Competing C-F activation pathways in the reaction of Pt(0) with fluoropyridines: Phosphine-assistance versus oxidative addition

Article abstract of DOI:10.1021/ja8046238

A survey of computed mechanisms for C-F bond activation at the 4-position of pentafluoropyridine by the model zero-valent bis-phosphine complex, [Pt(PH₃)(PH₂Me)], reveals three quite distinct pathways leading to square-planar Pt(II) products. Direct oxidative addition leads to cis-[Pt(F)(4-C₅NF₄)(PH₃)(PH₂Me)] via a conventional 3-center transition state. This process competes with two different phosphine-assisted mechanisms in which C-F activation involves fluorine transfer to a phosphorus center via novel 4-center transition states. The more accessible of the two phosphine-assisted processes involves concerted transfer of an alkyl group from phosphorus to the metal to give a platinum(alkyl)(fluorophosphine), trans-[Pt(Me)(4-C₅NF ₄)(PH₃)(PH₂F)], analogues of which have been observed experimentally. The second phosphine-assisted pathway sees fluorine transfer to one of the phosphine ligands with formation of a metastable metallophosphorane intermediate from which either alkyl or fluorine transfer to the metal is possible. Both Pt-fluoride and Pt(alkyl)(fluorophosphine) products are therefore accessible via this route. Our calculations highlight the central role of metallophosphorane species, either as intermediates or transition states, in aromatic C-F bond activation. In addition, the similar computed barriers for all three processes suggest that Pt-fluoride species should be accessible. This is confirmed experimentally by the reaction of [Pt(PR ₃)₂] species (R = isopropyl (iPr), cyclohexyl (Cy), and cyclopentyl (Cyp)) with 2,3,5-trifluoro-4-(trifluoromethyl)pyridine to give cis-[Pt(F){2-C₅NHF₂(CF₃)}(PR₃) ₂]. These species subsequently convert to the trans-isomers, either thermally or photochemically. The crystal structure of cis-[Pt(F){2-C ₅NHF₂(CF₃)}(PiPr₃)₂] shows planar coordination at Pt with r(F-Pt) = 2.029(3) A and P(1)-Pt-P(2) = 109.10(3)°. The crystal structure of frans-[Pt(F){2-C₅NHF ₂(CF₃)}(PCyp₃)₂] shows standard square-planar coordination at Pt with r(F-Pt) = 2.040(19) A.

Full text of DOI:10.1021/ja8046238

Published on Web 10/25/2008
Competing C-F Activation Pathways in the Reaction of Pt(0)
with Fluoropyridines: Phosphine-Assistance versus Oxidative
Addition
Ainara Nova,† Stefan Erhardt,‡, Naseralla A. Jasim,† Robin N. Perutz,*,†
Stuart A. Macgregor,*,‡ John E. McGrady,*,§ and Adrian C. Whitwood†
Department of Chemistry, UniVersity of York, Heslington, York YO10 5DD, United Kingdom,
School of Engineering and Physical Sciences, Heriot-Watt UniVersity, Edinburgh EH14 4AS,
United Kingdom, and WestCHEM, Department of Chemistry, UniVersity of Glasgow,
Glasgow G12 8QQ, United Kingdom
Received June 24, 2008; E-mail: S.A.Macgregor@hw.ac.uk; j.mcgrady@chem.gla.ac.uk; rnp1@york.ac.uk
Abstract: A survey of computed mechanisms for C-F bond activation at the 4-position of pentafluoropyridine
by the model zero-valent bis-phosphine complex, [Pt(PH3)(PH2Me)], reveals three quite distinct pathways
leading to square-planar Pt(II) products. Direct oxidative addition leads to cis-[Pt(F)(4-C5NF4)(PH3)(PH2Me)]
via a conventional 3-center transition state. This process competes with two different phosphine-assisted
mechanisms in which C-F activation involves fluorine transfer to a phosphorus center via novel 4-center
transition states. The more accessible of the two phosphine-assisted processes involves concerted transfer
of an alkyl group from phosphorus to the metal to give a platinum(alkyl)(fluorophosphine), trans-[Pt(Me)(4-
C5NF4)(PH3)(PH2F)], analogues of which have been observed experimentally. The second phosphine-
assisted pathway sees fluorine transfer to one of the phosphine ligands with formation of a metastable
metallophosphorane intermediate from which either alkyl or fluorine transfer to the metal is possible. Both
Pt-fluoride and Pt(alkyl)(fluorophosphine) products are therefore accessible via this route. Our calculations
highlight the central role of metallophosphorane species, either as intermediates or transition states, in
aromatic C-F bond activation. In addition, the similar computed barriers for all three processes suggest
that Pt-fluoride species should be accessible. This is confirmed experimentally by the reaction of [Pt(PR3)2]
species (R ) isopropyl (iPr), cyclohexyl (Cy), and cyclopentyl (Cyp)) with 2,3,5-trifluoro-4-(trifluorometh-
yl)pyridine to give cis-[Pt(F){2-C5NHF2(CF3)}(PR3)2]. These species subsequently convert to the trans-
isomers, either thermally or photochemically. The crystal structure of cis-[Pt(F){2-C5NHF2(CF3)}(PiPr3)2]
shows planar coordination at Pt with r(F-Pt) ) 2.029(3) Å and P(1)-Pt-P(2) ) 109.10(3)°. The crystal
structure of trans-[Pt(F){2-C5NHF2(CF3)}(PCyp3)2] shows standard square-planar coordination at Pt with
r(F-Pt) ) 2.040(19) Å.
[Ni(PEt3)2],4 and [Ni(iPr2Im)2] (iPr2Im ) 1,3-di(isopropyl)imi-
dazol-2-ylidene).5 The corresponding reactions of fluorinated
pyridines and pyrimidines at [Ni(PEt3)2] proceed even more
readily, and an extensive chemistry based on these reactions
has been developed, providing routes to a number of hitherto
inaccessible selectively fluorinated species.1b,4,6 Comparable
reactions of fluorinated pyridines and pyrimidines have also been
reported with [Pd(PR3)2] complexes, although the regioselec-
tivity of attack is somewhat different from the corresponding
1. Introduction
The group 10 metals have played a key role in the develop-
ment of the C-F bond activation chemistry of fluoroaromatics
and fluoroheteroaromatics.1,2 Intermolecular C-F oxidative
addition of hexafluorobenzene has been observed at a number
of nickel species, including [Ni(tBu2PCH2CH2PtBu2)],3
† University of York.
‡ Heriot-Watt University.
§ University of Glasgow.
Present Address: Max-Planck-Institut fu¨r Kohlenforschung, Kaiser-
Wilhelm-Platz 1, 45470 Mu¨lheim an der Ruhr, Germany.
(1) (a) Braun, T.; Perutz, R. N. Transition-Metal Mediated C-F Bond
Activation. In ComprehensiVe Organometallic Chemistry III; Crabtree,
R. H.; Mingos, D. M. P.; Elsevier: Amsterdam, 2006; Vol. 1, Chapter
26. (b) Braun, T.; Perutz, R. N. Chem. Commun. 2002, 2749.
(2) (a) Torrens, H. Coord. Chem. ReV. 2005, 249, 1957. (b) Burdeniuc,
J.; Jedlicka, B.; Crabtree, R. H. Chem. Ber. 1997, 130, 145. (c)
Murphy, E. F.; Murugavel, R.; Roesky, H. W. Chem. ReV. 1997, 97,
3425. (d) Kiplinger, J. L.; Richmond, T. G.; Osterberg, C. E. Chem.
ReV. 1994, 94, 373. (e) Jones, W. D. Dalton Trans. 2003, 3991.
(3) Bach, I.; Po¨rschke, K.-R.; Goddard, R.; Kopiske, C.; Kru¨ger, C.;
Rufinska, A.; Seevogel, K. Organometallics 1996, 15, 4959.
(4) (a) Cronin, L.; Higgitt, C. L.; Karch, R.; Perutz, R. N. Organometallics
1997, 16, 4920. (b) Archibald, S. J.; Braun, T.; Gaunt, J. F.; Hobson,
J. E.; Perutz, R. N. J. Chem. Soc., Dalton Trans., 2000, 2013. (c)
Braun, T.; Cronin, L.; Higgitt, C. L.; McGrady, J. E.; Perutz, R. N.;
Reinhold, M. New J. Chem. 2001, 25, 19. (d) Burling, S.; Elliott,
P. I. P.; Jasim, N. A.; Lindup, R. J.; McKenna, J.; Perutz, R. N.;
Archibald, S. J.; Whitwood, A. C. Dalton Trans. 2005, 3686.
(5) (a) Schaub, T.; Radius, U. Chem. Eur. J. 2005, 11, 5024–5030. (b)
Schaub, T.; Backes, M.; Radius, U. J. Am. Chem. Soc. 2006, 128,
15964.
(6) (a) Braun, T.; Foxon, S. P.; Perutz, R. N.; Walton, P. H. Angew. Chem.,
Int. Ed. 1999, 38, 3326. (b) Steffen, A.; Sladek, M. I.; Braun, T.;
Neumann, B.; Stammler, H.-G. Organometallics 2005, 24, 4057.
9
10.1021/ja8046238 CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 15499–15511 15499

A R T I C L E S
Nova et al.
