"This is the peer reviewed version of the following article: New vistas in transmetalation with discrete “AgCF3”
species: Implications in Pd-mediated trifluoromethylation reactions which has been published in final form at

https://onlinelibrary.wiley.com/doi/epdf/10.1002/chem.201802586
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

New vistas in transmetalation with discrete “AgCF3” species:
Implications in Pd-mediated trifluoromethylation reactions
Sara Martínez de Salinas,[a] Ángel L. Mudarra,[a],[b] Jordi Benet-Buchholz,[a] Teodor Parella,[c] Feliu
Maseras, [a],[d] Mónica H. Pérez-Temprano*[a]
Abstract: This work describes the employment of discrete “AgCF3”
complexes as efficient transmetalating agents to PdIIto surmount
overlooked challenges related to the transmetalation step in Pdcatalyzed trifluoromethylation processes. We report the participation
of a unique silver ate (Cs)[Ag(CF3)2] complex, under stoichiometric
and catalytic conditions, in the unprecedented one-pot formation of
PhCF3 using PhI as starting material. Moreover, we show that the
transmetalation step, which is often ignored in these transformations,
can also determine the success or failure of the coupling process.

Over the past few years, organosilver(I) intermediates have
demonstrated their potential as nucleophilic coupling partners in
Pd-catalyzed transformations.1 However, their ability as
transmetalating agents is far from being fully exploited, most likely
due to their instability (e.g. photosensitivity).2 Therefore, important
fundamental questions such as the scope of the transferred group
or the reactivity of silver(I) ate complexes remains essentially
unexplored. In this context, a particularly interesting test case is
the synergistic Ag/Pd cooperation for the transfer of a CF3 moiety.
This group is a prevalent structural motif in high-value molecules
and organometallic scaffolds due to its unique capability to modify
physicochemical and/or biological properties.3 A priori, the design
of new CF3 shuttles for their use in Pd0/II-catalyzed aryl
trifluoromethylation could seem unnecessary since the reductive
elimination step is considered the central problem associated with
these processes.4 However, a close look at the literature reveals
that the transmetalation step can also dramatically hamper the C–

CF3 bond-forming reaction (Figure 1). A slow nucleophilic
trifluoromethylation leads to undesired reactions by “mismatched”
group exchanges.4a-b,d-e,h,5 Furthermore, CF3– ions can displace
the stabilizing ligands on Pd, forming inactive PdII(CF3)n
species.4a, d-e, h-i,6
Intrigued by these overlooked challenges, we envisioned to
surmount these limitations by exploring “AgCF3” complexes as
selective and rapid CF3 shuttles to PdII. We support our
hypothesis on the well-known lability of the Ag–CF3 bond of in situ
generated “AgCF3” species which readily form a silver(I) ate
[Ag(CF3)2]– complex through a CF3 exchange reaction in polar
solvents.7 Herein, we reveal the exceptional transmetalating
activity of well-defined isolated “AgCF3” compounds to PdII metal
centers (Figure 1), including: 1) their relative reactivity to a
benchmark complex and their comparison to commercially
available nucleophilic reagents; and 2) the high efficiency of
[Ag(CF3)2]– species, only detected by NMR spectroscopy to date
and whose reactivity has been unrecognized for decades,7b-d in
one of the few productive PdII systems, in which the
transmetalation has been pointed out as a challenging step.

