Selective Aryl-Fluoride Reductive Elimination from a Platinum(IV) Complex

Article abstract of DOI:10.1002/anie.201503116

A difluoro(mesityl)platinum(IV) complex underwent highly selective reductive elimination of 2-fluoromesitylene upon heating in toluene. Kinetic analysis and DFT calculations suggest that the C-F coupling involves a five-coordinate Pt^IV transient intermediate resulting from the rate-limiting dissociation of the pyridine ligand.

Full text of DOI:10.1002/anie.201503116

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201503116
German Edition: DOI: 10.1002/ange.201503116
Electrophilic Fluorination
Selective Aryl–Fluoride Reductive Elimination from a Platinum(IV)
Complex**
Ina Dubinsky-Davidchik, Israel Goldberg, Arkadi Vigalok,* and Andrei N. Vedernikov*
Tos, but not OPh or OAc).[11] Again, no aryl F bond was
formed in those reactions. Finally, Mirica and co-workers
recently studied analogous PdIV complexes; they reported an
À
Abstract: A difluoro(mesityl)platinum(IV) complex under-
went highly selective reductive elimination of 2-fluoromesity-
lene upon heating in toluene. Kinetic analysis and DFT
À
À
calculations suggest that the C F coupling involves a five-
aryl OH reductive elimination from a hydroxopalladium(IV)
species, but no aryl–fluorine bond formation was observed for
coordinate PtIV transient intermediate resulting from the rate-
limiting dissociation of the pyridine ligand.
its fluoro analogue.[12] Herein, we present the first example of
À
exclusive C(aryl) F reductive elimination from an isolated
À
À
E
lectrophilic fluorination of organometallic complexes of
aryl platinum(IV) complex from which both aryl O and aryl
transition metals often involves the formation of high-valent
metal species, which in turn undergo reductive elimination.[1]
In recent years, such a reaction sequence has been widely used
F reductive-elimination reactions might be viable.
À
Previously, while probing the viability of aryl F bond
formation upon the electrophilic fluorination of diaryl
platinum(II) complexes with XeF2,[5b,13] we found that a com-
peting aryl–aryl elimination from the PtIV intermediates is by
for making new C X bonds (X = halogen, O, N, C).[2] Of these
À
reactions, the reductive elimination of carbon–fluorine bonds
is perhaps the most attractive because of both the importance
of organofluorine compounds in industry[3] and the scarcity of
À
far the predominant path. To exclude the possibility of a C C
coupling reaction, we designed and prepared the new cyclo-
metalated monoaryl PtII complexes 1a–c supported by a very
bulky anionic P, O chelating ligand (Scheme 1). We hypothe-
synthetic alternatives for constructing such bonds.[4] Aryl F
À
reductive elimination is a particularly challenging reaction
that has received considerable attention with regard to the
catalytic fluorination of non-activated aromatic com-
pounds.[5,6] With the notable exception of the Buchwald Pd0/
PdII catalytic cycle,[7] the majority of catalytic aromatic
À
sized that the C F reductive elimination of the aryl plati-
À
fluorination reactions involve aryl F reductive elimination
from a late transition metal in a high oxidation state and
usually rely on a directing group in the ortho position.[8]
Importantly, analysis of previously reported examples of
À
aryl F reductive elimination suggests that it is the least
kinetically competitive elimination reaction.[9] Recently, San-
Scheme 1. Synthesis of O-metalated aryl platinum complexes 1 and 2.
Ad=adamantyl, cod=1,5-cyclooctadiene, py=pyridine.
À
ford and co-workers showed the concurrent formation of C
3
3
2
IV
À
À
N, C(sp ) F, and C(sp ) C(sp ) bonds from a Pd complex
with alkyl, aryl, sulfonamide, and fluoro ligands.[10] Aryl
À
fluoride reductive-elimination products were not formed
num(IV) difluorides derived from 1a–c could be favored by
À
under the reported conditions. The same research group
the steric bulk of the ligand over their C O elimination.