Scheme 1
with C6H5F and other monofluorinated aromatics: the barriers
to concerted oxidative addition were found to be 40 kcal/mol,
reducing to 28 kcal/mol with 4-nitro substitution.15 The
introduction of an ortho carboxylate group can also promote
reaction.16 McGrady and Perutz have also compared the
reactivity of hexafluorobenzene with that of benzene at
[M(H2PCH2CH2PH2)] and [M(PH3)2] (M ) Ni, Pt).17 C-F
oxidative addition in these cases proceeds via a classic
3-centered transition state (A: see below) and is invariably
exothermic.
reactions at nickel.7,8 The facile oxidative addition of C-F bonds
at both nickel and palladium has recently been exploited in
catalytic cross-coupling reactions.9 A more comprehensive
introduction to C-F activation is given in the previous paper
in this issue.10
In contrast to the classic C-F intermolecular oxidative
addition chemistry accessible to nickel and palladium, such
reactions have been almost completely absent from platinum
chemistry (but intramolecular reactions are more common).11
Prior to the present work only a single example of Pt-fluoride
formation by intermolecular reaction has been reported via the
reaction of [Pt(tBu2PCH2PtBu2)] with C6F6.12 Instead, the
reaction of [Pt(PR3)2] species (R ) Cy, iPr) with pentafluoro-
pyridine yields a completely different outcome, in which a
fluorine atom migrates to one of the phosphine ligands,
displacing an alkyl substituent, R, which is transferred to Pt
(Scheme 1).7 The net result is a Pt(alkyl)(fluorophosphine),
trans-[Pt(R)(4-C5NF4)(PR3)(PR2F)], and not the Pt-fluoride
anticipated from conventional oxidative addition. Grushin et al.
have shown that analogous Pt-alkyl products are obtained in
the reaction of [Pt(PR3)2] with hexafluorobenzene.13 The fact
that oxidative addition of a C-F bond at platinum has only
been observed with the highly constrained [Pt(tBu2PCH2PtBu2)]
system with a small bite-angle chelating phosphine makes it
tempting to speculate that formation of Pt(alkyl)(fluorophos-
phine) products is the “normal” outcome of the reaction of
fluoroaromatics with bis-trialkylphosphine complexes of Pt(0).14
Previous computational papers on aromatic C-F activation
at group 10 metal centers concerned the reaction of [Pd(PH3)2]
In the preceding paper in this issue,10 Macgregor and Erhardt
discussed the reaction of hexafluorobenzene with the electron-
rich Ir(I) complex, [IrMe(PEt3)3], to form trans-[Ir(C6F5)-
(PEt3)2(PEt2F)].18 There it was shown that the phosphine ligands
can play an active role in a novel “phosphine-assisted” C-F
activation process which involves a 4-centered transition state
(B). In the present paper we build on these ideas in order to
establish a clear understanding of aromatic C-F bond activation
chemistry at platinum. In the computational section, we
characterize pathways leading to Pt-alkyl and Pt-fluoride
products which feature both 3-centered concerted oxidative
addition and 4-centered phosphine-assisted transition states.
These pathways are very finely balanced, suggesting that subtle
changes to the system may induce a switch from one pathway
to another. We then provide new experimental data in support
of this proposal, and show that platinum complexes with simple
monodentate phosphine ligands, [Pt(PR3)2], can be directed to
form Pt-fluoride products with the highly electron-deficient
fluoropyridine derivative, 2,3,5-trifluoro-4-(trifluoromethyl)py-
ridine, C5NHF3(CF3).
(7) Jasim, N. A.; Perutz, R. N.; Whitwood, A. C.; Braun, T.; Izundu, J.;
Neumann, B.; Rothfeld, S.; Stammler, H. G. Organometallics 2004,
23, 6140.
2. Results
2.1. Computational Studies on the Reactions of [Pt(PH3)-
(PH2Me)] with Pentafluoropyridine. 1. Choice of Model System.
In order to rationalize the formation of Pt(alkyl)(fluorophos-
phine) species, [Pt(R)(ArF)(PR3)(PR2F)], we have investigated
the potential energy surface for the C-F activation reactions
between pentafluoropyridine and the simple bis-phosphine model
complex [Pt(PH3)(PH2Me)]. This choice of model system
requires some comment. The experimental results involve a
diverse range of phosphine ligands, PR3, where R is any of iPr,
Cy or Cyp; the accurate modeling of such systems would
represent a major challenge due to their conformational flex-
(8) Braun, T.; Schorlemer, V.; Neumann, B.; Stammler, H.-G. J. Fluorine
Chem. 2006, 127, 367.
(9) (a) Braun, T.; Perutz, R. N.; Sladek, M. I. Chem. Commun. 2001,
2254. (b) Steffen, A.; Sladek, M. I.; Braun, T.; Neumann, B.;
Stammler, H. G. Organometallics 2005, 24, 4057. (c) Kiso, Y.; Tamao,
K.; Kumada, M. J. Organomet. Chem. 1973, 50, C12–C14. (d) Bo¨hm,
V. P. W.; Gsto¨ttmayr, C. W. K.; Weskamp, T.; Herrmann, W. A.
Angew. Chem., Int. Ed. 2001, 40, 3387. (e) Dankwardt, J. W. J.
Organomet. Chem. 2005, 690, 932. (f) Ackermann, L.; Born, R.; Spatz,
J. H.; Meyer, D. Angew. Chem., Int. Ed. 2005, 44, 7216. (g) Lamm,
K.; Stollenz, M.; Meier, M.; Gorls, H.; Walther, D. J. Organomet.
Chem. 2003, 681, 24. (h) Mongin, F.; Mojovic, L.; Guillamet, B.;
Trecourt, F.; Queguiner, G. J. Org. Chem. 2002, 67, 8991. (i)
Widdowson, D. A.; Wilhelm, R. Chem. Commun. 2003, 578. (j)
Widdowson, D. A.; Wilhelm, R. E. Chem. Commun. 1999, 2211–
2212. (k) Wilhelm, R.; Widdowson, D. A. J. Chem. Soc., Perkin Trans.
1 2000, 3808.
(14) The use of a chelating bis-phosphine ligands would be expected to
promote oxidative addition. See: (a) Kozuch, S.; Amatore, C.; Jutand,
A.; Shaik, S. Organometallics 2005, 24, 2319. (b) Su, M.; Chu, S.
Inorg. Chem. 1998, 37, 3400. (c) Sakaki, S.; Biswas, B.; Sugimoto,
M. J. Chem. Soc., Dalton Trans. 1997, 803.
(10) Erhardt, S.; Macgregor, S. A. J. Am. Chem. Soc. 2008, 130, 15490.
(11) Intramolecular oxidative addition of C-F bonds in Schiff base ligands
bearing fluoroaromatic substituents has been observed in Pt(0) and
Pt(II) complexes, and Pt-catalyzed cross coupling of such C-F bonds
has also been observed. See: (a) Crespo, M.; Granell, J.; Font-Badia,
M.; Solans, X. J. Organomet. Chem. 2004, 689, 3088. (b) Crespo,
M.; Martines, M.; Sales, J. Organometallics 1993, 12, 4297. (c) Wang,
T.; Alfonso, B. J.; Love, J. A. Org. Lett. 2007, 9, 5629.
(15) Jakt, M.; Johannissen, L.; Rzepa, H. S.; Widdowson, D. A.; Wilhelm,
R. J. Chem. Soc., Perkin Trans. 2, 2002, 576.
(16) Bahmanyar, S.; Borer, B. C.; Kim, Y. M.; Kurtz, D. M.; Yu, S. Org.
Lett. 2005, 7, 1011.
(17) Reinhold, M.; McGrady, J. E.; Perutz, R. N. J. Am. Chem. Soc. 2004,
126, 5268.
(12) Hofmann, P.; Unfried, G. Chem. Ber. 1992, 125, 659.
(13) Macgregor, S. A.; Roe, D. C.; Marshall, W. J.; Bloch, K. M.;
Bakhmutov, V. I.; Grushin, V. V. J. Am. Chem. Soc. 2005, 127, 15304.
(18) Blum, O.; Frolow, F.; Milstein, D. J. Chem. Soc., Chem. Commun.
1991, 258.
9
15500 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
Figure 1. Computed structures of the separated reactants, precursor adducts (1 and 2), and alternative products trans-3 and trans-4 with distances in Å. For
trans-4 selected experimental data for trans-[Pt(Cy)(4-C5NF4)(PCy3)(PCy2F)]7 are shown in italics.
ibility. Instead, we have chosen to consider the general factors
controlling C-F activation in [Pt(PR3)2]/C5NF5 systems. The
experimental products require alkyl group transfer from phos-
phine to metal, and so we have retained one phosphine methyl
group in order to model this appropriately.
fluorophosphine ligand. These trends are also seen in the
experimental structure of trans-[Pt(Cy)(4-C5NF4)(PCy3)(PCy2F)],
the key distances of which are included in italics in Figure 1.
The computed energies of trans-3 and trans-4 show that the
formation of both species is favorable compared to that of the
η2-arene precursor, 1. More significantly, the Pt-fluoride
species, trans-4, is computed to be 9.1 kcal/mol more stable
than Pt-alkyl trans-3, and this preference is retained even when
the full systems including the Cy groups are considered.19 This
indicates that the experimental observation of trans-[Pt(Cy)(4-
C5NF4)(PCy3)(PCy2F)]20 must be kinetic rather than thermo-
dynamic in origin. In the following section we therefore assess
the possible mechanisms of C-F activation and the barriers to
reaction, and consider the competition between conventional
oxidative addition and phosphine-assisted C-F activation.
2.2. Geometries of Complexes and Overall Reaction Energet-
ics. As an initial step, we computed the overall energy change
for the formation of the alternative Pt-fluoride and Pt(alkyl)-
(fluorophosphine) products. The reactant and product species
concerned are shown in Figure 1, where all energies are quoted
relative to those of the isolated reactants. Two distinct forms
of the initial adduct between [Pt(PH3)(PH2Me)] and C5NF5 were
located, the more stable of which is an η2-arene complex (1, E
) -8.8 kcal/mol). This exhibits an approximately trigonal-
planar geometry with a P-Pt-P angle of 109° and significant
elongation of the C3-C4 bond interacting with the Pt center
(1.48 Å). A noncovalently bound species (2, E ) -5.1 kcal/
mol) was also found, in which the geometries of the two
reactants are effectively unchanged from the isolated species.
