A novel reversible aryl exchange involving two organometallics: Mechanism of the gold(I)-catalyzed isomerization of trans-[PdR2L2] complexes (R = Aryl, L = SC4H8)

Article abstract of DOI:10.1021/om980283s

[AuR¹(tht)] (3a) (R¹ = 3,5-C₆Cl₂F₃, tht = tetrahydrothiophene) very efficiently catalyzes the isomerization of trans-[Pd(R¹)₂(tht)₂] (1a) to cis-[Pd(R₁)₂(tht)₂] in CDCl₃. The ¹⁹F NMR kinetic study leads to the first-order rate law r_iso = k_iso[1a] = (k_spo + k_cat[3a])[1a], where k_spo = (1.50 ± 0.03) × 10^-6 s^-1 and k_cat = a/(b + [tht]) with a = (1.32 ± 0.07) × 10^-4 s^-1 and b = (30 ± 0.2) × 10^-5 mol L^-1 (at 304.4 K). The reaction of 1a and [AuR²(tht)] (R² = C₆F₅) yields cis-[PdR¹R₂(tht)₂] and 3a, evidencing that the catalyzed isomerization takes place with aryl-group exchange between Pd(II) and Au(I). An associative mechanism passing through R-bridged intermediates [(tht)(R¹)₂Pd(μ-R²)Au(tht)] and a donor-acceptor activated . complex [(tht)(R¹)₂(R²)Pd→Au(tht)]? is proposed. The results suggest that the associative displacement of tht from 1a by the nucleophilic arylgold(I) complex to give [(tht)(R₁)₂Pd(μ-R²)Au(tht)] is the rate-determining step (k₁). This is supported by the typical bimolecular activation parameters that were found: Δ₁? = 56.4 ± 1.6 kJ mol^-1 and ΔS₁? = -46 ± 6 J K^-1 mol^-1.

Full text of DOI:10.1021/om980283s

Organometallics 1998, 17, 3677-3683
3677
A Novel Rever sible Ar yl Exch a n ge In volvin g Tw o
Or ga n om eta llics: Mech a n ism of th e Gold (I)-Ca ta lyzed
Isom er iza tion of tr a n s-[P d R2L2] Com p lexes
(R ) Ar yl, L ) SC4H8)
Arturo L. Casado and Pablo Espinet*
Departamento de Quı´mica Inorga´nica, Facultad de Ciencias, Universidad de Valladolid,
E-47005 Valladolid, Spain
Received April 14, 1998
[AuR1(tht)] (3a ) (R1 ) 3,5-C6Cl2F3, tht ) tetrahydrothiophene) very efficiently catalyzes
the isomerization of trans-[Pd(R1)2(tht)2] (1a ) to cis-[Pd(R1)2(tht)2] in CDCl3. The 19F NMR
kinetic study leads to the first-order rate law riso ) kiso[1a ] ) (kspo + kcat[3a ])[1a ], where kspo
) (1.50 ( 0.03) × 10-6 s-1 and kcat ) a/(b + [tht]) with a ) (1.32 ( 0.07) × 10-4 s-1 and b )
(3.0 ( 0.2) × 10-5 mol L-1 (at 304.4 K). The reaction of 1a and [AuR2(tht)] (R2 ) C6F5)
yields cis-[PdR1R2(tht)2] and 3a , evidencing that the catalyzed isomerization takes place
with aryl-group exchange between Pd(II) and Au(I). An associative mechanism passing
through R-bridged intermediates [(tht)(R1)2Pd(µ-R2)Au(tht)] and a donor-acceptor activated
complex [(tht)(R1)2(R2)PdfAu(tht)]q is proposed. The results suggest that the associative
displacement of tht from 1a by the nucleophilic arylgold(I) complex to give [(tht)(R1)2Pd(µ-
R2)Au(tht)] is the rate-determining step (k1). This is supported by the typical bimolecular
q
q
activation parameters that were found: ∆H1 ) 56.4 ( 1.6 kJ mol-1 and ∆S1 ) -46 ( 6 J
K-1 mol-1
.
In tr od u ction
catalytic conditions,4 which can affect the reaction
selectivity. Thus, the understanding of such exchange
mechanisms is fundamental for the control of the
coupling efficacy. Surprisingly, despite the importance
of the topic, rigorous and detailed kinetic studies to
support the mechanistic proposals are scarce, probably
due to the difficulty introduced by the instability and
reactivity of many transition-metal organometallics,
particularly, in our case, those of Pd.
