An aryl exchange reaction with full retention of configuration of the complexes: Mechanism of the aryl exchange between [PdR2L2] complexes in chloroform (R = pentahalophenyl, L = thioether)

Article abstract of DOI:10.1021/om970721f

The complexes [Pd(3,5-C₆Cl₂F₃)₂L₂] and [Pd(C₆F₅)₂L₂] (L = SC₄H₈ (tht), or SMe₂) react in CCl₃D to give a mixture also containing the heteroaryl products [Pd(3,5-C₆Cl₂F₃)(C₆F ₅)L₂], whereas for L = PPh₃, AsPh₃, 2-picoline, 4-picoline, or 1/2COD, this exchange is not observed. The ¹⁹F NMR kinetic study supports that the scrambling of the aryl groups takes place with retention of configuration at both Pd centers via a triply-bridged binuclear activated complex [L(C₆Cl₂F₃)Pd(μ-C₆Cl ₂F₃)(μ-C₆F₅)(μ-L)Pd(C ₆F₅)L]?. The aryl double bridge is supported by a bridging S-donor ligand facilitating an otherwise difficult exchange (not observed for other weak L ligands, such as picolines, lacking a second lone pair). This constitutes the first reported example of reversible migration of σ-C-bonded groups between bis-organo Pd(II) complexes proceeding without cis-trans isomerization. The isomerization process is also observed, occurring at a much slower rate.

Full text of DOI:10.1021/om970721f

5730
Organometallics 1997, 16, 5730-5736
An Ar yl Exch a n ge Rea ction w ith F u ll Reten tion of
Con figu r a tion of th e Com p lexes: Mech a n ism of th e Ar yl
Exch a n ge betw een [P d R2L2] Com p lexes in Ch lor ofor m
(R ) P en ta h a lop h en yl, L ) Th ioeth er )
Arturo L. Casado, J uan A. Casares, and Pablo Espinet*
Departamento de Quı´mica Inorga´nica, Facultad de Ciencias, Universidad de Valladolid,
E-47005 Valladolid, Spain
Received August 14, 1997X
The complexes [Pd(3,5-C6Cl2F3)2L2] and [Pd(C6F5)2L2] (L ) SC4H8 (tht), or SMe2) react in
CCl3D to give a mixture also containing the heteroaryl products [Pd(3,5-C6Cl2F3)(C6F5)L2],
1
whereas for L ) PPh3, AsPh3, 2-picoline, 4-picoline, or /2COD, this exchange is not observed.
The 19F NMR kinetic study supports that the scrambling of the aryl groups takes place
with retention of configuration at both Pd centers via a triply-bridged binuclear activated
complex [L(C6Cl2F3)Pd(µ-C6Cl2F3)(µ-C6F5)(µ-L)Pd(C6F5)L]q. The aryl double bridge is sup-
ported by a bridging S-donor ligand facilitating an otherwise difficult exchange (not observed
for other weak L ligands, such as picolines, lacking a second lone pair). This constitutes
the first reported example of reversible migration of σ-C-bonded groups between bis-organo
Pd(II) complexes proceeding without cis-trans isomerization. The isomerization process is
also observed, occurring at a much slower rate.
Sch em e 1
In tr od u ction
The alkyl or aryl transfer from main-group or transi-
tion-metal organometallics to transition-metal halo
complexes has been known for many years and is used
as a preparative method of σ-C organopalladium(II)
complexes.1 The process is also involved in the Pd-
assisted cross-coupling reactions of organic halides RX
and main-group organometallics MR′ (Scheme 1, main
cycle) via PdRR′L2 (II), which eventually reductively
eliminates the cross-coupling product R-R′.2
Aryl-for-aryl and aryl-for-alkyl exchanges have been
detected between PdRXL2 complexes (I) and arylating
main-group reagents.2e,3 These processes seem to be
responsible for the formation of undesired homocoupling
products RR or R′R′ in the cross-coupling reactions, as
they give rise to PdR2L2 or PdR′2L2 complexes (IV)
(Scheme 1, side path).
Other exchanges between organopalladium(II) inter-
mediates affecting the selectivity of the coupling process
have been reported. Thus, Yamamoto et al. have found
the exchange of R groups in the reaction between
[PdR2L2] and PdR'XL2 (L ) phosphine),2c,3b as well as
in the trans-cis spontaneous isomerization of PdMe2-
(PR3)2 complexes, which was studied by isotopic label-
ing.4 MgR2(ether)n reagents also exchange with [PdR2L2]
and simultaneously promote the isomerization of the
Pd(II) complex.5 In each case, a dissociation of a neutral
ligand is proposed prior to the formation of a binuclear
complex of the types V-VII (Chart 1).
The methyl-for-halogen exchange was studied by
Puddephatt et al. in a systematic study of the sym-
metrization reactions of PtX2L2 and PtR2L2.6 A mech-
anism without ligand dissociation involving intermedi-
ate VIII (Chart 2) was established for phosphine
X Abstract published in Advance ACS Abstracts, December 1, 1997.
(1) (a) Canty, A. L. In Comprehensive Organometallic Chemistry II;
Wilkinson, G., Stone, F. G., Abel, E. W., Eds.; Pergamon Press:
London, 1995; Vol. 9, p 238-242. (b) Maitlis, P. M.; Espinet, P.; Russell,
M. J . H. In Comprehensive Organometallic Chemistry; Wilkinson, G.,
Stone, F. G., Abel, E. W., Eds.; Pergamon Press: London, 1982; Vol.
6, p 306-311.
