.
(7) Hopkinson, M. N.; Tlahuext-Aca, A.; Glorius, F. Merging Visible
Light Photoredox and Gold Catalysis. Acc. Chem. Res. 2016, 49,
2261−2272.
(8) Witzel, S.; Hashmi, A. S. K.; Xie, J. Light in Gold Catalysis.
Chem. Rev. 2021, 121, 8868−8925.
(9) Chintawar, C. C.; Yadav, A. K.; Kumar, A.; Sancheti, S. P.; Patil,
N. T. Divergent Gold Catalysis: Unlocking Molecular Diversity
through Catalyst Control. Chem. Rev. 2021, 121, 8478−8558.
(10) Li, D.; Zang, W.; Bird, M. J.; Hyland, C. J. T.; Shi, M. Gold-
Catalyzed Conversion of Highly Strained Compounds. Chem. Rev.
2021, 121, 8685−8755.
(11) Rocchigiani, L.; Bochmann, M. Recent Advances in Gold(III)
Chemistry: Structure, Bonding, Reactivity, and Role in Homogeneous
Catalysis. Chem. Rev. 2021, 121, 8364−8451.
(12) Wang, T.; Hashmi, A. S. K. 1,2-Migrations onto Gold Carbene
Centers. Chem. Rev. 2021, 121, 8948−8978.
(13) Herrera, R. P.; Gimeno, M. C. Main Avenues in Gold
Coordination Chemistry. Chem. Rev. 2021, 121, 8311−8363.
(14) Webster, S.; O’Rourke, K. M.; Fletcher, C.; Pimlott, S. L.;
Sutherland, A.; Lee, A.-L. Rapid Iododeboronation with and without
Gold Catalysis: Application to Radiolabelling of Arenes. Chem.Eur.
J. 2018, 24, 937−943.
(15) Ciriminna, R.; Falletta, E.; Pina, C. D.; Teles, J. H.; Pagliaro, M.
Industrial Applications of Gold Catalysis. Angew. Chem., Int. Ed. Engl.
2016, 55, 14210−14217.
7219
Research Article
(16) Arcadi, A.; Giuseppe, S. Recent applications of gold catalysis in
organic synthesis. Curr. Org. Chem. 2004, 8, 795−812.
(17) Alshammari, A. S. Heterogeneous Gold Catalysis: From
Discovery to Applications. Catalysts 2019, 9, 402.
(18) Pflästerer, D.; Hashmi, A. S. K. Gold catalysis in total
synthesisrecent achievements. Chem. Soc. Rev. 2016, 45, 1331−
1367.
(19) Collado, A.; Nelson, D. J.; Nolan, S. P. Optimizing Catalyst and
Reaction Conditions in Gold(I) Catalysis-Ligand Development.
Chem. Rev. 2021, 121, 8559−8612.
(20) Wang, D.; Cai, R.; Sharma, S.; Jirak, J.; Thummanapelli, S. K.;
Akhmedov, N. G.; Zhang, H.; Liu, X.; Petersen, J. L.; Shi, X. “Silver
Effect” in Gold(I) Catalysis: An Overlooked Important Factor. J. Am.
Chem. Soc. 2012, 134, 9012−9019.
(21) Ranieri, B.; Escofet, I.; Echavarren, A. M. Anatomy of gold
catalysts: facts and myths. Org. Biomol. Chem. 2015, 13, 7103−7118.
(22) Jia, M.; Bandini, M. Counterion Effects in Homogeneous Gold
Catalysis. ACS Catal. 2015, 5, 1638−1652.
(23) Eisler, R. Silver Hazards to Fish, Wildlife, and Invertebrates: A
Synoptic Review; Contaminant Hazard Biological Report; US Depart-
ment of the Interior, National Biological Service: Laurel, MD, USA,
1996.
(24) Franchino, A.; Montesinos-Magraner, M.; Echavarren, A. M.