◼

Potential unproductive transmetalations in Pd0/II-catalyzed Ar−CF3 couplings
Ligand Displacement
Mismatched Transmetalation
“ CF3− ”

[LmPd(CF3)x(Ar)]
L
Ph−I

L

Pd

Ph
I

LnPd

(i)

Ar LnPdAr(CF3)
CF3
Ar
+ LnPd
LnPd
X
X
Ar
“AgCF3”
◼

[a]

[b]

[c]

[d]

Dr. S. Martínez de Salinas, A. L. Mudarra, Dr. J. Benet-Buchholz,
Prof. F. Maseras, Dr. M. H. Pérez-Temprano
Institute of Chemical Research of Catalonia (ICIQ)
The Barcelona Institute of Science and Technology (BIST)
Avgda. Països Catalans 16, 43007 Tarragona (Spain)
E-mail: mperez@iciq.es
A. L. Mudarra
Departament de Química Analítica i Química Orgànica, Universitat
Rovira i Virgili,
C/ Marcel·li Domingo s/n, 43007 Tarragona (Spain)
Dr. T. Parella
Servei de RMN, Facultat de Ciències
Universitat Autònoma de Barcelona, 08193 Bellaterra (Spain)
Prof. F. Maseras
Departament de Química, Facultat de Ciències
Universitat Autònoma de Barcelona, 08193 Bellaterra (Spain)
Supporting information for this article is given via a link at the end
of the document.

[LnPd0]

L L = dppp
Xantphos

(iii)
Ph−CF3

L
L

Pd

Ph

(ii)

This work (i, ii and iii)

Synthesis of well-defined “AgCF3” ,
including unique [Cat][Ag(CF3)2] salts
Relative reactivity towards a benchmark
system and further translation to
stoichiometric/catalytic coupling

CF3

Figure 1. Exploration of trifluoromethylsilver(I) nucleophiles as efficient CF3
shuttle to PdII systems.

We started our study by exploring the relative reactivity of “AgCF3”
complexes towards a PdII model system (Scheme 1). Several
considerations were taken into account for this initial
investigation. Firstly, (dppp)Pd(Ph)I (1; dppp = 1,3bis(diphenylphosphino)propane) was selected as benchmark
complex because it contains a strong coordinating ligand that

prevents the formation of inactive poly(trifluoromethyl)palladium
compounds and the resulting product (2) does not undergo Ph–
CF3
coupling.8
Secondly,
only
well-defined
isolable
trifluoromethylsilver(I) complexes were targeted to avoid potential
reproducibility issues associated with in situ generated species.
Finally, we pursued fast I-for-CF3 exchanges (< 30 minutes), to
minimize possible undesired by-products related to long reaction
times.

(a)
touchstone
Ph Ph
R3SiCF3/CsF
P
Ph
(2 /4 equiv)
Pd
30 min
P
CF3
Ph Ph 2
19% for R = Me
traces for R = Et

Ph Ph
P
Ph
Pd
P
I
Ph Ph 1

benchmark PdII system

THF, rt, Ar
< 30 min

Ph Ph
P
Ph
+ “AgI”
Pd
P
CF3
Ph Ph 2

well-defined isolable “AgCF3”

10 min

Me

Ph
N
N

SIPrAgCF3
(3, 1.5 equiv)
30 min

Ag

I
2

Me 7

Ph

traces of 2
(b)

Ph Ph
P
Ph
+ “AgCF3”
Pd
P
I
Ph Ph 1

2 (quantitative)
+

4 (1.5 equiv)

Me3SiCF3 (2 equiv)
bathocuproine (1 equiv)

AgF
(3 equiv)