3
3
À
À
also described competitive C(sp ) F and C(sp ) O reductive
elimination from aryl alkyl palladium(IV) complexes, pro-
Complexes 1a–c were characterized by multinuclear NMR
spectroscopic techniques and, in the case of 1a and 1c, also by
X-ray crystallography (see Figure S1 in the Supporting
Information).
À
À
vided the O ligand was a weak nucleophile (O NO2 or O
Complexes 1a–c reacted with XeF2 in CH2Cl2 to give the
corresponding platinum(IV) difluorides 2a–c (Scheme 1).
The fluoro ligands are trans to the phenoxo and aryl groups
(see below), thus giving very distinct signals in the 19F NMR
spectra of 2. In contrast to the selective formation of 2a,b, the
[*] I. Dubinsky-Davidchik, Prof. I. Goldberg, Prof. A. Vigalok
School of Chemistry, The Sackler Faculty of Exact Sciences
Tel Aviv University
Tel Aviv 69978 (Israel)
E-mail: avigal@post.tau.ac.il
Prof. A. N. Vedernikov
Department of Chemistry and Biochemistry
University of Maryland
College Park, MD 20742 (USA)
E-mail: avederni@umd.edu
À
formation of 2c was accompanied by minor benzylic C H
fluorination of the aryl ligand to give 3 and HF (Scheme 2), so
À
that 2c appeared as the HF adduct. The competing C H
fluorination could be fully suppressed by the use of CH3CN–
CH2Cl2 mixtures (1:5–1:1 ratio). Conversely, in C6H5Cl or
C6H5Cl–toluene mixtures, 2c·HF and the fluoride 3 were
formed in an approximately 1:1 ratio (Scheme 2).
[**] This research was supported by grant 2010119 from the US–Israel
Binational Science Foundation.
Supporting information for this article, including complete exper-
imental and computational details, is available on the WWW under
http://dx.doi.org/10.1002/anie.201503116.
Complex 2c was isolated and characterized by NMR
spectroscopy and X-ray crystallography (Figure 1). The
Angew. Chem. Int. Ed. 2015, 54, 12447 –12451
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12447

.
Angewandte
Communications
À
for the C H fluorination leading to 3 that we observed in the
reaction of 1c with XeF2 in CH2Cl2 or chlorobenzene solvents.
To our knowledge, the formation of 2-fluoromesitylene from
2c is unprecedented; it is the first example of well-defined
2
IV
À
C(sp ) F reductive elimination from an isolated Pt com-
plex.[15]
À
Our kinetics study of the Mes F reductive elimination
from 2c revealed the first-order behavior of the reaction,
which has the following reaction-activation parameters:
DH° = (25.3 Æ 1.3) kcalmolÀ1
and
DS° = (2.5 Æ 0.3)
calKÀ1 molÀ1 (Figure 2; see also Figure S3). The addition of
pyridine (20 equiv) did not decrease the reaction rate.
À
Scheme 2. Competitive oxidative addition–benzylic C H fluorination of
the mesityl complex 1c. Mes=mesityl.
Figure 2. Eyring plot for the reductive elimination of 2-fluoromesitylene
from complex 2c.
Figure 1. X-ray crystal structures of complexes 2c (left) and 4 (right).
Hydrogen atoms are omitted for clarity. Selected bond distances [Š]
and angles [8]: 2c: Pt1–F1 2.043(2), Pt1–F2 1.963(2), Pt1–O1 1.998(2),
Pt1–C32 2.084(4), Pt1–N1 2.119(3); F2-Pt1-F1 88.68(9), F2-Pt1-C32
89.50(12), O1-Pt1-P1 82.28(7); 4: Pt1–F1 1.987(4), Pt1–O1 1.990(5);
O1-Pt1-P1 86.02(14), F1-Pt1-N1 87.55(19).