(2) (a) Farina, V. In Comprehensive Organometallic Chemistry II;
Wilkinson, G., Stone, F. G., Abel, E. W., Eds.; Pergamon Press:
London, 1995; Vol. 12, p 161-240. (b) Mitchell, T. N. Synthesis 1992,
803-815. (c) Ozawa, F.; Kurihara, K.; Fujimori, M.; Hidaka, T.;
Toyoshima, T.; Yamamoto, A. Organometallics 1989, 8, 180-188. (d)
Broen, J . M.; Cooley, N. A. Chem. Rev. 1988, 88, 1031-1046. (e) Ozawa,
F.; Hidaka, T.; Yamamoto, T.; Yamamoto, A. J . Organomet. Chem.
1987, 330, 253-263. (f) Stille, J . K. Angew. Chem., Int. Ed. Engl. 1986,
25, 508-524. (g) Labadie, J . W.; Stille, J . K. J . Am. Chem. Soc. 1983,
105, 6129-6137. (h) Neghishi, F.; King, A. O.; Okukado, N. J . Org.
Chem. 1977, 42, 1821-1823.
(4) Ozawa, F.; Ito, T.; Nakamura, Y.; Yamamoto, A. Bull. Chem. Soc.
J pn. 1981, 54, 1868-1880.
(5) Ozawa, F.; Kurihara, K.; Yamamoto, T.; Yamamoto, A. J .
Organomet. Chem. 1985, 279, 233-243.
(3) (a) van Asselt, R.; Elsevier, C. J . Organometallics 1994, 13,
1972-1980 and references therein. (b) Ozawa, F.; Fujimori, M.;
Yamamoto, T.; Yamamoto, A. Organometallics 1986, 5, 2144-2149.
(6) (a) Scott, J . D.; Puddephatt, R. J . Organometallics 1983, 2, 1643-
1648. (b) Puddephatt, R. J .; Thompson, P. J . J . Chem. Soc., Dalton
Trans. 1977, 1219-1233. (c) Ibid. 1975, 1810-1814.
S0276-7333(97)00721-8 CCC: $14.00 © 1997 American Chemical Society

Aryl Exchange between [PdR2L2] Complexes
Organometallics, Vol. 16, No. 26, 1997 5731
Ch a r t 1
Ch a r t 2
complexes, whereas prior ligand dissociation was pro-
posed for SMe2 complexes. In the latter case, it was
not possible to decide between several probable inter-
mediates IX-XII.
In the course of our investigation of the dynamic
behavior in solution of bis(perhaloaryl)palladium(II)
complexes, we found that the complexes [Pd(C6-
Cl2F3)2(tht)2] (C6Cl2F3 ) 3,5-dichlorotrifluorophenyl; tht
) tetrahydrothiophene) and [Pd(C6F5)2(tht)2] react re-
versibly in CDCl3 solution to give [Pd(C6Cl2F3)(C6F5)-
(tht)2] without isomerization. The process is much
faster than the spontaneous trans-to-cis isomerization
found in [Pd(C6F5)2(tht)2] (for which a dissociative
mechanism has been proposed),7 suggesting a new type
of aryl exchange in the chemistry of d8 square-planar
complexes. Here, we report a detailed mechanistic
study on the aryl-exchange reaction between complexes
[PdR2L2] (R ) C6F5, C6Cl2F3). The stability of these
complexes toward other possible reactions, e.g., reduc-
tive elimination,8 facilitates this study.
F igu r e 1. Concentration-time plots for the reactions
between (a) cis-[Pd(C6Cl2F3)2(tht)2] (1a ) and cis-[Pd-
(C6F5)2(tht)2] (1b); (b) trans-[Pd(C6Cl2F3)2(tht)2] (2a ) and
trans-[Pd(C6F5)2(tht)2] (2b); and (c) 1a and 2b. Initial
concentration in each complex was 10.0 ( 0.7 mM (imposed
by the low solubility of the trans complexes), CDCl3, 320.1
( 0.2 K. The products are cis-[Pd(C6Cl2F3)(C6F5)(tht)2] (1c)
and trans-[Pd(C6Cl2F3)(C6F5)(tht)2] (2c).
Resu lts
The complexes [PdR2L2] (L ) tht) undergo three
processes in CDCl3: (i) neutral ligand exchange (within
seconds); (ii) aryl exchange (within hours); and (iii)
trans-cis isomerization (within days). This difference
in rate allowed us to carry out kinetic studies on the
second process using NMR techniques.
(1c) only (eq 1, henceforth R1 ) C6Cl2F3 and R2 ) C6F5).9
The equilibrium concentrations are practically those
E xch a n ge of R Gr ou p s b et w een Com p lexes
[P d R2L2]. Mixtures of [Pd(C6Cl2F3)2(tht)2] and [Pd-
(C6F5)2(tht)2] (whether cis or trans) evolve in CDCl3 to
a mixture containing also the heteroaryl products
[Pd(C6Cl2F3)(C6F5)(tht)2]. The most noticeable feature
in these reactions is that the aryl-exchange process
occurs with retention of the trans-cis configuration at
Pd. Figure 1 shows the concentration-time plot of three
independent experiments involving different configura-
tions of the starting complexes.
expected from statistical considerations: 25% of 1a +
25% of 1b + 50% of 1c.
The trans complexes 2a and 2b (Figure 1b and eq 2)
yield trans-[Pd(C6Cl2F3)(C6F5)(tht)2] (2c) first.9 As the
The reaction of a mixture of complexes cis-[Pd(C6-
Cl2F3)2(tht)2] (1a ) and cis-[Pd(C6F5)2(tht)2] (1b) (Figure
1a) shows the formation of cis-[Pd(C6Cl2F3)(C6F5)(tht)2]
(9) The structures of 1c and 2c have been assigned by comparison
of the 19F NMR chemical shift values with those of the cis (1a ,b) and
trans (2a ,b) complexes. Moreover, in the case of the cis complex 1c,
through-space 19F-19F couplings between the C6Cl2F3 and C6F5 rings
produce multiplet signals for the Fortho nuclei, which is not observed
in trans-2c. See: Albe´niz, A. C.; Casado, A. L.; Espinet, P. Organo-
metallics 1997, 16, 5416.
