Di- And trifluorobenzenes in reactions with Me2em (E = P, N; M = SiMe3, SnMe3, Li) reagents: Evidence for a concerted mechanism of aromatic nucleophilic substitution

Article abstract of DOI:10.1002/ejoc.200900880

Dimethyl(trimethylsilyl)phosphane (Me₃SiPMe₂) has been successfully used for the substitution of fluorine variously in 1,3- (1) or 1,2-difluorobenzene (2) or in 1,3,5- (3) or 1,2,3trifluorobenzene (4) by the Me₂P group at

Full text of DOI:10.1002/ejoc.200900880

FULL PAPER
DOI: 10.1002/ejoc.200900880
Di- and Trifluorobenzenes in Reactions with Me2EM (E = P, N; M = SiMe3,
SnMe3, Li) Reagents: Evidence for a Concerted Mechanism of Aromatic
Nucleophilic Substitution
Leonid I. Goryunov,[a] Joseph Grobe,*[b] Duc Le Van,[b] Vitalij D. Shteingarts,*[a]
Rüdiger Mews,[c] Enno Lork,[c] and Ernst-Ulrich Würthwein*[d]
Keywords: Phosphanes / Polyfluoroarenes / Aromatic substitution / Nucleophilic substitution / Ab initio calculations /
Bis(phosphane)palladium dichloride complexes
Dimethyl(trimethylsilyl)phosphane (Me3SiPMe2) has been
successfully used for the substitution of fluorine variously in
1,3- (1) or 1,2-difluorobenzene (2) or in 1,3,5- (3) or 1,2,3-
MP2 calculations predict that the gas-phase reactions should
each proceed by a concerted mechanism with a single transi-
tion state. These predictions are in good agreement with the
trifluorobenzene (4) by the Me2P group at 150–190 °C either experimental observations, especially because the structural
in benzene solution or without solvent to give 1-(dimethyl-
phosphanyl)-3-fluorobenzene (5) from 1, 1-(dimethylphos-
phanyl)-2-fluorobenzene (6) from 2, 1-(dimethylphosphanyl)-
3,5-difluorobenzene (7) from 3 or a 1:1 mixture of 1-(dimeth-
ylphosphanyl)-2,3-difluorobenzene (8) and 1-(dimethylphos-
phanyl)-2,6-difluorobenzene (9) from 4. The substrate selec-
tivities and regioselectivities exhibited by 4 with Me3SiPMe2
and in competitive reactions between 1 and 2 or 3 and 4 with
Me2PSiMe3, Me2PSnMe3 or Me2PLi indicate relative fluorine
substituent rate factors fo-F Ͼ fm-F, whereas for reactions that
proceed through a two-step SNAr mechanism the opposite
sequence is typical. High-level quantum-chemical DFT and
features of the Meisenheimer adduct are unfavourable for
the SNAr mechanism. This proposal is consistent with the ob-
servation of the opposite sequence (fm-F Ͼ fo-F) for the reac-
tions between the same substrates and MeONa in DMSO/
CH3OH solvent mixtures. The novel phosphanes were char-
acterized by spectroscopic (NMR) and spectrometric (MS) in-
vestigation, preparation of the thiophosphanes ArFPSMe2
10–12, their spectroscopic data and, in the case of 12, by its
X-ray structure. The phosphanes 5–9 were treated with
bis(benzonitrile)dichloropalladium(II) to afford the corre-
sponding bis(phosphane)palladium dichloride complexes
17–21 and 23 in isolated yields of up to 95%.
reactions were considered to be comparable with those be-
tween polyfluoroarenes and oxygen, sulfur or nitrogen nu-
cleophiles conventionally described in terms of the SNAr
mechanism. In a continuation of this approach, this paper
reports on synthetic applications and experimentally ob-
served substrate selectivities and regioselectivities, as well
as on computational modelling of the mechanism, for the
reactions between Me2EM (E = P, N; M = SiMe3, SnMe3,
Li) and the difluorobenzenes 1,3- (1) or 1,2-C6F2H4 (2) or
the trifluorobenzenes 1,3,5- (3) or 1,2,3-C6F3H3 (4) . The
potential to use the new ArFPMe2 derivatives as ligands
for transition metal complexes has been demonstrated by
treatment with bis(benzonitrile)dichloropalladium(II).
Introduction
We recently reported that the compounds Me2PM (M =
SiMe3, SnMe3) are effective, convenient and versatile rea-
gents for the substitution of fluorine atoms in polyfluoroar-
enes C6F5X (X = H, Cl, CF3, PMe2, AsMe2) by a PMe2
group in the position para to X.[1,2] Mechanistically, these
[a] N.N. Vorozhtsov Novosibirsk Institute of Organic Chemistry,
Siberian Division of RAS,
Lavrentiev Ave. 9, 630090 Novosibirsk, Russian Federation
Fax: +7-383-234-4752
E-mail: shtein@nioch.nsc.ru
[b] Institut für Anorganische und Analytische Chemie, Universität
Münster,
Corrensstraße 28/30, 48149 Münster, Germany
Fax: +49-251-83-36012
E-mail: grobe@uni-muenster.de
Results and Discussion
[c] Institut für Anorganische und Physikalische Chemie, Uni-
versität Bremen,
Leobener-Straße, NW2 C, 28334 Bremen, Germany
Fax: +49-251-83-36012
Preparation of (Dimethylphosphanyl)fluorobenzenes
Me2PC6FxH5–x (x = 1, 2)
E-mail: mews@chemie.uni-bremen.de
[d] Organisch-Chemisches Institut, Universität Münster,
Corrensstraße 40, 48149 Münster, Germany
Fax: +49-251-83-39772
Dimethyl(trimethylsilyl)phosphane (Me2PSiMe3) was
successfully used for substitution of fluorine by the Me2P
group in the precursors 1–4. Conditions and results are pre-
sented in Scheme 1. The reactions occur at temperatures be-
E-mail: wurthwe@uni-muenster.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900880.
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1111

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
Scheme 1.
tween 170 and 190 °C either in benzene solution or in the at 170 °C, rather than at 20–70 °C for pentafluorobenzenes
absence of solvent to give (NMR spectroscopic data) the C6F5X.[1,2] However, the di- and trifluorobenzenes are sig-
monosubstituted products 1-(dimethylphosphanyl)-3-fluo- nificantly more reactive than C6H5F, for which tempera-
robenzene (5), 1-(dimethylphosphanyl)-2-fluorobenzene (6) tures above 180 °C are necessary.[1]
or 1-(dimethylphosphanyl)-3,5-difluorobenzene (7) in yields
The relative substrate reactivities with Me2PM (M =
between 40–70%. In the case of 4, a 1:1 mixture of 1-(di- SiMe3, SnMe3) – 2 Ͼ 1 and 4 Ͼ 3 – were determined
methylphosphanyl)-2,3-difluorobenzene (8) and 1-(dimeth- through the competitive reactions shown in Scheme 2 and
ylphosphanyl)-2,6-difluorobenzene (9) had previously been Scheme 3, respectively. The product distributions observed
obtained.[3a]
for the Me2PM (M = SiMe3, SnMe3) reagents in benzene
The (dimethylphosphanyl)difluorobenzenes 8 and 9 had or cyclohexane indicate relative rate factors fo-F Ͼ fm-F in
also previously been prepared by treatment of 1-bromo-2,3- competitions between 2 and 1, and between 4 and 3 and in
and 1-bromo-2,6-difluorobenzene, respectively, with Me2- terms of the regioselectivity in the case of 4. In the more
1
PLi.[3b] The phosphanes 5–9 were characterized by H, 19F polar solvent CH3CN the relative substrate reactivities and
and 31P NMR spectroscopic investigations, by the prepara- regioselectivities for the latter competition experiment de-
tion of the corresponding bis(phosphane)palladium dichlo- crease, but are still in favour of position 2.
ride complexes (ArFMe2P)2PdCl2 17–21 (Scheme 6, below)
and also by the synthesis of the dimeric (phosphane)palla-
dium dichloride complex [(1-Me2P-3-FC6H4)PdCl2]2 (23)
from phosphane 5 (Scheme 7, below). Compounds 5, 8 and
9 were also characterized in the form of their corresponding
thiophosphanes ArFP(S)Me2 (10–12). The structure of 12
was determined by single-crystal X-ray diffraction studies
[selected bond lengths and bond angles are collected in
Table 5 (see below) and experimental and X-ray crystallo-
graphic data in Table 6; these data are similar to those for
dimethyl(2,3,5,6-tetrafluorophenyl)phosphane
Scheme 2.
sulfide[1]].
The spectroscopic and analytic data for compounds 5–7
and 10 are given in the Exp. Section; those for compounds
8, 9, 11 and 12 have already been published.[3]
If Me2PLi is used instead of Me2PM (M = SiMe3,
SnMe3) in C6D6 solution, the fo-F/fm-F ratio remains nearly
the same (Scheme 3). Obviously because of the higher po-
larity of the P–Li bond in relation to the P–Si and P–Sn
bonds, the reaction is in this case already complete within
0.5 h at temperatures between 5 and 20 °C.
Substrate Selectivities and Regioselectivities of Reactions
between Di- or Trifluorobenzenes and Me2PM (M =
SiMe3, SnMe3, Li)
These data establish the general sequence fo-F Ͼ fm-F
,
whereas for the same substrates with MeONa in DMSO/
The difluorobenzenes 1 and 2 and the trifluorobenzenes CH3OH the inverse correlation fm-F Ͼ fo-F was obtained
3 and 4 react with Me2PSiMe3 markedly less readily than (e.g., the ratio 1-MeO-2,3-F2C6H3/1-MeO-2,6-F2C6H3
=
pentafluorobenzenes C6F5X. Compounds 3 and 4 display 4:1 in the reaction of 4).[4] This reaction with an anionic
reasonable reaction rates only at 150 °C and 1 and 2 only nucleophile in a polar solvent thus shows greater activities
1112
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
ently with fluorine partial rate factors fm-F Ͼ fo-F. This is in
contrast with the relative substrate selectivities and regio-
selectivities observed for the phosphanyl-defluorinations
(Schemes 2 and 3), but analogous to the results observed
for the reactions of 3 and 4 or 1 and 2 with sodium meth-
oxide.