Some cationic monofluoropalladium(IV) complexes were
À
found to be significantly more reactive in aryl F elimination
than the corresponding neutral difluoropalladium(IV) ana-
logues owing to more readily generated coordinative unsatu-
ration.[15] Although facile FÀ dissociation from 2c appears
unlikely in nonpolar toluene, we decided to independently
prepare a cationic monofluoro analogue of 2c and study its
crystal structure revealed the PtIV center in an octahedral
À
À
environment with Pt F1 and Pt F2 distances of 2.043(2) and
1.963(2) Š, respectively. These data are in agreement with
a stronger trans influence of the aryl group as compared to the
phenoxide ligand.
À
reactivity toward C F elimination. The treatment of 1c with
Selectfluor (1 equiv) gave an unstable product lacking a signal
for platinum-bound fluorine in the 19F NMR spectrum. No
reaction was observed between 1c and the bulky N-fluoro-
2,4,6-collidinium cation. The reaction between N-fluoropyr-
idinium tetrafluoroborate and 1c gave a mixture of two
complexes that showed signals for the platinum-bound
fluorine atom, at À298.4 (JPt,F = 1149 Hz) and À310.7 ppm
(JPt,F = 1442 Hz), respectively, which were tentatively
assigned to cationic 5c-L, in which L is either pyridine or
acetonitrile (Scheme 4). Upon the addition of solid CsF, both
compounds were slowly converted into 2c. Heating of the
above mixture of 5c-L as a solution in acetonitrile/toluene or
Significantly, whereas 2a,b failed to produce fluoroarenes
at elevated temperatures, heating of the mesitylplatinum(IV)
difluoride 2c in toluene at 60–1008C led to clean elimination
of 2-fluoromesitylene and the formation of the PtII complex 4
(Scheme 3; see also Figure S2).[14] The identity of complex 4
was confirmed by NMR spectroscopy and X-ray crystal-
À
structure analysis (Figure 1). No aryl O elimination was
observed under these conditions. Furthermore, no benzylic
fluorination to form 3 took place, thus indicating that this
difluoroplatinum(IV) complex is not a reaction intermediate
Scheme 4. Synthesis and reactivity of the cationic fluoroplatinum(IV)
complex.
Scheme 3. Selective reductive elimination of 2-fluoromesitylene from
2c.
12448 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 12447 –12451

Angewandte
Chemie
the suspension of 5c-L in chlorobenzene led to decomposition
and a complex mixture of products containing neither 2-
fluoromesitylene nor 3. Thus, the cations 5c-L do not appear
five-coordinate transient intermediate 8 (path c), a product of
pyridine dissociation from 2c. Path c shows what we believe is
À
the most likely mechanism of the C F coupling of 2c.
À
to be intermediates in the aryl F reductive elimination or
According to this mechanism, 2c changes its conformation to
form 7 with the mesityl-ring arrangement suitable for the
À
benzylic C H fluorination.
2
À
À
To get further insight into the mechanism of the C(sp ) F
subsequent C F coupling with the fluorine ligand trans to the
elimination of 2c, we performed DFT calculations. The
resulting Gibbs energy profiles for three potential reaction
pathways a–c are given in Figure 3 along with a pictorial
representation of the structures of the species involved.
