Experimental Evaluation of (L)Au Electron-Donor Ability in Cationic Gold Carbene Complexes

Article abstract of DOI:10.1002/chem.201703820

²⁹Si NMR spectroscopy was employed to evaluate the electron donor properties of the (L)Au fragments in the cationic gold (β,β-disilyl)vinylidene complexes [(L)Au=C=CSi(Me)₂CH₂CH₂Si(Me)₂]⁺B(C₆F₅)₄^? [L=P(tBu)₂o-biphenyl or NHC] relative to the p-substituted aryl group in the α-aryl-(β,β-disilyl)vinyl cations [(p-C₆H₄X)-C= CSi(Me)₂CH₂CH₂Si(Me)₂]⁺B(C₆F₅)₄^?. Similarly, ¹⁹F NMR was employed to evaluate the σ- and π-electron donor properties of the (L)Au fragments in the neutral gold fluorophenyl complexes (L)Au(C₆H₄F) and in the cationic (fluorophenyl)methoxycarbene complexes [(L)AuC(OMe)(C₆H₄F)]⁺SbF₆^? [L=P(tBu)₂o-biphenyl or IPr] relative to the p-substituted aryl group of the protonated monofluorobenzophenones [(p-C₆H₄X)(C₆H₄F)COH]⁺OTf^?. The results of these studies indicate that relative to p-substituted aryl groups, the gold (L)Au fragments [L=P(tBu)₂o-biphenyl or NHC] are significantly more inductively electron donating and are comparable π-donors and for this reason, the extent of (L)Au→C1 electron donation in gold carbene complexes appears to exceed that provided by a p-(dimethyamino)phenyl group. Furthermore, the [L=P(tBu)₂o-biphenyl]Au fragment is a nominally stronger electron donor than the (IPr)Au fragment, and both are significantly more inductively electron donating than the (PPh₃)Au and [P(OMe)₃]Au fragments.

Full text of DOI:10.1002/chem.201703820

A Journal of
Accepted Article
Title: Experimental Evaluation of (L)Au Electron Donor Ability in
Cationic Gold Carbene Complexes
Authors: Ross Widenhoefer, Robert Carden, and nathan lam
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201703820
Link to VoR: http://dx.doi.org/10.1002/chem.201703820
Supported by

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
Experimental Evaluation of (L)Au Electron Donor Ability in
Cationic Gold Carbene Complexes
Robert G. Carden, Nathan Lam, and Ross A. Widenhoefer*
Abstract: 29Si NMR spectroscopy was employed to evaluate the
electron donor properties of the (L)Au fragments in the cationic gold
is similar to that provided by a methoxy group.[4] Straub
attributed the relatively short Au–C(Mes)2 bond of the structurally
characterized bis(mesityl)carbene complex 3 to the "significant,
but not predominant double bond character Au=CMes2".[5]
(,-disilyl)vinylidene
complexes
–
[(L)Au=C=CSi(Me)2CH2CH2Si(Me)2]+ B(C6F5)4
[L = P(tBu)2o-biphenyl
Similarly,
structural
analysis
of
the
gold
or NHC] relative to the p-substituted aryl group in the -aryl-(-
cyclopropyl(methoxy)carbene complex 4 suggested that the
extent of (L)Au  C electron donation was similar to that
provided by a cyclopropyl group.[6]
concluded that "there is only little back donation of electron
density from gold to the carbene center" in the structurally
characterized bis(anisyl)carbene complex 5.[7]
–
[(p-C6H4X)–C=CSi(Me)2CH2CH2Si(Me)2]+ B(C6F5)4
disilyl)vinyl cations
.
Similarly, 19F NMR was employed to evaluate the - and -electron
donor properties of the (L)Au fragments in the neutral gold
fluorophenyl complexes (L)Au(C6H4F) and in the cationic
(fluorophenyl)methoxycarbene complexes [(L)AuC(OMe)(C6H4F)]+
SbF6– [L = P(t-Bu)2o-biphenyl or IPr] relative to the p-substituted aryl
group of the protonated monofluorobenzophenones [(p-
C6H4X)(C6H4F)COH]+ OTf–. The results of these studies indicate that
relative to p-substituted aryl groups, the gold (L)Au fragments [L =
P(t-Bu)2o-biphenyl or NHC] are significantly more inductively
electron releasing and are comparable -donors and for this reason,
the extent of (L)Au  C1 electron donation in gold carbene
Conversely, Fürstner
complexes appears to exceed that provided by
a
p-
(dimethyamino)phenyl group. Furthermore, the [L = P(tBu)2o-
biphenyl]Au fragment is a nominally stronger electron donor than is
the (IPr)Au fragment while both are significantly more inductively
electron releasing than are the (PPh3)Au and [P(OMe)3]Au
fragments.
Figure 1. Cationic Gold Carbene Complexes.
Introduction
Embedded within the debate regarding the extent of (L)Au 
C electron donation in gold carbene complexes is the role of the
supporting ligand in modulating the electron donor ability of the
(L)Au fragment.[1,2] In particular, the computationally-derived
bonding model for gold carbene complexes proposed by Toste
and Goddard invokes an L–Au–C bonding network consisting of
a three-center, four-electron -hyperbond and two orthogonal -
bonds involving donation of electron density from filled metal 5d
orbitals to -acceptor orbitals on the ligand and carbene carbon
atom (Figure 2).[4] Owing to the nature of these orbitals and the
competition for electron density, the better the sigma donor is L,
the weaker is the sigma component of the Au–C1 bond, and
similarly, the more -acidic is L, the weaker is the Au  C1 back
donation.[4]
Cationic gold carbene complexes have been widely invoked as
intermediates in a diverse range of gold(I)-catalyzed reactions,
most notably enyne cycloisomerizations and alkene
cyclopropanation.[1,2]
Although the intermediacy of gold
carbenes in these transformations is supported by a wealth of
experimental and computational data, there is still much debate
regarding the nature of the gold-carbon bond, specifically the
extent of d  p back bonding and, more generally, the extent of
electron donation from the (L)Au fragment to the electron-
deficient C1 carbon atom.[2] For example, Fürstner concluded
that stabilization of the -dialkoxy allylic cation by the gold
phosphine fragment in complexes 1 was “marginal” on the basis
of C2–C3 rotational barriers (Figure 1).[3] However, DFT
analysis of complexes 2 by Toste and Goddard suggested that
stabilization of a tertiary allylic carbocation by a (Me3P)Au group
[a]
Robert G. Carden, Nathan Lam, and Prof. Ross A. Widenhoefer
Department of Chemistry
Duke University
French Family Science Center, Durham, North Carolina, USA
E-mail: rwidenho@chem.duke.edu
Supporting information for this article is given via a link at the end of
the document.
Figure 2. Proposed bonding model for gold carbene complexes.[4]
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
The pronounced effect of the supporting ligand on the
catalytic behavior of gold(I) carbene complexes is well
documented,[8] including notable differences between N-
heterocyclic carbene (NHC) ligands and dialkyl(o-biphenyl)
phosphine ligands, which are among the most important
supporting ligands employed in gold(I) catalysis and which are
both considered to be strong -donors and weak -acceptor
ligands.[8] For example, Echavarren has shown that the product
ratio formed in the cycloaddition/arylation of enyne 6 with indole
found to track inversely with the -donor ability of the (L)Au
fragment which increased in the order P(OR)3 < PPh3 < P2 < IPr.
