Empirical Estimation of the Molecular Weight of Gold Complexes in Solution by Pulsed-Field Gradient NMR

Article abstract of DOI:10.1021/acs.organomet.8b00709

Homogenous gold catalysis has emerged as a powerful tool in organic synthesis, but many mechanistic questions in that area remain unanswered. The ability of diffusion-ordered NMR spectroscopy (DOSY) to investigate interactions between substrates, to study

Full text of DOI:10.1021/acs.organomet.8b00709

Article
pubs.acs.org/Organometallics
Cite This: Organometallics XXXX, XXX, XXX−XXX
Empirical Estimation of the Molecular Weight of Gold Complexes in
Solution by Pulsed-Field Gradient NMR
Ghanem Hamdoun,† Christophe Bour,‡ Vincent Gandon,‡,§ and Jean-Nicolas Dumez*,†
†
́
́
Institut de Chimie des Substances Naturelles, CNRS UPR2301, Universite Paris Sud, Universite Paris-Saclay, 91190 Gif-Sur-Yvette,
France
‡
́
́
́
́
Institut de Chimie Moleculaire et des Materiaux d’Orsay, CNRS UMR 8182, Universite Paris Sud, Universite Paris-Saclay, 91405
Orsay France
§
́
́
Laboratoire de Chimie Moleculaire, CNRS UMR 9168, Ecole Polytechnique, Universite Paris-Saclay, Route de Saclay, 91128
Palaiseau cedex, France
S
* Supporting Information
ABSTRACT: Homogenous gold catalysis has emerged as a powerful tool in organic
synthesis, but many mechanistic questions in that area remain unanswered. The
ability of diffusion-ordered NMR spectroscopy (DOSY) to investigate interactions
between substrates, to study molecular assemblies, and to characterize reactive
intermediates makes it a powerful technique to get a better understanding of gold-
catalyzed processes. In this article, we show that a simple empirical approach makes it
possible to derive estimates of the molecular weight of gold complexes from their
translational diffusion coefficients measured with DOSY. This approach is based on
external calibration curves (ECC), modified to take into account the molecular
density of gold complexes. The estimated molecular weights in turn provide
information on interactions, as notably illustrated here with ion pairing, and have the
potential to provide key information to rationalize the reactivity of reagents in gold-
catalyzed reactions.
cationic gold catalysts can have a dramatic influence on the
selectivity13 or that some reactions actually require several
catalyst molecules to activate the substrates. In this context,
the arsenal of spectroscopy and spectrometry techniques at
the chemists’ disposal to analyze the mechanisms of gold-
catalyzed transformations without affecting the catalytic
process is still incomplete.
Nuclear magnetic resonance (NMR) spectroscopy is a
powerful method for the noninvasive analysis of molecules in
solution. Pulsed-field gradient (PFG) NMR and its
implementation DOSY (for diffusion-ordered NMR spectros-
copy)18 are a particularly relevant approach for the analysis of
organometallic compounds in solution. DOSY provides, for
each peak in an NMR spectrum, an estimate of the
translational diffusion coefficients of the corresponding
molecules. For the interpretation of these coefficients, several
techniques are available that provide information on molecular
sizes and shapes.19−21 Typically, the Stokes−Einstein equa-
tion,22 or a derivative thereof, is used to estimate hydro-
dynamic radii, and this approach has proven to be very
successful for the analysis of organometallic compounds with,
for example, the characterization of ion pairs.23−25 For gold
complexes, recent examples include the analysis of the
bonding mode of cationic gold(I) species in solution26 and
INTRODUCTION
■
Over the past 18 years, gold catalysis has opened new
perspectives in organic chemistry and has played a key role in
the synthesis of complex molecules from simple precur-
sors.1−17 While gold(I) and gold(III) complexes have been
considered to be inert for a long time, a tremendous number
of studies now clearly demonstrate the great interest of gold
derivatives in homogeneous catalysis. This synthetic work has
been supported by coordination chemistry efforts, which have
identified various families of neutral or cationic mono- or
digold complexes that are now broadly applied and easily
available from commercial sources. One of the most salient
features of gold complexes is their ability to activate CC π
bonds (alkynes, alkenes, or allenes) toward nucleophilic
attack. This means that the coordination of a soft CC π
bond to a soft gold-based Lewis acid renders the LUMO of
the substrate, such as an alkyne, more accessible to the
HOMO of a variety of nucleophiles. However, the choice of
ligands, counterions, solvents, and other experimental
parameters is still a tedious task. Although high levels of
selectivity are usually observed, anticipation and ration-
alization of the outcome of gold-catalyzed reactions also
remain challenging. To gain deeper insight into gold-catalyzed
transformations, intensive mechanistic work has been
provided. On several occasions, it has been shown that the
mechanisms are more complex than originally thought. For
instance, it is now clearly established that the counterions of
Received: October 3, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
the study of counterion effects.27 Alternatively, empirical
relationships can be established to obtain estimates of
molecular weights from measured diffusion coefficients. Such
relationships may also be derived from the Stokes−Einstein
equation.28,29 More accurate results are usually obtained with
power laws of the form D = KMWα,30 where K and α are
coefficients that are specific to a given class of compounds and
require a calibration.31−33 A first option is to derive an
internal calibration curve (ICC) by including multiple
reference molecules in the sample.31,33 Applications of this
approach have given information on the aggregation state of
organolithium complexes in solution31,34 and their degree of
solvation.35 Recently, Stalke and co-workers have proposed to
replace the use of multiple internal references by a single
internal reference, which may be the solvent, used together
with external calibration curves.36−39 The use of a reference
and a normalization step makes it possible to derive
calibration curves that are largely independent of the
reference, temperature, and other experimental parameters.