phosphine. This isomerization proceeds via the low-energy
transition state TSiso7. The subsequent rate-limiting dissoci-
ation of the pyridine ligand and the migration of a fluoride
ligand (TSd7, 23.0 kcalmolÀ1) generate the required transient
À
À
The transition state TSC-F2 for the direct C F elimination
intermediate 8. Since the following C F coupling of 8 is much
of 2c has a prohibitively high energy exceeding 70 kcalmolÀ1
(path a), a significant part of which is needed to arrange the
mesityl fragment perpendicular to the Pt-C-F plane. This
arrangement is disfavored by one of the 1-adamantyl groups
of the bulky P, O chelating ligand present in 2c. These highly
unfavorable interactions are avoided in 6, the isomer of 2c
with both fluorine atoms cis to the mesityl ligand (path b). In
6, the steric bulk of the P, O ligand favors the necessary
faster than pyridine recoordination to form 7, the dissociation
of pyridine from 7 is virtually irreversible.[16] Remarkably, the
Gibbs energy of TSd7 corresponding to the rate-limiting loss
of pyridine is a good match to the experimental Gibbs
À
activation energy for the C F coupling reaction in Scheme 3
((24.5 Æ 0.2) kcalmolÀ1).[17] The proposed mechanism (path c)
À
can also readily account for the lack of C F elimination
reactivity of 2a,b, the electron-poorer and less bulky ana-
logues of 2c. According to our calculations, the barriers for
pyridine dissociation from both 2a and 2b (TSd7a, TSd7b) are
2–3 kcalmolÀ1 higher than that for 2c.[16] Furthermore, their
À
conformation of the Pt–Mes fragment for the C F coupling
involving the fluorine ligand trans to the phosphorus atom.
The corresponding transition state TSC-F6 has a low energy of
22.1 kcalmolÀ1. In spite of these features, path b was excluded
from further analysis, since the Gibbs activation energy for
À
C F elimination barriers (TSC-F8a, TSC-F8b) are 8–9 kcal
molÀ1 higher than for 2c, thus reflecting a diminished steric
repulsion between their aryl ligands, which lack ortho
substituents, and other ligands present in 8. Finally, the lack
the direct isomerization of 2c to 6 is prohibitively high
À1
À
(40.4 kcalmol ). An even more facile C F coupling via the
transition state TSC-F8 (16.4 kcalmolÀ1) is possible from the
of the potentially competitive C O reductive elimination of
À
À
Figure 3. DFT-calculated Gibbs energy profile for potential pathways of the C F reductive elimination of 2c in the gas phase. The plane of the
mesityl ring is shown with a short bold line.
Angew. Chem. Int. Ed. 2015, 54, 12447 –12451
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12449

.
Angewandte
Communications
2c and 8 could be explained by the enormous steric strain that
would have to be imposed to achieve the required conforma-
tion of the Pt–Mes fragment, that is, to arrange the mesityl
ring perpendicular to the Pt-C-O plane. As a result of such
structural disparity, the transition states corresponding to the
with XeF2 occurs via a cationic intermediate “en route” to the
aryl platinum(IV) difluoride. Studies of the scope and
potential applications of the reaction are currently under way.
À
C O elimination of 2c or 8 could not be located.
Experimental Section
The synthesis and characterization of the new compounds are
described in the Supporting Information. In a typical kinetic experi-
ment, a solution of 2c (10 mg, 0.013 mmol) in toluene (0.5 mL) in
a polytetrafluoroethylene liner was heated inside the NMR probe at
60–1008C, and reaction progress was monitored by 19F NMR
spectroscopy by using an internal reference (1,4-difluorobenzene).
Only 4 and 2-fluoromesitylene were observed under these conditions.
The identity of 2-fluoromesitylene was confirmed by comparison with
an authentic sample and by GC.
The benzylic fluorination of 1c to form 3, an intriguing
reaction found during our attempts to convert the PtII
complex 1c into the PtIV difluoride 2c (Scheme 2), was also
studied computationally, albeit in less detail (Scheme 5). The
CCDC 1050345, 1050346, 1050347, and 1050348 contain the
supplementary crystallographic data for this paper. These data are
provided free of charge by The Cambridge Crystallographic Data
Centre.
Keywords: chemoselectivity · electrophilic fluorination ·
reaction mechanisms · platinum · reductive elimination
How to cite: Angew. Chem. Int. Ed. 2015, 54, 12447–12451
Angew. Chem. 2015, 127, 12624–12628
[1] A. Vigalok, Organometallics 2011, 30, 4802 – 4810.