(7) Minniti, D. Inorg. Chem. 1994, 33, 2631-2634.
(8) (a) Moravsky, A.; Stille, J . K. J . Am. Chem. Soc. 1981, 103, 4182-
4186. (b) Loar, M. K.; Stille, J . K. J . Am. Chem. Soc. 1981, 103, 4174-
4181. (c) Ozawa, F.; Ito, T.; Yamamoto, A. J . Am. Chem. Soc. 1980,
102, 6457-6463. (d) Gillie, A.; Stille, J . K. J . Am. Chem. Soc. 1980,
102, 4933-4941.

5732 Organometallics, Vol. 16, No. 26, 1997
Casado et al.
Ta ble 1. Secon d -Or d er Ra te Con sta n ts kex for th e Exch a n ge of Ar yl Gr ou p s betw een cis-[P d (C6Cl2F 3)2(th t)2]
(1a ) a n d cis-[P d (C6F 5)2(th t)2] (1b)a
entry
[1a ]/10-3 mol L-1
T/K
[tht]added/10-3 mol L-1
kex/10-3 L mol-1 s-1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
40.0 ( 0.4
40.0
40.0
40.0
40.0
40.0
40.0
40.0
40.0
30.0 ( 0.4
20.0 ( 0.3
15.0 ( 0.3
10.0 ( 0.3
10.0
330.8 ( 0.2
324.8
320.1
320.1
320.1
320.1
320.1
314.4
309.6
320.1
320.1
320.1
320.1
320.1
320.1
0
0
0
7.5 ( 0.4
3.65 ( 0.18
1.90 ( 0.09
0.096 ( 0.005
0.084 ( 0.004
0.055 ( 0.003
0.0350 ( 0.0017
1.04 ( 0.05
0.77 ( 0.04
3.4 ( 0.2
2.94 ( 0.09
3.43 ( 0.10
4.90 ( 0.13
7.8 ( 0.2
0
0
0
0
0
0
0
0
6.0 ( 0.3
12.0 ( 0.7
20.5 ( 1.4
11.8 ( 0.9c
4.2 ( 0.3d
10.0
Equimolecular mixtures in CCl3D. Estimated from the initial reaction rate between 1a and 2b. c The same for 2a and 2b.
a
b
shown in Table 1 (entries 13-15), the tht cis complexes
1 exchange faster than the trans complexes 2 and the
cis (1) + trans (2) mixture shows a rate almost halfway
between the other two.
The complexes cis-[Pd(C6Cl2F3)2(SMe2)2] (3a ) and cis-
[Pd(C6F5)2(SMe2)2] (3b) exchange their aryl groups
giving cis-[Pd(C6Cl2F3)(C6F5)(SMe2)2] (3c) at a rate
clearly faster than the tht complexes. On the contrary,
no aryl exchange was found between cis-[Pd(C6Cl2F3)2L2]
(L ) 2-pic, 4a ; 4-pic, 5a ; 1,5-cyclooctadiene (COD), 6a )
and cis-[Pd(C6F5)2L2] (L ) 2-pic, 4b; 4-pic, 5b; COD, 6b)
or between trans-[Pd(C6Cl2F3)2L2] (L ) 4-pic, 7a ; PPh3,
8a ; AsPh3, 9a ) and trans-[Pd(C6F5)2L2] (L ) 4-pic, 7b;
PPh3, 8b; AsPh3, 9b). This suggests a clear-cut influ-
ence of the ancillary ligand L on the aryl-exchange
process.
Ligand dissociation is likely to be needed in order to
allow for the formation of R bridges. In fact, the
addition of tht slowed down the reaction between 1a and
1b (Table 1). The ease of ligand dissociation was tested
by checking the ease of L exchange between neutral
complexes.
In the case of the thioether complexes, the neutral
ligand dissociation is clearly faster than the aryl groups
exchange. In reaction 6, the exchange of thioether,
which necessarily involves ligand dissociation, takes
place immediately after both reagents are mixed whereas
the aryl exchange between the same complexes needs
hours for completion. Also, the picoline complexes 4 and
F igu r e 2. 19F 282 MHz NMR spectra sequence (-116 to
-119 ppm region) for the reaction of cis-[Pd(C6Cl2F3)2(tht)2]
(1a , 10-2 M) and trans-[Pd(C6F5)2(tht)2] (2b, 10-2 M) in
CCl3D at 320.1 K.
reaction progresses, the initial trans mixture slowly
isomerizes and cis products (1a , 1b , and 1c) are
detected. This clearly shows that the isomerization is
a slower process that occurs independently from the aryl
exchange.7
Finally, a mixture of 1a (cis) and 2b (trans) generates
1c and 2c at similar rates at an early stage of the
reaction (Figure 1c, Figure 2). Later, the isomerization
products 1b and 2a are formed. Note that the trans
complex 2a is produced faster than expected from direct
cis-trans isomerization of 1a (this isomerization is
undetectable, as shown in Figure 1a). These observa-
tions suggest the occurrence of three concurrent ex-
changes with retention of configuration (eq 3-5). As
5 exchange their neutral ligands quite fast (eq 7, the
ca. 1:1:2 equilibrium is reached in about 1 h at room
temperature) but, as stated above, they do not exchange
aryl groups.