The reactions with sodium methoxide (MeONa) had pre-
viously been assumed to proceed through a SNAr mecha-
nism. Correspondingly, the generally observed para Ͻ ortho
Ͻ meta sequence for the fluorine substituent effect on the
rates of nucleophilic fluorine substitution in polyfluoroar-
enes would be explained by Meisenheimer-adduct-like tran-
sition states (TSs) with the ring a π-electron anionic pen-
tadienyl unit and increased electron density in the 1-, 3- and
5-positions.[5] The fluorine substituents ortho and para to
the site of nucleophile attack in the fluoroarene substrate
therefore exert a destabilizing p,π-repulsion effect in the TS,
which is overcompensated for the ortho-fluorine atom by its
electron-withdrawing inductive effect.[6] In these terms, the
substrate reactivity correlations 2 Ͻ 1 and 4 Ͻ 3 could be
expected, as well as the predominant F-1 replacement in 4,
consistently with the ortho-F Ͻ meta-F activating effect. As
shown above, this relation was indeed observed for the ni-
trogen reagent Me2NSiMe3, whereas the opposite meta Ͻ
ortho sequence was found for the phosphorus reagents
Me2PM (M = SiMe3 or SnMe3).
Scheme 3.
of substrates with mutual meta fluorine substituents than
in the isomers with adjacent fluorine atoms, so the data
obtained for the Me2PM reagents apparently suggest a flu-
orine effect different from those seen in SNAr-type reac-
tions.
These results also show that the observed selectivities are
qualitatively independent of the leaving group M in the
Me2PM reagent. However, the effect of the nucleophilic
centre could be very significant. To test this possibility we
studied the reactions of 3 and 4 with (dimethylamino)tri-
methylsilane (Me2NSiMe3).
Reactions of 3 and 4 with Me3SiNMe2
Quantum-Chemical Calculations and Discussion of the
Mechanism
The reactions of 3 and 4 with Me3SiNMe2 were carried
out in sealed ampoules. Conditions and results are pre-
sented in Schemes 4 and 5.
Quantum-chemical calculations were carried out for the
reactions between fluorobenzene or 1,2,3-trifluorobenzene
(4) and Me2ESiMe3 (E = P or N), including the TSs, in
order to elucidate the mechanistic features responsible for
the observed reactivity, with application of the DFT
method B3LYP and the basis sets 6-31+G(d,p) and 6-
311++G(2d,p) for geometry optimizations as implemented
into the program package GAUSSIAN 03.[7] For energy de-
terminations Grimme’s recently developed SCS-MP2
method [SCS-MP2/6-31+G(d,p)//B3LYP/6-31+G(d,p) and
SCS-MP2/6-311++G(d,p)//B3LYP/6-311++G(2d,p)] was
Scheme 4.
Table 1. Relative energies, with respect to the sum of the energies
of the starting materials, calculated by DFT [B3LYP/6-31+G(d,p)//
B3LYP/6-31+G(d,p)] and SCS-MP2 {6-31+D(d,p)//B3LYP/6-
31+G(d,p) [kcalmol–1]}.
Reacting system
Method
TS
E (sum of products)
Fluorobenzene +
Me2PSiMe3
Fluorobenzene +
Me2NSiMe3
4 (position 1) +
Me2PSiMe3
4 (position 2) +
Me2PSiMe3
4 (position 1) +
Me2NSiMe3
DFT
SCS-MP2
DFT
SCS-MP2
DFT
SCS-MP2
DFT
SCS-MP2
DFT
SCS-MP2
DFT
36.55
30.26
47.35
34.30
30.56
23.63
27.88
19.79
43.96
29.26
42.84
28.94
–41.00
–49.13
–31.21
–33.05
–46.74
–55.30
–49.12
–57.98
–35.29
–39.29
–34.69
–38.06
Scheme 5.
Both 3 and 4 react with Me3SiNMe2 less readily than
with the phosphane nucleophiles in Scheme 2, analogously
to pentafluorobenzene and its C6F5X derivatives.[2] The rel-
ative proportions of the products 13–15 were estimated
from the integrals of the 19F NMR signals (see Exp. Sect.).
The substrate 3 with meta fluorine atoms showed a higher
reactivity with Me2NSiMe3 than its isomer 4. In 4 the 1-
and 3-CF units exhibit the highest electrophilicity, consist-
4 (position 2) +
Me2NSiMe3
SCS-MP2
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1113

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
used.[8] All stationary points were subjected to frequency
Thus, according to the calculations, the reactions of these
analyses. Furthermore, the geometries obtained for the neutral nucleophiles with fluorobenzenes involving
a
transition states were used for IRC calculations in order to Me3Si-assisted substitution of the relatively weak leaving
ensure the correct trajectories between starting materials, group fluoride, as studied here, are concerted processes,
transition states and products. The data reported refer to leading to the Me2E-substituted aryl compounds with Me3-
0 K and include DFT zero-point energies. The calculated SiF as the second product.
relative energies, referenced to the sums of the energies (E)
An aromatic nucleophilic substitution mechanism of
such a type, involving a single TS (ANDN mechanism[10]),
had previously been reported by Williams et al. for the
transfer of 1,3,5-triazinyl groups from aryloxide ligands to
an anionic aryloxide or to amine and pyridine nucleo-
philes[10b,10c] and from a pyridinium ligand to amine or hy-
droxide nucleophiles[10d] in aqueous solution. These reac-
tions obey linear log knuc versus nucleophile pKa relation-
ships. By the same reasoning, this mechanism was also sug-
gested for the replacement of a pentafluorophenoxy group
by a fluoride anion in 2,3,5,6-tetrafluoro-4-(pentafluo-
rophenoxy)pyridine.[11] In the opinion of the authors of
ref.[10b] this mechanism is based on the weak basicities both
of the entering and of the leaving ligands. However, the abil-
ity of the transferred group, particularly of 1,3,5-triazinyl,
to allow efficient negative charge dispersion favours the
two-step SNAr mechanism with a discrete Meisenheimer-
adduct-like intermediate.[12–14] This conclusion is supported
by the observation of Meisenheimer complexes formed by
nitroarenes in the gas phase.[15] It is of fundamental impor-
tance that neither high basicity of an entering or leaving
ligand nor efficient resonance stabilization of a negative π-
charge in a ring are present in the compounds of our study.
These properties are unfavourable for the SNAr mechanism,
thus strongly supporting the ANDN mechanism.
of the corresponding substrates and nucleophiles (Erel
=
0.00 kcalmol–1), are presented in Table 1.
The Reaction Course
From the optimizations and from the IRC calculations
for the reactions of fluorobenzene and 4 with the neutral
nucleophiles Me2ESiMe3, similar energy profiles, each with
one transition state and two minima corresponding to
“loose” or van der Waals complexes (vWces) formed by the
starting components (starting vWc) and by the products (fi-
nal vWc), respectively, are obtained in all cases.[9] There is
no indication of Meisenheimer-adduct-like intermediates.
The single TSs correspond to the simultaneous formation
of the C–E bond and cleavage of the C–F and Si–E bonds
(see Table 1, Figures 1 and 2).
In our opinion, based on the calculations, another princi-
pal reason for the observed reaction (i.e., to escape the
SNAr mechanism) is the disfavouring of the Meisenheimer
adduct in the gas phase, due to charge separation. One may
argue that the concerted ANDN mechanism of the studied
reactions should reduce the charge separation in the TS in
relation to the SNAr mechanism.
From this point of view, the mechanism involving a Mei-
senheimer intermediate could seem to be especially favour-
able for reactions with anionic nucleophiles, because in this
case the formation of adducts should be accompanied by
negative charge delocalization. To check this assumption
Figure 1. Schematic energy profile for the reaction between fluoro-
benzene and Me2PSiMe3 [SCS-MP2/6-31+G(d,p)//B3LYP/6-
31+G(d,p)].
Figure 2. TS geometries for the reactions between: a) compound 4 (position 1) and Me2PSiMe3, b) compound 4 (position 2) and Me2P-
SiMe3, c) compound 4 (position 1) and Me2NSiMe3, d) compound 4 (position 2) and Me2NSiMe3, and e) fluorobenzene and Me2NSiMe3,
calculated at the B3LYP/6-31+G(d,p)//B3LYP/6-31+G(d,p) level.
1114
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
the anionic intermediate 16, resulting from attack of the gen substitution intermediate, it has been possible to ob-
methoxide ion at the 2-position of 4, was included in the serve a gem-difluoro Meisenheimer complex formed by pic-
computational study. Surprisingly, in 16 (Figure 3), which ryl fluoride and a fluoride anion by NMR spectroscopy.[16]
corresponds to a shallow local minimum, the C2–F bond is
In general, these results obviously indicate that a Mei-
enormously stretched (1.58 Å), whereas the C2–O bond is senheimer adduct intermediate with no strongly electron-
of normal length (1.43 Å). Moreover, the calculations pre- withdrawing groups and, moreover, formed by a neutral nu-
dict an extreme kinetic instability of this adduct, with only cleophile is inherently disfavoured in the gas phase and
a symbolic energy barrier of 1.3 kcalmol–1 to the escape of probably also in nonpolar media.
the fluoride to form a complex of 2,6-difluoroanisole and
F– (–42.8 kcalmol–1 better in energy than the sum of 4 and
–
OCH3 ), whereas ca. 17 kcalmol–1 would be needed to
stretch the C–OCH3 bond up to a length of 2.4 Å. Thus,
even in this special case the intermediate should be consid-
Reaction Energetics, TS Structures, Substrate Selectivities
and Regioselectivities
ered an elusive gas-phase complex of the substitution prod-
The data in Table 1 demonstrate that the experimentally
uct with a fluoride anion rather than a Meisenheimer ad-
observed relative substrate selectivities and regioselectivities
duct.
qualitatively correspond to the calculated reaction energies
for an ANDN mechanism. Because the reactions were car-
ried out in essentially nonpolar media, the calculations per-
formed for the gas phase seem to be fairly suitable to the
experimental conditions. From the calculated heats of reac-
tion (sums of energies of products) and activation barriers
one can derive that fluorobenzene is less reactive than 4
towards both nucleophiles. Both Me2ESiMe3 compounds
(E = P, N) kinetically prefer the C2 position of 4, by ca. 3
and 1 kcalmol–1, respectively, whereas thermodynamically
Figure 3. Calculated structure of intermediate 16 for the reaction
between compound 4 (position 2) and the methoxide ion [B3LYP/
this position is more attractive only for Me2PSiMe3, be-
6-311++G(2d,p)//B3LYP/6-311++G(2d,p)].