The computed energies indicate that 1 will be the dominant
species available in solution.
The two putative Pt(II) products, trans-[Pt(Me)(4-C5NF4)-
(PH3)(PH2F)] (trans-3) and trans-[Pt(F)(4-C5NF4)(PH3)(PH2-
Me)] (trans-4), both show square-planar geometries, with
trans-3 exhibiting a longer Pt-fluoropyridyl distance (2.09 Å
cf. 2.01 Å in trans-4) reflecting the much stronger trans influence
of an alkyl compared to a fluoride ligand. The other significant
difference is the short Pt-PFH2 distance computed in trans-3,
2.22 Å, ∼0.08 Å shorter than the other Pt-P distances and
indicative of a greater degree of π-back-donation to the
(19) Further test calculations on the full systems, trans-[Pt(Cy)(4-
C5NF4)(PCy3)(PCy2F)] (based on the experimentally determined
structure) and trans-[Pt(F)(4-C5NF4)(PCy3)2] (based on the structure
of its Pd analogue)7 show the preference for the Pt-F species to be
virtually unchanged. This key result was also insensitive to the choice
of functional and the use of larger basis sets including the incorporation
of diffuse functions on fluorine atoms.
(20) We have also considered the alternative regioisomeric products formed
through the C-F activation at the 2- or 3-positions of pentafluoro-
pyridine. For trans-3 the relative energies (kcal/mol) follow the trend,
4-C5NF4 (-24.1) < 3-C5NF4 (-23.6) , 2-C5NF4 (-12.1), while the
same trend is computed for trans-4, 4-C5NF4 (-33.2) < 3-C5NF4 (-
32.5) , 2-C5NF4 (-22.9). This order is consistent with the formation
of trans-[Pt(Cy)(4-C5NF4)(PCy3)(PCy2F)] as the favored product (see
Experimental Section), although the computed difference in energy
of the 4- and 3-isomers is small. See the Supporting Information for
details.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15501

A R T I C L E S
Nova et al.
Figure 2. Computed energy profile (kcal/mol) for phosphine-assisted C-F activation via Pathway 1. Energies are relative to the isolated [Pt(PH3)(PH2Me)]
and pentafluoropyridine reactants, and selected distances are given in Å.
2.3. Formation of Pt(Alkyl)(Fluorophosphine) Species. We
have characterized two distinct phosphine-assisted reaction
profiles, Pathways 1 and 2, for the C-F activation of pentafluoro-
pyridine at the 4-position with [Pt(PH3)(PH2Me)] that are
summarized in Figures 2 and 3, respectively. In both cases, the
reaction must necessarily be initiated by the close approach of
the pentafluoropyridine unit and the metal fragment to form
either 1 or 2. Along Pathway 1 the key C-F activation transition
state was found to link to 2 and not 1, and so the first step
would therefore be dissociation of pentafluoropyridine from 1.
This can readily occur via a low-energy transition state, TS
(1-2) (E ) -1.3 kcal/mol), the barrier for this process being
largely associated with the bending of the P-Pt-P angle.
From 2, a C-F activation transition state, TS(2-trans-3) (E
) +20.8 kcal/mol), was located in which the key C-F bond
elongates to 2.00 Å, the F moves toward the PH2Me group
(P· · ·F ) 2.14 Å) and a very short Pt-C distance of 2.03 Å
develops. Despite this, the Pt · · ·F separation remains long (2.72
Å), and so TS(2-trans-3) corresponds to a 4-centered transition
state in which the C-F bond effectively adds across the
Pt-PH2Me unit. At this point there is no significant movement
of the methyl group toward the Pt center, and TS(2-trans-3)
therefore resembles a metallophosphorane, featuring a formally
anionic phosphoranide {PFH2Me} moiety. Subsequent charac-
terization of this transition state shows that it collapses directly
to the Pt-(alkyl)(fluorophosphine) product, trans-3. Pathway
1 therefore involves a concerted activation of both the C-F
and P-C bonds. The concerted nature of this process reflects
the trans disposition of the {P-Pt-P} moiety, which maintains
a near-linear arrangement throughout the reaction. As a con-
sequence, the phosphoranide moiety develops cis to a vacant
site, and thus the transfer of fluorine onto the PH2Me ligand
induces the concerted release of the methyl group onto Pt.
The alternative phosphine-assisted C-F activation process,
Pathway 2, originates directly from the η2-arene complex, 1
(see Figure 3). From 1, a phosphine-assisted C-F activation
transition state, TS(1-5), was located with a computed relative
energy of +24.5 kcal/mol. TS(1-5) is closely related to TS(2
trans-3), with very similar distances within the 4-membered
Pt-C-F-P ring (Pt-C ) 2.01 Å C · · ·F ) 1.89 Å, P · · ·F )
2.17 Å, Pt · · ·F ) 2.80 Å). However, in contrast to Pathway 1,
the narrow P-Pt-P angle of only 110° in TS(1-5) means that
the developing vacant site is now trans to the {PFH2Me} moiety,
and as a result the concerted transfer of the Me group to the
metal center is blocked. Instead TS(1-5) links to a metallo-
phosphorane intermediate, 5, with an approximately T-shaped
Pt(II) center. Complex 5 is marginally more stable than the
separated reactants (E ) -2.0 kcal/mol) and features a
5-coordinate trigonal-bipyramidal phosphorus center in which
F and Me are in axial positions and the {Pt(PH3)(Ar)} moiety
is equatorial.
From metallophosphorane 5, the formation of the final
Pt(alkyl)(fluorophosphine) product, trans-3, can be effected by
transfer of the Me group from the phosphoranide ligand to the
Pt metal center. A transition state for this process was located
(TS(5-trans-3), E ) +8.3 kcal/mol) in which rotation about
9
15502 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
Figure 3. Computed energy profile (kcal/mol) for phosphine-assisted C-F activation via Pathway 2. From intermediate 5 Me transfer leads to trans-3
(Pathway 2a), and F transfer yields trans-4 (Pathway 2b). Energies are relative to the isolated [Pt(PH3)(PH2Me)] and pentafluoropyridine reactants, and
selected distances are given in Å.
the Pt-PFH2Me bond allows the Me group to approach the
metal center (Pt · · ·Me ) 2.77 Å) while at the same time the
P-Pt-P angle widens to 133° to accommodate the transferring
Me group. Formation of trans-3 via Pathway 2a is therefore a
two-step process in which the initial phosphine-assisted C-F
activation step via TS(1-5) is rate determining. Since TS(1-5)
is 3.7 kcal/mol higher than TS(2-trans-3), Pathway 1 should,
on this basis, be the favored route for the formation of trans-3.
The absolute barrier of 29.6 kcal/mol along Pathway 1 might
be considered rather high, given that the experimental C-F
2.4. Formation of Pt(Fluoride)(Phosphine) Species. In the
preceding discussion we considered only methyl transfer from
5 to form trans-3. Metallophosphorane 5, however, features both
an axial Me and an axial F group on the 5-coordinate phosphorus
center; thus, an analogous process involving F transfer leading
to the Pt-fluoride species, trans-4, should also be possible
(Pathway 2b), and a transition state for F transfer was indeed
located (TS(5-trans-4), E ) +3.6 kcal/mol). As in TS(5-trans-
3) the P-Pt-P angle widens (P-Pt-P ) 112°) to create a
vacant site at the metal, but in this case the rotation of the
Pt-PFH2Me bond is in the opposite sense such that the axial F
approaches the metal center (Pt · · ·F ) 2.50 Å) and ultimately
transfers to form trans-4. In fact, relative to the intermediate 5,
this F transfer process is much more accessible than methyl
transfer via TS(5-trans-3). However, it is important to recall
the high energy of the preceding transition state TS(1-5) that
means that both Pathways 2a and 2b are kinetically less
accessible than Pathway 1, the concerted formation of trans-3.
Nevertheless, the identification of species such as 5 as stable
intermediates on the potential energy surface does suggest that
activation
reaction
to
give
trans-[Pt(Cy)(4-
C5NF4)(PCy3)(PCy2F)] is performed at room temperature.
However, this reflects our choice of model system: with
[Pt(PMe3)2] (which represents the electronic contribution of the
trialkylphosphines more realistically) the barrier reduces to
+24.8 kcal/mol. The reduced barrier with more electron-
donating phosphine substituents is entirely consistent with our
view of phosphine-assisted C-F activation in terms of a
nucleophilic attack mechanism established in the preceding
paper.10,21
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15503

A R T I C L E S
Nova et al.
Figure 4. Computed energy profile for C-F oxidative addition along Pathway 3. Energies are given in kcal/mol relative to the isolated reactants, and
selected distances are in Å.
novel phosphine-assisted routes to the formation of M-fluoride
phosphine complexes may sometimes be accessible. Such a
process could be considered a phosphine-assisted C-F oxidative
addition.
of Pt-(alkyl)(fluorophosphine), trans-3, and of these two
alternatives, Pathway 1, which involves a concerted phosphine-
assisted process with a metallophosphorane transition state, lies
lower. Pathways 2b and 3 yield Pt-fluoride complexes, trans-4
for Pathway 2b and cis-4 for Pathway 3. Of these channels, the
concerted oxidative addition route, Pathway 3, is the more
accessible. Thus, we conclude that the preference for fluoride
or alkyl products in any given system will be determined by a
competition between Pathways 1 and 3. Their barriers differ
for our [Pt(PH3)(PH2Me)] model by only 1.3 kcal/mol.22 In light
of this small difference, it is reasonable to propose that subtle
perturbations to the system, introduced through changing (i) the
fluoroaromatic, (ii) the phosphine, or (iii) the conditions (solvent,
temperature) may direct the reaction down different channels.