More directly, Chen has evaluated the ligand-dependent
stabilization of cationic gold carbene complexes by measuring
the energy barriers for the formation of gold arylidene complexes
11a and 11b via collision-induced dissociation (CID) of the
corresponding gold sulfonyl-imidazolylidene precursors in the
gas phase (Scheme 4).[14] Quantitative CID threshold analysis
for the formation of 11a and 11b indicated that the IPr ligand
stabilized the carbene by 2.3 ± 2.7 kcal/mol relative to P1, which,
despite the error in this analysis, correlated well with the value of
1.8 kcal/mol predicted by DFT calculations.[14]
–
catalyzed by [(L)AuNCMe]+ SbF6 complexes changed from
80:20 favoring product 7a when L = P(t-Bu)2o-biphenyl (P1) to
75:25 favoring product 7b when L = IPr (Scheme 1).[9] Similarly,
Barriault documented the reversal of 6-endo/5-exo selectivity in
the gold(I)-catalyzed cycloisomerization of 1,6-enyne
8
employing either IPr or o-biphenylphosphine P2 as supporting
ligand (Scheme 2).[10]
Scheme 3
Scheme 1
Ph
Ph
Me
O
Ph
O
Me
TIPSO
Me
[LAuNCMe]SbF6 (5 mol%)
acetone
+
E
E
E
E
E
E
8
Scheme 4
P(Cy)2
iPr
L
endo:exo
yield
iPr
P2
63:37
13:87
89%
86%
iPr
IPr
We therefore sought to experimentally evaluate the ligand-
dependent electron donor ability of (L)Au fragments in cationic
gold carbene complexes, including an assessment of the relative
contributions - and -donation from the (L)Au fragment. Our
previous efforts directed toward the experimental evaluation of
the electron donor ability of the (L)Au fragment in cationic gold
carbene complexes analyzed the charge-dependent perturbation
of C–C bond lengths in the cyclopropyl group of the
cyclopropyl(methoxy) carbene complex 4 relative to protonated
cyclopropyl ketones (Figure 1).[6] However, this approach
suffered from the large experimental error in the C–C bond
lengths relative to the extent of perturbation and the necessity of
obtaining suitable crystals of both the carbene complexes and
oxocarbenium model compounds for analysis. To circumvent
these limitations, we sought to exploit the large dispersion and
charge-dependent chemical shifts of heteronuclear NMR to
evaluate charge distribution and (L)Au  C electron donation in
cationic gold carbene complexes.
P2
Scheme 2
In contrast to the analysis of ligand effects in gold(I) catalysis,
experimental quantification of the effect of supporting ligand on
the electron donor properties of (L)Au fragments, particularly
those in cationic gold carbene complexes, is largely absent.[11]
For example, Belpassi and co-workers measured the ligand-
dependent activation barriers for rotation about the C–
N(pyrrolidine) bond of gold NHC complexes 9 and correlated
these values to the computationally derived extent of d  
backbonding, although a direct comparison of alkyl phosphine
and NHC ligands was not feasible (Scheme 3).[12] Hashmi
investigated the ligand-dependent hydrolytic cleavage of -
dialkoxyl gold carbene complexes 10 and correlated these
reaction rates with the ligand-dependent LUMO energies of the
carbene complex (Scheme 3).[13] The rate of hydrolysis was
Here we report the 29Si NMR analysis of cationic gold (,-
disilyl)vinylidene complexes and the 19F NMR analysis of neutral
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
gold
fluorophenyl
complexes
and
cationic
gold
Table 1. Select 13C and 29Si NMR data for gold (-disilyl)vinylidene
complexes 13 and -aryl--disilyl vinyl cations 15.
flourophenyl(methoxy)carbene complexes in combination with
the spectroscopic analysis of relevant organic model compounds.
These data allowed comparisons to be made between the
electron donor ability of various (L)Au fragments relative to p-
substituted aryl groups. The key conclusions drawn from this
work are three-fold. (1) Relative to p-substituted aryl groups,
the (L)Au (L = P1, NHC) fragments are more inductively electron
releasing and are comparable -donors. (2) The net electron
donor ability of the (L)Au (L = P1, NHC) fragments is significant
and exceeds that of a p-(dimethylamino)phenyl group. (3) The
(P1)Au fragment is a nominally stronger electron donor than is
the (IPr)Au fragment, although both are significantly more
inductively electron releasing than are gold triarylphosphine or
triarylphosphite fragments.
Compound
-Substituent
P1
IPr
IMes
SIPr

13C (C1)
206

13C (C2)
112
111.8
112.3
114.2
115.2
83.6
84.5
84.0
85.9

29Si
 29Si[a]
+54.1
+56.7
+55.2
+57.0
+55.3
+70.2
+66.6
+65.0
+52.2
13a
13b
13c
13d
13e
15a
15b
15c
15d
33.5
35.8
34.6
36.2
34.5
56.0
52.3
50.2
37.3
203.5
202.8
198.5
202.8
183.9
185.8
187.2
194.9
SIMes
4-C6H4Me
4-C6H4OPh
4-C6H4OMe
4-C6H4NMe2
[a]  29Si refers to the chemical shift difference between  29Si of the
vinylidene complex or vinyl cation and the  29Si of the C(sp)-bound silicon
atom of the neutral acetylene precursor.
The propensity of a silicon atom to stabilize a carbenium ion
in the -position via hyperconjucation (-effect) is well
documented.[23] The -SiC hyperconjugation, in turn, leads to
the depletion of electron density from the silicon atom, which is
evidenced by a number of spectroscopic features, most notably
the deshielding of the silicon atom in the 29Si NMR spectrum.[15-
Results and Discussion
Cationic Gold (,-Disilyl)vinylidene Complexes. Guided by
the work of Müller,[15-19] we recently reported that low-
temperature hydride abstraction from the gold acetylide complex
12a with triphenylcarbenium tetrakis(pentafluorophenyl)borate
leads to selective formation of the thermally unstable cationic
gold (,-disilyl)vinylidene complex 13a (Scheme 5).[20,21] Key to
characterization of 13a were resonances at  = 206 and 112 in
the 13C NMR spectrum assigned to the C1 and C2 vinylidene
carbon atoms, respectively (Table 1). Also noteworthy was the
significant decrease in the one-bond C1–C2 coupling constant of
19,24-26]
Noteworthy therefore, was a single sharp resonance at 
= 33.5 in the 29Si NMR (–80 °C) spectrum of 13a that was
significantly deshielded ( = +54.1) relative to the 29Si NMR
resonance of the C(sp)-bound silicon atom of gold acetylide
complex 12a (
=
–20.6;
Table 1) owing to -SiC
hyperconjugation as represented by resonance structures C and
D (Figure 3). Importantly, all of the spectroscopy of 13a is
consistent with a symmetric Y-shaped or rapidly equilibrating
distorted Y-shaped ground state structure, both of which have
been validated for -substituted--disilyl vinyl cations.[15-19]
13a (1JCC = 60 Hz) relative to the acetylide precursor 13a (1JCC
=
91 Hz; 1JCC = 31 Hz) consistent with the diminished s-
character of the C2 atom of 13a relative to 12a.[22] Detection of
the C1 and C2 vinylidene carbon atoms was complicated by the
facile interconversion (G‡ = 9.7 kcal/mol) of these atoms,
presumably via the gold -disilacyclohexyne intermediate Ia.
Figure 3. Relevant resonance contributors for gold vinylidene complexes 13.
It has been shown that the extent of -SiC hyperconjugation
across a range of -silyl substituted carbenium ions is not
constant, but increases with the increasing electron demand of
the electron-deficient carbon atom.[15,17,24-26] Notably, Müller
quantified this relationship in the context of the -aryl--disilyl
vinyl cations 14 containing a six-membered disilacycle by
establishing a correlation between the 29Si NMR chemical shift
and the Hammett-Brown + parameter of the p-substituent of the
-aryl group (Chart I).[15] The Hammett-Brown + parameter has
been employed to good effect to characterize the electron
donor/acceptor properties of substituents that can effectively
delocalize positive change to the reaction center through
conjugation with more negative values indicating greater
stabilization of positive charge.[27] We therefore reasoned that
29Si NMR chemical shifts could be similarly employed to
evaluate the electron donor ability of (L)Au fragment in the
cationic gold (,-disilyl)vinylidene complex 13a and related
Scheme 5
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
NMR spectra of cations 15a - 15d displayed diagnostic
resonances at  = 184 - 195 and 84 - 86 assigned to the C1 and
C2 vinyl carbon atoms, respectively (Table 1). In the 29Si NMR
spectra of vinyl cations 15,  29Si depended strongly on the
electron donor ability of the -aryl group and ranged from  =
56.0 for the p-tolyl substituted vinyl cation 15a to  = 37.3 for the
p-(dimethylamino)phenyl substituted vinyl cation 15d, all of
which were significantly deshielded ( 29Si = 52.2 - 70.2)
relative to the C(sp)-bound silicon atom of aryl acetylene
precursors 16a-d ( = –14.1 - –14.9).[31] A plot of the Hammett-
Brown + parameter versus  29Si was linear (R2 = 0.99),[27]
which established a correlation between the 29Si NMR chemical
shift and the electron donor ability of the -aryl group of vinyl
cations 15 (Figure 4).
derivatives, provided that
a similar correlation could be
established between the electron donor ability of the -
substituent and the 29Si NMR chemical shift of the -aryl--
disilyl cations 15 containing a five-membered disilacycle (Chart
I).[28,29]
Chart I. -Aryl--disilyl cations 14 and 15.