cause for Me2NSiMe3 a slightly smaller exothermicity for
Similarly, the gas-phase calculations for anionic Mei- attack at C2 is calculated.
senheimer intermediates formed from fluorobenzene or 4
From the structural features of the stationary points of
(positions 1 and 2) and the fluoride anion (Table 2) also the calculated reaction path we conclude that the ring obvi-
result in local minima on the energy surface corresponding ously maintains a considerable degree of benzenoid reso-
to intermediates of quite low thermodynamic stability nance.[10] This is clearly seen by comparison of the geomet-
[–7.5 kcalmol–1 in the case of fluorobenzene and –16 to ric parameters of the ring skeleton of fluorobenzene or 4
–18 kcalmol–1 for the complexes of 4 at the SCS-MP2-6- with the TSs in their reactions with Me2ESiMe3 on the one
311++G(2d,p) level, relative to the sums of the starting ma- hand, and with the Meisenheimer adducts of fluorobenzene
terials]. Even in the cases of the fluorobenzene Mei- and 4 with the fluoride anion on the other. Thus, the calcu-
senheimer intermediate and that of 4 it requires only about lated C–C bond lengths in 4 amount to C1–C2 = C2–C3 =
2–4 kcalmol–1 to stretch one of the C–F bonds to 2.4 Å. In 1.3974 Å, C3–C4 = C1–C6 = 1.3897 Å, and C4–C5 = C5–C6
each case the global minimum corresponds to a hydrogen- = 1.3939 Å. The corresponding parameters for the Mei-
bonded complex with the fluoride ion situated at the para- senheimer adducts are presented in Table 2. In the (asym-
hydrogen atom of the aromatic ring (ca. 10–12 kcalmol–1 metric) TS for the reaction between 4 and Me2PSiMe3
lower in energy than the Meisenheimer adduct). One may slightly elongated C1–C2 and C2–C3 bonds (1.4274 and
argue that there are two main reasons for the calculated 1.4280 Å), slightly shortened C3–C4, C6–C1 bonds (1.3827
intermediacy of these special Meisenheimer adducts. The and 1.3801 Å) and again slightly shortened C4–C5, C5–C6
first is the capability to distribute the negative charge over bonds (1.4000 and 1.4011 Å) are found. This TS geometry
the ring system. The second is the enhanced stability of a is intermediate between those of the substrate and the Mei-
CF2 moiety. Thus, whereas a Meisenheimer adduct with a senheimer adducts, thus indicating significant aromatic
halogen atom at the sp3 carbon is in general an elusive halo- character in this TS.
Table 2. Calculated energies and geometrical parameters of the Meisenheimer adducts formed by fluorobenzene and 4 with F– [the atom
numbers correspond to those of the substrates; SCS-MP2/6-311++G(2d,p)//B3LYP/6-311++G(2d,p)].
Substrate
Energy[a]
Interatomic distances [Å]
C1–F
C2–F
C3–F
C1–C2
C2–C3
C3–C4
C4–C5
C5–C6
C1–C6
Fluorobenzene
4 (position 1)
4 (position 2)
–7.47
–17.71
–15.94
1.482
1.468
1.380
1.444
1.443
1.441
1.380
1.369
1.440
1.402
1.394
1.371
1.402
1.400
1.400
1.380
1.380
1.400
1.444
1.446
1.371
1.370
1.468
1.371
1.380
[a] kcalmol–1 relative to sum of substrate + F–.
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1115

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
In our opinion the preservation of substantial aromatic- conjugation between the dimethylamino group and the ben-
ity in the ANDN mechanism TSs provides a good basis for zene system, with 2,6-difluorodimethylaniline obviously be-
a qualitative explanation of the substrate selectivities and ing more sterically hindered than 2,3-difluorodimethylani-
the regioselectivity of 4 in favour of a preferred attack at line.
the 2-position in the reactions under investigation. Firstly,
In view of these arguments, the calculated internuclear
this diminishes the destabilizing p,π-repulsion effect of distances in the TS structures (Table 3, Figure 5) are of par-
ortho-F relative to a Meisenheimer-adduct-like TS. At the ticular interest. In the TSs for the Me2PSiMe3 reactions the
same time, the more electron-deficient the carbon atom of C–F and C–E contacts are less stretched (only by 12–18%)
a reaction centre is as a result of the combined fluorine than those of the reactions with Me2NSiMe3 (by 18–25%)
substituents’ inductive effects, the more stable the respective relative to standard bond lengths (C–F 1.35 Å, C–P 1.84 Å,
TS will be. This is in favour of an enhanced reactivity of 2 C–N 1.47 Å, Ph–N 1.42 Å).[18] This means that the phos-
in relation to 1 and of 4 in relation to 3. The same holds phorus reagent in the TS approaches the benzene skeleton
for position C2 in 4, centred between two CF groups, in more closely and that the C–F bond is less stretched than
relation to position C1. Secondly, because of the formation in the case of the nitrogen reagent. On the other hand, the
of a three-dimensional tetrahedral environment around the Si–F contacts in the TS with the nitrogen nucleophile are
reacting carbon centre in the TS, unfavourable interactions much shorter than in that of the phosphorus one, in the
of the neighbouring CF units in 2 and 4 are strongly dimin- former case being noticeably below the sum of the van der
ished. In the latter case this effect is more significant for C2 Waals radii of Si and F (3.5–3.6 Å). A loose Si–F coordina-
with the two remaining CF units at favourable 1,3-distances tion can therefore be assumed in the TS with Me2NSiMe3.
whereas for attack at C1 the two CF units remain at un- In contrast, in the TS with the phosphorus reagent the Si–
favourable 1,2-distances. These substituent effects are ac- F distance is not smaller than the sum of van der Waals
companied by measurements showing that ortho-difluo- radii, so the Me3SiF molecule is formed at the very end of
robenzene is higher in energy than its meta isomer by ca. the reaction course. Reaction path calculations indicate that
3.5 kcalmol–1 at 25–200 °C[17] and with the results of our the leaving fluoride ion moves along the Me2PSi subunit
calculations (SCS-MP2) revealing that 1,2,3-trifluoroben- with a rather short P–F distance (no minimum, however).
zene is significantly (7.7 kcalmol–1) destabilized in relation
to 1,3,5-trifluorobenzene.
Table 3. Calculated geometrical parameters of TSs formed by fluo-
robenzene and 4 with Me2ESiMe3 [B3LYP/6-31+G(d,p)].
Substrate
Me2ESiMe3
Internuclear distances in the TSs[Å]
Effects of the Natures of Nucleophilic E-Centres
E–C
C–F
E–Si
Si–F
All the reactivity differences discussed for the reactions
of Me2PSiMe3 are less pronounced for those of Me2N-
SiMe3, because here the calculated results predict similar 4 (position 2)
Fluorobenzene
4 (position 1)
2.057
2.083
2.077
1.706
1.686
1.681
1.552 2.305
1.518 2.307
1.524 2.310
1.640 1.910
1.607 1.933
1.621 1.933
3.485
3.486
3.438
2.574
2.566
2.505
Me2PSiMe3
Me2NSiMe3
Fluorobenzene
relative reactivities of fluorobenzene and 4 and a very small
4 (position 1)
reactivity difference between the 1- and 2-positions of 4.
4 (position 2)
Consequently, fluorine substitution at the 1-position should
be statistically predominant, as is observed experimentally
(with some small discrepancies, possibly caused by solvent
These properties can probably be attributed to the differ-
effects, between the calculated and experimentally observed ences in the energies and lengths of the C–N and C–P
replacement rates). Moreover, according to the calculations, bonds. Obviously, because of the larger C–N bond energy,
the formation of the Me2N product 13 is thermodynami- relative to the C–P bond, a smaller degree of C–N bond
cally favoured, whereas 14 is kinetically preferred (Table 1). formation is necessary in order to compensate for the en-
This may be attributed to the fact that the Me2P products ergy expenses of the C–F bond stretching and the out-of-
8 and 9 each prefer a conformation with a Me2P-position plane deviation in the TS of the reaction with Me2NSiMe3,
quasi-perpendicular to the benzene ring (Figure 4). In con- relative to the C–P bond formation in the TS of the reaction
trast, the anilines 13 and14, as would be expected, prefer with Me2PSiMe3. Therefore, the greater C–F bond stretch-
Figure 4. Calculated structures of 8, 9, 13 and 14 [B3LYP/6-3 l+G(d,p)//B3LYP/6-31+G(d,p)].
1116
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
Figure 5. Calculated structures of complexes and TS for the reaction between 4 (position 2) and Me2PLi.
ing and shorter E–Si distance in the nitrogen TS allow a
To check these expectations we performed SCS-MP2/6-
closer Si···F approach and a stronger Si–F linkage than in 31+G(d,p)//B3LYP/6-31+G(d,p) calculations for the reac-
the TS with the phosphorus reagent.
tion between 4 (2-position) and Me2PLi. As anticipated,
Interestingly, for both Me2ESiMe3 reagents the TSs of Me2PLi forms a rather strong Lewis acid/base complex with
the reactions with fluorobenzene correspond to a less pro- 4 (calculated coordination energy ca. 15 kcalmol–1). The re-
nounced synchronism of the process than those of the reac- action is calculated to proceed through a concerted ANDN
tions with 4. In the fluorobenzene TS some imbalance be- mechanism, leading in the gas phase to a strongly bound
tween bond formation and bond fission is observed: the C– LiF–(2,6-difluorophenyl)dimethylphosphane complex (ca.
F contact is stretched to a greater degree (by ca. 18% for E 20 kcalmol–1 lower in energy than the sum of the products).