In summary, whereas the neutral ligand exchange is
a process common to picoline and thioether complexes,

Aryl Exchange between [PdR2L2] Complexes
Organometallics, Vol. 16, No. 26, 1997 5733
F igu r e 4. Self-retardation effect of the total concentration
of the palladium complexes, [Pd], on the aryl-exchange rate
of cis-[Pd(C6Cl2F3)2(tht)2] (1a ) and cis-[Pd(C6F5)2(tht)2] (1b)
in CCl3D at 320.1 K. [Pd] ) [1a ]0 + [1b]0.
F igu r e 3. Retardation effect of the addition of tht on the
aryl-exchange rate cis-[Pd(C6Cl2F3)2(tht)2] (1a , 4 × 10-2 mol
L-1) and cis-[Pd(C6F5)2(tht)2] (1b, 4 × 10-2 mol L-1) in
CCl3D at 320.1 K.
Sch em e 2
Ch a r t 3
the aryl exchange occurs only in the latter. This
supports the idea that easy neutral ligand dissociation
is not the only requirement for the aryl exchange to
occur, but the intermolecular exchange of the aryl
groups requires the availability of a second electron pair
on the ancillary ligand (e.g. S-donors) in order to form
L-bridged A-like intermediates (Chart 3). This hypoth-
esis has been investigated further by carrying out a
kinetic study on reaction 1 between tht complexes.
Kin etics on Ar yl Exch a n ge betw een Com p lexes
1a a n d 1b. The cis complexes were chosen for the
kinetic studies. The reaction between cis-[Pd(C6Cl2F3)2-
(tht)2] (1a ) and cis-[Pd(C6F5)2(tht)2] (1b) in CCl3D (eq
1) was monitored by 19F NMR spectroscopy. It follows
a good second-order law until the equilibrium is reached
(see Experimental Section).10 The second-order rate
constants kex measured under different experimental
conditions are listed in Table 1.
Discu ssion
The aryl exchange is retarded by the addition of free
tht (entries 4-7), and the reaction follows an order
minus one with respect to the concentration of added
tht (the slope of the ln(kex) vs ln([tht]added) plot is -1.0).10
The study of mixtures of [Pd(C6F5)2L2] and [Pd(C6-
Cl2F3)2L2] has revealed the occurrence of a exchange
process unobservable on solutions of the pure complexes.
It consists of the intermolecular exchange of aryl groups,
which occurs between organopalladium(II) complexes
containing thioethers (tht or SMe2) as ancillary ligands
L, but not for other ligands such as COD, phosphines,
arsines, or picolines.
The mechanism given in Scheme 2 accounts for the
experimental kinetic results. The dissociation of tht
from complex 1a generates a three-coordinate interme-
diate 1a ′, which is quickly captured by a tetracoordinate
complex 1b giving intermediate A. Aryl scrambling in
A produces intermediate B, which is finally cleaved by
tht yielding two molecules of the product, 1c. Obviously
-1
A linear correlation between kex and [tht]added is then
obtained (Figure 3), the slope being (3.63 ( 0.09) × 106
s (for 4 × 10-2 mol L-1 solutions of both complexes at
320.1 K). On the other hand, the reaction rate constant
kex notably decreases with the total concentration of the
complexes, [Pd] ) [1a ]0 + [1b]0, when working in the
absence of added tht (entries 3, 10-13). The slope of
the ln(kex) vs ln([Pd]) plot is -1.7 (∼-3/2). Accordingly,
a linear correlation between kex and [Pd]-3/2 is obtained
(Figure 4), the slope being (6.1 ( 0.3) × 10-5 mol1/2 L-1/2
s-1 (at 320.1 K). Finally, the dependence on the
temperature (entries 1-3, 8-9) of 4 × 10-2 mol L-1
samples was also analyzed by means of an Eyring plot
(ln(kex/T) vs T-1), giving the following apparent activa-
tion parameters: ∆Hq ) 91 ( 7 kJ mol-1; and ∆Sq
(11) Since the absolute value of ∆Sq is small, this discounts a rate-
determining step purely associative (characterized by very negative
∆Sq, see ref 7) or purely dissociative (positive values ∆Sq, see ref 15)
ex
ex
and suggests that several steps contribute to the observed ∆Sq and
ex
) -15 ( 25 J K-1 mol-1 11
.
∆Hq values. Under the circumstances, these values are far from
ex
having
a clear interpretation (see ref 9a, Chapter 7.1). Strictly
(10) (a) Espenson, J . H. Chemical Kinetics and Reaction Mecha-
nisms, 2nd ed.; McGraw-Hill: Singapore, 1995. (b) Wilkins, R. G.
Kinetics and Mechanism of Reactions of Transition Metal Complexes,
2nd ed.; VCH: New York, 1991.
speaking, activation parameters can be calculated only for elemental
steps, and for this reason we refer to them as apparent ∆Sqex and ∆Hq
.
ex
Alternatively, the Arrhenius treatment gives the activation energy Ea
) 94 ( 7 kJ mol-1
.

5734 Organometallics, Vol. 16, No. 26, 1997
Casado et al.
Sch em e 3
the reaction can start at 1b, but the parallel steps have
not been drawn for simplicity. It should be noted that
the aryl exchange is a reversible process which also
takes place between molecules that are identical to each
other, but it can be detected only when different aryl
groups are involved.