Group transfer between organometallics is a critical
step in most organic reactions assisted by organotransi-
tion-metal complexes, such as the widely used pal-
ladium-catalyzed coupling of organic electrophiles RX
and organometallics MR′.1 In catalytic cycles, the
intermediate [PdRXL2], produced by fast oxidative ad-
dition of RX to [PdL2] species,2 can undergo two ex-
change processes with MR′. The X for R′ exchange (eq
1) leads, after reductive elimination, to the cross-
coupling product RR′, whereas the R for R′ exchange
(eq 2) seems to be responsible for the formation of
homocoupling products.3
In the past few years, we have shown the advantages
of pentafluorophenyl as a ligand to study palladium
complexes. This aryl group allows one to isolate some
otherwise unstable intermediates, slows down some
reactions, and facilitates the identification of the com-
pounds and the monitoring of the reactions using 19F
NMR in place of 1H NMR spectroscopy. We have
applied this to the study of the Heck reaction,5 Pd
migration processes,6 the oxidative addition of ArI to
Pd(0)2 and recently to prove the operation and mecha-
nisms of exchange processes involving some polydentate
ligands.7
[PdRXL2] + MR′ f [PdRR′L2] + MX
(1)
[PdRXL2] + MR′ f [PdR′XL2] + MR
(2)
Additionally, other group exchanges between orga-
nometallic intermediates have been detected under
* To whom correspondence should be addressed. E-mail:
espinet@cpd.uva.es.
(1) (a) Stanforth, S. P. Tetrahedron 1998, 54, 263-303. (b) Farina,
V.; Roth, G. P. Adv. Metalorg. Chem. 1996, 5, 1-53. (c) Curran, D. P.;
Hoshino, M. J . Org. Chem. 1996, 61, 6480-6481. (d) Mateo, C.;
Ca´rdenas, D. J .; Ferna´ndez-Ribas, C.; Echavarren, A. M. Chem. Eur.
J . 1996, 2, 1596-1606. (e) Roth, G. P.; Farina, V.; Liebeskind, L. S.;
Pen˜a-Cabrera, E. Tetrahedron. Lett. 1995, 36, 2191-2194. (f) Mitchell,
T. N. Synthesis 1992, 803-815. (g) Scott, W. J .; Stille, J . K. J . Am.
Chem. Soc. 1986, 108, 3033-3040. (h) Stille, J . K. Angew. Chem., Int.
Ed. Engl. 1986, 25, 508-524. (i) Beletskaya, I. P. J . Organomet. Chem.
1983, 250, 551-564.
(2) Casado, A. L.; Espinet, P. Organometallics 1998, 17, 954-959.
(3) (a) van Asselt, R.; Elsevier, C. J . Organometallics 1994, 13,
1972-1980. (b) Negishi, E.; King, A. O.; Okukado, N. J . Org. Chem.
1977, 42, 1821-1823.
(4) (a) Osakada, K.; Sakata, R.; Yamamoto, T. Organometallics 1997,
16, 5354-5364. (b) Osakada, K.; Sakata, R.; Yamamoto, T. J . Chem.
Soc., Dalton Trans. 1997, 1265-1266. (c) Ozawa, F.; Kurihara, K.;
Fujimori, M.; Hidaka, T.; Toyoshima, T.; Yamamoto, A. Organometal-
lics 1989, 8, 180-188. (d) Ozawa, F.; Hidaka, T.; Yamamoto, T.;
Yamamoto, A. J . Organomet. Chem. 1987, 330, 253-263. (e) Ozawa,
F.; Fujimori, M.; Yamamoto, T.; Yamamoto, A. Organometallics 1986,
5, 2144-2149. (f) Ozawa, F.; Kurihara, K.; Yamamoto, T.; Yamamoto,
A. J . Organomet. Chem. 1985, 279, 233-243. (g) Scott, J . D.; Pud-
dephatt, R. J . Organometallics 1983, 2, 1643-1648.
(5) Albe´niz, A. C.; Espinet, P.; J eannin, Y.; Philoche-Levisalles, M.;
Mann, B. E. J . Am. Chem. Soc. 1990, 112, 6594-6600.