Silver-Free Catalysis with Gold(I) Chloride Complexes. Bull. Chem.
Soc. Jpn. 2021, 94, 1099−1117.
(25) Sen, S.; Gabbaï, F. P. An ambiphilic phosphine/H-bond donor
ligand and its application to the gold mediated cyclization of
propargylamides. Chem. Commun. 2017, 53, 13356−13358.
(26) Devillard, M.; Nicolas, E.; Appelt, C.; Backs, J.; Mallet-Ladeira,
S.; Bouhadir, G.; Slootweg, J. C.; Uhl, W.; Bourissou, D. Novel
zwitterionic complexes arising from the coordination of an ambiphilic
phosphorus-aluminum ligand to gold. Chem. Commun. 2014, 50,
14805−14808.
(27) Fang, W.; Presset, M.; Guérinot, A.; Bour, C.; Bezzenine-
Lafollée, S.; Gandon, V. Silver- Free Two- Component Approach in
Gold Catalysis: Activation of [ LAuCl] Complexes with Derivatives of
Copper, Zinc, Indium, Bismuth, and Other Lewis Acids. Chem.Eur.
J. 2014, 20, 5439−5446.
(28) Desiraju, G. R.; Ho, P. S.; Kloo, L.; Legon, A. C.; Marquardt,
R.; Metrangolo, P.; Politzer, P.; Resnati, G.; Rissanen, K. Definition of
the halogen bond (IUPAC Recommendations 2013). Pure Appl.
Chem. 2013, 85, 1711−1713.
(29) Bulfield, D.; Huber, S. M. Halogen Bonding in Organic
Synthesis and Organocatalysis. Chem.Eur. J. 2016, 22, 14434−
14450.
(30) Bamberger, J.; Ostler, F.; Mancheño, O. G. Frontiers in
Halogen and Chalcogen-Bond Donor Organocatalysis. Chemcatchem
2019, 11, 5198−5211.
(31) Sutar, R. L.; Huber, S. M. Catalysis of Organic Reactions
through Halogen Bonding. ACS Catal. 2019, 9, 9622−9639.
(32) Wolf, J.; Huber, F.; Erochok, N.; Heinen, F.; Guérin, V.;
Legault, C. Y.; Kirsch, S. F.; Huber, S. M. Activation of a Metal-
Halogen Bond by Halogen Bonding. Angew. Chem., Int. Ed. Engl.
2020, 59, 16496−16500.
(33) Johansson, M. J.; Gorin, D. J.; Staben, S. T.; Toste, F. D.
Gold(I)-catalyzed stereoselective olefin cyclopropanation. J. Am.
Chem. Soc. 2005, 127, 18002−18003.
(34) Reiersølmoen, A. C.; Østrem, E.; Fiksdahl, A. Gold(III)-
Catalysed cis-to-trans Cyclopropyl Isomerization. Eur. J. Org. Chem.
2018, 2018, 3317−3325.
(35) Solvi, T. N.; Reiersølmoen, A. C.; Orthaber, A.; Fiksdahl, A.
Studies towards Pyridine-Based N,N,O-Gold(III) Complexes: Syn-
thesis, Characterization and Application. Eur. J. Org. Chem. 2020,
2020, 7062−7068.
(36) Reiersølmoen, A. C.; Solvi, T. N.; Fiksdahl, A. Au(III)
complexes with tetradentate-cyclam-based ligands. Beilstein J. Org.
Chem. 2021, 17, 186−192.
https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220
ACS Catalysis pubs.acs.org/acscatalysis Research Article

(37) Reiersølmoen, A. C.; Fiksdahl, A. Pyridine- and Quinoline- Cycloisomerization of 7-Aryl-1,6-Enynes. Organometallics 2014, 33,
Based Gold(III) Complexes: Synthesis, Characterization, and 6466−6473.