The theoretical rate equation, eq 8, has been obtained
under the steady-state assumption (see Appendix),10
[Pd] being the total concentration of the complexes ([Pd]
) [1a ] + [1b] + [1c] ) [1a ]0 + [1b]0).
rex ) kex[1a ][1b], with kex
)
2k1k2k3
(8)
(k2k4[Pd] + k-1k-2 + k-1k3)[tht] + k2k3[Pd]
Equation 8 agrees with the experimental order minus
one with respect to the concentration of tht shown in
Figure 3. The concentration of tht is never zero in eq
8. In effect, if it happened to be zero, then kex ) 2k1/
[Pd] and the aryl exchange would be controlled by the
tht dissociation step (k1) only, but we have found in the
experiments carried out on neutral ligand exchange (eqs
6 and 7) that the neutral ligand dissociation is much
faster than the aryl exchange. In fact, complexes 1
dissociate some tht ([tht]dis), and the apparent [Pd]-3/2
dependence of kex in the absence of added tht is
thioethers. The alternative paths would produce full
(Scheme 3, path c) or half isomerization (Scheme 3, path
b), which was not observed.
Yamamoto et al. have proposed an exchange of the
type shown in path 3b (L ) PPh3, R1 ) R2 ) Me) in the
auto-catalyzed cis-trans isomerization of [PdMe2-
(PPh3)2] complexes.4 Most likely, the fluorophenyl
complexes with ancillary ligands L other than thioethers
(phosphines, arsines, picolines) could react similarly but
this was not detected, meaning that their reaction is
extremely slow. This suggests that the initial formation
of a L-bridged intermediate A facilitates the formation
of R-bridges for R groups that otherwise would not react
at a reasonable rate by paths b or c in Scheme 3.
Perhaloaryl double bridges have been previously de-
scribed in stable organopalladium(II) and -platinum(II)
species.14 Bridging S-donor ligands are also well-
known.15 It is not unreasonable that other potential
bridging groups containing lone pairs, such as halogen
atoms,6 might be able to play a role similar to that of
tht.
The rate of exchange has a complex dependence on
several steps,16 as revealed by the appearance of several
ki in the expression for kex. The measured activation
parameters must then include contributions from these
steps.11 It is found that the aryl exchange is faster
between two cis complexes than for the cis and trans
case, whereas two trans complexes give the slowest
exchange. This can be related to the two main influ-
ences on the exchange rate. The tht dissociation (as-
sociated to k1) must be easier for the cis complexes
(because the tht ligand is trans to an R ligand having a
high trans influence).17-19 On the other hand, the
scrambling of the R groups in A (associated with k3)
must be controlled by the energy of the transition state
q3. The latter depends on the position of the ligands
attached to the pentacoordinate Pd(II) centers: It has
satisfactorily explained if [tht]dis ) m[Pd]1/2 12
Putting
.
this into eq 8 and assuming that the contribution of the
other terms can be neglected, we obtain eq 9, which
agrees with the experimental results in Figure 4.
2k1k3
mk4[Pd]3/2
kex
)
(9)
Analysis of Figure 4 using eq 9 gives m ) 3.6 × 10-4
mol1/2 L-1/2. This indicates that the concentration of tht
autodissociated from complexes 1 ([Pd] ) 8 × 10-2 mol
L-1) is 10-4 mol L-1, that is 0.13 mol %.
The [tht] and [Pd] terms in the denominator of eq 8
reflect two retarding effects, one due to [Pd] and the
other due to [tht]. The three-coordinate intermediate
1a ′ can be captured either by reagents 1a and 1b or by
product 1c at step k2 ([Pd] term). Only when the 1a ′ is
captured by 1b is the aryl exchange productive (via k3)
and generates 1c. In the other cases, 1c is not formed
and the capture of 1a ′ only reduces its concentration
(the same applies to intermediate 1b′). On the other
hand, the effect of the addition of tht is mostly to reduce
the concentration of intermediate A at step k4 ([tht]
term), before it rearranges to B (step k3).
The preservation of the geometry along the aryl-
exchange process is a very clear indication of the nature
of the activated complex at which the exchange takes
place. Scheme 3 shows the three different possible
activated complexes for the exchange of two aryl groups.
Only the first pathway, path a, passing through an
activated complex formed by two pentacoordinate Pd-
(II) centers sharing a face explains the full preservation
of the cis (or eventually the trans) geometry in both
organopalladium moieties.13 Moreover, this pathway
requires a good bridging ligand L, as is the case of
(13) For associative mechanisms on Pd(II) complexes, see: Cross,
R. J . Adv. Inorg. Chem. 1989, 34, 219-292.
(14) (a) Uso´n, R.; Fornie´s, J .; Falvello, L. R.; Toma´s, M.; Casas, J .
M.; Mart´ın, A.; Cotton, F. A. J . Am. Chem. Soc. 1994, 116, 7160-7165.
(b) Uso´n, R.; Fornie´s, J .; Toma´s, M.; Casas, J . M.; Cotton, F. A.;
Falvello, L. R.; Feng, X. J . Am. Chem. Soc. 1993, 115, 4145-4154. (c)
Uso´n, R.; Fornie´s, J .; Toma´s, M.; Casas, J . M.; Navarro, P. J . Chem.
Soc., Dalton Trans. 1989, 169-172. (d) Uso´n, R.; Fornie´s, J .; Toma´s,
M.; Casas, J . M. Organometallics 1988, 7, 2279-2285.
(15) Some examples of tht-bridged complexes are given, see: (a)
Uso´n, R.; Fornie´s, J .; Falvelo, L. R.; Toma´s, M.; Casas, J . M.; Martin,
A. Inorg. Chem. 1993, 32, 5212-5215. (b) Uso´n, R.; Fornie´s, J .; Toma´s,
M.; Ara, I. J . Chem. Soc., Dalton Trans. 1989, 1011-1016.
(16) Several definitions of the rate-determining (or controlling) step
(rds or rcs) have been proposed. A brief discussion is available in ref
9a, Chapter 4.5, or ref 9b, Chapter 2.3.