S0276-7333(98)00283-0 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 07/31/1998

3678 Organometallics, Vol. 17, No. 17, 1998
Casado and Espinet
Sch em e 1a
a
R1 ) C6Cl2F3, R2 ) C6F5, L ) tht.
F igu r e 1. 19F NMR (282 MHz) spectral sequence (12 min
invervals) of the isomerization of trans-[Pd(C6Cl2F3)2(tht)2]
(1a ) to cis-[Pd(C6Cl2F3)2(tht)2] (2a ) catalyzed by [Au(C6-
Cl2F3)(tht)] (3a ) in CDCl3 at 320.3 K. [1a ]0 ) (1.00 ( 0.03)
× 10-2 mol L-1, [3a ] ) (9.9 ( 0.5) × 10-6 mol L-1 (0.1 mol
%).
More recently, we have introduced the use of C6Cl2F3
(3,5-dichlorotrifluorophenyl) in connection with that of
C6F5. The availability of these two groups, chemically
very similar but spectroscopically very different, allows
us to carry out exchange studies similar to those
sometimes carried out by isotopic labeling.8 Thus, we
have reported a novel type of exchange between halo-
genoless organopalladium complexes: [Pd(C6Cl2F3)2-
(tht)2] (tht ) tetrahydrothiophene) and [Pd(C6F5)2(tht)2]
exchange their aryls reversibly, yielding [Pd(C6Cl2F3)-
(C6F5)(tht)2].9 The exchange takes place with full reten-
tion of the cis-trans configuration of the starting
complexes. This exchange occurs exclusively with neu-
tral ligands L possessing lone electron pairs (such as
thioethers, not with phosphines or pyridines) via the
formation of L-bridged intermediates, helping the for-
mation of the electron-deficient aryl bridges needed for
the aryl exchange to occur (Scheme 1, henceforth R1 )
C6Cl2F3 and R2 ) C6F5).
The cis-trans isomerization also occurs, but as an
independent and much slower process.10 However, we
have now discovered that this isomerization can be
catalyzed by Au(I) complexes. Moreover, the kinetic
study of this process reveals that this catalyzed isomer-
ization takes place by a reversible double-aryl exchange
between Pd(II) and Au(I). Thus, the reaction becomes
relevant in connection with synthetic processes cocata-
lyzed by two or more different metals (e.g., the Stille
reaction assisted by CuI).11
equilibrium is reached after days at room temperature
(eq 3, about 97% cis).10,12 However, the isomerization
(3)
is complete within seconds when 1 equiv of [Au(C6-
Cl2F3)(tht)] (3a ) is added, evidencing an extraordinary
catalytic effect of the Au(I) complex. Using smaller
amounts of 3a , the rate of isomerization can be con-
trolled to be adequate for its kinetic study by 19F NMR
(Figure 1).
The Au(I)-catalyzed isomerization of 1a followed good
first-order kinetics (eq 4) when the experiments were
carried out with addition of tht. In the experiments
[1a ]
ln
) kiso
t
(4)
(
)
[1a ]0
without added tht, a small retardation was observed as
the reaction proceeded, due to catalyst decomposition.13
Therefore, we measured the initial (up to 10% conver-
sion) isomerization rate of 1a (r0). The results obtained
under different conditions are given in Table 1.14
The reaction is first order with respect to the concen-
tration of catalyst 3a . The isomerization rate is pro-
portional to [3a ] as shown in Figure 2a, the slope being
(4.52 ( 0.11) × 10-2 s-1. Therefore, the rate law follows
eq 5, which shows two contributions: One accounts for
the spontaneous (kspo) and the other for the catalyzed
isomerization (kcat). The latter is by far the most
Resu lts
Isom er iza tion of tr a n s-[P d (C6Cl2F 3)2(th t)2] Ca ta -
lyzed by [Au (C6Cl2F3)(th t)]. The spontaneous isomer-
ization of trans-[Pd(C6Cl2F3)2(tht)2] (1a ) to cis-[Pd(C6-
Cl2F3)2(tht)2] (2a ) in CDCl3 is very slow, and the
(6) (a) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organometallics 1997,
16, 4138-4144. (b) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organome-
tallics 1997, 16, 4030-4032. (c) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S.