Applications of the ECC approach have been reported in
organolithium chemistry to provide a reliable and straightfor-
ward indication of the aggregation state of lithium
diisopropylamide in hydrocarbon solvents.40 The validity of
a given calibration curve, however, is still restricted to specific
classes of compounds.
In this article, we show that important pieces of information
on the interaction between cationic gold(I) complexes and
their counterion or the substrate can be obtained using
pulsed-field gradient NMR, together with external calibration
curves (ECC). The accuracy of the ECC approach is highly
dependent on the choice of model parameters, and new ECCs
are derived here using a set of gold-containing compounds.
These relationships are then used to estimate the molecular
weights of gold complexes in solution from their diffusion
coefficients. With the example of several cationic gold
catalysts, we show that PFG NMR with ECC gives
information on the type of ion pairing in solution with no
need to observe the counterion by NMR. The interactions of
gold complexes with several alkynes are also analyzed.
of 13 di- and monogold complexes, as listed in Table 1. For
this study, the experiments were performed in deuterated
Table 1. Gold Complexes Used to Derive External
a
Calibration Curves
MD
MW
(g/mol)
b
complex
g/(mol·m3)
a
b
c
d
e
f
g
h
i
(Et3P)AuCl
1.28 × 1030
9.38 × 1029
9.51 × 1029
7.49 × 1029
6.63 × 1029
7.51 × 1029
7.29 × 1029
9.85 × 1029
9.44 × 1029
8.59 × 1029
7.41 × 1029
8.52 × 1029
7.06 × 1029
350
434
494
621
879
769
602
963
1001
1188
1733
1145
1644
(tBu3P)AuCl
(Ph3P)AuCl
(IPr)AuCl
(phosphite)AuCl
(CDP)AuCl
(IPr)AuOH
(R,R)-(DIOP)-(AuCl)2
(CDP)-(AuCl)2
(R)-(DM-SEGPHOS)-(AuCl)2
(R)-(DTBM-SEGPHOS)-(AuBr)2
(R)-(TolBINAP)-(AuCl)2
(R)-(DTBM-SEGPHOS)-(AuCl)2
j
k
l
m
a
For each compound, the molecular density (MD) and the molecular
b
weight (MW) are given. See the SI for the structures of the
complexes.
dichloromethane, which is a very common solvent in gold
catalysis. The use of external calibration curves requires a
single internal standard. Common choices for the internal
standard include the residual, partially undeuterated solvent as
well as relatively inert molecules such as adamantane,
cyclopentane, tetramethylsilane, and naphthalene. Here, in
order to establish the external calibration curves, the measured
diffusion coefficients were normalized against adamantane.
The fixed values of the diffusion coefficient of the reference
(Dref,fix) were estimated by using the average log D value of 13
DOSY measurements of 33 mM solution at 25 °C (Table 2).
Table 2. Values of the Fixed Diffusion Coefficients for the
Two Compounds Used as Internal References
internal reference
log Dref,fix
RESULTS AND DISCUSSION
■
adamantane in CD2Cl2
CH2Cl2 in CD2Cl2
−8.82
−8.63
As mentioned above, external calibration curves have been
shown to be applicable for a specific range of molecular
density, where the molecular density can be defined as the
ratio of the molecular weight over the sum of the van der
Waals volume of the atoms in the molecule.36 Existing
calibration curves were derived from molecules composed of
carbon, hydrogen, and heteroatoms of the third period, with
molecular densities in the (4−6) × 1029 g/(mol·m3) range.
Unsurprisingly, when these calibration curves were applied to
gold complexes, large systematic errors were obtained. For
example, for the (JohnPhos)AuCl complex, which is widely
used in gold catalysis, with the existing ECC for dichloro-
methane solutions, the estimated molecular weight was 442 g/
mol, while the correct value is 531 g/mol, corresponding to a
relative error of 20.1%. With a van der Waals radius of 166 pm
and an atomic weight of 196 g/mol, the gold atom has a
molecular density of 1.03 × 1031 g/(mol·m3), far outside the
range covered by existing calibration curves.