Toward the objective of evaluating the electron donor ability
of the (L)Au fragment in gold (,-disilyl)vinylidene complexes,
we first sought to expand the scope of gold vinylidene
complexes with respect to supporting ligand. To this end, we
synthesized the thermally unstable gold (,-disilyl)vinylidene
–
[(L)Au=C=CSi(Me)2CH2CH2Si(Me)2]+ B(C6F5)4
complexes
[L = IPr
(13b), IMes (13c), SIPr (13d), SIMes (13e)] that contained an
NHC ligand employing procedures similar to that employed to
that used to synthesize 13a. Conversely, attempted synthesis of
gold (,-disilyl)vinylidene complexes containing less electron
donating ligands such as PPh3 proved unsuccessful. The
thermally unstable complexes 13b - 13e were characterized
without isolation by low temperature 1H, 13C, and 29Si NMR
spectroscopy (Table 1). The 13C NMR spectra of complexes
13b - 13e displayed vinylidene C1 and C2 resonances at  =
203 - 198 and  = 112 - 115, respectively (Table 1), that in
contrast to those of 13a, displayed no broadening at –80 °C.
This observation pointed to a higher energy barrier for C2/C1
interconversion via the corresponding -disilacyclohexyne
intermediates Ib-e relative to 13a. Indeed, spin saturation
transfer analysis of the IPr-supported gold (,-disilyl)vinylidene
complex 13b provided an energy barrier of G‡ = 11.4 ± 0.3
kcal/mol for interconversion of the C1 and C2 vinylidene carbon
atoms. Each of the complexes 13b - 13e displayed a single
sharp resonance in the 29Si NMR spectrum at –80 °C. The 29Si
NMR chemical shifts ( 29Si) showed only slight dependence on
the nature of the supporting ligand and ranged from  29Si = 34.5
for 13e (L = SIMes) to  29Si = 36.2 for 13d (L = SIPr), all of
which were slightly deshielded relative to phosphine derivative
13a ( = 33.5; Table 1).
Scheme 6
60
15a
55
15c
15b
50
45
40
35
30
15d
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
Figure 4. Plot of the Hammett-Brown + parameter versus  29Si for -
aryl,-disilyl vinyl cations 15a-15d where  29Si = (13.2 ± 0.9)+ + (59.9 ±
0.9), R = 0.99. Dotted lines represent the extrapolation of  29Si for gold
vinylidene complexes 13 onto this curve: from top to bottom 13d, 13b, 13d +
13e, 13a.
-Aryl--Disilyl Vinyl Cations.
We next sought to
establish a correlation between the 29Si NMR chemical shift (
29Si) and the electron donor ability of the -substituent in the -
aryl--disilyl vinyl cations 15. To this end, the thermally
Discussion of 29Si NMR Data. From the 29Si NMR data for
gold -disilyl vinylidene complexes 13 and -aryl--disilyl
vinyl cations 15, a number of conclusions can be drawn with
varying levels of confidence. Firstly, the correlation between +
and  29Si for -aryl,-disilyl cations 15 established that the
29Si NMR chemical shift is sensitive to and dependent on the
electron donor ability of the -substituent. Therefore, we can
reasonably conclude that the 29Si NMR chemical shift of gold
vinylidene complexes 13 likewise reflects with similar sensitivity
the electron-donor ability of the respective (L)Au fragment.
unstable
vinyl
cations
–
[(p-C6H4X)–C=CSi(Me)2CH2CH2Si(Me)2]+ B(C6F5)4
[R = Me (15a),
OPh (15b), OMe (15c), NMe2 (15d)] were synthesized via
treatment of the corresponding aryl acetylenes 16a-d with
triphenylcarbenium tetrakis(pentafluorophenyl)borate at –78 °C
in CD2Cl2 and were characterized without isolation by 1H, 13C,
and 29Si NMR spectroscopy at 0 °C (Scheme 6).[30] The 13C
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
Therefore, these data indicate that the electron donor ability of
the (L)Au fragments decreases in the order (P1)Au > (IMes)Au ≈
(SIMes)Au > (IPr)Au > (SIPr)Au, although the magnitude of the
difference between the most and least electron donating
fragment is small and less than the difference between a p-tolyl
group and a p-phenoxyphenyl group.
Taking the  29Si of vinylidene complexes 13 and vinyl
cations 15 as a measure of the electron donor ability of the -
substituent leads to the somewhat unexpected conclusion that
all of the (L)Au fragments investigated ( 29Si = 36.2 - 34.5) are
more electron donating than is the p-(dimethylamino)phenyl
group of 15d (29Si = 37.3). A number of considerations support
the validity of this comparison. Firstly, the spectroscopy of both
-donation from the -aryl group would be felt in both the silyl
alkyne moiety and the vinyl cation, whereas -donation from the
-aryl group would be felt only in the vinyl cation, the  29Si
values presumably reflects the -donor ability of the -aryl group
to a greater extent than does  29Si. However, because  29Si of
the silyl alkynes was largely invariant of the nature of the -aryl
group, similarly strong correlations exists between the Hammett-
Brown + parameter and either  29Si or  29Si for vinyl cations
14.
As was the case for vinyl cations 14 and their neutral silyl
alkyne precursors, the  29Si of the C(sp)-bound silicon atom of
aryl silyl alkynes 16 was largely invariant of the nature of the -
aryl group and -aryl,-disilyl cations 15 displayed equally
good correlation between the Hammett-Brown + and either 
29Si (R2 = 0.99) or  29Si (R2 = 0.99; Figure 5). Likewise, the 
29Si of the C(sp)-bound silicon atom of the gold acetylide
complexes 12 differed by ≤0.3 ppm. However, the  29Si of the
C(sp)-bound silicon atom of aryl silyl alkynes 16 are deshielded
by ~6 ppm relative to the  29Si of the C(sp)-bound silicon atom
of gold acetylide complexes 12. Because there is no net -
bonding in the Au–C bond of a gold acetylide complex,[34] these
differences in the  29Si of aryl acetylenes 16 relative to gold
acetylide complexes 12 can presumably be attributed to the
greater inductive electron releasing ability of the (L)Au
fragments of 12 relative to the p-substituted aryl groups of 16. It
therefore follows that -donation from the (L)Au fragment
likewise represents a significant component of the net (L)Au 
C1 electron donation in gold vinylidene complexes 13, as
reflected by the respective  29Si values. In this context,
comparison of the  29Si values for gold vinylidene complexes
13 and vinyl cations 15 is instructive (Table 1, Figure 5). In
particular, the  29Si for vinyl cation 15d (+52.2) is smaller than
is  29Si for any of the gold vinylidene complexes 13 (≥ +54.1),
suggesting that the p-(dimethylamino)phenyl is a stronger -
donor than are any of the (L)Au fragments.
gold vinylidene complexes 13
15 are
consistent with Y-shaped vinyl cation structures and, as such,
the structures of 13 and 15 differ only in the nature of the -
substituent. Secondly, although a number of factors in addition
to electron density are known to affect 29Si NMR chemical
shifts,[27] the similar structures and identical counterions of the
complexes and the identical medium for 29Si NMR analysis
suggests that these other factors are largely mitigated. Thirdly,
because the silicon atoms of 13 and 15 are far removed from the
-substituent, steric or through-space perturbation of  29Si by
the -substituent appears unlikely.