Organometallics 1996, 15, 5010-5017. (d) Albe´niz, A. C.; Espinet, P.;
Lin, Y.-S.; Orpen, A. G.; Mart´ın, A. Organometallics 1996, 15, 5003-
5009. (e) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. J . Am. Chem. Soc. 1996,
118, 7145-7152. (f) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organome-
tallics 1995, 14, 2977-2986. (g) Albe´niz, A. C.; Espinet, P. Organo-
metallics 1991, 10, 2987-2988. (h) Albe´niz, A. C.; Espinet, P.; Foces-
Foces C.; Cano, F. H. Organometallics 1990, 9, 1079-1085.
(7) (a) Casares, J . A.; Espinet, P.; Soulantica, K.; Pascual, I.; Orpen,
A. G. Inorg. Chem. 1997, 36, 5251-5256. (b) Casares, J . A.; Espinet,
P. Inorg. Chem. 1997, 36, 5428-5431.
riso ) kiso[1a ] ) (kspo + kcat[3a ])[1a ]
(5)
important contribution. For 0.1 mol % of 3a , the
isomerization mainly runs (97%) via the catalytic path-
(12) Casado, A. L.; Casares, J . A.; Espinet, P. Inorg. Chem., in press.
(13) Complex 3a decomposes in solution slowly to colloidal pink Au-
(0). This decomposition is inhibited by added tht. For other organogold-
(I) decompositions, see: Tamaki, A.; Kochi, J . K. J . Organomet. Chem.
1973, 61, 441-450.
(14) Since the equilibrium in eq 3 is shifted far toward the cis side
and the measurements have been made at the earlier stages of the
reaction, the r0 values given for the isomerization of 1a to 2a are hardly
affected by the opposite reaction, that of 2a to 1a .
(8) Ozawa, F.; Ito, T.; Nakamura, Y.; Yamamoto, A. Bull. Chem. Soc.
J pn. 1981, 54, 1868-1880.
(9) Casado, A. L.; Casares, J . A.; Espinet, P. Organometallics 1997,
16, 5730-5736.
(10) It has been studied in detail for C6F5 in place of C6Cl2F3, see:
Minniti, D. Inorg. Chem. 1994, 33, 2631-2634.
(11) Farina, F.; Kapadia, S.; Krishnan, B.; Wang, C.; Liebeskind,
L. S. J . Org. Chem. 1994, 59, 5905-5911.

Isomerization of trans-[PdR2L2] Complexes
Organometallics, Vol. 17, No. 17, 1998 3679
Ta ble 1. In itia l Ra tes for th e [Au (C6Cl2F 3)(th t)]
(3a )-Ca ta lyzed Isom er iza tion of
tr a n s-[P d (C6Cl2F 3)2(th t)2] (1a ) to
cis-[P d (C6Cl2F 3)2(th t)2] (2a ) in CDCl3
a
[3a ]/10-5
[tht]added/10-4
mol L-1
T/K
mol L-1
r0/10-6 mol L-1 s-1
304.4 ( 0.2
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
304.4
310.3
315.2
320.3
326.3
331.9
0
0
0
0.0150 ( 0.0003
0.54 ( 0.03
0.0596 ( 0.0008
1.00 ( 0.04
0.146 ( 0.004
1.71 ( 0.09
0.290 ( 0.003
2.6 ( 0.2
1.20 ( 0.02
0.824 ( 0.004
0.559 ( 0.007
0.431 ( 0.005
0.358 ( 0.004
3.54 ( 0.08
0.578 ( 0.005
4.6 ( 0.2
0.99 ( 0.05
0.99
1.97 ( 0.05
1.97
3.94 ( 0.05
3.94
5.91 ( 0.05
5.91
5.91
5.91
5.91
5.91
7.88 ( 0.06
7.88
9.85 ( 0.06
9.85
1.54 ( 0.04
0
1.54
0
1.54
0
F igu r e 2. Catalytic effect of [Au(C6Cl2F3)(tht)] (3a ) on the
isomerization of trans-[Pd(C6Cl2F3)2(tht)2] (1a ) to cis-[Pd(C6-
Cl2F3)2(tht)2] (2a ) in CDCl3 at 304.4 K: (a) without addition
of tht; (b) with [tht]added ) (1.54 ( 0.04) × 10-4 mol L-1
(1.6 mol %). [1a ]0 ) (1.00 ( 0.03) × 102 mol L-1. r0 is the
initial isomerization rate.