Figure 1 shows the normalized diffusion coefficients for a
set of 13 gold complexes, plotted against the known molecular
weight of each compound. Molecular weights actually range
from 350 to 1733 g/mol, and molecular densities range from
6.63 × 1029 to 12.8 × 1029 g/(mol·m3).
In order to assess the possibility to establish an empirical
relationship between the molecular weight and normalized
diffusion coefficient that would be applicable to gold catalysts,
a series of DOSY experiments was carried out on a selection
Figure 1. log D versus log MW in CD2Cl2. All compounds were
normalized to log Dref,fix (adamantane).
B
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
A qualitative analysis revealed that the data in Figure 1 may
not be described accurately with a single linear relationship.
However, when a separation was made between mono- and
digold complexes, a linear relationship was obtained within
each group. The parameters for these relationships are
summarized in Table 3. Importantly, they may be used to
Table 3. Parameters for the External Calibration Curves
Derived from Diffusion Coefficients of Gold Complexes in
CD2Cl2
−log K
−a
R2
monometallic gold complexes
bimetallic gold complexes
7.30
7.99
0.65
0.40
0.986
0.979
obtain estimates of molecular weights from measured diffusion
coefficients, with no need for an additional sample-specific
calibration procedure.
The difference in parameters for the calibration curves of
mono- and digold complexes may have two different origins.
A comparison of the molecular densities of the mono- and
digold species shows that the difference in calibration does not
arise from it; the average molecular densities for the two
groups of complexes differ by less than 3%. On the other
hand, the mono- and digold complexes have a different overall
shape. Qualitatively, monometallic complexes are mostly
spherical-like molecules, while digold complexes have a
more elongated shape. This can be appreciated in the three-
dimensional structure of the compound, as shown in the SI.
The validity of the proposed external calibration curves has
then been checked using the estimate of the molecular weights
of compounds that were left out of the data set for the linear
regressions shown in Figure 1. Table 4 shows the estimated
Table 4. Comparison between the Calculated and the
Estimated Molecular Weights for Two Gold Complexes
Using Adamantane as a Reference and the ECCs of ref 39
(classic-ECC) or the ECCs Introduced Here for Mono-
(G1) or Digold (G2) Complexes
MWest (g/mol)
classic-ECC
gold-ECC
MW
(g/mol) merged DSE
a
b
c
1
ED
G1
G2
Figure 2. H DOSY NMR of (a) [(JohnPhos)Au] + [OTf]-, (b)
(BINAP)-(AuCl)2, and (c) (Ph3P)AuMe complexes in CD2Cl2 with
internal references (adamantane).
(BINAP)-(AuCl)2
(Ph3P)AuMe
1087
474
751
377
664
351
561
347
864
492
1083
437
a
Merged calibration curves can be used if the molecular shape is
b
c
unknown. Dissipated spheres and ellipsoids. Expanded discs.
need to recalculate the calibration curve. Table 5 shows the
estimated molecular weights for the compounds analyzed
molecular weights for (Ph3P)AuMe as a monometallic
complex and R-(BINAP)-(AuCl)2 as a bimetallic complex,
as obtained from the diffusion measurements shown in Figure
2, with adamantane as an internal reference. For both
compounds, the estimated molecular weight differs from the
known value by 3.7 and 0.4% when using the calibration curve
derived from the corresponding number of gold atoms. In
contrast, using the calibration curve derived for low-density
compounds (classical ECC), errors of 25 to 45% were
observed, depending on the predicted shape (Table 4).
Conversely, large errors were also observed when the ECC
derived for monogold species was used for the digold species.
One of the proposed advantages of the ECC approach is
the possibility to use a variety of internal references, with no
Table 5. Comparison between the Calculated and the
Estimated Molecular Weights for Two Gold Complexes
Using CD2Cl2/CH2Cl2 as a Reference and the ECCs of Ref
39 (classic-ECC) or the ECCs Introduced Here for Mono-
(G1) or Digold (G2) Complexes
MWest (g/mol)
classic-ECC
gold-ECC
MW
(g/mol) merged DSE
ED
G1
G2
(BINAP)-(AuCl)2
(Ph3P)AuMe
1087
474
760
359
672
335
566
335
858
477
1037
430
C
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 6. Comparison between the Calculated and the Estimated Molecular Weights for Three Gold Complexes
estimated MW
(g/mol)
gold-ECC
mixture
expected compound
[(JohnPhos)Au]+[OTf]−
calculated MW (g/mol)
G1
G2
Δ%
(JohnPhos)AuCl
531
644
788
746
622
736
880
838
1424
494
610
754
718
561
611
770
786
638
699
833
1033
5.3
5.4
2.3
5.1
2.5
5
(JohnPhos)AuCl + AgOTf
(JohnPhos)AuCl + AgOTf + 1-pentyn-1-ylbenzene
(JohnPhos)AuCl + AgOTf + phenylacetylene
(IPr)AuCl
[(JohnPhos)Au(H7C3CC−C6H5)]+[OTf]−
[(JohnPhos)Au(HCC−C6H5)]+[OTf]−
(IPr)AuCl + AgOTf
[(IPr)Au]+[OTf]−
(IPr)AuCl + AgOTf + 1-pentyn-1-ylbenzene
(IPr)AuCl + AgOTf + phenylacetylene
[(IPr)Au (H7C3CC−C6H5)]+[OTf]−
[(IPr)Au (HCC−C6H5)]+,OTf−
[{(IPr)Au}2(η1,η2-CC−C6H5)]+[OTf]−
5.3
23
1444
1.4
1.4
1.8
1.5
2.6
(Ph3P)AuCl
501
621
743
737
(Ph3P)AuCl + AgOTf
(Ph3P)AuCl + AgOTf + 1-pentyn-1-ylbenzene
(Ph3P)AuCl + AgOTf + phenylacetylene
[(Ph3P)Au]+[OTf]−
[(Ph3P)Au(H7C3CC−C6H5)]+[OTf]−
[(Ph3P)Au(HCC−C6H5)]+[OTf]−
a
The estimated molecular weights were obtained from a normalized diffusion coefficient measured in CD2Cl2 using adamantane as an internal
reference and the gold-ECCs for mono- or digold complexes.