THF, rt, 72 h, Ar
80 %

(Bc)AgCF3
4a

[Cat][Ag(CF3)2]
4b

4a

▪   and 4b are observed by 19F NMR
4a

rapid transmetalation

Elucidation of the nature of the cation of 4b

Scheme 1. Requirements for the initial study

Ph
Ph
N

Before exploring the CF3 group transfer from Ag to (dppp)Pd(Ph)I,
we established as touchstone the trifluoromethylation of 1 with the
widely used nucleophilic trifluoromethyl sources R3SiCF3 (R =
Me,9 Et4b,h-i) in combination with CsF. After 30 minutes, we only
observed 2 in 19% and traces using Me3SiCF3 and Et3SiCF3,
respectively (Figure 2a).10 With these results as a reference, we
then focused on the activity of the scarce examples of isolated
trifluoromethylsilver(I) compounds reported in the literature to
date:
SIPrAgCF311
(SIPr
=
bis(1,3-bis(2,6diisopropylphenyl)imidazole-2-ylidene)
and
(bathophenanthroline)Ag(CF3).12 The treatment of (dppp)Pd(Ph)I
with 1.5 equiv of SIPrAgCF3 (3)13 resulted in <5% yield of
(dppp)Pd(Ph)(CF3) after 30 minutes (Figure 2a). We discarded
the exploration of the other known LAgCF3, bearing the
bathophenanthroline ligand,12 due to stability issues under our
reaction conditions.14 Notably, after some experimentation, using
bathocuproine, we were able to prepare 4 in THF, as a mixture of
(Bc)Ag(CF3) (4a; Bc = bathocuproine) in equilibrium with an ionic
[Ag(CF3)2]– species (4b) (Figure 2b). Initially, we hypothesized
that the structure of 4b was [(Bc)2Ag][Ag(CF3)2], by analogy to
related copper compounds.15 However, a detailed NMR
spectroscopic analysis, including DOSY experiments, confirmed
the absence of [(Bc)2Ag]+ as the cation of 4b, on the basis of the
higher hydrodynamic radius measured for [(Bc)2Ag](SbF6) (5)
(6.86 Å), compared to 4 (4.33 Å). Challenged by this unexpected
outcome, we performed computational studies which
suggested,16 as the most stable cation for 4b, a structure that
contains the silver center coordinated to a bathocuproine ligand
along with two molecules of THF, and a second Bc bound to the
system through stabilizing  interactions.17 In line with the DFT
calculations, we observed the formation of [(Bc)Ag(THF)](SbF6)
(6), characterized by X-ray diffraction, upon exposure of 5 to THF.
With the structural information of 4a/4b in hand, we assessed the
reactivity of this equilibrium mixture. Gratifyingly, the
trifluoromethylation of (dppp)Pd(Ph)I with 1.5 equiv of 4
proceeded cleanly and quantitatively in 10 minutes to afford 2 and
an iodo-bridged dimeric compound with Ag–Ag interactions (7)
(Figure 2a).

N
Ph

5

N

Ph

Ph
Ph

N

Ag
N

N
Ph

Initial proposal
(discarded by 1H DOSY NMR)

N
Solv

N
Ag
Solv

Cation of 6

Our current proposal
(supported by DFT and experimental evidences)

Figure 2. (a) Reactivity of Me3SiCF3, Et3SiCF3, 3 and 4 towards 1. (b) Synthesis
and characterization of 4.

Inspired by the extraordinary transmetalating ability of 4, and
prompted by its different behavior from 3, we wondered whether
[Ag(CF3)2]– could be a non-innocent spectator and participate as
CF3 shuttle.7 To unravel this key question, we targeted the
synthesis of two well-defined (Cat)[Ag(CF3)2] salts, Cat = NBu4
(8NBu4) or Cs (8Cs) to evaluate their relative stability and reactivity.
As shown in Figure 3, the reaction of AgOAc with 4 equiv of
Me3SiCF3, in THF at room temperature, in the presence of 4 equiv
of KF and 1 equiv of NBu4OAc afforded a white crystalline solid,
8NBu4, in 83% isolated yield. Following a similar synthetic route but
using 2 equiv of CsF instead of the combination KF/NBu4OAc, we
synthesized (Cs)[Ag(CF3)2] in 85% yield as a yellow solid. The
structures of both salts, that can be stored for months at –30 ºC
under inert atmosphere in the dark,18 were unambiguously
confirmed by NMR spectroscopy, ESI-MS and single crystal X-ray
diffraction. It is worth mentioning the different bonding situation
between these ionic species. The X-ray structure of 8NBu4 shows
a linear bis(trifluoromethyl)argentate paired together with the
NBu4 cation. In sharp contrast, 8Cs presents a rather unique
structure, with the silver atoms forming linear chains, and the
cesium cations interacting with twelve different fluorine atoms.19
Having synthesized and fully-characterized these singular ionic
species, we next investigated their efficiency as nucleophilic
trifluoromethyl sources. To our delight, the reaction of
(dppp)Pd(Ph)I with 0.75 equiv of either of the two silver salts
resulted in the quantitative formation of 2 in 10 minutes, when
using 8Cs, and in 93% yield for 8NBu4.20