As was noted by Müller, the -substituent of an -substituted
-disilyl vinyl cation stabilizes the C1 carbon atom through a
combination of inductive effects and resonance electron
donation.[15] The strong correlation between + and  29Si for -
aryl cations 15 indicates that for these compounds, the C1
stabilization from the aryl group is predominantly that of -
donation. This does not, however, require or imply that the
electronic stabilization of the C1 atom in the gold vinylidene
complexes 13 by the (L)Au fragment is also predominantly that
of -donation. Here it is important to note that sp carbenium
ions are more sensitive to inductive effects than are sp2
carbenium ions owing to the greater s character and higher
electronegativity of the sp carbenium ion relative to the sp2
carbenium ion.[32,33] For example, ab initio calculations suggest
that whereas a single -CH3 group stabilizes an alkyl cation by
~1.3 kcal/mol more than does an -trimethylsilyl group, the -
trimethylsilyl group stabilizes a vinyl cation by ~8.3 kcal/mol
more than does an -methyl group.[32] In the former case, the
superior hyperconjugation of the C–H bonds outweighs the
lower electronegativity of the silicon atom whereas in the latter
case, electronegativity differences take precedence.
75
15a
70
15b
15c
65
60
55
Given the sensitivity of an sp carbenium ion to inductive
stabilization, we considered that the apparent greater electron
donor ability of the (L)Au fragments in gold vinylidene complexes
13 relative to the p-(dimethylamino)phenyl group of vinyl cation
15d might be due to the greater inductive electron releasing
ability of the (L)Au fragments relative to the p-
(dimethylamino)phenyl group. Here it should be noted that in his
analysis of -SiC hyperconjugation in -aryl,-disilyl cations 14,
Müller established a correlation between the Hammett-Brown +
parameter and  29Si, where  29Si represents the difference
between  29Si of 14 and that of the C(sp)-bound silicon atom of
the corresponding neutral aryl silyl alkyne precursor.[15] Because
15d
50
-2.0
-1.5
-1.0
-0.5
0.0
Figure 5. Plot of the Hammett-Brown + parameter versus  29Si for -
aryl,-disilyl vinyl cations 15a-15d where  29Si = (12.6 ± 0.9)+ + (73.9 ±
0.9), R = 0.99. Dotted lines represent the extrapolation of  29Si for gold
vinylidene complexes 13 onto this curve: from top to bottom 13d, 13b, 13d +
13e, 13a.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
Neutral Gold Fluorophenyl Complexes. Our analysis of
the 29Si NMR spectroscopy of gold vinylidene complexes 13,
vinyl cations 15, and their neutral acetylenic precursors pointed
to the greater inductive electron releasing ability of the (L)Au
fragments vis-à-vis p-substituted aryl groups. To gain greater
insight into the ligand-dependent - and -donor properties of
the (L)Au fragment, we sought to determine the inductive (i)
and resonance (r) substituent parameters of (L)Au fragments
employing the 19F NMR method developed by Taft.[35] The 19F
NMR chemical shifts of m- and p-substituted fluorobenzenes
Table 2. 19F NMR chemical shift data (m and p) and inductive (i) and
resonance (r) parameters for gold m- and p-fluorophenyl complexes 17.
[a]
[a]
cmpd
17a
L
m
p
i
r
P1
3.72
3.74
2.43
3.65
2.10
5.71
5.85
3.51
4.51
2.95
–0.44
–0.44
–0.26
–0.43
–0.21
–0.07
–0.07
–0.04
+0.01
–0.02
17b
17c
IPr
PPh3
17d
PtBu3
––[b]
P(OPh)3
(FC6H4X) represent
a
sensitive probe of the electron
For a
donor/acceptor properties of the substituent X.[35]
substituent X, the 19F chemical shifts of m-FC6H4X (m) and p-
FC6H4X (p) relative to fluorobenzene internal standard are
related empirically to the inductive (i) and resonance (r)
[a]Chemical shift relative to C6H5F, with positive increments indicating more
negative chemical shifts. [b]Taken from ref 39.
substituent parameters according to the relationships i = (m
–
Gold
Fluorophenyl(methoxy)carbene
Carbene
0.60)/–7.1 and r = (p – m)/–29.5. The inductive parameter
iis a reflection of the through-space and inductive electron
releasing/accepting properties of the substituent whereas the
resonance parameter is a reflection -donor/acceptor properties
of the substituent. A negative i or r value indicates that the
substituent is electron donating relative to hydrogen. This
technique has been employed to determine the inductive and
resonance donor/acceptor properties of hundreds of substituent
groups[36] and has been extended to include numerous transition
Complexes. 19F NMR analysis of gold fluorophenyl complexes
17 revealed the strong inductive electron releasing ability and
modest -donor properties of the (L)Au (L = IPr, P1) fragments.
We were concerned, however, that the relatively low electron
demand of the fluorophenyl group in compounds 17 might mask
both the -donor potential of the (L)Au fragments and subtle
differences between the / electron donor properties of the
(P1)Au and (IPr)Au fragments that might be revealed under the
conditions of higher electron demand found in cationic carbene
metal fragments[37] including
a
handful of gold(I)-ligand
complexes.
Toward an evaluation of the - and -electron
fragments. However absent are data for the (PR3)Au and
(NHC)Au fragments germane to gold(I) catalysis.[38,39]
donor properties of the (P1)Au and (IPr)Au fragments in cationic
gold carbene complexes, we first determined the inductive and
resonance parameters for the (L)AuC(OMe) (L = P1, IPr)
substituent
fluorophenyl(methoxy)carbene
[(L)AuC(OMe)(C6H4F)]+ SbF6 [L = P1 (m/p-18a), IPr (m/p-18b)]
via 19F NMR analysis. We targeted the methoxycarbene
complexes 18 for analysis because our previous analysis of the
gold methoxy(cyclopropyl)carbene complex 4 established that
the C1 methoxy group provided needed stabilization to the
complex without masking the electron donor effects of the (L)Au
fragment.[6]
To determine the inductive (i) and resonance (r)
substituent parameters for (L)Au fragments, we synthesized an
eight-membered family of gold m- and p-fluorophenyl complexes
(L)Au(C6H4F) [m-17/p-17; L = P1 (a), IPr (b) PPh3 (c), P(tBu)3
(d)] via transmetallation of m- or p-fluorophenylboronic acid with
(L)AuCl (Table 2).[40] The 19F NMR chemical shifts (m and p)
and the associated inductive (i) and resonance parameters (r)
of complexes 17 and values for L = P(OPh3)3 taken from the
groups
of
the
cationic
gold
complexes
–
literature are compiled in Table 2.
In accord with our
expectations, these data indicate that the gold fragments (P1)Au
and (IPr)Au, in particular, are strongly inductively electron
releasing and are modest -donors. Notably, the (P1)Au and
(IPr)Au fragments are much more inductively electron releasing
than are the triarylphosphine and triarylphosphite fragments
(Ph3P)Au and [P(OPh)3]Au and are likewise much more
inductively electron releasing than is a simple aryl group. As a
point of comparison, the inductive and resonance substituent
parameters for a phenyl group are i = 0.14 and r = 0.01.[35]
Also worth noting is that these data reveal no detectable
difference between the - and -electron donor/acceptor
properties of (P1)Au and (IPr)Au.