above, now using the measured diffusion coefficient of the
solvent as a reference. Strikingly, the results are comparable to
those obtained using adamantane, with errors of 0.6 and 4.8%
for the estimated molecular weights.
obtained with two other examples, with a phosphine and an
NHC (N-heterocyclic carbene) ligand, as shown in Table 6
(mixtures in rows six and ten).
A further step can be taken with the analysis of mixtures
containing a cationic gold complex and an excess of alkyne,
which would be representative of the numerous gold-catalyzed
alkyne transformations reported in the literature.41−43 Table 6
also shows the diffusion coefficients measured from the ligand
1H signals of gold complexes after the addition of an alkyne
for three complexes and two alkynes. In five cases, the
estimated molecular weights obtained with the G1-ECC curve
reveal that the cationic gold complex strongly associates with
the alkyne, with a deviation of less than 6% between the
estimated and the calculated molecular weight of the adducts.
This corresponds to the expected formation of gold π-alkyne
complexes,44−46 which is central to a number of important
gold-catalyzed reactions. This process, however, also depends
on alkyne substitution and on the nature and structure of the
ligand.
One of the main potential applications of our ECCs is in
the analysis of molecular interactions in gold-catalyzed
reactions. The estimated molecular weight of a chemical
species gives information on its association with other
compounds. This is illustrated here with a cationic gold
complex, obtained in situ by reacting (JohnPhos)AuCl with
AgOTf in CD2Cl2 and filtering the AgCl precipitate. Using the
ECC derived with monogold species and adamantane as an
internal reference, the estimated molecular weight is 611 g/
mol, as shown in Table 6. Compared to the actual molecular
weight of [(JohnPhos)Au]+, 495 g/mol, and that of the ion
pair formed with the [(JohnPhosAu)]+[OTf]− counterion,
644 g/mol, the best agreement clearly is with the ion pair.
This is a firm indication of a strong association in solution.
This observation can be validated by performing a 19F PFG
NMR experiment. Figure 3 shows that the measured diffusion
coefficient obtained from the [OTf]− ion 19F nuclei is close to
that obtained for the JohnPhos 1H signals. This example
illustrates the usefulness of external calibration curves, with
which the type of pairing can be characterized by observing
just one of the partners. Similar results and information were
Interestingly, the diffusion coefficient and estimated
molecular weight obtained for the mixture of complex
[(IPr)Au]+[OTf]− and phenylacetylene with G1-ECC is
1033 g/mol, which does not correspond to a simple gold π-
alkyne complex with formula [(IPr)Au(HCC−
C6H5)]+[OTf]−, which has a molecular weight of 796 g/
mol. This is, however, fully consistent with the observation
made by Widenhoeffer et al. Indeed, when [(IPr)Au]+[OTf]−
and phenylacetylene were reacted in CD2Cl2 at 25 °C, the
formation of a gold acetylide was observed, in which the
alkyne moiety still serves as a π ligand for a second gold ion,
as in [((IPr)Au)2-η1,η2-CC−C6H5]+[OTf]− (Scheme 1, 2),
as evidenced by the disappearance of the acetylenic proton at
1
δ = 4.56 ppm in the H NMR spectrum as well as by X-ray
characterization.44 Using G2-ECC, an estimated molecular
weight of 1444 g/mol was obtained, which is consistent with
the calculated molecular weight of 1424 g/mol for 2. This
example clearly illustrates the usefulness of the ECC for
analyzing the structure of gold π-alkyne complexes.
The use of empirical relationships between diffusion
coefficients as measured by PFG NMR becomes more
complicated when multiple species coexist and interconvert.47
Figure 3. 19F and 1H DOSY for the [(JohnPhos)Au]+[OTf]− catalyst
in CD2Cl2.