0.386 ( 0.010
0.772 ( 0.016
1.16 ( 0.03
1.54
1.93 ( 0.04
0
1.54
0
1.54
0.769 ( 0.007
0.80 ( 0.3
1.07 ( 0.05
1.59 ( 0.09
2.42 ( 0.12
3.81 ( 0.06
0.99 ( 0.05
0.99
0.99
0.99
0
0
0
0
0
0.99
a
Up to 10% conversion. [1a ]0 ) (1.00 ( 0.03) × 10-2 mol L-1
.
way, whereas more than 1 mol % of catalyst leads to
isomerization rates too fast for NMR monitoring at
304.4 K.
A similar behavior was obtained in the presence of
added tht, but the rates were much slower. The slope
F igu r e 3. Retardation effect by addition of tht on the [Au-
(C6Cl2F3)(tht)] (3a )-catalyzed isomerization of trans-[Pd(C6-
Cl2F3)2(tht)2] (1a ) to cis-[Pd(C6Cl2F3)2(tht)2] (2a ) in CDCl3
at 304.4 K. [1a ]0 ) (1.00 ( 0.03) × 10-2 mol L-1, [3a ] )
(5.91 ( 0.05) × 10-5 mol L-1 (0.6 mol %). r0 is the initial
isomerization rate.
in Figure 2b is (7.8 ( 0.2) × 10-3 s-1 for [tht]added
)
(1.54 ( 0.04) × 10-4 mol L-1. Clearly, tht retards the
Au-catalyzed isomerization. Experiments carried out
with 0.6 mol % of catalyst 3a reveal that the reaction is
minus first order with respect to the concentration of
tht.15 The plot of r0 vs [tht]added is a straight line
-1
Ta ble 2. Tem p er a tu r e Dep en d en ce of th e
Equ ilibr iu m Con sta n t Keq for th e Isom er iza tion of
tr a n s-[P d (C6Cl2F 3)2(th t)2] (1a ) to
(Figure 3) whose slope is (1.27 ( 0.04) × 1010 L2 s mol-2
-1
and the intercept is very close to the value of r0
a
measured without addition of tht. The dependence of
kcat on the concentration of added tht is given in eq 6,
with a ) (1.32 ( 0.07) × 10-4 s-1 and b ) (3.0 ( 0.2) ×
10-5 mol L-1, at 304.4 K.
cis-[P d (C6Cl2F 3)2(th t)2] (2a ) in CDCl3
T/K
Keq
332.5
326.0
320.6
316.5
311.9
306.0
300.4
296.1
22.0
23.5
25.2
27.1
30.2
32.7
36.0
37.4
a
kcat
)
(6)
b+[tht]
Equations 5 and 6 are consistent with the first-order
kinetics observed in the experiments carried out with
added tht. In the absence of added tht, the catalyst
decomposition makes the term [3a ] in eq 5 decrease with
time, diminishing kiso along the isomerization.13
a
Catalyzed by [Au(C6Cl2F3)(tht)] (3a ). [1a ]0 ) [3a ] ) (1.00 (
0.03) × 10-2 mol L-1
.
Ar yl-Gr ou p Scr a m blin g betw een Au (I) a n d P d -
(II) d u r in g th e Au (I)-Ca ta lyzed Isom er iza tion .
Experiments carried out mixing C6Cl2F3 and C6F5
complexes have revealed that the Au(I)-catalyzed isomer-
ization takes place with aryl exchange between Au(I)
and Pd(II). For example, the reaction between trans-
[Pd(C6Cl2F3)2(tht)2] (1a ) and [Au(C6F5)(tht)] (3b) (1:1)
produces cis-[Pd(C6Cl2F3)2-n(C6F5)n(tht)2] (n ) 0 (2a ), 1
(2c), 2 (2b)) and [Au(C6Cl2F3)(tht)] (3a ). Concentra-
tion-time profiles of this reaction, retarded by addition
of tht to make 19F NMR monitoring possible, are shown
in Figure 4.
The temperature dependence of the isomerization rate
has been studied in the absence of added tht and is
analyzed later (see Discussion). Finally, the tempera-
ture dependence of the equilibrium constant Keq ) [2a ]/
[1a ] in CDCl3 has been also measured using 3a -catalysis
in order to reach equilibrium within minutes. The
results are given in Table 2 and afford the following
thermodynamic parameters: ∆H° ) -12.7 ( 0.5 kJ
mol-1 and ∆S° ) -12.7 ( 1.6 J K-1 mol-1
.