In particular, although previously unobserved, conventional
oxidative addition of a C-F bond at a Pt(0) bis-trialkylphos-
phine complex should be possible. In the next section, we report
Finally we consider the conventional route to M-fluoride
complexes via concerted oxidative addition of a C-F bond.
With [Pt(PH3)(PH2Me)] this would be expected to form cis-
[Pt(C5F4N)(F)(PH3)(PH2Me)], cis-4, which could then undergo
cis/trans isomerization to give trans-4 (Pathway 3 in the
following). Starting from the η2-arene species 1 a 3-centered
oxidative addition transition state, TS(1-cis-4), was located at
+19.5 kcal/mol, and product formation is strongly exothermic
(∆E ) -20.8 kcal/mol, see Figure 4). The single imaginary
frequency at the transition state is associated with the C-F
stretching mode, and the C-F bond is significantly elongated
relative to those in free C5F5N (1.55 Å vs 1.35 Å). Moreover,
the P-Pt-P angle of 114° accommodates the developing Pt-C
and Pt-F bonds. This oxidative addition process is closely
related, both structurally and energetically, to that computed
for the activation of C6F6 discussed previously.17 Several
standard pathways could be conceived for subsequent cis/trans
isomerization from cis-5. We shall not consider these here,
although we shall return to the significance of the cis/trans
isomerization below.
(21) The barrier to phosphine-assisted C-F activation by [Pd(PH3)(PH2Me)]
alongPathway1risesto+35.3kcal/mol(relativeto[Pd(PH3)(PH2Me)(η2-
C5NF5)]). This increased barrier is again consistent with a view of
phosphine-assisted C-F activation proceeding via nucleophilic attack
and this process being less accessible with the less electron-rich Pd
center.
(22) Use of the [Pt(PMe3)2] model resulted in a reduction in the barrier
along Pathway 3 to 24.8 kcal/mol. As essentially the same barrier
was computed with this model along Pathway 1, this indicates that
the key conclusion, i.e. that Pathways 1 and 3 are close in energy, is
not affected by the choice of model.
To summarize the computational results thus far, the calcula-
tions have revealed the presence of four quite distinct C-F
activation pathways. Pathways 1 and 2a result in the formation
9
15504 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
Scheme 2
The trans-isomer trans-6(iPr) shows a triplet of doublets with
platinum satellites at δ -314.5 (JPF ) 17.7, JFF ) 9.1, JPtF
)
251 Hz) (Figure 5b). These high-field resonances with large
Pt-F coupling constants are characteristic of fluorine bonded
to the platinum center. Complex cis-6(iPr) also shows a triplet
resonance at δ -57.8 (JFF ) 20.0 Hz) assigned to the CF3 group
of the aromatic ring with coupling to two fluorine nuclei ortho
to the CF3 group. A quartet resonance with platinum satellites
δ -105.6 (JFF ) 21.0 Hz, JPtF ) 169 Hz) was assigned to the
aromatic fluorine ortho to the platinum center, and a quartet
resonance at δ -141.0 (JFF ) 21.0 Hz) assigned to the fluorine
para to the platinum (Figure S2 (Supporting Information (SI)).
(Note that coupling between the aromatic hydrogen and aromatic
fluorine is negligible and there is no resonance in the region δ
∼-85 characteristic of fluorine ortho to nitrogen.) The reso-
nances are consistent with the fluoropyridyl ring bonded to
platinum at the 2-position. The fluoropyridyl resonances of
trans-6(iPr) are similar (Figure S5 (SI)).
The 31P NMR spectrum of cis-6(iPr) shows two resonances,
a doublet of doublets with platinum satellites at δ 18.2 (JPF
)149.7, JPP ) 10, JPtP ) 4010 Hz) and a second doublet of
doublets at δ 30.4 (JPP ) 9.9, JPF ) 33.9, JPtP ) 1852 Hz, see
Figure S3 (SI)). The resonance with the larger P-F and P-Pt
coupling constants is assigned to the phosphorus trans to the
fluorine. The very low trans influence of fluorine results in an
exceptionally large coupling constant for Pt(II).24 The second
resonance is assigned to the phosphorus trans to the fluoropyridyl
group; its value of JPF is small compared to the first due to the
cis orientation of the fluorine. Its Pt-P coupling constant is
also small compared to the first one due to the much larger
trans influence of the fluoropyridyl group. The 31P NMR
spectrum of trans-6(iPr) shows a doublet with platinum
satellites at δ 33.5 (JPF ) 17.7, JPtP ) 2830 Hz, Figure S6 (SI)).
The value of JPtP is close to those of the trans platinum fluoride
new experimental data showing that replacing pentafluoropy-
ridine with 2,3,5-trifluoro-4-(trifluoromethyl)pyridine does just
this, redirecting the reaction from platinum alkyl to platinum
fluoride products.
2.5. Experimental Study of the Reaction of [Pt(PR3)2] with
2,3,5-Trifluoro-4-(trifluoromethyl)pyridine. A hexane solution
(close to saturated) of freshly synthesized [Pt(PiPr3)2] was treated
with ∼2 equiv 2,3,5-trifluoro-4-(trifluoromethyl)pyridine,
C5NHF3(CF3), at room temperature. The reaction mixture was
left to react for 2 days in the glovebox, resulting in conversion
to [Pt(F){2-C5NHF2(CF3)}(PiPr3)2], 6(iPr), giving colorless
crystals in yields up to 58%. The yield could be increased to
82% by leaving for a further week. The reaction was also
conducted on an NMR scale in benzene-d6 as solvent and
monitored regularly by 19F and 31P NMR spectroscopy. Conver-
sion to 6(iPr) was quantitative with no evidence for the
formation of the alternative platinum(alkyl)(fluorophosphine)
complex or any other platinum fluoride complexes. By far the
dominant product was cis-6(iPr) with only traces of trans-
6(iPr). In the dark, cis-6(iPr) proved to be stable for at least a
month in solution at room temperature, but it is converted over
the course of a few days to the trans-isomer by exposure to
sunlight, or more rapidly by photolysis (see Scheme 2). cis/
trans Isomerization can also be achieved by heating in benzene
or THF (no decomposition is observed up to 70 °C).
complexes, trans-[Pt(H)(FHF)(PiPr3)2] JPtP ) 2860 Hz, JPF
)
9 Hz, and trans-[Pt(H)(F)(PCy3)2] JPtP ) 2938 Hz, JPF ) 11
Hz.23
The 1H NMR spectrum for 2,3,5-trifluoro-4-(trifluoromethyl)-
pyridine, run for reference, shows a broad singlet at δ 6.99.
The 1H NMR spectrum of cis-6(iPr) shows a singlet at δ 8.32
assigned to the fluoropyridyl proton and a complicated multiplet
at δ 1.37 assigned to the isopropyl CH3 protons. Two multiplets
at δ 2.23 and δ 2.57 simplify to two septets on 31P decoupling,
characteristic of the isopropyl CH protons (Figure S1 (SI)). The
{1H-31P} correlation spectrum shows that the proton resonance
at δ 2.23 couples to the 31P resonance at δ 18.2, while the
resonance at δ 2.57 couples to the 31P resonance at δ 30.4. The
1H NMR spectrum of trans-6(iPr) shows two quartets for the
methyl protons (δ 1.28 and 1.33) that simplify to doublets on
31P decoupling (Figure S4 (SI)). There is a single multiplet for
the CH protons that simplifies to a septet on 31P decoupling.
These patterns can be understood if there is restricted rotation
about the Pt-P and P-C bonds that ensures that the two methyl
groups of any isopropyl unit remain inequivalent.
The corresponding reactions with [Pt(PCy3)2] or [Pt(PCyp3)2]
are slower, requiring up to 14 days (rate: Cyp > Cy), and yield
colorless crystals of trans-[Pt(F){2-C5NHF2(CF3)}(PCy3)2] trans-
6(Cy), and trans-[Pt(F)(2-C5NHF2CF3)(PCyp3)2] trans-6(Cyp)
in yields up to 83% and 69%, respectively. When the reactions
were monitored daily as described above, a mixture of cis and
trans products was observed in solution, the proportion of trans
increasing with time. Once again, conversion to 6(Cyp) and
6(Cy) was quantitative with no evidence for the formation of
the alternative platinum(alkyl)(fluorophosphine) complex or any
other platinum fluoride complexes. These complexes are much
more readily isolated than the hydride fluoride complex, trans-
[Pt(H)(F)(PCy3)2], reported previously.23
The NMR data summarized in the previous paragraphs secure
the assignments of cis-6(iPr) and trans-6(iPr) as cis- and trans-
isomers of [Pt(F){2-C5NHF2(CF3)}(PiPr3)2]. These new experi-
ments therefore demonstrate unambiguously that the platinum
system can be switched from Pt-alkyl formation to Pt-fluoride
2.6. NMR Spectra of 6. The 19F NMR spectrum for 2,3,5-
trifluoro-4-(trifluoromethyl)pyridine was analyzed for reference,
and relevant data are listed in Table 1. The 19F NMR spectrum
of cis-6(iPr) shows a doublet of doublets with platinum satellites
at δ -253.4 (JPF ) 149.0 and 34.0, JPtF ) 330 Hz) (Figure 5a).
(24) (a) Yahav, A.; Goldberg, I.; Vigalok, A. J. Am. Chem. Soc. 2003,
125, 13634. (b) Yahav, A.; Goldberg, I.; Vigalok, A. Inorg. Chem.
2005, 44, 1547.