D
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 1. Ar = 2,6-Diisopropylphenyl and Y = OTf or SbF6
a
See ref 15a.
on the reference compound under identical experimental conditions.
These quantities are usually manipulated in log form, and the
resulting definition of the normalized diffusion coefficient Dx,norm is
When chemical exchange is fast on the NMR and diffusion-
encoding time scales, the apparent diffusion coefficient will be
a compromise that depends on the diffusion coefficients and
exchange rates of the interconverting species. As a result, an
unexpected value of the estimated molecular weight for an
organometallic species may also be an indication that chemical
exchange is present.48 The effects of dynamics on apparent
diffusion coefficients can be limited by working at low
temperature, as was done in analyses of organolithium
compounds.49,50
The proposed method was also assessed for concentrations
of the compounds that are closer to those found in catalytic
reactions. As shown in the Supporting Information, com-
parable estimates of molecular weights are obtained at a
concentration 10 times smaller than for the cationic gold
complexes analyzed here along with their reaction with an
alkyne substrate. Overall, the ECC approach proves very
useful from an experimental point of view because it can
provide accurate results without the need to add potentially
reactive compounds to the solution.
log(Dx,norm) = log(D ) − log(D ) + log(D
)
(1)
x
ref
ref,fix
The calibration curve approach then relies on the empirical
assumption that a power law relates the normalized diffusion
coefficient to the molecular weight of the analyte:
log(Dx,norm) = log(k) + α log(MW)
(2)
Parameters k and α are obtained from a linear regression between
log(Dx,norm) and the known molecular weight for a series of
compounds. Once ECC parameters k and α are determined, the
molecular weight of an unknown compound can be estimated with
the relationship
x,norm)−log(k)/α)
= 10(log(D
MW
(3)
estimated
EXPERIMENTAL SECTION
■
Sample Preparation for Neutral Gold Complexes LAuX or
L-(AuX)2 (X = Cl, Br, OH). For each gold complex, a solution was
prepared in 600 μL of CD2Cl2 at a concentration of 33 mM. In each
solution, adamantane was added at a concentration of 33 mM (2.7
mg) to be used as a reference. Undeuterated dichloromethane
(CH2Cl2) was also added, at a concentration of 33 mM (1.7 mg), for
a separate application. In the following sections, the resonances of
CH2Cl2 and CDHCl2 were integrated simultaneously.
CONCLUSIONS
■
We have shown that pulsed-field gradient (PFG) NMR
together with external calibration curves can be used to obtain
information on the interaction of gold(I) complexes with their
counterion and substrates. While the original calibration
curves derived for compounds with low molecular density are
not applicable to gold complexes, a new set of parameters
provide estimates of molecular weight with much-improved
accuracy. This is illustrated with several cationic gold
complexes, counterions, and complexation with alkynes. The
proposed approach is straightforward and has the potential to
aid in the understanding of the mechanism of gold-catalyzed
reactions.
Sample Preparation for Cationic Gold Complexes
[LAu]+[OTf]−. The gold chloride precursor ((JohnPhos)AuCl, 10.6
mg; (IPr)AuCl, 12.4 mg; or (Ph3P)AuCl), 9.9 mg) was weighed into
a glass vial. CD2Cl2 (600 μL) was then added using a syringe. Silver
triflate (5.14 mg, 1 equiv) was added, and the vial was sonicated for 3
min at room temperature. The content of the vial was filtered over
Celite, and the clear solution was transferred into the NMR tube
previously charged with adamantane (5 mg). Undeuterated dichloro-
methane (CH2Cl2) was also added (3 μL) in the case of
[(JohnPhos)Au]+[OTf]−.
NMR Experiments. All of the experiments were performed on a
600 MHz Bruker Avance spectrometer using a 5 mm TXI probe with
actively shielded triple-axis gradients at a nominal temperature of 298
K. Sample spinning was deactivated. All of the DOSY experiments
were performed using a double stimulated echo sequence with
bipolar gradient pulses with convection compensation. The diffusion
time ranged from 150 to 230 ms. The delay for gradient recovery was
0.2 ms, and the eddy current delay was 5 ms. For each DOSY-NMR
experiment, a series of 8 spectra of 32 768 data points were collected
over a total experiment time of 22 min. The gradient amplitude was
increased in increments from 6.5 to 45.5 G cm−1 for the maximum
gradient strength in a linear ramp. Diffusion coefficients were
calculated by a monoexponential fit with the T1/T2 analysis module
of the topspin software or the Dynamics Center software (Bruker,
Billerica, U.S.A.).
METHODS
■
The external calibration curve (ECC) approach is briefly described
here. For a more detailed explanation, see ref 36. Diffusion
coefficients obtained from pulsed-field gradient (PFG) NMR
experiments are highly sensitive to experimental conditions and
parameters. In order to attenuate the influence of sources of
systematic errors (e.g., gradient calibration, temperature effects...),
the diffusion coefficient measured for the compound of interest Dx is
divided by that of a reference compound Dref present in the solution.