The requisite flourophenyl(methoxy)carbene complexes 18
were isolated as thermally unstable yellow microcrystalline
solids in 32-52% yield via metathesis of the chromium
fluorophenyl(methoxy)carbene
complexes
(CO)5CrC(OMe)(C6H4F) (m/p-19) with a 1:1 mixture of (L)AuCl
and AgSbF6 in CH2Cl2 at room temperature followed by
recrystallization from CH2Cl2/pentane at –20 °C (Scheme 7). As
was the case with the cationic gold vinylidene complexes 13,
efforts to synthesize the corresponding triphenylphosphine
carbene
complexes
(PPh3)AuC(OMe)(C6H4F)
proved
unsuccessful. Complexes 18 were characterized by 1H, 13C, and
19F spectroscopy. Notably, the C1 carbene resonances of
complexes 18 appeared at  = 282 - 290 in the 13C NMR spectra,
which is typical of gold Fischer carbene complexes.[41] 19F NMR
analysis of complexes 18 provided inductive and resonance
parameters for the respective (L)AuC(OMe) substituent groups
of i = +0.60 and r = +0.61 for L = P1 and i = +0.62 and r =
+0.62 for L = IPr (Table 3). The large positive values indicate
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
that the (L)AuCOMe substituent group is highly electron
withdrawing, as would be expected owing to the net positive
charge on the carbene C1 atom.
monofluorobenzophenones (p-C6H4X)(C6H4F)CO m/p-20 with
excess triflic acid (≥10 equiv) in CD2Cl2 (Scheme 8).[45]
Formation of oxocarbenium ions 20•HOTf was established by
the significant deshielding of the oxygen-bound carbon atom (
≈ 23) of 20•HOTf relative to 20 in the 13C NMR spectra.
Complete conversion of 20 to 20•HOTf was established by
titrating 20 with HOTf until no further change in the NMR spectra
was observed. The 19F NMR chemical shifts (m and p) and the
associated inductive (i) and resonance parameters (r) for the
(p-C6H4X)COH substituent groups of oxocarbenium ions
20•HOTf are shown in Table 3. The inductive parameters for the
(p-C6H4X)(COH) substituents ranged from i = +1.04 for X = Br
to i = +0.90 for X = OPh, all of which are significantly more
positive than are the inductive parameters for the (L)AuC(OMe)
substituents (i = +0.60 - +0.62). The resonance parameters for
the (p-C6H4X)(COH) substituents were more sensitive to the
nature of the p-C6H4X group and ranged from r = +0.80 for X =
H to r = +0.55 for X = OMe, as compared to values of r =
+0.61 - +0.62 for the (L)AuC(OMe) substituents.
Scheme 7
Table 3. 19F NMR chemical shift data and inductive (i) and resonance (r)
parameters for gold fluorophenyl(methoxy)carbene complexes 18 and
protonated monofluorobenzophenones 20•HOTf.
[a]
[a]
cmpd
18a
18b
20aHOTf
20bHOTf
20cHOTf
20dHOTf
20eHOTf
substituent
(P1)AuC(OMe)
(IPr)AuC(OMe)
(Ph)C(OH)
(p-C6H4Br)C(OH)
(p-C6H4Me)C(OH)
(p-C6H4OPh)C(OH)
(p-C6H4OMe)C(OH)
m
p
i
r
–3.69
–3.78
–6.55
–6..77
–6.22
–5.82
–5.85
–21.58
–21.92
–30.11
–29.58
–27.27
–23.71
–22.05
+0.60
+0.62
+1.01
+1.04
+0.96
+0.90
+0.91
+0.61
+0.62
+0.80
+0.77
+0.71
+0.61
+0.55
[a]Chemical shift relative to C6H5F, with positive increments indicating more
negative chemical shifts
Scheme 8
Protonated Monofluorobenzophenones. The substituent
parameters determined for the (L)AuC(OMe) fragments of gold
carbene complexes 18 reflect the electron demand of the C1
carbene atom, which is stabilized by electron donation from both
the (L)Au fragment and the OMe moiety. To evaluate the
contribution of the (L)Au fragment to the electronic stabilization
of the C1 carbon atom of complexes 18, we determined the
inductive and resonance parameters for the (p-C6H4X)COH
substituent groups of the protonated monofluorobenzophenone
derivatives [(p-C6H4X)(C6H4F)COH]+ OTf– (m/p-20•HOTf) [X = H
(a), Br (b), Me (c), OPh (d), OMe (e)] via 19F NMR analysis.[42,43]
Although the OH group of oxocarbenium ions 20•HOTf
represents an obvious deviation from the OMe group of
fluorophenyl(methoxy)carbene complexes 18, Olah has
previously shown there is no significant difference in the C=O
bond polarization of protonated and methylated ketones,[44] and
potential hydrogen bonding between OTf– and the protonated
ketone becomes insignificant at high HOTf concentration such
as those employed in these reactions.[45] Therefore, we
reasoned that if correlations could be established between the
inductive electron releasing ability and -donor ability of the C1
aryl groups with the i and r parameters, respectively, of the (p-
C6H4X)COH substituents, then meaningful comparisons could
be made between the inductive electron releasing ability and -
donor ability of the (L)Au fragments of 18 relative to p-
substituted aryl groups of 20•HOTf.
Interpretation of the r parameter for the (p-C6H4X)COH
groups of 20•HOTf and comparison to those of the (L)AuC(OMe)
groups of 18 requires consideration of the structures of
protonated benzophenones. It has been shown that the log(pKa)
of protonated monosubstituted benzophenone derivatives
display
a
linear correlation with the Hammett-Brown +
parameter ( = 1.09),[46] but the reaction constant is attenuated
relative to those obtained for the correlation of log(pKa) with +
for protonated benzaldehyde ( = 1.86)[47] and acetophenone
( = 2.17)[48] derivatives. These differences can be attributed in
part to twisting of the aryl rings of the benzophenones to
minimize the unfavorable van der Walls interactions of the ortho
hydrogen atoms,[49] which attenuates the -overlap between the
aryl groups and the electron deficient carbonyl carbon atom.
Nevertheless, the much lower r values relative to i for (p-
C6H4X)COH substituents and the greater sensitivity of r (r =
0.25) relative to i (i = 0.14) suggests that the electron
deficient carbonyl carbon atom of 20•HOTf is stabilized primarily
via -donation.
A plot of the Hammett  parameter for aryl substituent X
versus i for the (p-C6H4X)(COH) substituent of protonated
monofluorobenzophenones 20•HOTf established a correlation
between these parameters [i = (0.26 ± 0.03) + (0.99 ± 0.01);
R2 = 0.96] (Figure 6) that was clearly superior to the correlation
between the Hammett-Brown + parameter and i (R2 = 0.88).
Similarly, a plot of the Hammett-Brown + parameter for aryl
substituent X versus r for the (p-C6H4X)(COH) substituents
The requisite protonated monofluorobenophenones 20•HOTf
were generated in situ via protonation of the corresponding
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
established a modest correlation between these parameters [r
= (0.27 ± 0.5)+ + (0.77 ± 0.02); R2 = 0.92; Figure 6], whereas
overlap of the aryl rings in 20•HOTf, these comparisons suggest
that the (L)Au groups are modest -donors relative to p-
substituted aryl groups.
no correlation was observed between  and r.
These
correlations indicate that differences in i of the (p-C6H4X)COH
substituent are determined primarily by the inductive electron
releasing ability of the p-C6H4X aryl group and, likewise, that
differences in r of the (p-C6H4X)COH substituent are
determined primarily by the -donor ability of the p-C6H4X aryl
group.
Conclusions
We have analyzed the spectroscopy of cationic gold (,-
disilyl)vinylidene complexes 13, neutral gold fluorophenyl
complexes 17, cationic gold fluorophenyl(methoxy)carbene
complexes 18, and relevant organic model compounds to
evaluate the ligand-dependent electron donor ability of the (L)Au
fragment in gold carbene complexes relative to p-substituted aryl
groups. The first key conclusion drawn from these experiments
is that the (P1)Au fragment is a nominally stronger electron
donor in cationic gold carbene complexes than is the (IPr)Au
fragment. For example, 29Si NMR analysis of the cationic gold
vinylidene complexes 13 suggests that the (P1)Au fragment is
more electron donating than is (IPr)Au, although the difference is
less than that between a p-tolyl group and a p-phenoxyphenyl
1.1
1.0
0.9
si/sr
0.8
0.7
0.6
0.5
group.