(23) Jasim, N. A.; Perutz, R. N. J. Am. Chem. Soc. 2000, 122, 8685.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15505

A R T I C L E S
Nova et al.
Table 1. NMR Data for C5NHF3(CF3) and Complexes cis-6 and trans-6 at 300 K in C6D6, δ (J/Hz)
product
1H
19F
31P{H}
-135.6 (m), F4
C5NHF3CF3
6.99 (broad s)
-129.4 dqd (JFF ) 30, 19, 2.7), F3
-85.9 (dd, JFF ) 30, 28 Hz), F2
-57.9 (dd JFF ) 20.9 and 19.3), CF3
-253.4 (dd, 1F, JPF ) 149.0 and 34.0,
JPtF ) 330), Pt-F
cis-6(iPr)
1.37, m (6H)
18.2 (dd, JPF ) 149.7, JPP ) 10.0, JPtP ) 4010
P trans to F
2.23, m (1H) septet when 31P decoupled,
(J ) 6.8)
30.4 dd (JPF ) 33.9, JPP ) 9.9, JPtP ) 1852),
-141.0 (q, 1F, JFF ) 21.0), F5
P trans to fluoropyridyl
2.57 m (1H) septet when 31P decoupled,
(J ) 6.8)
-105.6 (q, 1F, JFF ) 21.0, JPtF )169), F3
8.32 s (1H)
-57.8 (t, 3F, JFF ) 20.0), CF3
-314.5 (td, 1F, JPF ) 17.7, JFF ) 9.1,
JPtF ) 251), Pt-F
trans-6(iPr)
1.28 (q, 6H, JHH ) JHP ) 8), 1.33 (q, 6H,
33.5 (d, JPF )17.7, JPtP ) 2830)
JHH ) JHP ) 8) (each is doublet
a
when 31P decoupled)
-143.0 (q, 1F, JFF ) 18.0), F5
-101.0 (qd, 1F, JFF ) 18.0 and 8.6,
JPtF ) 292), F3
2.24 (2H) (J ) 7)
8.39 s (1H)
-56.8 (t, 3F, JFF) 18.0), CF3
-252.9 (dd, 1F, JPF ) 150.0 and 32.0,
JPtF ) 338), Pt-F
cis-6(Cy)
1.25-2.1 (m, 66H)
10.1 (dd, JPF ) 148.6, JPP ) 10.1, JPtP ) 4004),
P trans to F
21.0 (dd, JPF ) 32.5, JPP ) 10.2 JPtP ) 1834),
8.52 (s, 1H)
-140.1 (q, 1F, JFF ) 23.0), F5
P trans to fluoropyridyl
-107.0 (q, 1F, JFF ) 20.0, JPtF ) 172), F3
-56.6 (t, 3F, JFF ) 20.0), CF3
trans-6(Cy)
cis-6(Cyp)
1.0-2.5 (m, 66H)
-317.9 (t, 1F, JPF )17.01 JPtF ) 227), Pt-F
25.5 (d JPF )17.3, JPtP ) 2822)
-144.2 (q, 1F, JFF ) 18.0), F5
8.54 (s, 1H)
-101.1 (q, 1F, JFF ) 18.0, JPtF ) 299), F3
-56.5 (t, 3F, (JFF ) 22.1), CF3
-261.3 (dd, 1F, JPF ) 154, JPF ) 36,
JPtF ) 173), Pt-F
1.5-2.0 (m, 54H)
10.1 (dd, JPF ) 154, JPP ) 10.5, JPtP ) 4036),
P trans to F
-143.0 (q, 1F, JFF ) 23.0), F5
8.50 (s, 1H)
20.4 (dd JPF ) 33.36, JPP ) 10.3 JPtP ) 1858),
P trans to fluoropyridyl
-108.3 (q, 1F, JFF ) 18.0, JPtF ) 181), F3
-56.5 (t, 3F, JFF ) 20.0), CF3
trans-6(Cyp)
0.5-2.0 (m, 54H)
8.44 (s, 1H)
-321.8 (t, 1F, JPF ) 17.4 JPtF ) 229), Pt-F
24.8 (d JPF )17.5, JPtF ) 2828)
-146 (q, 1F, JFF ) 19.0), F5
-101.7 (q, 1F, JFF ) 18.0, JPtF ) 305), F3
-60.1 (t, 3F, JFF ) 20.0), CF3
a Proved to arise from two separate resonances by comparing spectra at 500 and at 700 MHz.
group, activation occurs at the 2-position and is also selective
for the C-F over the C-H bond. The NMR spectra of cis-
6(Cy) and cis-6(Cyp) are similar to those of cis-6(iPr), and
those of trans-6(Cy) and trans-6(Cyp) are similar to those of
trans-6(iPr) (Table 1).
2.7. Crystal Structures of cis-6(iPr) and trans-6(Cyp). Crys-
tals of cis-6(iPr) were obtained by slow crystallization of the
reaction mixture in hexane solution and characterized by X-ray
crystallography (Figure 6a, Table 2). The molecular structure
shows a planar geometry at Pt with distortion from standard
angles imposed by the cis disposition of the phosphine ligands
(P(1)-Pt(1)-P(2) ) 109.10(3)°, F(1)-Pt(1)-C(1) ) 82.44(11)°,
F(1)-Pt(1)-P(1) ) 80.11(6)°). The Pt-C(fluoropyridyl) dis-
tance is 2.039(3) Å, while the Pt-F distance is 2.029(2) Å. As
expected, the Pt-P distances are very different, 2.3884(8) Å
for Pt-P(1) and 2.2258(9) Å for Pt-P(2), with the shorter
distance for the phosphorus trans to fluorine. The difference of
0.163(1) Å reflects the very strong trans influence of the
fluoropyridyl ligand and is fully consistent with the large
difference between JPtPA and JPtPB (>2100 Hz). The CF3 group
exhibits rotational disorder and was refined with two different
positions differing by a 60° rotation with occupancies of 85%
and 15%. The maximum deviation from the best plane formed
by C(1), F(1), P(1), P(2) is 0.080(2) Å, and the platinum atom
lies 0.022 Å from this plane. The plane of the fluoropyridyl
ring is rotated by 65.47(12)° relative to the best plane formed
by C(1), F(1), P(1), and P(2). The nearest intermolecular contact
to the fluoride bound to platinum is a hydrogen of a fluoropyridyl
group at 2.44 Å.
Figure 5. 19F NMR spectra (470.4 MHz) in [2H6]benzene showing the
Pt-F region of (a) cis-6(iPr) and (b) trans-6(iPr) (cis- and trans-
[Pt(F){C5NHF2(CF3)}(PiPr3)2]).
formation with a small change of fluoropyridine. With the
4-position blocked by the highly electron-withdrawing CF3
9
15506 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
of trans-6(Cy); details are described in the Supporting Informa-
tion (Figures S7, S8 and Tables S1 and S2).
The differences between the bond lengths of cis-6(iPr) and
trans-6(Cyp) are summarized in Table 2: the Pt-C bond length
is 0.051(4) Å shorter in trans-6(Cyp), while the Pt-F distance
does not change significantly. The Pt-P distances in trans-
6(Cyp) are equal and lie between the two extreme values
observed in cis-6(iPr). We have located only four previous
structures of Pt(II) fluoride complexes, all of which contain
phosphine ligands. Three contain fluoride trans to phosphine:
cis-[PtF2(PPh3)2] (mean r(Pt-F)
[Pt(F){CH(CF3)2}(PPh3)2] (r(Pt-F)
)
)
2.008(2) Å), cis-
2.03(1) Å), and
[PtF(PEt3)3]BF4 (r(Pt-F) ) 2.043(7) Å).24-26 The fourth
contains a fluoride ligand trans to phenyl, trans-[Pt(F)(Ph)-
(PPh3)2] (r(Pt-F) ) 2.117(3) Å).27 It is noteworthy that cis-
[PtF{CH(CF3)2}(PPh3)2] and [PtF(PEt3)3]BF4 show much smaller
differences in Pt-P bond lengths compared to cis-6(iPr),
reiterating the very high trans influence of the unusual {2-
C5NHF2(CF3)} ligand. The distortions from square-planar
geometry of cis-6(iPr) are comparable to those for cis-
[Pt(SiPh2H)(H)(PCy3)2] for which the P(1)-Pt-P(2) angle is
113.55(3)°.
The reactions above provide the first evidence for C-F
oxidative addition of fluoropyridines by zero-valent platinum.
Indeed, the only previous example of intermolecular C-F
oxidative addition at platinum involved a phosphine designed
to lower barriers by constraining the P-Pt-P bite angle to be
much less than 90°.12 The preference for oxidative addition
represents a stark contrast with the chemistry of closely related
substrates: pentafluoropyridine reacts with [Pt(PR3)2] via a
Figure 6. Molecular structures (50% ellipsoids, hydrogen atoms omitted)
of cis-6(iPr) (above) and trans-6(Cyp) (below).
phosphine-assisted
pathway
to
yield
trans-[PtR(4-
C5NF4)(PR2F)(PR3)] (reaction with C6F6 proceeds similarly but
requires a higher temperature),7,13 while 2,3,5,6-tetrafluoro-
pyridine reacts via C-H bond activation to give cis-[PtH(4-
C5NF4)(PR3)2].7,28 Thus, we have three quite distinct classes of
reaction for a series of closely related substrates: (i) phosphine-
assisted C-F bond activation, (ii) C-H oxidative addition, and
now (iii) C-F oxidative addition.
The reasons for the switch to oxidative addition are not
obvious from a simple consideration of structure. The CF3 group
certainly prevents access to the 4-position of the fluoropyridine,
where nucleophilic attack typically occurs.29 However, there is
evidence that the role of the CF3 group is more complicated
and that the strongly electron-withdrawing character of this
substituent may be significant, since no analogous reaction is
seen with 2,3,5,6-tetrafluoro-4-methylpyridine under the same
conditions used above. Solvent polarity may also play a role.