Then, the resulting relative diffusion coefficient is multiplied by a so-
called fixed value of the diffusion coefficient of the reference
compound Dref,fix so that the same calibration curve can be used
irrespective of the choice of the reference compound. This fixed
value is obtained by means of a series of measurements carried out
E
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
The diffusion coefficients of 13 gold complexes were measured in
deuterated dichloromethane solvent, and adamantane was used as an
internal reference to establish the calibration curves. Undeuterated
dichloromethane was also employed to test the possibility to use
another internal reference, using ECCs calculated with adamantane.
The fixed diffusion coefficients of adamantane and dichloromethane
(Dref, fix) were determined by averaging the values obtained in 13
separate experiments.
Molecular Geometry. The geometries of the gold complexes (a,
b, c, d, e, g, i, and l) were obtained from the CCDC X-ray structure
database when available. All other structures for complexes f, h, j, k,
and m were optimized using the Gaussian 09 software package at the
B3LYP level of density functional theory (DFT). The gold and the
bromine atoms were described by the LANL2DZ ECP basis set. All
other atoms were described by the 6-31G(d,p) basis set. (See the SI
for details.)
(8) Fu
̈
rstner, A. From Understanding to Prediction: Gold- and
Platinum-Based π-Acid Catalysis for Target Oriented Synthesis. Acc.
Chem. Res. 2014, 47, 925−938.
(9) Zhang, Y.; Luo, T.; Yang, Z. Strategic innovation in the total
synthesis of complex natural products using gold catalysis. Nat. Prod.
Rep. 2014, 31, 489−503.
(10) Dorel, R.; Echavarren, A. M. Gold(I)-Catalyzed Activation of
Alkynes for the Construction of Molecular Complexity. Chem. Rev.
2015, 115, 9028−9072.
(11) Joost, M.; Amgoune, A.; Bourissou, D. Reactivity of Gold
Complexes towards Elementary Organometallic Reactions. Angew.
Chem., Int. Ed. 2015, 54, 15022−15045.
(12) Pflasterer, D.; Hashmi, A. S. K. Gold catalysis in total synthesis
- recent achievements. Chem. Soc. Rev. 2016, 45, 1331−1367.
(13) Zi, W.; Dean Toste, F. Recent advances in enantioselective
gold catalysis. Chem. Soc. Rev. 2016, 45, 4567−4589.
́
(14) Miro, J.; del Pozo, C. Fluorine and Gold: A Fruitful
ASSOCIATED CONTENT
* Supporting Information
■
Partnership. Chem. Rev. 2016, 116, 11924−11966.
S
(15) Wei, Y.; Shi, M. Divergent Synthesis of Carbo- and
Heterocycles via Gold-Catalyzed Reactions. ACS Catal. 2016, 6,
2515−2524.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.8b00709.
(16) Stathakis, C. I.; Gkizis, P. L.; Zografos, A. L. Metal-catalyzed
cycloisomerization as a powerful tool in the synthesis of complex
sesquiterpenoids. Nat. Prod. Rep. 2016, 33, 1093−1117.
(17) Li, Y.; Li, W.; Zhang, J. Frontispiece: Gold-Catalyzed
Enantioselective Annulations. Chem. - Eur. J. 2017, 23, 467.
(18) Johnson, C. S. Diffusion ordered nuclear magnetic resonance
spectroscopy: principles and applications. Prog. Nucl. Magn. Reson.
Spectrosc. 1999, 34, 203−256.
Calculation of the molecular van der Waals density
MDW, 3D geometries and chemical structures of the
model compounds used for ECCs, influence of
concentration, 1H 1D NMR spectra of compounds
LAuX, and abbreviations (PDF)
Coordinates of the computed structures (XYZ)
(19) Macchioni, A.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.
Determining accurate molecular sizes in solution through NMR
diffusion spectroscopy. Chem. Soc. Rev. 2008, 37, 479−489.
AUTHOR INFORMATION
Corresponding Author
*E-mail: jeannicolas.dumez@cnrs.fr.
■
́
(20) Pregosin, P. S.; Kumar, P. G. A.; Fernandez, I. Pulsed Gradient
Spin−Echo (PGSE) Diffusion and 1H,19F Heteronuclear Over-
hauser Spectroscopy (HOESY) NMR Methods in Inorganic and
Organometallic Chemistry: Something Old and Something New.
Chem. Rev. 2005, 105, 2977−2998.
(21) Edward, J. T. Molecular volumes and the Stokes-Einstein
equation. J. Chem. Educ. 1970, 47, 261−270.
(22) Einstein, A. Uber die von der molekularkinetischen Theorie
der Warme geforderte Bewegung von in ruhenden Flussigkeiten
suspendierten Teilchen. Ann. Phys. (Berlin, Ger.) 1905, 322, 549−
560.