◼ Synthesis
(NBu4)[Ag(CF3)2]

Me3SiCF3 (4 equiv)
KF (4 equiv)
NBu4OAc (1 equiv)

AgOAc

THF, rt, 22 h, Ar

8NBu4 (83%)

Me3SiCF3 (4 equiv)
CsF (2 equiv)

(Cs)[Ag(CF3)2]

THF, rt, 19 h, Ar
Cs

≣
8NBu4

8Cs (85%)
Cs

Cs

Cs
Cs
F
F
F FF
F
F
C F
C
C
F
Ag
Ag
Ag

8Cs
◼ Reactivity
Ph Ph
P
Ph (Cat)[Ag(CF3)2] (8, 0.75 equiv)
Pd
THF, rt, 10 min, Ar
P
I
Ph Ph 1

Ph Ph
P
Ph + (Cat)[CF3AgI]
Pd
+ (Cat)[AgI2]
P
CF3
Ph Ph 2
93% for 8NBu4
100% for 8Cs

Figure 3. Synthesis, Characterization and Reactivity of (Cat)[Ag(CF3)2] (Cat =
NBu4, Cs)

Next, we aimed at evaluating the real potential of our silver
nucleophiles, focusing our attention on one of the few PdII
systems which affords relatively facile PhCF3 coupling. In 2006,
Grushin et al. reported the first example of C–CF3 bond-forming
reductive elimination from an isolated Xantphos-based PdII
derivative, synthetized by treatment of (Xantphos)Pd(Ph)F with
Me3SiCF3.4a In this work and subsequent elegant mechanistic
studies,4d the authors explained in detail, not only the challenges
associated to the reductive elimination from this system, but also
the appealing difficulties related to unfruitful attempts to achieve
the nucleophilic trifluoromethylation of (Xantphos)Pd(Ph)I (9)
using Me3SiCF3/CsF, such as: i) ligand displacement by CF3–;
and/or ii) the formation of unproductive Ph–Ph homocoupling.
Encouraged by our previous results, we envisioned that our silver
nucleophiles could overcome these shortcomings and provide the
unprecedented formation of PhCF3 using 9 as starting material
(Scheme 2). Following the same strategy used for 1, we first
defined as touchstone the reactivity of 9 with R3SiCF3/F– (R = Me,
Et), under comparable reaction conditions to those reported
previously for the high-yielding formation of PhCF 3 from
(Xantphos)Pd(Ph)(CF3) (10).4a,d As expected, we observed the
formation of the coupling product in low yields in the presence of
CsF (14% and 20%, for Me3SiCF3 and Et3SiCF3 respectively).
Then, we examined the reactivity of 9 with our most efficient Ag–
CF3 sources, 4a,21 8NBu4 and 8Cs. We were pleased to observe that
all these transformations led to the targeted product in moderate
to excellent yields (4: 70%,22 8NBu4: 42%, and 8Cs: 84%). The lower
reactivity of (Bc)Ag(CF3) and (NBu4)[Ag(CF3)2] can be ascribed to
unproductive pathways: 23 ligand scrambling between both metals
which affords (Bc)Pd(Ph)(CF3) for 4, and the formation of
poly(trifluoromethyl)complexes and decomposition to AgIII for
8NBu4.18,23 For 8Cs, we corroborated the rapid and selective CF3
transfer by observing full conversion of 9 into 10 in less than 10
minutes in C6H6 at room temperature.