Both the C-H oxidative addition with 2,3,5,6-tetrafluoropyridine
and the C-F oxidative addition with 2,3,5-trifluoro-4-(trifluo-
romethyl)pyridine with [Pt(PR3)2] (R ) iPr, Cy) were performed
in hexane, and for [Pt(PCy3)2] it was confirmed that C-F
oxidative addition also occurs in THF (and at a somewhat faster
rate). In contrast, the phosphine-assisted C-F bond activation
observed with pentafluoropyridine does require the more polar
Table 2. Selected bond lengths (Å) and angles (deg) of cis-6(iPr)
and trans-6(Cyp)
difference (cis-6(iPr) -
trans-6(Cyp))
cis-6(iPr)
length (Å)
trans-6(Cyp)
bond length (Å)
bond
length (Å)
C(1)-Pt(1) 2.039(3)
F(1)-Pt(1) 2.029(2)
P(1)-Pt(1) 2.3884(8) P(1)-Pt(1) 2.3224(8)
P(2)-Pt(1) 2.2258(9) P(2)-Pt(1) 2.3225(7)
C(2)-F(2) 1.350(4)
C(1)-Pt(1) 1.988(3)
+0.051(4)
-0.011(3)
+0.066(1)
-0.097(1)
-0.016(6)
F(1)-Pt(1) 2.0402(19)
C(2)-F(2) 1.366(4)
bond
angle (deg)
bond
angle (deg)
F(1)-Pt(1)-C(1)
F(1)-Pt(1)-P(1)
C(1)-Pt(1)-P(1)
F(1)-Pt(1)-P(2)
C(1)-Pt(1)-P(2)
P(1)-Pt(1)-P(2)
82.44(11)
80.11(6)
161.78(10)
170.55(6)
88.57(10)
109.10(3)
C(1)-Pt(1)-F(1)
C(1)-Pt(1)-P(1)
F(1)-Pt(1)-P(1)
C(1)-Pt(1)-P(2)
F(1)-Pt(1)-P(2)
P(1)-Pt(1)-P(2)
174.27(12)
95.69(9)
84.51(6)
92.11(9)
87.42(6)
171.62(3)
Colorless crystals of complex trans-6(Cyp) were obtained
from a saturated hexane solution of the reaction mixture left in
a glovebox at room temperature to crystallize overnight. The
crystal structure was determined by X-ray diffraction at low
temperature (Figure 6, Table 2). The molecular structure shows
the expected trans disposition of the phosphine ligands with
square-planar coordination at platinum. The Pt-C and Pt-F
bond lengths are 1.988(3) and 2.0402(19) Å, respectively. The
fluoropyridyl ring is rotated by 90.00(13)° relative to the C(1),
P(2), P(2), F(1) best-fit plane. The nearest intermolecular contact
to the fluoride bound to platinum is a hydrogen of a cyclopentyl
group at 2.51 Å. We have also determined the crystal structure
(25) Howard, J.; Woodward, P. J. Chem. Soc., Dalton Trans. 1973, 1840.
(26) Russell, D. R.; Mazid, M. A.; Tucker, P. A. J. Chem. Soc., Dalton
Trans. 1980, 1737.
(27) Nilsson, P.; Plamper, F.; Wendt, O. F. Organometallics 2003, 22, 5235.
(28) Solvent effects may also play a role in selectivity, see below.
(29) (a) Brooke, G. M. J. Fluorine Chem. 1997, 86, 1. (b) Coe, P. L.; Rees,
A. J. J. Fluorine Chem. 2000, 101, 45. (c) Adamson, A. J.; Jondi,
W. J.; Tipping, A. E. J. Fluorine Chem. 1996, 76, 67. (d) Benmansour,
H.; Chambers, R. D.; Hoskin, P. R.; Sandford, G. J. Fluorine Chem.
2001, 112, 133.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15507

A R T I C L E S
Nova et al.
THF. The initial product of C-F oxidative addition is cis-
[Pt(F){2-C5NHF2(CF3)}(PR3)2]], and this is shown to be
thermodynamically unstable with respect to trans-[Pt(F){2-
C5NHF2(CF3)}(PR3)2]. This evidence is consistent with con-
ventional oxidative addition (Pathway 3) and inconsistent with
phosphine-assisted C-F oxidative addition (Pathway 2b) which
should yield the trans-isomer directly.
via R/F exchange in trans-[PtF(4-C5NF4)(PR3)2].8 This was
based on the precedent of Grushin’s F/Ph exchange reaction in
[RhF(PPh3)3];13 however, this possibility may now be ruled out
on the basis of the energetics reported here. A corollary of these
results is that C-F bond formation via reductive elimination
from 4-coordinate Pt(II) of the type [PtF(R)L2] will be prob-
lematic. This is a significant result given the considerable
discussion of this point in the recent literature for related Pd(II)
complexes,30-33 and calculations here show trans-[PdF(4-
C5NF4)(PH3)(PH2Me)] to be 25.8 kcal/mol more stable than free
[Pd(PH3)(PH2Me)] and pentafluoropyridine.
The different outcomes of C-F bond activation at Pt(0)
species are kinetic in origin, and we have shown here that the
barriers to phosphine-assisted C-F activation and C-F oxida-
tive addition are very close in energy. Both processes involve
concerted mechanisms but with very different transition states.
The phosphine-assisted C-F activation involves a 4-center
transition state with a relatively large extension of the C-F
bond, substantial Pt-fluoropyridyl bond formation and transfer
of the released fluoride onto a phosphine that lies cis to the
developing M-fluoropyridyl bond. In contrast, in oxidative
addition the phosphine ligands are merely spectators, and
Pt-fluoride formation involves a 3-center transition state with
less C-F elongation and Pt-C bond formation. The difference
in transition states indicates that the phosphine-assisted mech-
anism may overcome some of the electron-electron repulsion
between fluorine and metal characteristic of C-F oxidative
addition.17,34
3.2. Overview of Phosphine-Assisted C-F Bond Activa-
tion. Three systems demonstrating formation of M-alkyl/
fluorophosphine species from M-phosphine complexes have
now been studied in detail by computational methods. Two of
these processes involve intermolecular reactions with fluoro-
aromatics: the reaction of [Pt(PR3)2] species (R ) iPr, Cy) with
pentafluoropyridine discussed here and that of [IrMe(PR3)3] with
C6F6 described in the preceeding paper.10 In both cases C-F
activation proceeds via very similar transition states that exhibit
a metallophosphorane structure. For the iridium system a
metallophosphorane intermediate is also located, although this
corresponds to a very shallow minimum which reacts further
to [Ir(C6F5)(PR3)2(PR2F)], C2H4, and CH4. Phosphine-assisted
C-F bond activation can be described in terms of the nucleo-
philic attack of an electron-rich metal center on an electron-
deficient fluoroaromatic substrate. This process is therefore
promoted by more nucleophilic metal centers, and the lower
barrier computed for [IrMe(PH3)2(PH2Me)] (13 kcal/mol)
compared to [Pt(PH3)(PH2Me)] (25 kcal/mol) is consistent with
the lower ionization potential computed for the former (6.4 eV
vs 7.4 eV). Phosphine-assisted C-F bond activation proceeds
without a dienyl intermediate characteristic of the classical
Meisenheimer mechanism but is instead characterized by a
single, very late transition state in which the M-fluoropyridyl
bond is almost fully formed. The strength of this bond therefore
determines the selectivity of reaction; in particular, we have
shown for iridium that the activation barrier decreases in size
2.8. Computational Study of the C-F Activation Reaction
with 2,3,5-trifluoro-4-(trifluoromethyl)pyridine. In light of the
new experimental data obtained with 2,3,5-trifluoro-4-(trifluo-
romethyl)pyridine, we have returned to computation to assess
the impact of altering the substrate on the key stationary points
of Pathways 1 and 3. Optimized minima are very similar to
those reported above in the pentafluoropyridine study and the
computed metal-ligand bonds reproduce the trends in the
experimental structures very well (see Supporting Information).
The computed relative energies of the three products, trans-
3(CF3) (E ) -10.3 kcal/mol), cis-4(CF3) (E ) -18.1 kcal/
mol) and trans-4(CF3) (E ) -21.8 kcal/mol) also reproduce
the trends seen previously and compare with relative energies
of -12.1, -18.8, and -22.9 kcal/mol, respectively, for activa-
tion at the 2-position of pentafluoropyridine. Moreover, the
trans-Pt-fluoride product remains significantly more stable than
the cis-isomer, consistent with the cis/trans isomerization
observed experimentally. The key phosphine-assisted C-F
activation transition state along Pathway 1 (E ) +24.1 kcal/
mol) and the oxidative addition transition state along Pathway
3 (E ) +22.0 kcal/mol) also remain very close in energy. This
difference of 2.1 kcal/mol is 0.7 kcal/mol greater than that
computed for the activation of pentafluoropyridine at the
4-position. This marginal relative stabilization of Pathway 3 is
consistent with a shift in favor of oxidative addition for 2,3,5-
trifluoro-4-(trifluoromethyl)pyridine, although the change is
perhaps too small to be significant. The more secure conclusion
is that the delicate balance between these two mechanisms
remains.
3. Discussion
3.1. Competing Bond Activation Processes at Pt(0). The
accumulated experimental evidence on the reactions of bis-
trialkylphosphine Pt(0) species with fluoropyridines indicates
that three processes are in competition. As discussed here,
phosphine-assisted C-F activation at platinum can form Pt(alkyl)-
(fluorophosphine) complexes, while C-F oxidative addition
leads to Pt-fluorides. It has also been shown that C-H oxidative
addition may occur if C-H bonds are present on the pyridine.7
The reactions of [Pt(PR3)2] species with 2,3,5-trifluoro-4-
(trifluoromethyl)pyridine reported here are the first examples
of intermolecular C-F oxidative addition at a platinum complex
of monodentate phosphines and the first examples of selectivity
for C-F over C-H bonds in an intermolecular reaction at Pt.