(23) Kumar, P. G. A.; Pregosin, P. S.; Schmid, T. M.; Consiglio, G.
PGSE diffusion, 1H−19F HOESY and NMR studies on several
[Rh(1,5-COD)(Biphemp)]X complexes: detecting positional anion
effects. Magn. Reson. Chem. 2004, 42, 795−800.
(24) Kumar, P. G. A. PGSE Diffusion NMRAn Emerging
Technique for Inorganic/Organometallic Chemists. Aust. J. Chem.
2006, 59, 78−78.
(25) Pregosin, P. S. Ion pairing using PGSE diffusion methods.
Prog. Nucl. Magn. Reson. Spectrosc. 2006, 49, 261−288.
(26) Abadie, M. A.; Trivelli, X.; Duhal, F. M. N.; Kouach, M.;
Linden, B.; Vandewalle, E. G. M.; Capet, F.; Roussel, P.; Rosal, I. D.;
Maron, L.; Agbossou-Niedercorn, F.; Michon, C. Gold(I)-Catalysed
Asymmetric Hydroamination of Alkenes: A Silver- and Solvent-
Dependent Enantiodivergent Reaction. Chem. - Eur. J. 2017, 23,
10777−10788.
(27) Rocchigiani, L.; Jia, M.; Bandini, M.; Macchioni, A. Assessing
the Role of Counterion in Gold-Catalyzed Dearomatization of
Indoles with Allenamides by NMR Studies. ACS Catal. 2015, 5,
3911−3915.
(28) Evans, R.; Dal Poggetto, G.; Nilsson, M.; Morris, G. A.
Improving the Interpretation of Small Molecule Diffusion Coef-
ficients. Anal. Chem. 2018, 90, 3987−3994.
(29) Evans, R.; Deng, Z.; Rogerson, A. K.; McLachlan, A. S.;
Richards, J. J.; Nilsson, M.; Morris, G. A. Quantitative Interpretation
of Diffusion-Ordered NMR Spectra: Can We Rationalize Small
ORCID
Christophe Bour: 0000-0001-6733-5284
Vincent Gandon: 0000-0003-1108-9410
Jean-Nicolas Dumez: 0000-0002-5394-8001
̈
Notes
̈
̈
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by the LabEx Charmmmat
(ANR-11-LABX-0039) and the Agence Nationale de la
Recherche (ANR-16-CE29-0012). The authors thank Arnaud
Voituriez for providing some of the gold complexes.
REFERENCES
■
(1) Hashmi, A. S. K. Gold-Catalyzed Organic Reactions. Chem. Rev.
2007, 107, 3180−3211.
(2) Li, Z.; Brouwer, C.; He, C. Gold-Catalyzed Organic
Transformations. Chem. Rev. 2008, 108, 3239−3265.
́
́
̃
ez, E.; Echavarren, A. M. Gold-Catalyzed Cyclo-
(3) Jimenez-Nun
isomerizations of Enynes: A Mechanistic Perspective. Chem. Rev.
2008, 108, 3326−3350.
(4) Furstner, A. Gold and platinum catalysis-a convenient tool for
̈
generating molecular complexity. Chem. Soc. Rev. 2009, 38, 3208−
3221.
(5) Huang, H.; Zhou, Y.; Liu, H. Recent advances in the gold-
catalyzed additions to C−C multiple bonds. Beilstein J. Org. Chem.
2011, 7, 897−936.
(6) Wang, Y.-M.; Lackner, A. D.; Toste, F. D. Development of
Catalysts and Ligands for Enantioselective Gold Catalysis. Acc. Chem.
Res. 2014, 47, 889−901.
(7) Obradors, C.; Echavarren, A. M. Gold-Catalyzed Rearrange-
ments and Beyond. Acc. Chem. Res. 2014, 47, 902−912.
F
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Molecule Diffusion Coefficients? Angew. Chem., Int. Ed. 2013, 52,
3199−3202.
LiCl on the Schlenk-Equilibrium. J. Am. Chem. Soc. 2016, 138,
4796−4806.
(49) Reich, H. J. Role of Organolithium Aggregates and Mixed
Aggregates in Organolithium Mechanisms. Chem. Rev. 2013, 113,
7130−7178.
(30) Chen, A.; Wu, D.; Johnson, C. S. Determination of Molecular
Weight Distributions for Polymers by Diffusion-Ordered NMR. J.
Am. Chem. Soc. 1995, 117, 7965−7970.
́
(50) Barozzino-Consiglio, G.; Hamdoun, G.; Fressigne, C.;
(31) Li, D.; Keresztes, I.; Hopson, R.; Williard, P. G. Character-
ization of Reactive Intermediates by Multinuclear Diffusion-Ordered
NMR Spectroscopy (DOSY). Acc. Chem. Res. 2009, 42, 270−280.