(15) With 0.6 mol % of 3a , the catalyzed pathway contributes 99.4%
to the total isomerization rate, then kiso ≈ kcat.

3680 Organometallics, Vol. 17, No. 17, 1998
Casado and Espinet
Sch em e 2
F igu r e 4. Concentration-time plots for the reaction
between trans-[Pd(C6Cl2F3)2(tht)2] (1a ) with [Au(C6F5)(tht)]
(3b) in the presence of tht (6 × 10-2 mol L-1) in CDCl3 at
322.6 K. The products are cis-[Pd(C6Cl2F3)2-n(C6F5)n(tht)2]
(n ) 0 (2a ), 1 (2c), 2 (2b)), trans-[Pd(C6Cl2F3)(C6F5)(tht)2]
(1c), and [Au(C6Cl2F3)(tht)] (3a ).
Reaction s between th t an d Au (I) Catalyst. [Au(C6-
Cl2F3)(tht)] (3a ) was checked for possible reactions with
tht which could inhibit its catalytic activity. The
addition of free tht to 3a (2:1 mol, in CDCl3) produced
Ta ble 3. P r od u ct Distr ibu tion a t Equ ilibr iu m of
th e Rea ction in F igu r e 4
complex
cfound/10-3 mol L-1
ccalcd/10-3 mol L-1
1a
1b
1c
2a
2b
2c
3a
3b
0.18
∼0
0.17
0.04
0.17
4.3
1.1
4.3
1
fluxionality in the H NMR spectrum, indicating a fast
free-for-coordinated tht exchange, but not in the 19F
NMR spectrum. This result discards the formation of
any [Au(C6Cl2F3)(tht)n] (n > 1) in appreciable amount.
Thus, the inhibition observed under catalytic conditions
(where the concentration of free tht is much lower) is
not due to the formation of three-coordinate gold(I)
species.
0.17
4.4
1.2
4.1
6.6
3.4
6.7
3.3
Complexes 2c and 3a are initially formed by the aryl
exchange shown in eq 7.16 Once some 3a has been
Ar yla t ion of Ca t ion ic Au (I) b y An ion ic P d (II)
Com p lexes. To support a step in our mechanistic
proposal, the reaction of (NBu4)[Pd(C6Cl2F3)3(tht)] (NBu4‚
4) and [Au(tht)2](ClO4) (5‚ClO4) in CDCl3 was studied.
As expected from closely related studies in the litera-
ture,18 the anionic Pd complex arylated the cationic Au-
(I) one (eq 9).
(7)
formed, it can react with 1a to give 2a (as in eq 8). The
(9)
(8)
Discu ssion
rest of the products are generated by similar exchange
processes. Thus, from this experiment it is clear that
the Au(I)-catalyzed isomerization in Pd(II) complexes
takes place with simultaneous aryl exchange between
both organometallics.
At the end of the reaction in Figure 4, the concentra-
tions of the complexes correspond to a statistical dis-
tribution of the aryls between Pd(II) and Au(I) (Table
3),17 as expected from the chemical similarity of the C6-
Cl2F3 and C6F5 groups.
The spontaneous isomerization of trans-[Pd(C6F5)2-
(tht)2] (1b) to cis-[Pd(C6F5)2(tht)2] (2b) has been reported
and is slow. A dissociative mechanism (Scheme 2) has
been proposed from the retardation effect of the addition
of tht and from the values of the activation param-
eters: ∆Hq ) 137 ( 6 kJ mol-1 and ∆Sq ) 83 ( 19 J
K-1 mol-1 10 The very positive ∆Sq value agrees with
.
an increase in the number of particles through the
isomerization. Nevertheless, as we have recently
shown,12 the high value of ∆Hq is mainly contributed
by the topomerization step of the three-coordinate
intermediate.19
The Au(I)-catalysis of this cis-trans isomerization is
an unreported process. Its most significant features are
(i) The Au(I)-catalyzed isomerization involves aryl ex-
change between Pd(II) and Au(I); and (ii) there is a
retardation effect by addition of neutral ligand. The
(16) The cis-trans configuration of heteroaryl Pd(II) complexes 1c
and 2c have been assigned by comparison of their 19F NMR chemical
shifts with those of the homoaryl complexes cis (1a ,b) and trans (2a ,b).