= P and by ca. 25% for E = N with respect to the standard The calculated structure of the TS is depicted in Figure 5;
bond lengths) in relation to the C–E contacts (by ca. 12% its calculated relative energy (the energy of the starting vWc
for E = P and by ca. 20% for E = N). On the other hand, is taken as reference; Erel = 0.0 kcalmol–1) and geometry
the stretchings of both C–F and C–E bonds in the TS of 4 parameters are presented in Table 4.
are nearly balanced (ca. 12–13% for E = P and ca. 18–20%
for E = N). In terms of the More O’Ferral–Jencks dia-
Table 4. Calculated relative energies and interatomic distances of
grams,[19] all TSs related to 4 are obviously positioned near the complexes and TS for the reaction between compound 4 (posi-
tion 2) and Me2PLi [SCS-MP2/6-311++G(2d,p)//B3LYP/6-
311++G(2d,p)] relative to the sum of the energies of Me2PLi and
4.
the tightness diagonal, the phosphorus TSs being somewhat
more shifted toward the top-left (Meisenheimer adduct)
corner than the nitrogen TSs (presuming very similar curva-
tures of the energy surfaces for all these reactions along the
reaction coordinate). Relative to these TSs, the TSs involv-
ing fluorobenzene deviate somewhat from the tightness di-
agonal toward the top-right (products) corner. This corre-
sponds to somewhat “later” TSs in reactions of fluoroben-
zene than in the case of those of 4, consistently with the
reactivity correlation of these substrates.
Energy
[kcalmol–1]
–30.90
Interatomic distances [Å]
Li–P Li–F1 Li–F2 C2–P C–F1 C–F2
Starting
complex
TS
2.35 2.07 2.10
1.36 1.36
1.43
–24.80
–105.98
2.33 2.04 1.92 2.95
2.56 2.05 1.62 1.86
Final
complex[a]
[a] Sum of the energies of the products: –88.44 kcalmol–1.
Me2PLi as a Model Anionic Nucleophile
These data show that in the TS neither the Li–P nor the
As specified above, the reactions between di- or trifluoro- Li–F1 contact are markedly stretched in relation to the
benzenes and Me2PLi exhibit the same substrate selectivi- starting complex; however, considerable changes are ob-
ties and regioselectivities as their reactions with the Me2PM served in the approach of the Me2P group to C2, leading to
(M = SiMe3, SnMe3) reagents. The same concerted ANDN a C2–P contact of 2.97 Å, much longer than in the TS for
mechanism can therefore also be assumed for this case. Be- the Me2PSiMe3 nucleophile. Dramatic changes accompany
cause of the nonpolar character of the medium, Me2PLi is the generation of the final complex, with practically the
subject to a strong ionic association and can obviously be standard value (1.85 Å) for the C2–P bond and a very short
approximated as reacting in form of a tight ion pair. As a Li–F2 bond (1.61 Å). Together with an extremely low acti-
consequence, the essential prerequisite for the SNAr mecha- vation energy and high exothermicity of the reaction, these
nism – the charge separation in the Meisenheimer adduct results are the characteristic features of an early (reagent-
and in the Meisenheimer adduct-like TS – is not fulfilled, like) TS, so one can conclude that the transformation of the
whereas in the ANDN TS the incipient fluoride anion will benzene ring into the anionic cyclohexadienyl system is also
be stabilized by coordination to the lithium cation.
energetically unfavourable in this case and that the principal
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1117

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
stabilizing interactions in the TS are the Li···P and Li···F
coordinations (Li–F2 1.93 Å is close to the average between
those in the starting and final complexes).
Preparation of Fluorinated (Dimethylphenylphosphane)
palladium Dichlorides
The application of the fluorinated dimethylphenylphos-
phanes ArFMe2P for the preparation of transition metal
complexes has been demonstrated by the syntheses of the
palladium complexes 17–22 in 71–95% isolated yields as
shown in Scheme 6. The reactions were carried out as re-
ported for the ligand Ph2PCH=CH2.[20a]
Figure 7. X-ray structure of 19.
Scheme 6.
The structures of 17–22 were assigned on the basis of
1
their H, 19F and 31P NMR and elemental analysis data
and those of compounds 18, 19, 21 and 22 were also sup-
ported by single-crystal X-ray diffraction studies (see Fig-
ures 6, 7, 8, and 9; selected bond lengths and bond angles
are collected in Table 5 and experimental and X-ray crystal-
lographic data in Table 6). In addition, the molecular geo-
metries in the solid state were determined by far-infrared
(FIR) spectroscopy (Exp. Section). The presence of three
νPd–Cl valence bands in the spectra of 18 and 19, and proba-
bly also those of 17 and 21, indicate the formation of cis
and trans isomer mixtures.[20] The presence of two νPd–Cl
bands in the spectra of 20 and 22 indicates the trans config-
uration. The X-ray analysis of randomly selected crystals
characterized 18 and 22 as their trans isomers and 19 as a
cis isomer. Because the complexes 17–21 each exhibit two
signals in their 31P{1H} NMR spectra (see Exp. Section),
they obviously exist in CDCl3 solution as mixtures of cis
and trans isomers. Compound 22 shows only one singlet
resonance, which can be ascribed to the trans isomer.
Figure 8. X-ray structure of 21.
Figure 9. X-ray structure of 22.
The reaction between trans-[(PhCN)2PdCl2] and 5 (ratio
1:1) was carried out under conditions (Scheme 7) similar to
those of Scheme 6 and led to the binuclear complex 23 in
83% yield. The structure of 23 [di-µ-chloro-dichlorobis{(3-
fluorophenyl)dimethylphosphane}dipalladium(II)] was as-
1
signed on the basis of its H, 19F and 31P NMR, X-ray
(Figure 10) and elemental analysis data.
Conclusions
Dimethyl(trimethylsilyl)phosphane (Me3SiPMe2) has
been shown to act as a convenient and effective reagent for
dimethylphosphanyl-defluorination reactions of the rela-
tively weakly electrophilic di- and trifluorobenzenes. The
Figure 6. X-ray structure of 18.
1118
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
Table 5. Selected bond lengths [pm] and angles [°] in compounds 12, 18, 19 and 21–23.
E–C bonds
12[a]
18[a]
19
21
22[a]
23
Pd–E
Pd(1)–P(1): 232.63(4)
Pd(1)–P(1): 225.9(2)
Pd(1)–P(1): 226.23(9)
Pd(1)–P(1): 230.73(17) Pd(1)–P(1): 221.47(11)
bonds
Pd(1)–P(1)#1: 232.63(4) Pd(1)–P(2): 225.42(19) Pd(1)–P(2): 226.48(10) Pd(1)–Cl(1):
Pd(1)–Cl(1):
229.29(15)
228.99(10)
Pd(1)–Cl(1): 230.60(5)
Pd(1)–Cl(1):
Pd(1)–Cl(1): 234.40(9)
Pd(1)–Cl(2):
235.63(19)
246.70(12)
Pd(1)–Cl(1)#1:
230.60(5)
Pd(1)–Cl(2):
236.68(18)
Pd(1)–Cl(2):
235.45(10)
Pd(1)–Cl(3):
233.36(10)
P–C
Bonds
P(1)–C(1): 180.4(2)
P(1)–C(2): 180.00(18) P(1)–C(2): 181.24(17)
P(1)–C(3): 182.64(16) P(1)–C(3): 181.55(16)
P(1)–C(1): 181.88(17)
P(1)–C(1): 180.8(7)
P(1)–C(2): 182.2(7)
P(1)–C(3): 182.4(7)
C(3)–C(4): 139.8(10)
P(1)–C(1): 181.9(3)
P(1)–C(2): 180.9(3)
P(1)–C(3): 182.9(3)
C(3)–C(4): 138.6(4)
P(1)–C(1): 181.5(6)
P(1)–C(2): 181.1(6)
P(1)–C(3): 183.4(6)
C(3)–C(4): 137.8(8)
P(1)–C(1): 181.0(4)
P(1)–C(2): 180.2(4)
P(1)–C(3): 182.0(3)
C(3)–C(4): 139.3(5)
Ring
C(3)–C(4): 138.9(2)
C(3)–C(4): 139.0(2)
bonds
C(3)–C(8): 139.7(2)
C(4)–C(5): 137.8(2)
C(5)–C(6): 137.8(3)
C(6)–C(7): 139.1(3)
C(7)–C(8): 137.3(3)
P(1)–S(1): 195.76(7)
C(3)–C(8): 140.0(2)
C(4)–C(5): 138.2(2)
C(5)–C(6): 138.7(3)
C(6)–C(7): 138.3(3)
C(7)–C(8): 139.3(2)
C(3)–C(8): 138.9(10)
C(4)–C(5): 137.2(11)
C(5)–C(6): 133.60(13) C(5)–C(6): 137.30(6)
C(6)–C(7): 134.7(13)
C(7)–C(8): 139.5(11)
C(3)–C(8): 138.6(4)
C(4)–C(5): 137.6(5)
C(3)–C(8): 139.2(8)
C(4)–C(5): 138.1(9)
C(5)–C(6): 139.8(8)
C(6)–C(7): 138.2(8)
C(7)–C(8): 137.0(8)
C(6)–C(9): 149.3(9)
C(3)–C(8): 138.8(5)
C(4)–C(5): 137.8(5)
C(5)–C(6): 136.8(3)
C(6)–C(7): 137.2(6)
C(7)–C(8): 138.8(6)
C(6)–C(7): 138.2(6)
C(7)–C(8): 137.3(5)
Other
bonds
C–F
bonds
(av.)