(12) For a given dissociation reaction, C ) C′ + L, the equilibrium
constant is Kdis ) x2/([C]0 - x). If x is low, it reduces to Kdis ) x2/[C]0
and [L] ) x )
Kdis[C]0. Then, the proposed square root dependence
of [tht]dis on [Pd] is acceptable.
x

Aryl Exchange between [PdR2L2] Complexes
Organometallics, Vol. 16, No. 26, 1997 5735
CHN microanalyzer. 1H and 19F NMR spectra were recorded
on a Bruker ARX 300 instrument equipped with a VT-100
variable-temperature probe. The temperature (( 0.2 K) was
calibrated using MeOH and ethylene glycol 1H NMR standard
methods. Chemical shifts are reported in ppm from tetram-
ethylsilane (1H) or CCl3F (19F) in CCl3D solutions at 293 K.
The complexes cis-[PdR2(tht)2] (R ) C6Cl2F3, 1a ; C6F5, 1b),
trans-[PdR2(tht)2] (R ) C6Cl2F3, 2a; C6F5, 2b), cis-[PdR2(SMe2)2]
(R ) C6Cl2F3, 3a ; C6F5, 3b), cis-[PdR2(COD)] (R ) C6Cl2F3, 6a ;
C6F5, 6b), trans-[PdR2(PPh3)2] (R ) C6Cl2F3, 8a ; C6F5, 8b), and
trans-[Pd(C6F5)2(AsPh3)2] (9b) were prepared as previously
reported.21,22 NMR data (some not previously reported) are
as follows. 1a : 1H NMR δ 2.87 (m, SCH2), 1.86 (m, CCH2);
19F NMR δ -89.98 (s, o-CF), -118.23 (s, p-CF). 1b: 1H NMR
δ 2.90 (m, SCH2), 1.87 (m, CCH2); 19F NMR δ -116.46 (m,
o-CF), -160.20 (t, p-CF), -162.93 (m, m-CF). 2a : 1H NMR δ
2.66 (m, SCH2), 1.85 (m, CCH2); 19F NMR δ -92.65 (s, o-CF),
-117.16 (s, p-CF). 2b: 1H NMR δ 2.68 (m, SCH2), 1.85 (m,
CCH2); 19F NMR δ -118.41 (m, o-CF), -158.86 (t, p-CF),
-161.64 (m, m-CF). 3a : 1H NMR δ 2.15 (s, SCH3); 19F NMR
δ -90.70 (s, o-CF), -118.04 (s, p-CF). 3b: 1H NMR δ 2.17 (s,
SCH3); 19F NMR δ -117.00 (m, o-CF), -159.86 (t, p-CF),
-162.56 (m, m-CF). 6a : 1H NMR δ 5.79 (s, CH), 2.75 (s, CH2);
19F NMR δ -91.20 (s, o-CF), -117.30 (s, p-CF). 6b: 1H NMR
δ 5.83 (s, CH), 2.76 (s, CH2); 19F NMR δ -117.40 (m, o-CF),
-159.20 (t, p-CF), -162.40 (m, m-CF). 8a : 1H NMR δ 7.5-
7.2 (m, CH); 19F NMR δ -88.10 (t, J PF ) 4.8 Hz, o-CF), -118.23
(t, J PF ) 2 Hz, p-CF); 31P{1H} NMR δ 22.54 (m). 8b: 1H NMR
δ 7.5-7.2 (m, CH); 19F NMR δ -113.91 (m, o-CF), -162.38
(m, p-CF), -163.41 (m, m-CF); 31P{1H} NMR δ 22.73 (m). 9b:
1H NMR δ 7.4-7.1 (m, CH); 19F NMR δ -113.87 (m, o-CF),
-161.39 (m, p-CF), -163.20 (m, m-CF).
Ch a r t 4
been shown that in trigonal-bipyramidal d8 pentacoor-
dinate transition metal complexes, the stronger σ donors
prefer the axial sites.20 Hence, the transition states q3
(Chart 4) with the R groups (stronger donors than tht)
in the axial positions must be lower in energy and more
easily accessible. Thus, both effects contribute in the
same sense to the reaction rates observed: cis + cis >
cis + trans > trans + trans
cis-[P d R2L2] (R ) C6Cl2F 3, L ) 2-p ic (4a ) a n d 4-p ic (5a );
R ) C6F 5, L ) 2-p ic (4b) a n d 4-p ic (5b)). To a solution of
cis-[PdR2(COD)] (9) (0.2 mmol) in CH2Cl2 (20 mL) was added
an excess of L (0.5 mmol), and the mixture was stirred for 30
min. The resulting colorless solution was evaporated to
dryness to give white cis-[PdR2L2] (4 and 5), which was washed
with n-hexane (3 × 4 mL) and air-dried (quantitative yields).
4a : IR 1050 vs, 1030 vs, 774 vs, 758 vs, 699 s, 688 s; 1H NMR
δ 8.86 (m, H, CH), 7.57 (dt, H, CH), 7.16 (m, 2H, CH), 3.10 (s,
3H, CH3); 19F NMR δ -91.19 (fluxional, o-CF), -120.08 (s,
p-CF). Anal. Calcd for C24H14Cl4F6N2Pd: C, 41.62; H, 2.04;
N, 4.05. Found: C, 41.30; H, 2.22; N, 4.01. 4b: IR 1059 vs,
Con clu sion s
The presence of ligands with two electron pairs (such
as tht or SMe2) seems to open a novel lower energy
pathway for aryl exchange, which occurs with retention
of configuration at both Pd centers differently from other
pathways previously proposed. This occurs because the
ancillary ligand helps in the formation of a triply-
bridged binuclear activated complex where the aryl
double bridge is supported by a bridging S-donor ligand.