Application. Eur. J. Org. Chem. 2020, 2020, 2867−2877. (57) Couce-Rios, A.; Kovács, G.; Ujaque, G.; Lledós, A. Hydro-
(38) Haraguchi, R.; Hoshino, S.; Sakai, M.; Tanazawa, S.-g.; Morita, amination of C-C Multiple Bonds with Hydrazine Catalyzed by N-
Y.; Komatsu, T.; Fukuzawa, S.-i. Bulky iodotriazolium tetrafluor- Heterocyclic Carbene-Gold(I) Complexes: Substrate and Ligand
oborates as highly active halogen-bonding-donor catalysts. Chem. Effects. ACS Catal. 2015, 5, 815−829.
Commun. 2018, 54, 10320−10323. (58) Pazderski, L. 15N NMR coordination shifts in Pd(II), Pt(II),
(39) Kumar, M.; Jasinski, J.; Hammond, G. B.; Xu, B. Alkyne/ Au(III), Co(III), Rh(III), Ir(III), Pd(IV), and Pt(IV) complexes with
alkene/allene-induced disproportionation of cationic gold(I) catalyst. pyridine, 2,2′-bipyridine, 1,10-phenanthroline, quinoline, isoquino-
Chem.Eur. J. 2014, 20, 3113−3119. line, 2,2′-biquinoline, 2,2′:6′, 2′-terpyridine and their alkyl or aryl.
(40) Guérinot, A.; Fang, W.; Sircoglou, M.; Bour, C.; Bezzenine- Magn. Reson. Chem. 2008, 46, S3−S15.
Lafollée, S.; Gandon, V. Copper Salts as Additives in Gold(I)- (59) Solum, M. S.; Altmann, K. L.; Strohmeier, M.; Berges, D. A.;
Catalyzed Reactions. Angew. Chem., Int. Ed. Engl. 2013, 52, 5848− Zhang, Y.; Facelli, J. C.; Pugmire, R. J.; Grant, D. M. 15N Chemical
5852. Shift Principal Values in Nitrogen Heterocycles. J. Am. Chem. Soc.
(41) Mora, M.; Gimeno, M. C.; Visbal, R. Recent advances in gold- 1997, 119, 9804−9809.
NHC complexes with biological properties. Chem. Soc. Rev. 2019, 48, (60) Tomás-Mendivil, E.; Toullec, P. Y.; Borge, J.; Conejero, S.;
447−462. Michelet, V.; Cadierno, V. Water-Soluble Gold(I) and Gold(III)
(42) Carlsson, A.-C. C.; Mehmeti, K.; Uhrbom, M.; Karim, A.; Complexes with Sulfonated N-Heterocyclic Carbene Ligands: Syn-
Bedin, M.; Puttreddy, R.; Kleinmaier, R.; Neverov, A. A.; thesis, Characterization, and Application in the Catalytic Cyclo-
Nekoueishahraki, B.; Gräfenstein, J.; Rissanen, K.; Erdélyi, M. isomerization of gamma-Alkynoic Acids into Enol-Lactones. ACS
Substituent Effects on the [N-I-N]+ Halogen Bond. J. Am. Chem. Catal. 2013, 3, 3086−3098.
Soc. 2016, 138, 9853−9863. (61) Sak, H.; Mawick, M.; Krause, N. Sustainable Gold Catalysis in
(43) Dunn, E. J.; Purdon, J. G.; Bannard, R. A. B.; Albright, K.; Water Using Cyclodextrin-tagged NHC-Gold Complexes. Chem-
Buncel, E. Correlation of P-31 Nuclear Magnetic-Resonance CatChem 2019, 11, 5821−5829.
Chemical-Shifts in Aryl Phosphinates with Hammett Substituent (62) Fernandez, G. A.; Dorn, V. B.; Chopa, A. B.; Silbestri, G. F.