C(4)–F(1): 135.6(2)
C(8)–F(2): 135.1(2)
C(4)–F(1): 135.86(19)
C(5)–F(1): 137.6(10)
C(7)–F(2): 135.3(10)
C(4)–F(1): 134.9(4)
C(8)–F(2): 136.3(4)
C(4)–F(1): 134.0(6)
C(5)–F(1): 135.7(5)
C(5)–F(2): 134.4(6)
C(9)–F(7): 132.1(6)
P(1)#1–Pd(1)–P(1):
180.0
P(1)–Pd(1)–Cl(1):
94.51(6)
Angles at
Pd and P
C(1)–P(1)–C(2):
103.70(11)
C(1)–P(1)–C(3):
104.63(10)
P(1) #1–Pd(1)–
P(1):180.000(14)
P(1)–Pd(1)–Cl(1):
87.015(15)
P(1)–Pd(1)–P(2):
94.12(7)
P(1)–Pd(1)–Cl(1):
172.25(7)
P(1)–Pd(1)–P(2):
100.34(4)
P(1)–Pd(1)–Cl(1):
172.83(3)
P(1)–Pd(1)–Cl(1):
88.19(4)
P(1)–Pd(1)–Cl(3):
93.87(4)
C(2)–P(1)–C(3):
107.70(9)
P(1) #1–Pd(1)–
Cl(1):92.985(15)
Cl(1) #1–Pd(1)–Cl(1):
180.0
P(2)–Pd(1)–Cl(1):
86.35(7)
P(1)–Pd(1)–Cl(2):
90.26(7)
P(2)–Pd(1)–Cl(1):
84.81(4)
P(1)–Pd(1)–Cl(2):
85.50(4)
P(1)#1–Pd(1)–Cl(1):
85.49(6)
Cl(1)#1–Pd(1)–Cl(1):
180.00(7)
Cl(1)–Pd(1)–Cl(3):
177.34(3)
P(1)–Pd(1)–Cl(2):
173.32(3)
C(1)–P(1)–C(2):
102.87(9)
P(2)–Pd(1)–Cl(2):
171.84(7)
P(2)–Pd(1)–Cl(2):
173.42(3)
C(1)–P(1)–C(2):
103.1(3)
Cl(1)–Pd(1)–Cl(2):
91.97(4)
C(1)–P(1)–C(3):
106.11(8)
Cl(1)–Pd(1)–Cl(2):
90.24(7)
Cl(1)–Pd(1)–Cl(2):
89.65(4)
C(1)–P(1)–C(3):
106.8(3)
Cl(2)–Pd(1)–Cl(3):
85.79(4)
C(2)–P(1)–C(3):
104.10(8)
C(1)–P(1)–C(2):
103.7(4)
C(1)–P(1)–C(2):
103.70(17)
C(2)–P(1)–C(3):
106.8(3)
Pd(1)–Cl(2)–Pd(2):
92.24(4)
C(1)–P(1)–Pd(1):
113.95(6)
C(1)–P(1)–C(3):
107.6(4)
C(1)–P(1)–C(3):
99.80(15)
C(1)–P(1)–Pd(1):
114.8(2)
Pd(1)–Cl(3)–Pd(2):
92.97(4)
C(2)–P(1)–Pd(1):
112.69(6)
C(2)–P(1)–C(3):
100.6(4)
C(2)–P(1)–C(3):
108.15(15)
C(2)–P(1)–Pd(1):
114.2(2)
C(1)–P(1)–C(2):
104.19(19)
C(3)–P(1)–Pd(1):
115.85(5)
C(1)–P(1)–Pd(1):
116.5(3)
C(1)–P(1)–Pd(1):
112.74(12)
C(3)–P(1)–Pd(1):
110.47(18)
C(1)–P(1)–C(3):
106.68(17)
C(2)–P(1)–Pd(1):
113.6(3)
C(2)–P(1)–Pd(1):
111.70(12)
C(2)–P(1)–C(3):
107.89(17)
C(3)–P(1)–Pd(1):
113.3(2)
C(3)–P(1)–Pd(1):
119.13(10)
C(1)–P(1)–Pd(1):
114.06(13)
C(2)–P(1)–Pd(1):
115.26(13)
C(3)–P(1)–Pd(1):
108.29(11)
Angles
(int.) ring
C(4)–C(3)–C(8):
113.90(15)
C(4)–C(3)–C(8):
116.75(14)
C(4)–C(3)–C(8):
120.6(7)
C(4)–C(3)–C(8):
114.2(3)
C(4)–C(3)–C(8):
114.9(5)
C(4)–C(3)–C(8):
119.7(3)
C(5)–C(4)–C(3):
124.19(16)
C(5)–C(4)–C(3):
123.67(15)
C(5)–C(4)–C(3):
118.2(8)
C(5)–C(4)–C(3):
123.7(3)
C(5)–C(4)–C(3):
122.3(5)
C(5)–C(4)–C(3):
118.1(3)
C(6)–C(5)–C(4):
118.90(19)
C(6)–C(5)–C(4):
118.08(15)
C(6)–C(5)–C(4):
122.9(9)
C(6)–C(5)–C(4):
118.9(3)
C(6)–C(5)–C(4):
121.9(5)
C(6)–C(5)–C(4):
123.4(4)
C(5)–C(6)–C(7):
120.19(18)
C(5)–C(6)–C(7):
120.45(16)
C(7)–C(6)–C(5):
117.9(8)
C(5)–C(6)–C(7):
120.7(3)
C(5)–C(6)–C(7):
116.1(5)
C(5)–C(6)–C(7):
117.8(3)
C(6)–C(7)–C(8):
118.21(18)
C(6)–C(7)–C(8):
120.25(16)
C(6)–C(7)–C(8):
123.5(9)
C(6)–C(7)–C(8):
117.6(3)
C(6)–C(7)–C(8):
121.0(5)
C(6)–C(7)–C(8):
121.2(4)
C(3)–C(8)–C(7):
124.58(19)
C(3)–C(8)–C(7):
120.79(16)
C(3)–C(8)–C(7):
117.0(8)
C(3)–C(8)–C(7):
124.9(3)
C(3)–C(8)–C(7):
123.8(5)
C(3)–C(8)–C(7):
119.7(4)
[a] Symmetry transformations used to generate equivalent atoms: –x + 1, –y + 1, –z + 1.
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1119

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
Table 6. Experimental and X-ray crystallographic data for compounds 12, 18, 19 and 21–23.
12 18 19 21
Empirical formula C16H17F2NOP2S2 C16H20Cl2F2P2Pd C16H18Cl2F4P2Pd C16H18Cl2F4P2Pd·CHCl3C18H12Cl2F14P2Pd C16H20Cl4F2P2Pd2
22
23
Formula weight
Temperature [K]
Radiation [Å]
Crystal system
Space group
Cell dimensions
a [pm]
b [pm]
c [pm]
α [°]
β [°]
403.37
172(2)
Mo-Kα (0.71073)
monoclinic
P21/n
489.56
173(2)
Mo-Kα (0.71073)
monoclinic
P21/c
525.54
173(2)
Mo-Kα (0.71073)
triclinic
¯
P1
644.91
173(2)
Mo-Kα (0.71073)
monoclinic
P21/c
733.52
173(2)
Mo-Kα (0.71073)
monoclinic
P21/c
666.86
173(2)
Mo-Kα (0.71073)
monoclinic
P21/n
804.50(10)
1132.2(3)
1980.9(3)
90
95.790(10)
90
715.60(10)
963.60(10)
1408.40(10)
90
103.870(10)
90
0.94285(18)
2
1.724
796.8(2)
901.4(10)
1586.3(3)
85.47(10)
80.62(2)
63.86(10)
1.0091(3)
2
956.45(19)
2842.9(6)
942.00(19)
90
111.74(3)
90
2.3791(8)
4
1.801
623.6(3)
748.0(3)
2543.6(15)
90
90.83(3)
90
1.1862(10)
2
2.054
1364.4(3)
1127.2(2)
1571.6(3)
90
114.37(3)
90
2.2016(8)
4
2.012
γ [°]
Volume [nm3]
Z
1.7951(6)
4
d (calcd.) [mg/m3] 1493
1.730
1.375
µ [mm–1]
498
1.450
1.510
1.260
2.280
Crystal size [mm3] 0.70ϫ0.50ϫ0.20 0.60ϫ0.20ϫ0.10 0.90ϫ0.20ϫ0.02 0.40ϫ0.20ϫ0.20
0.80ϫ0.60ϫ0.50
2.84 to 27.50
–1 Յ h Յ 8
–1 Յ k Յ 9
–33ՅlՅ32
3975
0.40ϫ0.20ϫ0.10
2.44 to 26.05
–16 Յ h Յ 16
–13 Յ k Յ 13
–19ՅlՅ19
14946
Theta range [°]
Range in h k l
2.65 to 27.51
–1 Յ h Յ 10
–14 Յ k Յ 14
–25ՅlՅ25
5509
2.59 to 27.50
–9 Յ h Յ 9
–12 Յ k Յ 12
–18ՅlՅ18
8562
2.52 to 24.99
–9 Յ h Յ 1
–10 Յ k Յ 10
–18ՅlՅ18
4454
2.29 to 26.01
–11 Յ h Յ 11
–34 Յ k Յ 34
–11ՅlՅ11
25010
Refl. collected
Independent refl. 4082, Rint = 0.0394 2176, Rint = 0.0292 3551, Rint = 0.0335 4360, Rint = 0.0496
Refl. I Ͼ 2σ(I) 2615
2705, Rint = 0.0403 4280, Rint = 0.0342
Restr. / parameters 0 / 225
R1[a] [IϾ2σ(I)]
0.0573
0.1033
0.1350
0.1524
0.995
0.0177
0.0435
0.0198
0.0444
0.0542
0.1274
0.0873
0.1274
0.0274
0.0758
0.0346
0.0786
0.0515
0.1311
0.0585
0.1396
0.0235
0.0499
0.0375
0.0556
[b]
wR2 [IϾ2σ(I)]
R1[a] (all data)
[b]
wR2 (all data)
2
GooF[c] (Fo )
1.076
0.983
1.080
1.156
0.992
Largest diff. peak 0.529 [eÅ–3]
Largest diff. hole –0.479 [eÅ–3]
0.376 [eÅ–3]
–0.434 [eÅ–3]
0.812 [eÅ–3]
–1.462 [eÅ–3]
0.467 [eÅ–3]
–0.484 [eÅ–3]
1.021 [eÅ–3]
–1.358 [eÅ–3]
0.450 [eÅ–3]
–0.413 [eÅ–3]
[a] R1 = Σ||Fobsd.| – |Fcalcd.||/Σ|Fobsd.|. [b] wR2 = {Σ[w(Fobsd. – Fcalcd.2)2]/Σ[w(Fobsd.2)2]}1/2. [c] GooF = {Σ[w(Fobsd. – Fcalcd.2)/(n – p)}1/2
.
2
2
Meisenheimer SNAr mechanism. Quantum-chemical calcu-
lations, however, predict that the reactions under study
should each occur through a concerted ANDN mechanism
with a single TS. Dimethyl(polyfluorophenyl)phosphanes
have been shown to be suitable ligands for the preparation
of phosphanepalladium dichloride complexes.
Scheme 7.
Experimental Section
General: Because of the air- and moisture-sensitivities of the Me3-
MEMe2 compounds (M = Si, Sn; E = As, P, N) and the resulting
(polyfluoroaryl)phosphanes, as well as their expected toxicities, all
experiments were carried out with use of high-vacuum and Schlenk
techniques. The glassware used was thoroughly heated and evacu-
ated. The NMR spectra were recorded with a Bruker AC 200 (1H:
200.13 MHz, 19F: 188.31 MHz, 31P: 81.02 MHz) or Bruker AV 300
(1H: 300.13 MHz, 19F: 282.39 MHz, 31P: 121.49 MHz) spectrome-
ter. Mass spectra of the new compounds were obtained with a Var-
ian MAT 212 mass spectrometer. The fragmentation data are given
together with the preparative procedure for the individual products.