Moreover, in the case of the cis complex 2c, 19F-19F through-space
couplings between the C6Cl2F3 and C6F5 rings produce a typical fine
structure in the Fortho resonances, see: Albe´niz, A. C.; Casado, A. L.;
Espinet, P. Organometallics 1997, 16, 5416-5423.
(17) The statistical distribution in Table 3 has been calculated by
resolving
a system with eight variables (concentration of eight
complexes) and eight relationships (first, balance of Pd; second, balance
of Au; third, balance of C6Cl2F3 group; fourth to sixth, Kiso ) [cis]/[trans]
) 97/3, for each pair of cis-trans isomers; and seventh to eighth, Kex
) [hetero]2/([homo-C6F5][homo-C6Cl2F3]) ) 4, for cis and trans
configurations.
(18) Uso´n, R.; Fornie´s, J .; Toma´s, M.; Ara, I.; Casas, J . M.; Mart´ın,
A. J . Chem. Soc., Dalton Trans. 1991, 2253-2264.
(19) Tatsumi, K.; Hoffmann, R.; Yamamoto, A.; Stille, J . K. Bull.
Chem. Soc. J pn. 1981, 54, 1857-1867.

Isomerization of trans-[PdR2L2] Complexes
Organometallics, Vol. 17, No. 17, 1998 3681
Sch em e 3
F igu r e 5. Simplified drawings for the X-ray difraction
structures of (a) [(PEt3)2C6F5)Pt(µ-H)Au(PPh3)]+ and (b)
[(tht)(C6Cl5)(C6F5)2PtfAg(PPh3)]. From refs 22 and 18,
respectively.
F igu r e 6. Eyring representation of k1 values derived from
the [Au(C6Cl2F3)(tht)] (3a )-catalyzed isomerization of trans-
[Pd(C6Cl2F3)2(tht)2] (1a ) to cis-[Pd(C6Cl2F3)2(tht)2] (2a ).
[2a ]0 ) (1.00 ( 0.03) × 10-2 mol L-1, [3a ] ) (9.9 ( 0.5) ×
10-6 mol L-1 (0.1%), CDCl3.
Also, the activated complex B has an even closer model
characterized by X-ray diffraction, the complex [(tht)-
(C6Cl5)(C6F5)2PtfAg(PPh3)] (Figure 5b), where there is
a donor-acceptor PtfAg bond.18 For both kinds of
complexes, a greatly reduced stability is expected when
Pt is substituted for Pd and in fact the gold complex 5+
is arylated by the anionic tris-arylated Pd complex 4-,
as shown in eq 9.
Ch a r t 1
The mechanism in Scheme 3 leads to the theoretical
rate law given in eq 10 (see Appendix). Comparing both
k1k2[3a ]
kcat
)
(10)
k2 + k-1[tht]
mechanism outlined in Scheme 3 is consistent with
these observations. It starts with a nucleophilic attack
of the electron-rich complex 3a to 1a (Chart 1), to
produce an L-for-R2 associative substitution, leading to
an aryl-bridged intermediate A (step k1).20 In the
coordination sphere of Pd, the fragment Au(tht) can
slide from one R2 to any R1 (step k2), most probably
through a high-energy intermediate or activated com-
plex B, to give C, which is finally cleaved by tht (step
k3) producing the net effect of a change of the geometry
at the Pd center with aryl exchange.
The existence of some kind of intermediate A (and C,
both in undetectable concentration) is required by the
kinetic results.21 The structure proposed is identical to
that found by X-ray diffraction for the hydrido-bridged
cation [(PEt3)2(C6F5)Pt(µ-H)Au(PPh3)]+ (Figure 5a).22
equations, the following values are obtained: k1 ) 4.4
( 0.3 mol-1 L s-1 and k2/k-1 ) (3.0 ( 0.2) × 10-5 mol
L-1 in CDCl3 at 304.4 K.
In the absence of added tht and neglecting the small
contribution of the uncatalyzed isomerization,15 the
reaction rate can be simplified to eq 11, where the
isomerization rate seems to be controlled by the bimo-
lecular interaction between Au and Pd complexes (step
k1 in Scheme 3). Applying eq 12, values of k1 can be
estimated from those of r0 measured in the absence of
added tht (Table 1).
riso ) kcat[1a ] ) k1[3a ][1a ]
(11)
r0
k1 )
(12)
(20) Associative ligand substitutions on Pd(II) species take usually
place with preservation of the geometry, see: Cross, R. J . Adv. Inorg.