[2] a) A. Vigalok, Acc. Chem. Res. 2015, 48, 238 – 247; b) K. M.
Engle, T.-S. Mei, X. Wang, J.-Q. Yu, Angew. Chem. Int. Ed. 2011,
50, 1478 – 1491; Angew. Chem. 2011, 123, 1514 – 1528.
[3] For recent reviews, see: a) J. Wang, M. Sµnchez-Roselló, J. L.
AceÇa, C. del Pozo, A. E. Sorochinsky, S. Fustero, V. A. Solo-
shonok, H. Liu, Chem. Rev. 2014, 114, 2432 – 2506; b) G.
Theodoridis, Adv. Fluorine Sci. 2006, 2, 121 – 175; c) F. Babudri,
G. M. Farinola, F. Naso, R. Ragnia, Chem. Commun. 2007,
1003 – 1022.
[4] For general reviews, see: a) S. D. Taylor, C. C. Kotoris, G. Hum,
Tetrahedron 1999, 55, 12431 – 12477; b) R. D. Chambers, Fluo-
rine in Organic Chemistry, CRC, Boca Raton, FL, 2004; c) T.
Hiyama, Organofluorine Compounds, Springer, Berlin, 2000;
d) T. Liang, C. Neumann, T. Ritter, Angew. Chem. Int. Ed. 2013,
52, 8214 – 8264; Angew. Chem. 2013, 125, 8372 – 8423.
[5] a) T. Furuya, T. Ritter, J. Am. Chem. Soc. 2008, 130, 10060 –
10061; b) A. W. Kaspi, A. Yahav-Levi, I. Goldberg, A. Vigalok,
Inorg. Chem. 2008, 47, 5 – 8; c) N. D. Ball, M. S. Sanford, J. Am.
Chem. Soc. 2009, 131, 3796 – 3797.
[6] V. V. Grushin, Acc. Chem. Res. 2010, 43, 160 – 171.
[7] a) D. A. Watson, M. Su, G. Teverovskiy, Y. Zhang, J. Garcia-
Fortanet, T. Kinzel, S. L. Buchwald, Science 2009, 325, 1661 –
1664; b) H. G. Lee, P. J. Milner, S. L. Buchwald, J. Am. Chem.
Soc. 2014, 136, 3792 – 3795.
[8] a) K. L. Hull, W. Q. Anani, M. S. Sanford, J. Am. Chem. Soc.
2006, 128, 7134 – 7135; b) X. Wang, T.-S. Mei, J.-Q. Yu, J. Am.
Chem. Soc. 2009, 131, 7520 – 7521; c) K. S. L. Chan, M. Wasa, X.
Wang, J.-Q. Yu, Angew. Chem. Int. Ed. 2011, 50, 9081 – 9084;
Angew. Chem. 2011, 123, 9247 – 9250; d) S.-J. Lou, D.-Q. Xu, Z.-
Y. Xu, Angew. Chem. Int. Ed. 2014, 53, 10330 – 10335; Angew.
Chem. 2014, 126, 10498 – 10503; e) for a recent review, see: Y. Li,
Y. Wu, G.-S. Li, X.-S. Wang, Adv. Synth. Catal. 2014, 356, 1412 –
1418.
À
Scheme 5. Proposed mechanism of the benzylic C H fluorination of
1c with XeF2 in CH2Cl2. The DFT-calculated standard Gibbs energies of
the reactants are shown in blue in kcalmolÀ1 for a solution in CH2Cl2.
IV
À
proposed key intermediate in this case is the C H agostic Pt
complex 9, which is deprotonated by a “naked” fluoride anion
to form the benzylplatinum(IV) complex 10. The latter is
converted into the final benzyl fluoride 3 through the loss of
a fluoride anion (perhaps in the form of HF2À and assisted by
HF) and its subsequent SN2 attack at the benzylic carbon
atom of the cationic PtIV transient intermediate 11. We also
found that the alternative intermediate 12 is unlikely to be
À
involved in the C F elimination; the Gibbs activation energy
for this reaction is prohibitively high at 40.5 kcalmolÀ1. The
calculations are in agreement with our experimental data,
which show a significant solvent effect in the formation of 3.