Similarly, 19F NMR analysis of the cationic gold
-0.8 -0.6 -0.4 -0.2 0.0
0.2
fluorophenyl(methoxy)carbene complexes 18 suggests that
(P1)Au is both more inductive electron releasing ability anda
better -donor than is (IPr)Au, although the differences are less
than the differences between a phenyl group and a p-
s/s+
Figure 6. Plots of  versus i () and + versus r (O) for the (p-
C6H4X)(COH) substituents of 20•HOTf where i = (0.26 ± 0.03) + (0.99 ±
0.01); R2 = 0.96 and r = (0.27 ± 0.5)+ + (0.77 ± 0.02); R2 = 0.92.
bromophenyl group.
19F NMR analysis of neutral gold
fluorophenyl complexes 17 revealed no detectable difference
between either the inductive electron releasing ability or -donor
ability of the (P1)Au and (IPr)Au fragments. Although the
contention that the (P1)Au fragment is more inductively electron
releasing than is the (IPr)Au fragment, albeit minimally, appears
to contradict the general perception of IPr as a strong donor
ligand to gold,[50] both our analysis of the kinetics of gold(I)-
catalyzed allene racemization[51] and Belpassi's computational
analysis of the electronic structure of cationic gold carbonyl
complexes[52] support this conclusion.
Regarding the electron donor ability of the (L)Au fragments
relative to p-substituted aryl groups, 29Si NMR analysis of the
gold (,-disilyl)vinylidene complexes 13 and the -aryl-,-
disilyl vinyl cations 15 suggests that the electron donor ability of
the (L)Au (L = P1, NHC) fragments exceeds that of p-
Discussion of substituent constants for 18 and 20•HOTf.
From the observed correlations between  and i and between
+ and r for the (p-C6H4X)COH substituents of protonated
monofluorobenzophenones 20•HOTf, it follows that any
significant differences in the i and r values of the (L)AuC(OMe)
(L = P1, IPr) fragments of 18 or between the (L)AuC(OMe) and
(p-C6H4X)COH substituents can be attributed to differences in
the inductive electron releasing ability and -donor ability,
respectively, of the corresponding (L)Au and/or p-C6H4X groups.
Therefore, several conclusions can be reasonably drawn from
comparison of these data. Firstly, (P1)Au is a nominally greater
inductive electron releasing ability and -donor ability than does
(IPr)Au, although the differences are less than those between a
phenyl group and a p-bromophenyl group. Secondly, the
significantly larger (more positive) inductive parameters of the
(p-C6H4X)COH substituents (i = +1.04 - 0.90) relative to those
of the (L)AuC(OMe) substituents (i = 0.60 - 0.62) further
supports the contention that the (L)Au gold fragments are
significantly more inductively electron releasing than are p-
(dimethyamino)phenyl group.
Given the sensitivity of sp
carbenium ions to inductive effects,[29,30] comparative 29Si NMR
analysis of the neutral gold acetylide complexes 12 and aryl
acetylenes 16 suggests that -donation represents a major
component of the net (L)Au  C electron donation in gold
vinylidene complexes 13. This hypothesis was corroborated
through 19F NMR analysis of gold fluorophenyl complexes 17
and comparative 19F NMR analysis of the gold
fluorophenyl(methoxy)carbene complexes 18 and the protonated
monoflurobenzophenone derivatives 20•HOTf. These analyses
revealed that the inductive electron releasing ability of the (L)Au
(L = P1, IPr) fragments significantly exceeds that of a p-
(dimethyamino)phenyl group, whereas the -electron donor
ability of the (L)Au fragment was equal to or less than that
provided by a p-phenoxyphenyl group.
substituted aryl groups.
Indeed, the observed correlation
between i and  (Figure 6) predicts a i value of +0.78 for the
hypothetical (p-C6H4NMe2)(COH) substituent [(NMe2) = –0.83],
which suggests that the (L)AuC fragments are significantly more
inductively electron releasing than is a p-(dimethylamino)phenyl
group. Thirdly, comparison of the resonance parameters for the
(p-C6H4X)COH and (L)AuC(OMe) substituents indicates that the
-donor ability of the (L)Au fragments is lower than that of a p-
methoxyphenyl group but comparable to that of
a
p-
phenoxyphenyl group. Given the potential attenuation of -
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
[9]
C. H. M. Amijs, V. López-Carrillo, M. Raducan, P. Pérez-Galán, C.
Taken in aggregate, a picture emerges from these studies of
(L)Au (L = P1, IPr) fragments as strongly inductively electron
releasing and modest -donors, with very similar electron donor
properties between the two fragments. These subtle differences
in electron donor ability, coupled with the different steric profiles
of the ligands, is apparently sufficient to lead to the sometimes
disparate catalytic behavior observed for (P1)Au and (IPr)Au
complexes.[8-10] The recognition of these (L)Au fragments as
strongly inductively electron releasing is significant because
extant discussions of (L)Au  C electron donation in the context
of cationic gold carbene complexes has focused nearly
exclusively on d  p backbonding as the mechanism for
stabilization of the electron-deficient carbene carbon atom.
Rather, our studies strongly suggest that d  p backbonding
represents a minor component of the total (L)Au  C electron
donation in gold carbene complexes. It therefore follows that
any evaluation of the electron donor ability of (L)Au fragments in
cationic gold carbene complexes that considers only d  p
backbonding likely significantly underestimates the full extent of
(L)Au  C electron donation.
Relevant to the discussion of the inductive electron releasing
ability of the (L)Au fragment is the ambiguity regarding the
electronegativity of gold. On the Pauling scale, gold is the most
electronegative of the transition metals, with a value of 2.54
relative to 2.55 for carbon. However, other measures of
electronegativity suggest that gold is considerably more
electropositive than is carbon, as is suggested by our studies.
For example, on the Mulliken-Jaffe scale, gold has a value of
1.87 versus 2.66 for sp2 hybridized carbon and 2.28 for
silicon.[53]
Ferrer, A. M. Echavarren, J. Org. Chem. 2008, 73, 7721–7730.
[10] F. Barabé, P. Levesque, I. Korobkov, L. Barriault, Org. Lett. 2011, 13,
5580–5583.
[11] For relevant computational studies see: a) K. M. Azzopardi, G. Bistoni,
G. Ciancaleoni, F. Tarantelli, D. Zuccacciab, L. Belpassi, Dalton Trans.
2015, 44, 13999-14007; b) L. Nunes dos Santos Comprido, J. E. M. N.
Klein, G. Knizia, J. Kästner, A. S. K. Hashmi, Chem. Eur. J. 2016, 22,
2892 – 2895; c) L. Nunes dos Santos Comprido, J. E. M. N. Klein, G.
Knizia, J. Kästner, A. S. K. Hashmi, Angew. Chem. Int. Ed. 2015, 54,
10336 –10340.
[12] G. Ciancaleoni, L. Biasiolo, G. Bistoni, A. Macchioni, F. Tarantelli, D.
Zuccaccia, L Belpassi, Chem. Eur. J. 2015, 21, 2467 – 2473.
[13] R. Döpp, C. Lothschütz, T. Wurm, M. Pernpointner, S. Keller, F.
Rominger, A. S. K. Hashmi, Organometallics 2011, 30, 5894–5903.
[14] D. H. Ringger, I. J. Kobylianskii, D. Serra, P. Chen, Chem. Eur. J. 2014,
20, 14270-14281.
[15] T. Müller, D. Margraf, Y. Syha, J. Am. Chem. Soc. 2005, 127, 10852-
10860.
[16] T. Müller, M. Juhasz, C. A. Reed, Angew. Chem. Int. Ed. 2004, 43,
1543 -1546.
[17] A. Klaer, T. Müller, J. Phys. Org. Chem. 2010, 23, 1043-1048.
[18] A. Klaer, Y. Syha, H. R. Nasiri, T. Müller, Chem. Eur. J. 2009, 15, 8414-
8423.