To date all the computational evidence indicates that
Pt-fluorides are the most favored products thermodynamically.
This is seen in the greater stability of both isomers of [PtF(4-
C5NF4)(PR3)2] over trans-[PtR(4-C5NF4)(PR2F)(PR3)] in the
present study and in the more exothermic reaction computed
for C-F oxidative addition of C6F6 to [Pt(dhpe)] compared to
the C-H oxidative addition of C6H6.17 Thus C-F oxidative
addition invariably yields the most stable products. Moreover,
trans-[PtF(4-C5NF4)(PR3)2] is energetically more stable than cis-
[PtF(4-C5NF4)(PR3)2]. When we first reported the formation of
the [PtR(ArF)(PR3)(PR2F)] species, we postulated its formation
(30) (a) Grushin, V. V. Chem. Eur. J. 2002, 8, 1006. (b) Grushin, V. V.
Organometallics 2000, 19, 1888.
(31) Yandulov, D. V.; Tran, N. T. J. Am. Chem. Soc. 2007, 129, 1342.
(32) Grushin, V. V.; Marshall, W. J. Organometallics 2007, 26, 4997.
(33) Hull, K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc. 2006,
128, 7134.
(34) Bosque, R.; Clot, E.; Fantacci, S.; Maseras, F.; Eisenstein, O.; Perutz,
R. N.; Renkema, K. B.; Caulton, K. G. J. Am. Chem. Soc. 1998, 120,
12634.
9
15508 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
as the number of ortho-fluorine substituents increases. For both
Pt- and Ir-M(alkyl)(fluorophosphine) species are significantly
less stable than the M(fluoride)(phosphine) alternatives, indicat-
that a change to aryl substituents can reduce the stability of the
fluoride complexes markedly.
ing the bond strength relationship: DM-F + DP-R > DM-R
DP-F
+
5. Experimental Section
.
5.1. General procedures. All operations were performed under
an argon atmosphere, either on a high-vacuum line (10-4 mbar)
using modified Schlenk techniques, on standard Schlenk (10-2
mbar) lines or in a glovebox. Solvents for general use (hexane,
benzene, toluene) were of AR grade, dried by distillation over
standard reagents, or with a solvent purification system, and stored
under Ar in ampoules fitted with Young’s PTFE stopcocks.
Deuterated solvents were dried by stirring over potassium, and were
distilled under high vacuum into small ampoules with potassium
mirrors.
The third example of M(alkyl)(fluorophosphine) formation
is the reversible isomerization of [RhF(PPh3)3] to trans-
[RhPh(PPh3)2(PFPh2)], and this differs from the above cases in
that it is an intramolecular process.13 The mechanism involves
F transfer to produce a shallow metallophosphorane intermedi-
ate, followed by Ph transfer to Rh. The reversibility of the
reaction and the small energy difference between reactants and
products imply the bond energy relation: DRh-F + DP-Ph
≈
DRh-Ph + DP-F. This differs from the relationship established
for platinum and iridium above and may reflect the fact that
M-alkyl bonds are generally weaker than M-aryl bonds. So
far R/F isomerization has only been observed in the rhodium
system, and this may reflect the less electron-rich character of
the PPh3 ligand in this case, which makes it more prone to
intramolecular nucleophilic attack by the neighboring fluoride.
For the platinum and iridium systems featuring more electron-
rich trialkylphosphines the calculations predict that, although
R/F isomerization is thermodynamically favored, exchange via
metallophosphorane species may suffer from high barriers (cf.
Figure 3 above and Figure 3 in the preceding paper10). Thus,
for trans-[PtR(4-C5NF4)(PR2F)(PR3)] R/F exchange (R ) iPr,
Cy) has yet to be observed experimentally. Instead these
electron-donating phosphines confer sufficient nucleophilicity
upon the metal center to induce the phosphine-assisted inter-
molecular C-F activation. As shown in the preceding paper10
(and predicted previously13), the transferring fluoride becomes
so highly nucleophilic in the transition state of this process that
it can attack even a relatively electron-rich phosphorus center.
In this context, it appears that the two distinct types of chemistry
involving metallophosphorane species (C-F bond activation and
isomerization) may be mutually exclusive.13
Photochemical reactions, at room temperature, were performed
in glass NMR tubes fitted with Young’s PTFE stopcocks, using a
Philips 125 W medium-pressure mercury vapor lamp with a water
filter (5 cm).
All NMR spectra were recorded on Bruker AMX500 spectrom-
eters in the same type of tube. All 1H NMR spectra were recorded
at 500.2 MHz; chemical shifts are reported in ppm (δ) relative to
tetramethylsilane and are referenced using the chemical shifts of
residual protio solvent resonances (benzene, δ 7.16). The 31P{1H}
NMR spectra were recorded at 202.5 MHz and are referenced to
external H3PO4. 19F NMR spectra were recorded at 470.5 MHz
and referenced to external CFCl3 at δ 0 or internal C6F6 at δ -162.9.
Mass spectra were recorded by the University of York analytical
services either on a VG Auto-Spec mass spectrometer in FAB mode,
or on a Waters GCT instrument fitted with a Linden LIFDI probe.
The value of m/z is listed for 194Pt unless otherwise stated. Infrared
spectra were recorded as KBr disks (prepared in the glovebox) on
a Mattson research series FTIR spectrometer fitted with a KBr beam
splitter. The sample chamber was purged with dry, CO2-free air.
Chemicals were obtained from the following sources: potassium
tetrachloroplatinate from Aldrich (or recovered from platinum
residues by a standard procedure); PiPr3, PCy3, and PCyp3 from
Strem, 2,3,5-trifluoro-4-trifluoromethylpyridine from Fluorochem.
5.2. Synthesis of [Pt(PR3)2]. trans-[PtCl2(PR3)2] (1 g) was
placed in a 50 mL Schlenk flask containing PR3 (0.13 g) and a
magnetic stirring bar and suspended in THF (2 mL). A 0.33 mol
dm-3 THF solution of sodium naphthalenide, freshly prepared from
freshly cut sodium (0.5 g) and naphthalene (2.2 g in 50 mL of
THF), was added dropwise to the suspension. The addition was
halted at the point when the dark green color of the naphthalenide
persisted. After 2 h stirring the solvent was removed under vacuum
and the excess naphthalene removed by sublimation (10-4 mbar).35
The dark-brown solid residue was extracted with hexane in the
glovebox and the extract transferred into a funnel and filtered
through paper to yield a yellow solution of [Pt(PR3)2]. This solution
was either employed directly in the next step or pumped to dryness
before use. The following accounts describe synthesis using solid
trans-[PtCl2(PR3)2].
5.3. Synthesis of cis-[Pt(F){2-C5NHF2(CF3)}(PiPr3)2], cis-
6(iPr). [Pt(PiPr3)2] (0.4 g, 0.78 mmol) was dissolved in hexane
(∼5 mL) in the glovebox, and 2,3,5-trifluoro-4-trifluoromethyl-
pyridine (0.3 g, 1.5 mmol) was added. The solution was left in the
glovebox at room temperature for 2 days; colorless crystals of cis-
6(iPr) formed. The solvent was decanted off, and the crystals were
dried in vacuo. Yield 0.32 g, 0.45 mmol (58%). If the solution
was left for a further week and the hexane was allowed to reduce
in volume by evaporation, the yield could be increased to 82%.
IR, KBr disk, cm-1: 2983 (s), 2964 (m), 2918 (s sh), 1568 m, 1543
(w), 1517 (w), 1472 (s, sh), 1461 (s), 1370 (s), 1332 (s, b), 1258
(m), 1167(s), 1089 (w), 1070 (w),1010 (m), 1002 (w), 910 (w),
4. Conclusions
We have used both computational and experimental methods
to rationalize the diverse reactivity of [Pt(PR3)2] species toward
fluoropyridines. The calculations have shown that a novel
phosphine-assisted C-F activation mechanism is possible with
these electron-rich precursors. The barrier associated with this
process is, however, very similar to that for conventional
oxidative addition and we have shown experimentally that the
balance between these processes is extremely delicate. Our
experiments provide the first examples of conventional oxidative
addition of aromatic C-F bonds at Pt(0) species with mono-
dentate phosphines.
Phosphine-assisted C-F activation is emerging as a novel
mechanism for C-F bond activation where an electron-rich
organometallic attacks an electron-deficient substrate with
transfer of the released fluoride onto a phosphine that lies cis
to the developing M-C bond. The phosphine-assisted pathway
has general implications for the reactivity of metal phosphine
complexes toward attack by fluorocarbons in C-F activation
processes. We have noted that it can provide routes to metal
alkyls with fluorophosphine ligands or even to metal fluoride
complexes. Low-valent MF(PR3) species are often quite stable
thermodynamically and are therefore not prone to reductive
elimination or conversion to metal fluorophosphine alkyls.
However, the reversible isomerization of [RhF(PPh3)3] signals
(35) (a) Yoshida, T.; Otsaka, S. Inorg. Synth. 1979, 19, 105. (b) Immirzi,
A.; Musco, A. Inorg. Chim. Acta 1975, 12, L23.
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15509

A R T I C L E S
Nova et al.