(32) Viel, S.; Capitani, D.; Mannina, L.; Segre, A. Diffusion-
Ordered NMR Spectroscopy: A Versatile Tool for the Molecular
Weight Determination of Uncharged Polysaccharides. Biomacromole-
cules 2003, 4, 1843−1847.
Harrison-Marchand, A.; Maddaluno, J.; Oulyadi, H. A Combined
1H/6Li NMR DOSY Strategy Finally Uncovers the Structure of
Isopropyllithium in THF. Chem. - Eur. J. 2017, 23, 12475−12479.
(33) Li, D.; Kagan, G.; Hopson, R.; Williard, P. G. Formula Weight
Prediction by Internal Reference Diffusion-Ordered NMR Spectros-
copy (DOSY). J. Am. Chem. Soc. 2009, 131, 5627−5634.
(34) Li, D.; Sun, C.; Liu, J.; Hopson, R.; Li, W.; Williard, P. G.
Aggregation Studies of Complexes Containing a Chiral Lithium
Amide and n-Butyllithium. J. Org. Chem. 2008, 73, 2373−2381.
(35) Hamdoun, G.; Sebban, M.; Cossoul, E.; Harrison-Marchand,
A.; Maddaluno, J.; Oulyadi, H. 1H Pure Shift DOSY: a handy tool to
evaluate the aggregation and solvation of organolithium derivatives.
Chem. Commun. 2014, 50, 4073−4075.
(36) Neufeld, R.; Stalke, D. Accurate molecular weight determi-
nation of small molecules via DOSY-NMR by using external
calibration curves with normalized diffusion coefficients. Chem. Sci.
2015, 6, 3354−3364.
(37) Bachmann, S.; Neufeld, R.; Dzemski, M.; Stalke, D. New
External Calibration Curves (ECCs) for the Estimation of Molecular
Weights in Various Common NMR Solvents. Chem. - Eur. J. 2016,
22, 8462−8465.
(38) Kreyenschmidt, A. K.; Bachmann, S.; Niklas, T.; Stalke, D.
Molecular Weight Estimation of Molecules Incorporating Heavier
Elements from van-der-Waals Corrected ECC-DOSY. ChemistrySelect
2017, 2, 6957−6960.
(39) Bachmann, S.; Gernert, B.; Stalke, D. Solution structures of
alkali metal cyclopentadienides in THF estimated by ECC-DOSY
NMR-spectroscopy (incl. software). Chem. Commun. 2016, 52,
12861−12864.
(40) Neufeld, R.; John, M.; Stalke, D. The Donor-Base-Free
Aggregation of Lithium Diisopropyl Amide in Hydrocarbons
Revealed by a DOSY Method. Angew. Chem., Int. Ed. 2015, 54,
6994−6998.
(41) Wang, Z. J.; Benitez, D.; Tkatchouk, E.; Goddard Iii, W. A.;
Toste, F. D. Mechanistic Study of Gold(I)-Catalyzed Intermolecular
Hydroamination of Allenes. J. Am. Chem. Soc. 2010, 132, 13064−
13071.
(42) Zhdanko, A.; Maier, M. E. The Mechanism of Gold(I)-
Catalyzed Hydroalkoxylation of Alkynes: An Extensive Experimental
Study. Chem. - Eur. J. 2014, 20, 1918−1930.
́
́
(43) Kovacs, G.; Ujaque, G.; Lledos, A. The Reaction Mechanism
of the Hydroamination of Alkenes Catalyzed by Gold(I)−Phosphine:
The Role of the Counterion and the N-Nucleophile Substituents in
the Proton-Transfer Step. J. Am. Chem. Soc. 2008, 130, 853−864.
(44) Brown, T. J.; Widenhoefer, R. A. Cationic Gold(I) π-
Complexes of Terminal Alkynes and Their Conversion to Dinuclear
σ,π-Acetylide Complexes. Organometallics 2011, 30, 6003−6009.
(45) Brooner, R. E. M.; Widenhoefer, R. A. Cationic, Two-
Coordinate Gold π Complexes. Angew. Chem., Int. Ed. 2013, 52,
11714−11724.
(46) Jones, A. C. Gold π-Complexes as Model Intermediates in
Gold Catalysis. In Homogeneous Gold Catalysis; Slaughter, L. M., Ed.;
Springer International Publishing: Cham, 2015; pp 133−165.
(47) Kharlamov, S. V.; L, S. K. Modern diffusion-ordered NMR
spectroscopy in chemistry of supramolecular systems: the scope and
limitations. Russ. Chem. Rev. 2010, 79 (8), 635−653.
(48) Neufeld, R.; Teuteberg, T. L.; Herbst-Irmer, R.; Mata, R. A.;
Stalke, D. Solution Structures of Hauser Base iPr2NMgCl and
Turbo-Hauser Base iPr2NMgCl·LiCl in THF and the Influence of
G
DOI: 10.1021/acs.organomet.8b00709
Organometallics XXXX, XXX, XXX−XXX