[19] a) A. Klaer, W. Saak, D. Haase, T Müller J. Am. Chem. Soc. 2008, 130,
14956-14957; b) T. Müller, R. Meyer, D. Lennartz, H.-U. Siehl, Angew.
Chem. Int. Ed. 2000, 39, 3074-3077.
[20] R. J. Harris, R. A. Widenhoefer, Angew. Chem. Int. Ed. 2015, 54, 6867-
6869.
[21] For examples of cationic gold vinylidene complexes proposed as
intermediates in gold(I)-catalyzed transformations see: a) Y. Xia, A. S.
Dudnik, Y. Li, V. Gevorgyan, Org. Lett. 2010, 12, 5538-5541; b) V.
Lavallo, G. D. Frey, S. Kousar, B. Donnadieu, G. Bertrand, Proc. Natl.
Acad. Sci. U.S.A. 2007, 104, 13569-13573; c) H. Rabaâ, B. Engels, T.
Hupp, A. S. K. Hashmi, Int. J. Quantum Chem. 2007, 107, 359-365; d)
V. Mamane, P. Hannen, A. Fürstner, Chem. Eur. J. 2004, 10, 4556-
4575; e) E. Soriano, J. Marco-Contelles, Organometallics 2006, 25,
4542- 4553; f) I. V. Seregin, V. Gevorgya, J. Am. Chem. Soc. 2006,
128, 12050-12051. g) A. Gómez-Suárez, S. P. Nolan, Angew. Chem.
Int. Ed. 2012, 51, 8156 – 8159; h) L.; Ye, Y. Wang, D. H. Aue, L.
Zhang, J. Am. Chem. Soc. 2012, 134, 31-34; i) A. S. K. Hashmi, M.
Wieteck, I Braun, M. Rudolph, F. Rominger, Angew. Chem. Int. Ed.
2012, 51,10633-10637; j) A. S. K. Hashmi, M. Wieteck, I. Braun, P.
Nösel, L. Jongbloed, M. Rudolph, F. Rominger, Adv. Synth. Catal. 2012,
354, 555-562; k) M. M. Hansmann, M. Rudolph, F. Rominger, A. S. K.
Hashmi, Angew. Chem. Int. Ed. 2013, 52, 2593-2598; l) I. Braun, A. M.
Asiri, A. S. K. Hashmi, ACS Catal. 2013, 3, 1902−1907; m) P. Nösel, T.
Lauterbach, M. Rudolph, F. Rominger, A. S. K. Hashmi, Chem. Eur. J.
2013, 19, 8634-8641; n) M. M. Hansmann, S. Tsupova, M. Rudolph, F.
Rominger, A. S. K. Hashmi, Chem. Eur. J. 2014, 20, 2215-2223; i) A. S.
K. Hashmi, Acc. Chem. Res. 2014, 47, 864-876; jo J. Bucher, T. Wurm,
K. S. Nalivela, M. Rudolph, F. Rominger, A. S. K. Hashmi, Angew.
Chem. Int. Ed. 2014, 53, 3854-3858; p) H. M. Vilhelmsen, A. S. K.
Hashmi, Chem. Eur. J. 2014, 20, 1901-1908; q) J. Xie, C. Pan, A.
Abdukadera, C. Zhu, Chem. Soc. Rev. 2014, 43, 5245-5256; r) O. J. S.
Pickup, I. Khazal, E. J. Smith, A. C. Whitwood, J. M. Lynam, K. Bolaky,
T. C. King, B. W. Rawe, N. Fey, Organometallics 2014, 33, 1751−1761;
s) Y. Tokimizu, M. Wieteck, M. Rudolph, S. Oishi, N. Fujii, A. S. K.
Hashmi, H. Ohno, Org. Lett. 2015, 17, 604-607; t) P. Nösel, V. Müller,
S. Mader, S. Moghimi, M. Rudolph, I. Braun, F. Rominger, A. S. K.
Hashmi, Adv. Synth. Catal. 2015, 357, 500-506; i) J. Bucher, T.
Stößer, M. Rudolph, F. Rominger, A. S. K. Hashmi, Angew. Chem.
2015, 127, 1686-1690; Angew. Chem. Int. Ed. 2015, 54, 1666-1670;
v) P. Morán-Poladura, E. Rubio, J. M. González, Angew. Chem. Int. Ed.
2015, 54, 3052-3055.
Acknowledgements
We acknowledge the NSF (CHE-1465209) for support of this
research. RGC was supported through a GAANN fellowship
(P200A150114).
Keywords: gold • vinylidenes • carbenes • Hammett analysis •
sigma stabilization
[1]
[2]
D. Qian, J. Zhang, Chem. Soc. Rev. 2015, 44, 677-698.
a) R. J. Harris, R. A. Widenhoefer, Chem. Soc. Rev. 2016, 45, 4533 -
4551; b) Y. Wang, M. E. Muratore, A. M. Echavarren, Chem. Eur. J.
2015, 21, 7332 - 7339.
[3]
[4]
[5]
[6]
G. Seidel, R. Mynott, A. Fürstner, Angew. Chem. Int. Ed. 2009, 48,
2510-2513.
D. Benitez, N. D. Shapiro, E. Tkatchouk, Y. Wang, W. A. Goddard III, F.
D. Toste Nat. Chem. 2009, 1, 482-486.
M. W. Hussong, F. Rominger, P. Kramer, B. F. Straub, Angew. Chem.
Int. Ed. 2014, 53, 9372-9375.
R. E. M. Brooner, R. A. Widenhoefer, Chem. Commun. 2014, 50, 2420-
2423.
[7]
[8]
G. Seidel, A. Fürstner, Angew. Chem. Int. Ed. 2014, 53, 4807-4811.
a) D. J. Gorin, B. D. Sherry, F. D. Toste, Chem. Rev. 2008, 108, 3351–
3378; b) T. Wurm, A. M. Asiri, A. S. K. Hashmi, in N-Heterocyclic
Carbenes: Effective Tools for Organometallic Synthesis, First Edition. S.
P. Nolan, Ed. 2014 Wiley-VCH.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
[22] a) P. E. Hansen, Vi. Wray, Org. Mag. Res. 1981, 15, 102-103; b) K.
Frei, H. J. Bernstein, J. Chem. Phys. 1963, 38, 1216-1226.
[37] a) M. J. Church, M. J. Mays, J. Chem. Soc. (A) 1968, 3074-3078; b) G.
W. Parshall J. Am. Chem. Soc. 1966, 88, 704-708; c) H. A. O. Hill, K.
G. Morallee, F. Cernivez, G. Pellizer, J. Am. Chem. Soc. 1972, 94, 277-
278; d) R. P. Stewart, P. M. Treichel, J. Am. Chem. Soc. 1970, 92,
2710-2718; e) E. S . Bolton, G. R. Knox, C. G. Robertson, Chem.
Commun. 1969, 664-665; f) W. Adcock, B. F. Hegarty, W. Kitching, A.
J. Smith, J. Organomet. Chem. 1968, 12, 21-23; g) W. Kitching, W.
Adcock, B. F. Hegarty, Aust. J. Chem. 1968, 21, 2411-2416; h) A.
Cairncross, W. A. Sheppard, J. Am. Chem. Soc., 1968, 90, 2186-2187.
[38] L. G. Vaughan, W. A. Sheppard, J. Am. Chem. Soc. 1969, 91, 6151-
6156.
[23] J. B. Lambert, Tetrahedron 1990, 46, 2671-2689.
[24] a) H.-U. Siehl, Pure & Appl. Chem. 1995, 67, 769-775; b) W. Zhang, J.
A. Stone, M. A. Brook, G. A. McGibbon, J. Am. Chem. Soc. 1996, 118,
5764-5771; c) X. Li, J. A. Stone, J. Am. Chem. Soc. 1989, 111, 5586-
5592; d) H.-U. Siehl, in Dicoordinated Carbocations; Rappoport, Z.,
Stang, P. J., Eds.; Wiley: New York, 1997, pp 189.