ConstantsInductive Versus Resonance Interactions and Relevance Sulphonated NHC-gold(I) complexes in aqueous catalysis. An
to P-Pi-D-Pi Bonding. Can. J. Chem. 1988, 66, 3137−3142. experimental and DFT computational analysis. 19th Int. Electronic
(44) Becke, A. D. Density-functional exchange-energy approxima- Conf. on Synth. Org. Chem., 2015.
tion with correct asymptotic behavior. Phys. Rev. A: At., Mol., Opt. (63) Minkler, S. R. K.; Lipshutz, B. H.; Krause, N. Gold Catalysis in
Phys. 1988, 38, 3098−3100. Micellar Systems. Angew. Chem., Int. Ed. Engl. 2011, 50, 7820−7823.
(45) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping (64) Baron, M.; Biffis, A. Gold(I) Complexes in Ionic Liquids: An
function in dispersion corrected density functional theory. J. Comput. Efficient Catalytic System for the C-H Functionalization of Arenes
Chem. 2011, 32, 1456−1465. and Heteroarenes under Mild Conditions. Eur. J. Org. Chem. 2019,
(46) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- 2019, 3687−3693.
Salvetti correlation-energy formula into a functional of the electron (65) Marinelli, F.; Ambrogio, I.; Arcadi, A.; Cacchi, S.; Fabrizi, G.
density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. Gold-catalyzed synthesis of 2-substituted, 2,3-Disubstituted and 1,2,3-
(47) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence, trisubstituted Indoles in [Bmim]BF4. Synlett 2007, 2007, 1775−1779.
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(48) Grimme, S. Supramolecular binding thermodynamics by
dispersion-corrected density functional theory. Chem.Eur. J. 2012,
18, 9955−9964.
(49) Jensen, J. H. Predicting accurate absolute binding energies in
aqueous solution: thermodynamic considerations for electronic
structure methods. Phys. Chem. Chem. Phys. 2015, 17, 12441−12451.
(50) Komiya, S.; Albright, T. A.; Hoffmann, R.; Kochi, J. K.
Reductive elimination and isomerization of organogold complexes.
Theoretical studies of trialkylgold species as reactive intermediates. J.
Am. Chem. Soc. 1976, 98, 7255−7265.
(51) Colburn, C. B.; Hill, W. E.; Mcauliffe, C. A.; Parish, R. V. P-31
Nmr-Study of Tertiary Phosphine Complexes of Gold(I). Chem.
Commun. 1979, 218−219.
(52) Dickson, P. N.; Wehrli, A.; Geier, G. Coordination Chemistry
of Gold(I) with Cyanide and 1-Methylpyridine-2-ThioneKinetics
and Thermodynamics of Ligand-Exchange at Gold(I) in Aqueous-
Solution. Inorg. Chem. 1988, 27, 2921−2925.
(53) Komiya, S.; Iwata, M.; Sone, T.; Fukuoka, A. Isolation of
Highly Nucleophilic Gold(I) Alkoxides Having a Tertiary Phosphine
Ligand. Chem. Commun. 1992, 16, 1109−1110.
(54) Zhdanko, A.; Maier, M. E. Quantitative Evaluation of the
Stability of gem-Diaurated Species in Reactions with Nucleophiles.
Organometallics 2013, 32, 2000−2006.
(55) Xi, Y.; Wang, Q.; Su, Y.; Li, M.; Shi, X. Quantitative kinetic
investigation of triazole-gold(I) complex catalyzed [3,3]-rearrange-
ment of propargyl ester. Chem. Commun. 2014, 50, 2158−2160.
(56) Brooner, R. E. M.; Robertson, B. D.; Widenhoefer, R. A.
Mechanism of 1,3-Hydrogen Migration in a Gold Bicyclo[3.2.0]-
heptene Complex: The Role of Bronsted Acid in the Gold-Catalyzed

7220 https://doi.org/10.1021/acscatal.2c01864
ACS Catal. 2022, 12, 7210−7220