11 (%)b

Entry
O
Ph2P

“CF3” (XX equiv)
Xantphos (1 equiv)
I

Pd
Ph

9

PPh2

11

Me3SiCF3/CsF (2/4)

14

Et3SiCF3/CsF (2/4)

20

3

C6H6, 90 ºC
3 h, Ar

CF3 source (equiv)

1
2

CF3

Scheme 2. Thermolysis of 9 with different “CF3–” sources. aReaction conditions:
9 (0.006 mmol), Xantphos (0.006 mmol), C6H6 (0.01 M) under Ar. b 19F NMR
analysis using fluorobenzene or 4,4’-difluorobiphenyl as internal standards.

4a (1.5)

70

4

8 NBu4 (0.75)
8Cs (0.75)

I

Stoichiometric conditions
Pd(dba)2/Xantphos (1/1 equiv)
(Cs)[Ag(CF3)2] (0.75 equiv)
Toluene, 95 ºC, 3 h, Ar
91%a

(excess)

Pd(dba)2/Xantphos (0.2/0.2 equiv)
(Cs)[Ag(CF3)2] (1 equiv, slow addition)

CF3

Toluene/THF (3:1), 95 ºC, 5 h, Ar
56%a
Catalytic conditions

Scheme 3. Stoichiometric and catalytic reactions.
fluorobenzene as internal standard.

a19

11

F NMR analysis using

In summary, this work presents the potential of discrete “AgCF3”,
including a unique (Cs)[Ag(CF 3)2] salt, as CF3 shuttle to PdII
systems. Our results, not only provide the first reported example
of stoichiometric and catalytic one-pot formation of Ph–CF3
starting from PhI, but also confirm the crucial role of the
nucleophile in the transmetalation step, which can be decisive,
enabling or preventing the product formation. Further work
towards unravelling the potential of silver nucleophiles as
transmetalating agents is currently ongoing in our laboratory.

Acknowledgements
We thank the CERCA Programme/Generalitat de Catalunya and
the Spanish Ministry of Economy, Industry and Competitiveness
(MINECO: CTQ2017-87792-R, CTQ2016-79942-P, CTQ201564436-P, AIE/FEDER, EU, and Severo Ochoa Excellence
Accreditation 2014-2018, SEV-2013-0319) for the financial
support. S. M. S. thanks Severo Ochoa Excellence Accreditation
for post-doctoral contract. A. L. M. thanks La Caixa-Severo Ochoa
programme for a predoctoral grant. We thank to the Research
Support Area of ICIQ. The authors also thank Prof. Rubén Martín
and Dr. Alex Shafir for useful discussions.

42

5

Based on these promising data, we next examined the
compatibility of 8Cs with all elementary steps involved in the
catalytic cycle, under stoichiometric conditions in the presence of
excess of PhI. As previously described, the accumulation of the
oxidative addition product, (Xantphos)Pd(Ph)I, could favor the
mismatched transmetalation shown in Figure 1. Delightfully, using
Pd(dba)2 as Pd0 source and 30 equiv of PhI, we observed the
desired product (11) in slightly higher yield (91%) when compared
to the entry 5 of Scheme 2 (Scheme 3), along with the
regeneration of the oxidative addition product.24 This result points
out the capability of 8Cs for precluding the side-product formation
that could potentially lead to dead-end routes under catalytic
conditions using Xantphos as ligand. Indeed, as a proof-ofconcept, preliminary results show the formation of 11 in 56% yield
by slow addition of 8Cs under catalytic conditions in the presence
of 60 equiv of PhI (Scheme 3).25

84

Conflict of interest

The authors declare no conflict of interest.