In the presence of coordinating CH3CN, the formation of an
agostic complex 9 should be disfavored, thus suppressing the
À
C H fluorination pathway. In turn, in more weakly coordi-
nating media, the C H agostic intermediate 9 would prevail,
thus leading to 3, as we observed experimentally.
À
In summary, we presented herein the first example of
IV
À
selective aryl F reductive elimination from an isolated Pt
complex. In contrast to previously reported similar PdIV
2
À
systems, the C(sp ) F elimination outcompetes the poten-
tially available C(sp ) O pathway, which is disfavored as
a result of the enormous steric bulk imposed by a new
chelating 2-bis(1-adamantyl)phosphinophenoxide ligand at-
tached to the PtIV center. Experimental and computational
studies suggest that the rate-limiting step of the aryl F
elimination involves the formation of a neutral five-coordi-
nate metal species with subsequent fast C F coupling. In turn,
benzylic C H fluorination of aryl platinum(II) complexes
2
À
À
[9] We did not find reported examples of aryl F reductive
elimination reactions that compete successfully with other
À
À
elimination pathways, including alkyl F reductive elimination;
see, for example: J. M. Racowski, B. G. Gary, M. S. Sanford,
Angew. Chem. Int. Ed. 2012, 51, 3414 – 3417; Angew. Chem. 2012,
124, 3470 – 3473.
À
À
12450 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 12447 –12451

Angewandte
Chemie
IV
À
[10] M. H. PØrez-Temprano, J. M. Racowski, J. W. Kampf, M. S.
Sanford, J. Am. Chem. Soc. 2014, 136, 4097 – 4100.
[11] N. M. Camasso, M. H. PØrez-Temprano, M. S. Sanford, J. Am.
Chem. Soc. 2014, 136, 12771 – 12775.
[15] For aryl F elimination from an isolated Pd complex, see: T.
Furuya, D. Benitez, E. Tkatchouk, A. E. Strom, P. Tang, W. A.
Goddard III, T. Ritter, J. Am. Chem. Soc. 2010, 132, 3793 – 3807.
[16] See the Supporting Information for more details.
[17] An alternative reaction sequence leading from 2c to 8 that
involves the rate-limiting dissociation of the pyridine ligand from
2c first, with subsequent isomerization of the resulting five-
coordinate PtIV transient intermediate, was also analyzed (see
the Supporting Information). Our estimate of the Gibbs
[12] F. Qu, J. R. Khusnutdinova, N. P. Rath, L. M. Mirica, Chem.
Commun. 2014, 50, 3036 – 3039.
[13] a) A. Yahav, I. Goldberg, A. Vigalok, J. Am. Chem. Soc. 2003,
125, 13634 – 13635; b) A. W. Kaspi, I. Goldberg, A. Vigalok, J.
Am. Chem. Soc. 2010, 132, 10626 – 10627; c) I. S. Dubinsky-
Davidchik, I. Goldberg, A. Vigalok, A. N. Vedernikov, Chem.
Commun. 2013, 49, 3446 – 3448.
activation energy for this reaction sequence is 31.0 kcalmolÀ1
.
[14] The steric bulk of the aryl ligand is important for the formation
Received: April 5, 2015
Revised: May 20, 2015
Published online: June 19, 2015
À
of a C F bond: S. B. Zhao, R. Y. Wang, H. Nguyen, J. J. Becker,
M. R. GagnØ, Chem. Commun. 2012, 48, 443 – 445.
Angew. Chem. Int. Ed. 2015, 54, 12447 –12451
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12451