HRMS were obtained with a GC/MS-System [GCToF, Mikromass;
column HPS 25 m; temperature changes: starting at 40 °C (1 min),
increase 10 °Cmin–1, finishing at 300 °C]. Far-infrared spectra were
recorded in the form of polyethylene pellets with a Perkin–Elmer
Figure 10. X-ray structure of 23.
substrate selectivities and regioselectivities exhibited by
these substrates with Me2PSiMe3, Me2PSnMe3 or Me2PLi
in C6D6 or cyclohexane indicate the fluorine substituent ef-
fects sequence fo-F Ͼ fm-F, which is not typical for nucleo-
philic replacement of fluorine in polyfluoroarenes, conven-
tionally considered to proceed through
a
two-step 1800 FT infrared spectrometer.
1120
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
Structural Investigations: The crystallographic data for compounds
12, 18, 19 and 22 were collected with a SIEMENS P4 dif-
fractometer and the data for 21 and 23 were collected with a STOE
IPDS diffractometer; both were fitted with graphite monochroma-
tors. For the data for 12, 19, 21, 22 and 23 an absorption correction
was applied with the aid of the program DIFABS.[21] All structures
were solved by direct methods and refined based on F2 by use of
the SHELX program package.[22] All hydrogen atoms were placed
on calculated positions and allowed to ride on their corresponding
carbon atoms with isotropic thermal parameters of 1.5 times the
value for Ueq of the bonding atom in case of the methyl protons
and of 1.2 times the value for Ueq of the bonding atom for all other
hydrogen atoms. All non-hydrogen atoms were refined anisotropi-
cally.
sulting solution was diluted with pentane (1 mL), and the precipi-
tated white crystals were filtered off, washed with pentane and
1
dried in vacuo to give 10 (0.07 g, 58%); m.p. 45–46 °C. H NMR
(200 MHz, C6D6): δ = 7.19 (m, 1 H, CH), 7.47 (m, 1 H, CH), 7.54–
2
7.74 (2 H, CH), 1.97 [dm, J(P,H) = 13.2 Hz, 6 H, CH3] ppm. 19F
NMR (188 MHz, C6D6): δ = –112.1 (m) ppm. 31P{1H} NMR
(81 MHz, C6D6): δ = 33.9 [d, 4J(P,F) = 5 Hz] ppm. MS (70 eV, EI):
m/z (%) = 188 (100) [M]+, 173 (85) [M – Me]+, 155 (12) [M – S –
H]+. HRMS: calcd. for C8H10FPS [M] 188.0225; found 188.0238.
Dichlorobis[(3-fluorophenyl)dimethylphosphane]palladium
(17):
Phosphane 5 (0.074 g, 0.48 mmol) and C6D6 (1 mL) were con-
densed into an evacuated reaction vessel that was charged with
bis(benzonitrile)-trans-dichloropalladium(II) (0.091 g, 0.24 mmol)
and cooled with liquid nitrogen. The resulting solution was stirred
at 20 °C for 2 h. The volatile components were distilled off in vacuo
(0.7 mbar), and the residue was dissolved in CH2Cl2 (2 mL) and
filtered. The resultant solution was diluted with EtOH (1 mL), and
the precipitated pale yellow crystals were filtered off, washed with
EtOH and dried in vacuo to give 17 (0.111 g, 95%); m.p. 168–
169 °C (for a small amount of the crystals obtained) and 173–
175 °C (for the bulk of the product). 1H NMR (300 MHz, CDCl3):
δ = 6.6–7.6 (CH; 4 H cis- and 4 H trans-17), 1.78 [d, 2J(P,H) =
11.3 Hz, 6 H, CH3, cis-17], 1.74 [t, 2J(P,H) ≈ 4J(P,H) ≈ 3.6 Hz, 6
H, CH3, trans-17] ppm. 19F NMR (282 MHz, CDCl3): δ = –111.2
(m, cis-17), –112.5 (m, trans-17) ppm. 31P{1H} NMR (11 MHz,
CCDC-737603 (for 12), -737605 (for 18), -737606 (for 19), -737604
(for 21), -737601 (for 22) and -737602 (for 23) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Preparations and Characterization of (Dimethylphosphanyl)fluoro-
benzene Ligands and Selected Palladium Complexes
1-(Dimethylphosphanyl)-3-fluorobenzene (5): Compound 1 (0.34 g,
3.0 mmol) and Me3SiPMe2 (0.101 g, 0.75 mmol) were condensed
into a liquid-nitrogen-cooled evacuated glass ampoule. The reactor
was sealed and kept at 190 °C for 70 h. The volatile components
(Me3SiF and 1) were distilled off under argon while the bath tem-
perature was increased to 130 °C, and 5 was obtained as a colour-
less liquid by vacuum distillation (0.7 mbar) at bath temperatures
increasing from 20 to 35 °C (yield 0.074 g, 63%). 1H NMR
(200 MHz, C6D6): δ = 6.4–7.2 (4 H, CH), 1.06 [dm, 2J(P,H) =
3.1 Hz, 6 H, CH3] ppm. 19F NMR (188 MHz, C6D6): δ = –111.4
(m) ppm. 31P NMR (81 MHz, C6D6): δ = –43.3 (br. s) ppm.
4
2
CDCl3): δ = 6.5 [d, J(P,F) = 3.7 Hz, cis-17, 38%], –3.9 [t, J(P,P)
4
≈ J(P,F) ≈ 2.1 Hz, trans-17, 62%] ppm. IR (polyethylene, νPd–Cl):
ν = 351, 309, 281 cm–1. C H Cl F P Pd (489.61): calcd. C 39.25,
˜
16 20
2 2 2
H 4.12; found C 39.13, H 4.15.
Dichlorobis[(2-fluorophenyl)dimethylphosphane]palladium
(18):
Phosphane 6 (0.078 g, 0.5 mmol) and C6D6 (1 mL) were condensed
into an evacuated reaction vessel, charged with bis(benzonitrile)-
trans-dichloropalladium(II) (0.096 g, 0.25 mmol) and cooled with
liquid nitrogen. The resulting solution was stirred at 20 °C for 1 h.
The volatile components were distilled off in vacuo (0.7 mbar) and
the residue was dissolved in CH2Cl2 (2 mL) and filtered. The result-
ant solution was diluted with EtOH (1 mL) and the precipitated
pale yellow crystals were filtered off, washed with EtOH and dried
in vacuo to give 18 (0.116 g, 95%); m.p. 189–193 °C. 1H NMR
(300 MHz, CDCl3): δ = 6.85–8.0 (CH, 4 H cis- and 4 H trans-18),
1.87 [d, 2J(P,H) = 11.9 Hz, 6 H, CH3, cis-18], 1.83 [t, 2J(P,H) ≈
4J(P,H) ≈ 3.5 Hz, 6 H, CH3, trans-18] ppm. 19F NMR (282 MHz,
CDCl3): δ = –102.8 (br. s, cis-18), –103.7 (br. s, trans-18) ppm.
31P{1H} NMR (11 MHz, CDCl3): δ = 2.1 [d, 3J(P,F) = 13.4 Hz,
1-(Dimethylphosphanyl)-2-fluorobenzene (6): Compound 2 (0.23 g,
2.0 mmol) and Me3SiPMe2 (0.13 g, 1.0 mmol) were condensed into
a liquid-nitrogen-cooled evacuated glass ampoule. The reactor was
sealed and kept at 170 °C for 72 h. The volatile components (Me3-
SiF and 2) were distilled off under argon while the bath tempera-
ture was increased to 130 °C, and 6 was obtained as a colourless
liquid by vacuum distillation (0.7 mbar) at bath temperatures in-
creasing from 20 to 35 °C (yield 0.11 g, 70%). 1H NMR (200 MHz,
C6D6): δ = 6.7–7.5 (4 H, CH), 1.20 [dm, 2J(P,H) = 3.8 Hz, 6 H,
3
CH3] ppm. 19F NMR (188 MHz, C6D6): δ = –102.3 [dm, J(P,F) =
3
34.3 Hz] ppm. 31P NMR (81 MHz, C6D6): δ = –51.4 [dm, J(P,F)
= 34.3 Hz] ppm.
cis-18, 58%], –4.3 [t, 2J(P,P)
≈ ≈ 2.5 Hz, trans-18,
3J(P,F)
42%] ppm. IR (polyethylene, νPd–Cl): ν = 358, 309, 294 cm–1.
˜
1-(Dimethylphosphanyl)-3,5-difluorobenzene (7): Compound
3
C16H20Cl2F2P2Pd (489.61): calcd. C 39.25, H 4.12; found C 39.46,
H 4.11.
(0.26 g, 2.0 mmol) and Me3SiPMe2 (0.13 g, 1.0 mmol) were con-
densed into a liquid-nitrogen-cooled evacuated glass ampoule. The
reactor was sealed and kept at 190 °C for 24 h. The volatile compo-
nents (Me3SiF and 3) were distilled off under argon while the bath
temperature was increased to 130 °C, and 7 was obtained as colour-
less liquid by vacuum distillation (0.7 mbar) at bath temperatures
increasing from 20 to 35 °C (yield 0.07 g, 42%). 1H NMR
(200 MHz, C6D6): δ = 6.8 [tm, 3J(P,H) ≈ 3J(F,H) ≈ 6.1 Hz, 2 H,
Dichlorobis[(3,5-difluorophenyl)dimethylphosphane]palladium (19):
Phosphane 7 (0.063 g, 0.36 mmol) and C6D6 (0.7 mL) were con-
densed into an evacuated reaction vessel, charged with bis(benzon-
itrile)-trans-dichloropalladium(II) (0.069 g, 0.18 mmol) and cooled
with liquid nitrogen. The resulting solution was stirred at 20 °C for
1 h. The volatile components were distilled off in vacuo (0.7 mbar)
and the residue was dissolved in CH2Cl2 (1 mL) and filtered. The
resultant solution was diluted with EtOH (2 mL) and the precipi-
tated pale yellow crystals were filtered off, washed with EtOH and
dried in vacuo to give 19 (0.084 g, 88%); m.p. 209–213 °C. 1H
NMR (300 MHz, CDCl3): δ = 7.15–7.33 (2 H, CH), 6.85 [tt,
3J(F,H) = 8.8, 3J(H,H) = 2.3 Hz, 1 H, CH, trans-19), 7.15–7.32,
3
4
CH], 6.5 [tt, J(F,H) = 8.9, J(H,H) = 2.0 Hz, 1 H, CH], 0.94 [d,
2J(P,H) = 3.3 Hz, 6 H, CH3] ppm. 19F NMR (188 MHz, C6D6): δ
= –111.3 (m) ppm. 31P NMR (81 MHz, C6D6): δ = –40.4 (m) ppm.