[11]

Keywords: organometallic synthesis • silver • structural
elucidation • transmetalation • trifluorometalation

[12]
[13]

[1]

[2]

[3]

[4]

[5]

[6]
[7]

[8]
[9]
[10]

a) L. J. Gooßen, P. P. Lange, N. Rodríguez, C. Linder, Chem. Eur. J.
2010, 16, 3906; b) Y. Gu, X. Leng, Q. Shen, Nat. Commun. 2014, 5,
5405; c) S. Y. Lee, J. F. Hartwig, J. Am. Chem. Soc. 2016, 138, 15278;
d) D. Whitaker, J. Burés, I. Larrosa, J. Am. Chem. Soc. 2016, 138, 8384;
e) M. D. Lotz, N. M. Camasso, A. J. Canty, M. S. Sanford,
Organometallics 2017, 36, 165.
H. Schmidbaur, A. Bayler, Synthesis and uses of organosilver
compounds, in The Chemistry of Organic Derivatives of Gold and Silver,
S. Patai. Z. Rappoport, eds., Wiley, Chichester, UK, 1999, pp. 211.
For selected reviews, see: a) M. A. McClinton, D. A. McClinton,
Tetrahedron, 1992, 48, 6555; b) M. Schlosser, Angew. Chem. Int. Ed.
2006, 45, 5432; Angew. Chem. 2006, 118, 5558; c) D. L. Orsi, R. A.
Altman, Chem. Commun. 2017, 53, 7168.
a) V. V. Grushin, W. J. Marshall, J. Am. Chem. Soc. 2006, 128, 12644;
b) E. J. Cho, T. D. Senecal, T. Kinzel, Y. Zhang, D. A. Watson, S. L.
Buchwald, Science 2010, 328, 1679; c) P. Anstaett, F. Schoenebeck,
Chem. Eur. J. 2011, 17, 12340; d) V. I. Bakhmutov, F. Bozoglian, K.
Gómez, G. González, V. V. Grushin, S. A. Macgregor, E. Martin, F. M.
Miloserdov, M. A. Novikov, J. A. Panetier, L. V. Romashov,
Organometallics 2012, 31, 1315; e) M. C. Nielsen, K. J. Bonney, F.
Schoenebeck, Angew. Chem. Int. Ed. 2014, 53, 5903; f) K. Natte, R. V.
Jagadeesh, L. He, J. Rabeah, J. Chen, C. Taeschler, S. Ellinger, F.
Zaragoza, H. Neumann, A. Brückner, M. Beller, Angew. Chem. Int. Ed.
2016, 55, 2782; Angew. Chem. 2016, 128, 2832. g) J. del Pozo, E.
Gioria, P. Espinet, Organometallics 2017, 36, 2884; h) D. M. Ferguson,
J. R. Bour, A. J. Canty, J. W. Kampf, M. S. Sanford, J. Am. Chem. Soc.
2017, 139, 11662; i) S. T. Keaveney, F. Schoenebeck, Angew. Chem.
Int. Ed. 2018, 57, 4073; Angew. Chem. 2018, 130, 4137.
Other
potential
“mismatched”
group
exchanges
involve
disproportionation of (Ln)Pd(Ar)X and/or (Ln)Pd(Ar)(CF3), for
representative examples, see: a) V. V. Grushin, J. William, W. J.
Marshall, J. Am. Chem. Soc. 2009, 131, 918; b) M. S. Remy, T. R.
Cundari, M. S. Sanford, Organometallics, 2010, 29, 1522; c) S.-L. Zhang,
Z.-Q. Deng, Phys. Chem. Chem. Phys. 2016, 18, 32664 and ref. 3d and
3g.
D. Naumann, N. V. Kirij, N. Maggiarosa, W. Tyrra, Y. L. Yagupolskii, M.
S. Wickleder, Z. Anorg. Allg. Chem. 2004, 630, 746.
a) R. R. Burch, J. C. Calabrese, J. Am. Chem. Soc. 1986, 108, 5359; b)
D. Naumann, W. Wessel, J. Hahn, W. Tyrra, J. Organomet. Chem. 1997,
547, 79; c) W. E. Tyrra, J. Fluorine Chem. 2001, 112, 149; d) W. Tyrra,
Heteroat. Chem. 2002, 13, 561.
V. V. Grushin, W. J. Marshall, J. Am. Chem. Soc. 2006, 128, 4632.
X. Liu, C. Xu, M. Wang, Q. Liu, Chem. Rev. 2015, 115, 683.
We also carried out the reaction of complex 1 with (Phen)CuCF3, other
commercially available nucleophilic trifluoromethylating reagent. This
copper compound only afforded 2 in 12% yield. See SI for further details.