[25] a) J. B. Lambert, Y. Zhao, J. Am. Chem. Soc. 1996, 118, 7867-7868;
b) J. B. Lambert, Y. Zhao, H. Wu, J. Org. Chem. 1999, 64, 2729-2736.
[26] C. A. Reed, Acc. Chem. Res. 1998, 31, 325-332.
[27] a) M. L. Stock, H. C. Brown, Adv. Phys. Org. Chem. 1963, 1, 35-154;
b) J. E. Leffler, E. Grunwald, Rates and Equilibria of Organic Reactions,
Wiley, 1963.
[39] D. I. Nichols, J. Chem. Soc. (A) 1970, 1216-1217.
[40] A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, J. Organomet. Chem.
2008, 694, 692-597.
[28] We were unable to establish the formation of a ground-state gold
vinylidene complex containing a six-membered disilacycle via hydride
[41] a) L.-P. Liu, B. Xu, M. S. Mashuta, G. B. Hammond, J. Am. Chem. Soc.
2008, 130, 17642-17643; b) M. M. Hansmann, F. Rominger, A. S. K.
Hashmi, Chem. Sci. 2013, 4, 1552-1559; c) U. Schubert, K.
Ackermann, R. Aumann, Cryst. Struct. Commun. 1982, 11, 591-594; d)
G. Seidel, B. Gabor, R. Goddard, B. Heggen, W. Thiel, A. Fürstner,
Angew. Chem. Int. Ed. 2014, 53, 879 - 882; e) M. N. Fañanás-Mastral,
F. Aznar, Organometallics 2009, 28, 666-668.
abstraction
from
the
gold
acetylide
complex
(P1)Au[1-
CCSi(Me)2(CH2)3SiMe2H], perhaps owing to formation of a ground-
state -disilacycloheptyne complex.[29] Furthermore, the 29Si NMR
chemical shifts of 14 and 15 are not directly comparable owing to the
influence of ring strain on the  29Si.[15]
[29] P. Schulte, U. Behrens, Chem. Commun. 1998, 1633-1634.
[30] Müller reported that -aryl--disilyl vinyl cations 14 decomposed in
CD2Cl2,[15] whereas we observed no such decomposition. Conversely,
whereas cations 14 were isolable, vinyl cations 15 decomposed upon
attempted isolation.
[42] Despite literature precedents,[43] we were unable to generate the
necessary family of methylated monofluorobenzophenone cations of
the form [(p-C6H4X)(C6H4F)COMe]+.
[43] G. A. Olah, J. R. DeMember, Y. K. Mo, J. J. Svoboda, P. Shilling, J. A.
Olah. J. Am. Chem. Soc. 1974, 96, 884-892.
[31] The  29Si of -aryl--disilyl cations 15 containing a five-membered
disilacycle appeared at higher frequency ( = 37 - 56) than did -aryl-
-disilyl cations 14 containing a six-membered disilacycle ( = 19 - 25)
and were also ~twice as sensitive to the electron donor ability of the -
[44] G. A. Olah, A. L. Berrier, G. K. S. Prakash, J. Am. Chem. Soc. 1982,
104, 2373-2376.
[45] C. M. Burba, N. M. Rocher, R. Frech, J. Phys. Chem. B 2009, 113,
11453–11458.
aryl substituent (+) than were cations 14:  29Si = (7.1 ± 0.5)+
(38.9 ± 0.1), R = 0.99.[15]
+
[46] a) T. G. Bonner, J. Phillips, J. Chem. Soc. (B) 1966, 650-654; b) O.
Kysel, I. Jány, Chem. zvesti 1974, 28, 70-76.
[32] K. van Alem, G. Lodder, H. Zuilhof, J. Phys. Chem. A 2002, 106,
10681-10690.
[47] R. Stewart, K. Yates, J. Am. Chem. Soc. 1958, 80, 6355-6359.
[48] K. Yates, R. Stewart, Can. J. Chem. 1959, 87, 664-671.
[49] G. A. Olah, M. I. Watkins, Proc. Natl. Acad. Sci. USA 1980, 77, 703-707.
[50] a) D. Gatineau, J.-P. Goddard, V. Mouris-Mansuy, L. Fensterbank, Isr.
J. Chem. 2013, 53, 892-900; b) S. P. Nolan, Acc. Chem. Res. 2011, 44,
91-100; c) S. Díez-Gonzalez, N. Marion, S. P. Nolan, Chem. Rev.
2009, 109, 3612-3676; d) J. C. Y. Lin, R. T. W. Huang, C. S. Lee, A.
Bhattacharyya, W. S. Hwang, I. J. B. Lin, Chem. Rev. 2009, 109, 3561-
3598; e) N. Marion, S. P. Nolan, Chem. Soc. Rev. 2008, 37, 1776-
1782; f) P. Klahn, S. F. Kirsch, ChemCatChem 2011, 3, 649-652.
[51] H. Li, R. J. Harris, K. Nakafuku, R. A. Widenhoefer, Organometallics
2016, 35, 2242−2248.
[33] a) Y. Apeloig, P. v. R. Schleyer, J. A. Pople, J. Am. Chem. Soc. 1977,
99, 1291-1296; b) F. Bernardi, A. Bottoni, A. Venturini, J. Am. Chem.
Soc. 1986, 108, 5395-5400; c) J. A. Pople, Y. Apeloig, P. v. R.
Schleyer, Chem. Phys. Lett. 1982, 85, 489-492; d) C. Eaborn, F.
Feichtmayr, M. Horn, J. N. Murrell, J. Organomet. Chem. 1974, 11, 39-
43; e) S. G. Wierschke, J. Chandrasekhar, W. L. Jorgensen J. Am.
Chem. Soc. 1985, 107, 1496-1500.
[34] H.-T. Liu, X.-G. Xiong, P. D. Dau, Y.-L. Wang, D.-L. Huang, J. Li, L.-S.
Wang, Nat. Commun. 2013, 4, 2223.
[35] a) R. W. Taft, E Price, I. R. Fox, I. C. Lewis, K. K. Andersen, G. T.
Davis, J. Am. Chem. Soc. 1963, 85, 3146-3156, and references therein;
b) J. W. Rakshys, R. W. Taft, W. A. Sheppard, J. Am. Chem. Soc.
1968, 90, 5236-5243; c) P. R. Wells, S. Ehrenson, R. W. Taft, Prog.
Phys. Org. Chem. 1968, 6¸147-322; d) R. W. Taft, E. Price, I. R. Fox,
I, C. Lewis, I. K. Andersen, G. T. Davis, J. Am. Chem. Soc. 1963, 85,
709-724; e) R. W. Taft, J. Am. Chem. Soc. 1957, 79, 1045-1049.
[36] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1997, 97, 165-195.
[52] G. Ciancaleoni, N. Scafuri, G. Bistoni, A. Macchioni, F. Tarantelli, D.
Zuccaccia, L. Belpassi, Inorg. Chem. 2014, 53, 9907-9916.
[53] a) J. E. Huheey, E. A. Keiter, R. L Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, 4th edition. New York:
HarperCollins College Publishers, 1993; p. 190; b) S. G. Bratsch. J.
Chem. Ed. 1988, 65, 34 - 41.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703820
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Robert G. Carden, Nathan Lam, and
Ross A. Widenhoefer*
Page No. – Page No.
Experimental Evaluation of (L)Au
Electron Donor Ability in Cationic
Gold Carbene Complexes
Spectroscopic analysis of cationic gold (,-disilyl)vinylidene complexes, gold
fluorophenyl complexes, cationic gold flourophenyl(methoxy)carbene complexes,
and relevant organic model compounds indicate that (L)Au fragments such as
(IPr)Au and [P(tBu)2o-biphenyl)]Au are strongly inductively electron releasing and
modest -donors and that the net electron donor ability of these (L)Au
fragments appears to exceed that provided by a p-(dimethylamino)phenyl group.
This article is protected by copyright. All rights reserved.