1-[Bis(methylthio)phosphanyl]-3-fluorobenzene (10): A mixture of 5
(0.10 g, 0.64 mmol), S8 (0.025 g, 0.78 mmol) and benzene (0.6 mL)
was heated at 90–100 °C for 1 h. After the mixture had cooled to
room temperature, C6D6 was distilled off in vacuo (0.7 mbar) and
the residue was dissolved in CHCl3 (0.2 mL) and filtered. The re-
2
4
6.70–6.82 (3 H, CH, cis-19), 1.76 [t, J(P,H) ≈ J(P,H) ≈ 3.4 Hz, 6
H, CH3, trans-19], 1.82 [d, 2J(P,H) = 11.6 Hz, 6 H, CH3, cis-19]
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1121

J. Grobe, V. D. Shteingarts, E.-U. Würthwein et al.
FULL PAPER
ppm. 19F NMR (282 MHz, CDCl3): δ = –106.8 (m, cis-19), –108.7
(1 mL) were condensed into an evacuated reaction vessel, charged
(m, trans-19) ppm. 31P{1H} NMR (11 MHz, CDCl3): δ = –1.9 (m, with
bis(benzonitrile)-trans-dichloropalladium(II)
(0.086 g,
trans-19, 88%), 8.0 [t, 4J(P,F) = 4.7 Hz, cis-19, 12%] ppm. IR (poly-
0.224 mmol) and cooled with liquid nitrogen. The resulting solu-
tion was stirred at 20 °C for 2 h. The volatile components were
distilled off in vacuo (0.7 mbar) and the residue was dissolved in
CH2Cl2 (2 mL) and filtered. The resultant solution was diluted with
EtOH (1 mL) and the precipitated red crystals were filtered off,
washed with EtOH and dried in vacuo to give 23 (0.062 g, 83%);
m.p. 217–218 °C (became dark at 157–160 °C). 1H NMR
(300 MHz, CDCl3): δ = 7.4–7.6 (3 H, CH), 7.22 (m, 1 H, CH),
ethylene, νPd–Cl): ν = 358, 309, 288 cm–1. C H Cl F P Pd
˜
16 18
2 4 2
(525.59): calcd. C 36.56, H 3.45; found C 36.85, H 3.36.
Dichlorobis[(2,3-difluorophenyl)dimethylphosphane]palladium (20):
Phosphane 8 (0.020 g, 0.115 mmol) and C6D6 (0.5 mL) were con-
densed into an evacuated reaction vessel, charged with bis(benzon-
itrile)-trans-dichloropalladium(II) (0.022 g, 0.057 mmol) and co-
oled with liquid nitrogen. The resulting solution was stirred at
20 °C for 2 h. The volatile components were distilled off in vacuo
(0.7 mbar) and the residue was dissolved in CH2Cl2 (1 mL) and
filtered. The resultant solution was diluted with EtOH (2 mL) and
the precipitated pale yellow crystals were filtered off, washed with
EtOH and dried in vacuo to give 20 (0.026 g, 87%); m.p. 192–
200 °C. 1H NMR (300 MHz, CDCl3): δ = 7.0–7.7 (CH; 3H cis- and
3H trans-20), 1.98 [d, 2J(P,H) = 11.8 Hz, 6 H, CH3, cis-20], 1.89
[dt, 2J(P,H) ≈ 4J(P,H) ≈ 3.7, 5J(F,H) = 1.0 Hz, 6 H, CH3, trans-
20] ppm. 19F NMR (282 MHz, CDCl3): δ = –129.0 (m), –136.4 (m,
2
1.82 [d, J(P,H) = 13.2 Hz, 6 H, CH3] ppm. 19F NMR (282 MHz,
CDCl3): δ = –111.1 (m) ppm. 31P{1H} NMR (121 MHz, CDCl3):
δ = 14.9 [d, 4J(P,F) = 4.2 Hz] ppm. C16H20Cl4F2P2Pd2 (666.93):
calcd. C 28.82, H 3.02; found C 28.98, H 2.92.
Competitive Reactions of 1 and 2 or 3 and 4 with Me3EPMe2 (E =
Si, Sn): Me3EPMe2 (E = Si or Sn, 0.3–0.5 mmol), an equimolar
mixture of di- or trifluorobenzenes (0.6–1.5 mmol) and the solvent
(0.5 mL) were condensed into an evacuated NMR tube cooled with
liquid nitrogen. The tube was sealed under vacuum, allowed to
warm to ambient temperature and then kept at the temperatures
specified in Schemes 2 and 3 while NMR spectra were periodically
recorded.
3
cis-20), –129.6 [dm, J(F,F) = 21.8 Hz], –138.1 (m, trans-20) ppm.
31P{1H} NMR (11 MHz, CDCl3): δ = –3.2 (m, trans-20, 69%), 2.7
3
2
4
[dt, J(P,F) = 12.3, J(P,P) ≈ J(P,F) ≈ 5.24 Hz, cis-20, 31%] ppm.
IR (polyethylene, νPd–Cl): ν = 363, 308 cm–1. C H Cl F P Pd
˜
16 18
2 4 2
Reaction between
4 and Me3SiNMe2: Me3SiNMe2 (0.059 g,
(525.59): calcd. C 36.56, H 3.45; found C 36.34, H 3.28.
0.5 mmol), 4 (0.132 g, 1.0 mmol) and C6H6 (0.3 mL) were con-
densed into an evacuated NMR tube cooled with liquid nitrogen.
The tube was sealed under vacuum, allowed to warm to ambient
temperature and then kept at the temperature specified in Scheme 4
while NMR spectra were periodically recorded. In the 19F NMR
spectrum, the known isomer 14[23] leads to the signal at δF = –122.0
(s, 2 F) ppm; the newly prepared isomer 13 gives δF = –141.0 (m,
1 F), –151.4 (m, 1 F) ppm. After 40 h at 190 °C the reaction mix-
ture contained 85 and 15% of the compounds 13 and 14, respec-
tively. The volatile components (Me3SiF, C6H6 and 4) were distilled
off and a mixture of 13 and 14 was obtained as a colourless liquid
by vacuum distillation (0.7 mbar) at bath temperatures increasing
up to 50 °C (yield 0.057 g, 72%). Two thirds of the mixture were
distilled off to give 13 of 96% purity. 1H NMR (200 MHz, CDCl3):
Dichlorobis[(2,6-difluorophenyl)dimethylphosphane]palladium (21):
Phosphane 9 (0.056 g, 0.32 mmol) and C6D6 (0.5 mL) were con-
densed into an evacuated reaction vessel, charged with bis(benzon-
itrile)-trans-dichloropalladium(II) (0.061 g, 0.16 mmol) and cooled
with liquid nitrogen. The resulting solution was stirred at 20 °C for
2 h. The volatile components were distilled off in vacuo (0.7 mbar)
and the residue was dissolved in CH2Cl2 (1 mL) and filtered. The
resultant solution was diluted with EtOH (2 mL) and the precipi-
tated pale yellow crystals were filtered off, washed with EtOH and
dried in vacuo to give 21 (0.060 g, 71%); m.p. 205–209 °C. 1H
NMR (30 MHz, CDCl3): δ = 7.28–7.44 (CH; 1 H cis- and 1 H
3
3
4
trans-21), 6.90 [tt, J(F,H) = 8.5, J(H,H) ≈ J(P,H) ≈ 1.5 Hz, 2 H,
3
CH, trans-21], 6.75 [dt, 3J(F,H) = 8.6, J(H,H) = 2.8 Hz, 2 H, CH,
cis-21], 1.90 [hept, 5J(F,H) ≈ 6J(H,H) ≈ 2J(P,H) ≈ 4J(P,H) ≈ 1.8 Hz,
6 H, CH3, trans-21, 56%], 2.02 [dt, 2J(P,H) = 11.8, 5J(F,H) =
2.1 Hz, 6 H, CH3, cis-21, 44%] ppm. 19F NMR (282 MHz, CDCl3):
δ = –99.8 (br. s, cis- and trans-21) ppm. 31P{1H} NMR (121 MHz,
CDCl3): δ = –7.2 (m, trans-21), 2.4 (m, cis-21) ppm. IR (polyethyl-
5
δ = 6.83–6.99 (1 H, CH), 6.50–6.74 (2 H, CH), 2.86 [d, J(F,H) =
0.6 Hz, 6 H, CH3] ppm. 19F NMR (188 MHz, CDCl3): δ = –140.3
(m, 1 F), –150.8 [dt, 3J(F,F) = 18.8, 4J(F,H) = 9.3 Hz, 1 F] ppm.
HRMS: calcd. for C8H9NF2 [M] 157.07030; found 157.07045.
ene, νPd–Cl): ν = 366, 305, 292 cm–1. C H Cl F P Pd (525.59):
˜
16 18
2 4 2
Competitive Reactions of 3 and 4 with Me3SiNMe2: Me3SiNMe2
(0.059 g, 0.5 mmol), 3 (0.132 g, 1.0 mmol), 4 (0.132 g, 1.0 mmol)
and C6H6 (0.5 mL) were condensed into an evacuated NMR tube
cooled with liquid nitrogen. The ampoule was sealed under vac-
uum, allowed to warm to ambient temperature and then kept at
the temperature specified in Scheme 5 while NMR spectra were
periodically recorded. In the 19F NMR spectrum the known isomer
15[24] is characterized by the signal at δF = –112.7 (s) ppm.
calcd. C 36.56, H 3.45; found C 36.54, H 3.44.