[14]

[15]

[16]

[17]

[18]
[19]

[20]
[21]
[22]

[23]
[24]

[25]

SIPrAgCF3 (3) was prepared using a modified procedure from B. K. Tate,
A. J. Jordan, J. Bacsa, J. P. Sadighi, Organometallics 2017, 36, 964. See
SI for further details.
Z. Weng, R. Lee, W. Jia, Y. Yuan, W. Wang, X. Feng, K.-W. Huang,
Organometallics 2011, 30, 3229.
CCDC 1588501 (3), CCDC 1566841 (4a), CCDC 1566842 (5), CCDC
1566843 (6), CCDC 1566844 (S1), CCDC 1566845 (7), CCDC 1566839
(8Cs) and CCDC 1588502 (8NBu4) contain the supplementary
crystallographic data for this paper. These data are provided free of
charge by The Cambridge Crystallographic Data Centre.
We observed the rapid decomposition of (bathophenanthroline)Ag(CF3)
to HCF3 and other by-products in polar coordinating solvents such as
THF or DMF. See SI for further details.
a) G. G. Dubinina, J. Ogikubo, D. A. Vicic, Organometallics 2008, 27,
6233; b) H. Morimoto, T. Tsubogo, N. D. Litvinas, J. F. Hartwig, Angew.
Chem. Int. Ed. 2011, 50, 3793; Angew. Chem. 2011, 123, 3877.
axD calculations with a double− plus polarization basis set in
solvent. Full computational details provided in the Supporting
Information; b) A dataset collection of computational results is available
in the ioChem-BD repository M. Alvarez-Moreno, C. de Graaf, N. Lopez,
F. Maseras, J. M. Poblet, C. Bo, J. Chem. Inf. Model. 2015, 55, 95.
The proposed 4b-cation has a free energy more than 5 kcal/mol below
than any of the computed cationic alternatives. Indeed, the comparison
between the calculated/experimental ratio for 4a:4b, 92:8/98:2 in THF
and 58:42/66:34 in DMF, also supports the assignment for 4b-cat. See
SI for further details.
For a complete description about the stability of 8Cs and 8NBu4, see SI p
S11-16.
For representative examples of high Cs···F bonding contacts, see: a) D.
Pollak, R. Goddard, K.-R. Pörschke, J. Am. Chem. Soc. 2016, 138, 9444;
b) L. Carreras, L. Rovira, M. Vaquero, I. Mon, E. Martin, J. BenetBuchholz, A. Vidal-Ferran, RSC Advances 2017, 7, 32833.
Since we only added 0.75 equiv of 8, the quantitative formation of 2
suggests that (Cat)[(CF3)AgI] is also capable of acting as CF3 shuttle.
In benzene, the equilibrium is totally shifted towards 4a, see p. S17.
This experimental result, along with those obtained in THF for the
benchmark system, suggest the participation of 4a and 4b in the CF3
group exchange.
See supporting information for further details on the thermolysis of 9 with
the different CF3.
Grushin observed the formation of the mismatched product
(Xantphos)Pd(CF3)I during the thermolysis of (XantPhos)Pd(Ph)(CF3)
and PhI during 8 h at 70 ºC. See SI p 5-7 from reference 4a.
The formation of 11 in moderate yield (56% yield or catalyst turnover
number of 5.6) is not due to the well-precedented unproductive reactions
discussed along the text, but for practical challenges related to the use
of 8Cs as transmetalating agent under catalytic conditions. The product
yield was calculated taking as limiting reagent 8Cs and considering the
transmetalation of both CF3 groups. See SI pS84 for further details.