The possibility for halogen ligands to be playing a
similar third-bridge role in exchanges between halogen-
containing organometallic complexes should be consid-
ered. Also, it might be worth testing the possibility to
induce exchanges with retention of configuration in
other systems by the use of lone-pair-containing ligands
in place of phosphines.
1
958 vs, 795 s, 783 s, 765 s; H NMR δ 8.90 (m, H, CH), 7.59
(dt, H, CH), 7.20 (m, 2H, CH), 3.10 (s, 3H, CH3); 19F NMR δ
-117.65 (fluxional, o-CF), -161.37 (t, p-CF), -164.37 (m,
m-CF). Anal. Calcd for C24H14F10N2Pd: C, 45.99; H, 2.25; N,
4.47. Found: C, 45.77; H, 2.27; N, 4.34. 5a : IR 1069 s, 1050
1
s, 1040 s, 813 s, 700 m, 690 m; H NMR δ 8.35 (m, 2H, CH),
7.09 (m, 2H, CH), 2.33 (s, 3H, CH3); 19F NMR δ -91.21 (s,
o-CF), -120.10 (s, p-CF). Anal. Calcd for C24H14Cl4F6N2Pd:
C, 41.62; H, 2.04; N, 4.05. Found: C, 41.76; H, 2.10, N, 3.87.
1
5b: IR 1056 s, 1040 s, 956 vs, 812 s, 808 s, 796 s, 786 s; H
NMR δ 8.35 (m, 2H, CH), 7.09 (m, 2H, CH), 2.34 (s, 3H, CH3);
19F NMR δ -117.61 (m, o-CF), -162.00 (t, p-CF), -164.34 (m,
m-CF). Anal. Calcd for C24H14F10N2Pd: C, 45.99; H, 2.25; N,
4.47. Found: C, 45.75; H, 2.17; N, 4.34.
Exp er im en ta l Section
All reactions involving organolithium reagents were carried
out under N2. Commercial 2-pic (2-picoline) and 4-pic (4-
picoline) were used without further purification. Solvents were
distilled using standard methods; deuteriochloroform for the
kinetic samples was treated with anhydrous MgSO4 and Na2-
CO3. Infrared spectra (cm-1, Nujol-polyethylene) were re-
corded on a Perkin-Elmer FT 1720 X spectrophotometer.
Combustion CHN analyses were made on a Perkin-Elmer 2400
tr a n s-[P d R2L2] (R ) C6Cl2F 3, L ) 4-p ic (7a ) a n d AsP h 3
(9a ); R ) C6F 5, L ) 4-p ic (7b)). To a solution of trans-
[PdR2(tht)2] (2; 0.2 mmol) in CH2Cl2 (10 mL) was added an
excess of L (1.0 mmol), and the mixture was stirred for 60 min.
The resulting colorless solution was evaporated to dryness to
give white complexes 7 or 9a , which were washed with diethyl
ether (3 × 2 mL) and air-dried (quantitative yields). 7a : IR
1039 s, 1022 m, 816 m, 774 s, 672 m; 1H NMR δ 8.54 (m, 2H,
CH), 7.03 (m, 2H, CH), 2.30 (s, 3H, CH3); 19F NMR δ -97.16
(s, o-CF), -119.45 (s, p-CF). Anal. Calcd for C24H14Cl4F6N2-
(17) For dissociative mechanisms in Pd(II) complexes, see: (a) Frey,
U.; Helm, L.; Merbach, A.; Romeo, R. J . Am. Chem. Soc. 1989, 111,
8161-8165. (b) Casares, J . A.; Coco, S.; Espinet, P.; Lin, Y.-S.
Organometallics 1995, 14, 3058-3067 and references therein.
(18) Hartley, F. R. Chem. Soc. Rev. 1973, 2, 163-179.
(19) Tatsumi, K.; Hoffmann, R.; Yamamoto, A.; Stille, J . K. Bull.
Chem. Soc. J pn. 1981, 54, 1857-1867.
(20) Rossi, A. R.; Hoffmann, R. Inorg. Chem. 1975, 14, 365-374.
(21) Uso´n, R.; Fornie´s, J .; Mart´ınez, F.; Toma´s, M. J . Chem. Soc.,
Dalton Trans. 1980, 888-893.
(22) Espinet, P.; Mart´ınez-Ilarduya, J . M.; Pe´rez-Briso, C.; Casado,
A. L.; Alonso, M. A. J . Organomet. Chem., in press.

5736 Organometallics, Vol. 16, No. 26, 1997
Casado et al.
Pd: C, 41.62; H, 2.04; N, 4.05. Found: C, 41.26; H, 2.09, N,
4.00. 7b: IR 1064 s, 1043 s, 950 vs, 812 vs, 768 s, 504 s; H
∂[1a ′]
∂t
1
) k1[1a ] + k-2[A] - k-1[1a ′][tht] - k2[1a ′][1a ] -
NMR δ 8.53 (m, 2H, CH), 7.01 (m, 2H, CH), 2.18 (s, 3H, CH3);
19F NMR δ -122.50 (m, o-CF), -161.20 (t, p-CF), -163.10 (m,
m-CF). Anal. Calcd for C24H14F10N2Pd: C, 45.99; H, 2.25; N,
4.47. Found: C, 45.72; H, 2.43; N, 4.25. 9a : IR (KBr) 1590
s, 1484 s, 1436 vs, 1428 vs, 1397 vs, 1045 vs, 1032 vs, 777 vs,
k2[1a ′][1b] - k2[1a ′][1c] ) 0
∂[1a ′]
∂t
) k1[1a ] + k-2[A] - k-1[1a ′][tht] - k2[1a ′][Pd] ) 0
k1[1a ] + k-2[A]
1
746 vs, 738 vs, 693 vs, 677 s, 480 vs, 472 s, 462 s; H NMR δ
7.4-7.2 (m, CH); 19F NMR δ -88.13 (s, o-CF), -119.98 (s,
p-CF). Anal. Calcd for C48H30As2Cl4F6Pd: C, 51.53; H, 2.70.