Table 3. Crystal and Structure Refinement Data
892 (m), 801 (w), 773 (w), 717 (w), 669 (s), 653 (w), 599 (m sh),
552 (s), 533 (m), 486 (m) and 447 (m).
complex
cis-6(iPr)
trans-6(Cyp)
C36H55F6N1P2Pt1
872.84
110(2)
monoclinic
P21/c
Mass spectrometry, LIFDI+ (m/z): C24H43F6NP2Pt 715.3 (M+)
100%.
empirical formula
formula weight
temperature (K)
crystal system
space group
C
24H43F6N1P2Pt1
716.62
115(2)
monoclinic
P21/n
Anal. calcd for C24H43F6NP2Pt: C, 40.22; H, 6.05; N, 1.95.
Found: C, 40.35; H, 6.45; N, 1.70.
5.4. Synthesis of trans-[Pt(F){2-C5NHF2(CF3)}(PCyp3)2], trans-
6(Cyp). [Pt(PCyp3)2], (0.2 g, 0.30 mmol) was dissolved in hexane
(∼5 mL) in the glovebox, and 2,3,5-trifluoro-4-trifluoromethyl-
pyridine (0.11 g, 0.55 mmol) was added. The solution was left in
the glovebox at room temperature for 2 weeks. Colorless crystals
of trans-6(Cyp) formed. The solvent was decanted off, and the
product was dried under vacuum. Residual naphthalene was
removed by sublimation at room temperature. Yield 0.22 g, 0.25
mmol (83%).
unit cell dimensions
a (Å)
b (Å)
15.9928(11)
10.8838(8)
16.4567(11)
100.3800(10)
2817.6(3)
4
1.689
5.146
1424
10.5548(6)
19.6695(10)
17.3098(9).
100.8160(10)
3529.8(3)
4
1.642
4.125
1760
c (Å)
ꢀ (deg)
V (Å3)
Z
density (calcd, Mg/m3)
absorption coeff (mm-1
F(000)
)
IR, KBr disk, cm-1: 2964 (s), 2864 (s), 1486 (w), 1448 (m),
1320(w), 1300 (m), 1259 (m), 1232 (w), 1223 (vw), 1116 (s), 1107
(s), 1069 (m), 1053 (m), 1030 (m), 943 (w), 902 (m), 805 (w), 702
(vw), 596 (b w), 531 (m sh), 523 (s b), 510 (m, sh), 491 (w), 478
(m sh), 458 (w) and 444 (w).
crystal size(mm3)
θ range (deg)
reflections collected
independent reflections
0.22 × 0.16 × 0.03
1.63 to 28.32
28466
0.27 × 0.14 × 0.06
1.58 to 30.00.
38910
7007
10154
[R(int) ) 0.0319]
7007/21/329
1.034
[R(int) ) 0.0307]
10154/0/416
1.035
data/restraints/params
Mass spectrometry, FAB+ (m/z): C36H55F6NP2Pt 872.3 (M +
H)+ 7%, 871.3 (M+) 7%, 852.3 (M+ - F) 100%, 669.3 [Pt(PCyp3)2
- H]+ 10%.
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
CCDC number
R1 ) 0.0272,
wR2 ) 0.0610
R1 ) 0.0374,
wR2 ) 0.0644
684870
R1 ) 0.0285,
wR2 ) 0.0673
R1 ) 0.0374,
wR2 ) 0.0711
684871
5.5. Synthesis of trans-[Pt(F){2-C5NHF2(CF3)}(PCy3)2], trans-
6(Cy). [Pt(PCy3)2] (0.2 g, 0.26 mmol) was dissolved in hexane (∼5
mL) in the glovebox, and 2,3,5-trifluoro-4-trifluoromethylpyridine
(0.10 g, 0.50 mmol) was added. The solution was left in the
glovebox at room temperature for 2 weeks. Colorless crystals of
trans-6(Cy) formed. The solvent was decanted off, and the product
was dried under vacuum. Residual naphthalene was removed by
sublimation at room temperature. Yield 0.15 g, 0.16 mmol (59%).
Another experiment was left for 3 weeks and yield was 0.175 g
(69%).
squares methods using SHELXL-97 (Sheldrick, 1997).36 Hydrogen
atoms were placed using a “riding model” and were included in
the refinement at calculated positions. For cis-6(iPr), disorder of
the CF3 group was modeled as two positions differing by a 60°
rotation. Their occupancies were optimized as ∼85% a form and
15% b form. The C-F bond lengths of the CF3 group were
restrained to be the same, and the spacings between the fluorine
atoms were also restrained to be the same. The anisotropic
displacement parameters of “opposite” fluorine atoms were con-
strained to be the same. For trans-6(Cyp), there was evidence of
disorder in some of the cyclopentyl rings. However, attempts to
model this using a two-site model failed, probably because the two
sites are too close to separate. This disorder leads to an apparent
shortening of some of the C-C bond lengths. Crystallographic data
are given in Table 3. Details for trans-6(Cy) are provided in the
Supporting Information.
5.8. Computational Methods. All density functional theory
(DFT) calculations described in this paper were performed using
the Gaussian0337 package and employed the BP8638 functional.
SDD pseudopotentials and the associated basis sets were used to
describe Pt,39 while 6-31G(d,p) basis sets were used for all other
atoms.40 Geometry optimizations were performed without symmetry
constraints, and stationary points were characterized as minima or
transition states by vibrational analysis. Transition states were
further characterized by IRC calculations and subsequent geometry
optimizations to confirm the nature of the minima involved in each
step. All energies include a correction for zero-point energy.
IR, KBr disk, cm-1: 2931 (s), 2870 (s), 1575 (m), 1559 (w sh),
1490 (s), 1356 (s), 1335 (m), 1319 (s), 1275 (m), 1250 (w), 1155
(s), 1094 (m), 1019 (m), 1017 (m), 785 (w), 738 (m), 733 (m sh),
711 (m), 670 (vw), 660 (w), 651 (w sh), 576 (m), 562 (w), 545
(w), 537 (m), 517 (m), 490 (w), 464 (m), 455 (w) and 442 (vw).
Mass spectrometry: LIFDI+, (m/z): C42H67F6NP2Pt 955.4 (M+)
50%.
Anal. calcd for C42H67F6NP2Pt: C, 52.71; H, 7.06; N, 1.46.
Found: C, 51.8; H, 7.9; N, 1.5.
5.6. Monitoring of Reactions by NMR. The reactions of
Pt(PR3)2 with 2,3,5-trifluoro-4-trifluoromethylpyridine in benzene-
d6 were monitored in situ in an NMR tube by 19F and 31P NMR
spectroscopy.
R ) iPr. The reaction was monitored daily for the first few days.
The major product was cis-6(iPr) with traces of trans-6(iPr). The
reaction was complete after 2 days, yielding the cis-isomer with
traces of the trans-isomer. Provided that the solution was kept in
the dark, this solution was stable for at least 1 month (monitoring
every 2 days approx.). There was no trace of platinum(alkyl)-
(fluorophosphine) products at any stage, nor any other platinum
fluoride complexes.
R ) Cyp or Cy. The reaction was monitored in the NMR tube
every 2 days. Both cis and trans products were observed after 2
days. The reaction was complete after about 2 weeks. At this stage
trans-6(Cyp) and trans-6(Cy), respectively, were the only products.
There was no trace of platinum(alkyl)(fluorophosphine) products
at any stage, nor any other platinum fluoride complexes.
5.7. X-ray Cystallography. Diffraction data were collected on
a Bruker Smart Apex diffractometer with Mo KR radiation (λ )
0.710 73 Å) using a SMART CCD camera. Diffractometer control,
data collection, and initial unit cell determination were performed
using “SMART” (v5.625 Bruker-AXS). Frame integration and unit-
cell refinement were carried out with “SAINT+” (v6.22, Bruker
AXS). Absorption corrections were applied using SADABS (v2.03,
Sheldrick). The structure was solved by direct methods using
SHELXS-97 (Sheldrick, 1997) and refined by full-matrix least-
(36) Sheldrick, G. M. SHELXS-97, Program for Structure Solution;
University of Go¨ttingen: Go¨ttingen, Germany, 1997. Sheldrick, G. M.
SHELXL-97: Program for the Refinement of Crystal Structures;
University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(37) Frisch, M. J.; et al. Gaussian 03, ReVision C.02; Gaussian, Inc.:
Wallingford CT, 2004.
(38) (a) Becke, A. D. Phys. ReV. A 1988, 38, 3098. (b) Becke, A. D.
J. Chem. Phys. 1993, 98, 1372. (c) Becke, A. D. J. Chem. Phys. 1993,
98, 5648. (d) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980,
58, 1200. (e) Becke, A. D. J. Chem. Phys. 1986, 84, 4524. (f) Perdew,
J. P. Phys. ReV. B 1986, 33, 8822.
(39) Andrae, D.; Ha¨usserman, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor.
Chim. Acta 1990, 77, 123.
(40) (a) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56,
2257. (b) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213.
9
15510 J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008

Competing C-F Activation Pathways at Platinum(0)
A R T I C L E S
Acknowledgment. A.N.’s stay at the University of York was
funded by the Ministerio de Ciencia y Tecnolog´ıa of Spain (Project
BQU2002-04110-C02-02). The support of the EPSRC for a
postdoctoral fellowship (S.E.) through Grant GR/T28539/01 is
gratefully acknowledged.
for trans-6(Cy), Cartesian coordinates for all the structures
calculated, and complete list of authors for ref 37, X-ray
crystallographic files for cis-6(iPr), trans-6(Cyp), and trans-
6(Cy) (CIF). This material is available free of charge via the
Internet at http://pubs.acs.org.
1
Supporting Information Available: H, 19F, and 31P NMR
spectra of cis-6(iPr) and trans-6(iPr). X-ray crystal structure
JA8046238
9
J. AM. CHEM. SOC. VOL. 130, NO. 46, 2008 15511