Dichlorobis{dimethyl[2,3,4,5-tetrafluoro-4-(trifluoromethyl)phen-
yl]phosphane}palladium (22): Dimethyl[2,3,4,5-tetrafluoro-4-(tri-
fluoromethyl)phenyl]phosphane (0.443 g, 1.56 mmol) and C6H6
(5 mL) were condensed into an evacuated reaction vessel, charged
with bis(benzonitrile)-trans-dichloropalladium(II) (0.295 g,
0.77 mmol) and cooled with liquid nitrogen. The resulting solution
was stirred at 20 °C for 0.5 h, filtered and concentrated in vacuo to
a volume of 2 mL. The resultant solution was diluted with pentane
(7 mL) and the precipitated pale yellow crystals were filtered off,
washed with pentane and dried in vacuo to give 22 (0.500 g, 88%);
Competitive Reactions of 3 and 4 with Me2PLi: Me2PH (0.12 g,
2.0 mmol) was condensed into an evacuated reaction vessel,
charged with 0.5 mL of a 2.7 molar solution of nBuLi (1.35 mmol)
in heptane and cooled with liquid nitrogen. The mixture was al-
lowed to warmed up to ambient temperature and stirred at 20 °C
for 1 h. The volatile components were distilled off in vacuo, the
remaining white Me2PLi was cooled with liquid nitrogen, and an
equimolar mixture of the trifluorobenzenes 3 and 4 (2.7 mmol) and
C6D6 (0.5 mL) were condensed into the reaction vessel. The mix-
ture was allowed to warm to ambient temperature, stirred for 0.5–
2 h and analysed by 31P and 19F NMR spectroscopy and CMS (see
Scheme 4).
1
m.p. 154–157 °C. H NMR (200 MHz, CDCl3): δ = 1.98 (m) ppm.
3
19F NMR (188 MHz, CDCl3): δ = –57.9 [t, J(F,F) = 17.3 Hz, 3 F,
CF3], –128.1 (m, 2 F, CF), –139.9 (m, 2 F, CF) ppm. 31P{1H} NMR
(81 MHz, CDCl3): δ = 2.4 (s) ppm. IR (polyethylene, νPd–Cl): ν =
˜
357, 307 cm–1. C18H12Cl2F14P2Pd (733.55): calcd. C 29.47, H 1.65;
found C 29.74, H 1.57.
Di-µ-chloro{dichlorobis[(3-fluorophenyl)dimethylphosphane]}-
dipalladium(II) (23): Phosphane 5 (0.035 g, 0.224 mmol) and C6D6
1122
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1111–1123

Evidence for a Concerted Aromatic Nucleophilic Substitution Mechanism
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W.
Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision C.01,
Gaussian, Inc., Wallingford CT, 2004.
Supporting Information (see also the footnote on the first page of
this article): Quantum chemical results (calculated energies, struc-
tures and lowest frequencies for transition states)
[8] S. Grimme, J. Chem. Phys. 2003, 118, 9095.
[9] The natures of these van der Waals complexes were further
investigated by optimizations with Grimme’s recently devel-
oped B97-D DFT functional (S. Grimme, J. Comp. Chem.
2006, 27, 1787) with application of the basis set def2-TZVP as
implemented in TURBOMOLE 6.0 (TURBOMOLE: R.
Ahlrichs, M. Bär, H.-P. Baron, R, Bauernschmitt, S. Böcker,
M. Ehring, K. Eichkorn, S. Elliot, F. Furche, F. Haase, M.
Häser, H. Horn, C. Huber, U. Huniar, M. Kattannek, C.
Kölmel, M. Kollwitz, K. May, C. Ochsenfeld, H. Ohm, A.
Schäfer, U. Schneider, O. Treutler, M. von Arnim, F. Weigend,
P. Weis, H. Weiss, University of Karlsruhe, 2005; see also:
http://wwu.turbomole.com) The results obtained are in good
agreement with the SCS-MP2 values reported in the text, un-
derlining the excellent suitability of both the SCS-MP2 the and
DFT-D approach for investigations of weakly bound com-
plexes.
[10] a) A. Hunter, M. Renfrew, J. A. Taylor, J. M. J. Whitmore, A.
Williams, J. Chem. Soc. Perkin Trans. 2 1993, 1703–1704; b) A.
Hunter, M. Renfrew, D. Rettura, J. A. Taylor, J. M. J. Whit-
more, A. Williams, J. Am. Chem. Soc. 1995, 117, 5484–5491;
c) J. Shakes, C. Raymond, D. Rettura, A. Williams, J. Chem.
Soc. Perkin Trans. 2 1996, 1553–1557; d) N. R. Cullum, D.
Rettura, J. M. J. Whitmore, A. Williams, J. Chem. Soc. Perkin
Trans. 2 1996, 1559–1563.
Acknowledgments
The authors are grateful to the Stiftung Volkswagenwerk for finan-
cial support of the Cooperation Project between the Institutes of
Inorganic and Organic Chemistry of the University of Münster
(Germany) and the N. N. Vorozhtsov Institute of Organic Chemis-
try, Siberian Division of RAS, Novosibirsk (Russia), which laid
the basis for the work described here. Financial support from the
Deutsche Forschungsgemeinschaft (DFG) (six-months stipend to
L. I. G. for carrying out experimental work in Münster and Bremen
and a three-months stipend to V. D. S., together with expenses for
chemicals and laboratory equipment) is especially gratefully ac-
knowledged. Last but not least, special thanks are expressed to Mr.
Peter Brockmann for valuable technical assistance.
[1] L. I. Goryunov, J. Grobe, V. D. Shteingarts, B. Krebs, A. Linde-
mann, E.-U. Würthwein, C. Mück-Lichtenfeld, Chem. Eur. J.
2000, 6, 4612–4622.
[2] L. I. Goryunov, V. D. Shteingarts, J. Grobe, B. Krebs, M. U.
Triller, Z. Anorg. Allg. Chem. 2002, 628, 1770–1779.
[3] a) J. Grobe, L. I. Goryunov, V. D. Shteingarts, Russ. J. Org.
Chem. 2005, 41, 1744–1745; b) L. I. Goryunov, J. Grobe, V. D.
Shteingarts, R. Mews, E.-U. Würthwein, Russian, J. Org.
Chem. (in press).
[11] V. M. Vlasov, V. V. Aksenov, P. P. Rodionov, I. V. Beregovaya,
L. N. Shchegoleva, Russ. J. Org. Chem. 2002, 38, 115–125.
[12] M. Renfrew, J. A. Taylor, J. M. J. Whitmore, A. Williams, J.
Chem. Soc. Perkin Trans. 2 1994, 2383–2384.
[13] N. R. Cullum, A. Hunter, M. Renfrew, D. Rettura, J. A. Taylor,
J. M. J. Whitmore, A. Williams, J. Am. Chem. Soc. 1995, 117,
9200–9205.
[4] R. Bolton, J. P. B. Sandall, J. Fluorine Chem. 1982, 21, 459–
467.
[5] a) Fluorine in Organic Chemistry (Ed.: R. D. Chambers), J.
Wiley & Sons, New York, 1973, chapter 9; b) L. S. Kobrina,
in: Fluorine Chemistry Reviews (Eds.: P. Tarrant, M. Dekker),
New York, 1974, vol. 7, pp. 1–114; c) G. M. Brooke, J. Fluorine
Chem. 1997, 86, 1–76.
[6] J. Burdon, Tetrahedron 1965, 21, 3373; R. D. Chambers,
W. K. R. Musgrave, J. S. Waterhouse, D. L. H. Williams, J. Bur-
don, W. B. Hollyhead, J. C. Tatlow, J. Chem. Soc., Chem. Com-
mun. 1974, 239–240; R. D. Chambers, D. Close, W. K. R. Mus-
grave, J. S. Waterhouse, D. L. H. Williams, J. Chem. Soc. Perkin
Trans. 1 1977, 1774–1778; R. D. Chambers, D. Close, D. L. H.
Williams, J. Chem. Soc. Perkin Trans. 2 1980, 778–780; R. D.
Chambers, M. J. Seabury, D. L. H. Williams, N. Hughes, J.
Chem. Soc. Perkin Trans. 1 1988, 251–254; R. D. Chambers,
M. J. Seabury, D. L. H. Williams, N. Hughes, J. Chem. Soc.
Perkin Trans. 1 1988, 255–257.
[7] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pom-
elli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
[14] A. Williams, Chem. Soc. Rev. 1994, 23, 93–100.
[15] B. Chiavarino, M. E. Crestoni, S. Fornarini, F. Lanucara, J.
Lemaire, P. Maitre, Angew. Chem. Int. Ed. 2007, 46, 1995–1998
(and references cited therein).
[16]
F. Terrier, G. Ah-Kow, M.-J. Pouet, M.-P. Simonnin, Tetrahe-
dron Lett. 1976, 227–228.
[17] D. R. Stull, E. F. Westrum Jr., G. C. Sinke, The Chemical Ther-
modynamics of Organic Compounds, John Wiley & Sons, New
York, 1969.
[18] International Tables for Crystallography, vol. C (Ed.: A. J. C.
Wilson), Kluwer Academic Publishers, Dordrecht, Boston,
London, 1995.
[19] W. P. Jencks, Acc. Chem. Res. 1980, 13, 161–169.
[20] a) J. A. Rahn, M. S. Holt, M. O’Neil-Johnson, J. H. Nelson,
Inorg. Chem. 1988, 27, 1316–1320; b) E. A. Allen, W. Wilkin-
son, Spectrochim. Acta Ser. A 1974, 30, 1219–1224.
[21] SAINT Plus, version 6.02, Bruker AXS Inc., Madison, WI
1999.
[22] G. M. Sheldrick, SHELX-97, University of Göttingen, Ger-
many, 1997.
[23] R. J. Abraham, J. F. Busby, M. A. Cooper, Org. Magn. Reson.
1974, 6, 612–613.
[24] A. S. Cantrell, P. Engelhardt, M. Hoegberg, S. R. Jaskunas,
N. G. Johansson, C. L. Jordan, J. Kangasmetsä, M. D. Kin-
nick, P. Lind, J. M. Morin Jr., M. A. Muesing, R. Noreén, B.
Öberg, P. Pranc, C. Sahlberg, R. J. Ternansky, R. T. Vasileff, L.
Vrang, S. J. West, H. Zhang, J. Med. Chem. 1996, 39, 4261–
4274.
Received: August 3, 2009
Published Online: January 12, 2010
Eur. J. Org. Chem. 2010, 1111–1123
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1123