Found: C, 51.33; H, 2.78.
[1a ′] )
k-1[tht] + k2[Pd]
Kin etics on th e Ar yl-Gr ou p Exch a n ge betw een [P d R 2-
(th t)2] (Ta ble 1). NMR tubes (5 mm) were charged with
equimolar solid mixtures of complexes and aliquots of a tht
solution in CCl3D (30.7 ( 0.7 × 10-3 mol L-1). The samples
were dissolved at room temperature in CCl3D to a fixed volume
of 600 ( 5 µL and placed into a thermostated probe. The
evolution of selected signals of the complexes (in the range
from -116 to -119 ppm, see Figure 2) was monitored by 19F
NMR. Conversion-time data were fitted to an integrated
reversible second-order equation ln[1 - ([1c]/[1a ]0)] ) -2kex-
[1a ]0t,10 giving kex (standard deviations are also given). 19F
NMR data for heteroaryl products 1c: δ -89.83 (m, o-C6-
Cl2F2F), -116.34 (m, o-C6F2F3), -118.20 (s, p-C6Cl2F2F),
-160.18 (t, p-C6F4F), -162.84 (m, m-C6F2F3). 2c: δ -92.67
(s, o-C6Cl2F2F), -117.10 (s, p-C6Cl2F2F), -118.34 (m, o-C6F2F3),
-158.90 (t, p-C6F4F), -161.63 (m, m-C6F2F3).
where [Pd] is the total amount of the complexes ([Pd]
) [1a ] + [1b] + [1c] ) [1a ]0 + [1b]0).
Similarly, for [1b′]
k1[1b] + k-2[A]
[1b′] )
k-1[tht] + k2[Pd]
Only the intermediates A formed by one moiety of 1a
and one moiety of 1b can yield aryl exchange:
∂[A]
∂t
) k2[1a ′][1b] + k2[1a ][1b′] - k-2[A] - k4[A][tht] -
Rea ction s of [P d R2L2] Com p lexes. Appropriate mixtures
of complexes in CCl3D (10.0 ( 0.7 × 10-3 mol L-1 in each one)
were placed into a thermostated bath ((1 K). The evolution
of the reaction mixtures was examined by 1H and 19F NMR.
The NMR data for products are given as follows. cis-[Pd(C6-
Cl2F3)(C6F5)(SMe2)2] (3c): 1H NMR δ 2.14 (s, SCH3), 2.16 (s,
SCH3); 19F NMR δ -90.70 (m, 2F, o-C6Cl2F2F), -116.09 (m,
2F, o-C6F2F3), -118.03 (s, F, p-C6Cl2F2F), -159.86 (t, F,
p-C6F4F), -162.50 (m, 2F, m-C6F2F3). cis-[Pd(C6F5)2(tht)-
(SMe2)] (10): 1H NMR δ 2.91 (m, 2H, SCH2), 2.17 (s, 3H, CH3),
1.88 (m, 2H, CCH2). cis-[Pd(C6F5)2(2-pic)(4-pic)] (11): 1H NMR
δ 8.93 (m, 1H, CH), 8.26 (m, 2H, CH), 7.61 (t, 1H, CH), 7.15
(m, 4H, CH), 3.08 (s, 3H, CH3), 2.32 (s, 3H, CH3); 19F NMR δ
-117.95 (br, 2F, o-CF), -161.89 (t, 1F, p-CF), -162.10 (t, 1F,
p-CF), -164.30 (m, 2F, m-CF), -164.50 (m, 2F, m-CF).
k3[A] ) 0
k2[1a ′][1b] + k2[1a ][1b′]
[A] )
k4[tht] + k-2 + k3
2k1k2[1a ][1b]
[A] )
(k4[tht] + k3)(k-1[tht] + k2[Pd]) + k-1k-2[tht]
Then the reaction rate for the forward direction rex is
rex ) k3[A] )
Er r or An a lysis. Errors σx of directly measured magni-
i
2k1k2k3
tudes xi were taken as determined by the error of the
apparatus. Errors of magnitudes extracted from least-squares
linear regression were taken as the standard deviation of
respective coefficient. Errors σy of mathematically calculated
[1a ][1b]
(k4[tht] + k3)(k-1[tht] + k2[Pd]) + k-1k-2[tht]
(10)
values y ) f(xi) were estimated as (σy)2 ) ∑i [(∂y/∂xi)σx ]2.
i
This equation can be simplified considering that [tht]2
in the denominator is very small:
Ack n ow led gm en t. Financial support by the Direc-
cio´n General de Investigacio´n Cient´ıfica y Te´cnica
(Project No. PB96-0363) and the J unta de Castilla y
Leo´n (Project No. VA40-96) and a fellowship to A.L.C.
from the Ministerio de Educacio´n y Ciencia are very
gratefully acknowledged.
2k1k2k3
rex
)
[1a ][1b]
(k2k4[Pd] + k-1k-2 + k-1k3)[tht] + k2k3[Pd]
Ap p en d ix. Der iva tion of Ra te Equ a tion 8. For
simplicity, we have considered that complexes 1a and
1b react at similar rates in Scheme 2. This seems a
reasonable approximation considering their chemical
similarity. Although the aryl exchange is reversible, we
will develop the steady-state approximation for the
forward reaction only (kex). The steady-state concentra-
tions of intermediates 1′ and A are
and the second-order rate constant is
2k1k2k3
kex
)
(8)
(k2k4[Pd] + k-1k-2 + k-1k3)[tht] + k2k3[Pd]
OM970721F