Organometallics
Communication
product 14b was formed in trace amounts, as revealed by its
diagnostic signals in the 1H NMR and 19F NMR spectra (Table
1, entry 4). Since the low yields obtained could be traced back to
experimental errors during the addition of the base due to the
small scale of the reaction, in the subsequent experiments the
amount of both the base and TMSCF2H were increased to 500
mol %. Fine-tuning of the experimental conditions for the
workup and the handling of the gold complexes (see Supporting
Information) led to almost full conversion to the product using
IPrAuCl (13a) (Table 1, entry 6) and to full conversion for
SIPrAuCl (14a) (Table 1, entry 5). Purification of both gold
complexes by column chromatography, precipitation, or
crystallization to give accurate isolated yields was not possible.
Due to this drawback, we turned our attention to quantitative
NMR for the determination of reaction yields with the addition
of an internal standard. Dibromomethane was selected for this
purpose and allowed to determine the yield for the IPrAu−
CF2H (13b) (92%) and SIPrAu−CF2H (14b) (quantitative)
gold complexes.
Next, we moved to the phosphine gold(I) complexes.
Unfortunately, when the conditions developed for the
preparation of (NHC)Au−CF2H were tested on the electron-
rich [tris(p-methoxyphenyl)phosphine]gold(I) chloride (1a),
the desired product was obtained in trace amounts (Table 1,
entry 7). Gratifyingly, the use of 200 mol % of NaOtBu and 500
mol % of TMSCF2H delivered desired compound 1b in
excellent yield (Table 1, entry 8).
With two set of conditions in hand (conditions A and B) the
full scope of this transformation was evaluated (Scheme 1).
Additional gold(I) complexes bearing electron-rich phosphines
could be converted into their corresponding Au−CF2H
complexes in yields ranging from very good to excellent (4b,
5b, 6b, and 7b). The neutral triphenylphosphinegold(I)
chloride (9a) was likewise a suitable substrate for this
transformation, since corresponding Au−CF2H complex 9b
could be generated in an excellent yield. To further expand the
scope for NHC−Au complexes, analogues 15b and 16b were
prepared following protocol A. To our delight, we were able to
obtain suitable crystals for single-crystal X-ray diffraction
analysis, which unambiguously confirmed the identity of
compounds 6b, 7b, 13b, and 14b. Surprisingly, the more
challenging electron-deficient (p-CF3−Ph)3PAu-CF2H (10b)
and (p-F-Ph)3PAu-CF2H (11b) complexes could be synthesized
in good and excellent yields, respectively (for scope limitations,
Gold(I) complex 10c was obtained as a byproduct in the
formation of 10b. To shed light on the origin of this compound
and the general mechanism for the formation of the LAu−CF2H
complexes, the experiments outlined in Scheme 2 were
performed. A TSMCF2H-free experiment with JohnPhosAu−
Cl 7a in the presence of NaOH (eq 1, Scheme 2) revealed the
partial conversion of the starting gold complex to gold(I)
hydroxide 7c and, very likely, to triaurated oxonium cation 7d,18
after 3 h of reaction. The fact that 7a, 7c, and 7d were present in
the crude mixture almost in the same ratio after 14 h of reaction
seems to indicate the establishment of an equilibrium between
the three species, which is eventually shifted to the right when
the difluoromethylating agent is present in the reaction media to
react with 7c and furnish product 7b. Similar results could be
extrapolated to the NaOtBu base (see the Supporting
Information for further details). The analogous control
experiment with 10a led to the isolation of 10c in 75% yield
(eq 2). Secondary phosphine oxide−gold(I) complex19 10d and
Figure 1. Starting [LAuCl] complexes.
could stabilize Au(I)−fluoroalkyl complexes (13a−16a)16 were
included in this study.
The Si-to-Au “transmetalation” experiments aimed to
determine the “matching transmetalation couples” from the
pool of gold(I) chloride derivatives and the commercially
available TMSCF2H. The optimization of the reaction
conditions began with the treatment of all the starting gold(I)
complexes with 200 mol % of the nucleophilic difluoromethylat-
ing agent in THF for 3 h without other additives. As expected, no
formation of the product was observed for any of them (Table 1,
Table 1. Optimization of the Reaction Conditions for the Si-
to-Au Transmetalation
TMSCF2H
(mol %)
a b
,
entry [LAu-Cl]
base (mol %)
yield
1
2
1a−16a
1a−16a
200
200
c
NaOAc, CsOAc, CsF,
c
NaF, n-Bu4NF
c
(200)
3
4
5
14a
14a
14a
200
200
500
NaOtBu (200)
NaOH (200)
NaOH (500)
trace amounts
quantitative
(conditions A)
d
6
13a
500
NaOH (500)
92%
(conditions A)
trace amounts
7
8
1a
1a
500
500
NaOH (500)
NaOtBu (200)
94%
(conditions B)
a
b
1
Reaction scale: 0.030−0.040 mmol. Yield calculated by H NMR
c
using CH2Br2 as internal standard. Complex 1a was used. MeOH or
CH3CN were used as cosolvents. Concentration: [0.03M].
d
entry 1). The use of bases such as acetates or fluorides did not
allow to obtain the desired CF2H−gold(I) complexes (Table 1,
entry 2). Next, we focused our attention on NHC−Au(I)
complex 14a with the aim of investigating the effect of other
bases in the reaction outcome. On the basis of the protocols
employed in the synthesis of the analogous silver(I)7b and
copper(I)5d complexes, NaOtBu was the first base selected.
Unfortunately, the addition of 200 mol % of this additive led to
the recovery of the starting complex (Table 1, entry 3).
Nevertheless, when the same amount of NaOH was used,
B
Organometallics XXXX, XXX, XXX−XXX