Chem. 1989, 34, 219-292.
(21) A direct transformation through a single activated complex (for
instance a species with the Au and Pd centers bridged by the two
exchanging groups) would yield a rate law independent of [L].
(22) Albinati, A.; Lehner, H.; Venanzi, L. M.; Wolfer, M. Inorg. Chem.
1987, 26, 3933-3939.
[3a ][1a ]0
The activation parameters derived from a Eyring
treatment of these k1 values (Figure 6) are ∆H1q ) 56.4
( 1.6 kJ mol-1 and ∆S1q ) -46 ( 6 J K-1 mol-1. They
are in sharp contrast with those found for the spontane-

3682 Organometallics, Vol. 17, No. 17, 1998
Casado and Espinet
1.01 (t, CH3); 19F NMR (282 MHz, CDCl3) δ -81.56 (q,
ous isomerization and clearly indicate a bimolecular
step, k1, as postulated in Scheme 3. A classical associa-
tive substitution of tht in 1a (k1) through a pentacoor-
dinate activated complex (Chart 1) is compatible with
the negative activation entropy and the low activation
enthalpy observed.23
through-space
through-space
J
) 6.8 Hz, 2F, o-CF), -83.13 (t,
J
)
FF
FF
6.8 Hz, 4F, o-CF), -118.99 (s, 2F, p-CF), -120.52 (s, F, p-CF).
Anal. Calcd for C38H44Cl6F9PdNS: C, 44.02; H, 4.28; N, 1.35.
Found: C, 44.23; H, 4.32; N, 1.41.
Kin etics on th e [Au (C6Cl2F 3)(th t)]-Ca ta lyzed Isom er -
iza tion of tr a n s-[P d (C6Cl2F 3)2(th t)2]. NMR tubes (5 mm)
were charged with 1a (4.1 ( 0.1 mg, 6.00 ( 0.15 µmol) and
suitable aliquots of two CDCl3 solutions, one with tht ((5.80 (
0.10) × 10-4 mol L-1) and the other with 3a ((5.91 ( 0.03) ×
10-4 mol L-1). Both solutions had been previously titrated by
1H NMR using naphthalene as the internal standard. The
mixtures were dissolved, in CDCl3 at room temperature (293
K), to a fixed volume of 600 ( 5 µL and placed into a
thermostated probe. Concentration vs time data were then
acquired by comparing the 19F NMR signal areas of complexes
1a and cis-[Pd(C6Cl2F3)2(tht)2] (2a ) (Figure 1). Attempts at
fitting the data points to a first-order law, eq 4, were satisfac-
tory only for experiments carried out with free tht. Alterna-
tively, initial reaction rates (in mol L-1 s-1) were measured
by fitting the initial data points (up to 10% of conversion) to a
Taylor equation [1a ] ) a0 + a1t + a2t2, from which the initial
reaction rate is r0 ) (∂[1a ]/∂t)t)0 ) a1. Measurements of the
equilibrium constant, Keq ) [2a ]/[1a ], were carried out on a
Con clu sion s
In this paper, we have described a novel aryl exchange
between [AuRL] and [PdR2L2] complexes (R ) perh-
alophenyl, L ) neutral ligand), which takes place with
cis-trans isomerization of the latter. The mechanism
involves associative substitution of the neutral ligand
L in trans-[PdR2L2] by the nucleophilic Au(I) complex
and formation of an aryl-bridged intermediate trans-
[LR2Pd(µ-R)AuL]. The subsequent AuL-fragment mi-
gration to terminal R groups leads to cis-[LR2Pd(µ-
R)AuL], which is finally cleaved by free L yielding cis-
[PdR2L2]. This novel isomerization pathway is enabled
by the use of the electron-rich Au(I) center, which makes
its R group nucleophilic enough to initiate the reaction
via nucleophilic attack to the electrophilic Pd center.
samples with [1a ]0 ) [3a ] ) (1.00 ( 0.03) × 10-2 mol L-1
.
Rea ction s of tr a n s-[P d (C6Cl2F 3)2(th t)2] (1a ) w ith [Au -
(C6F 5)(th t)] (3b). A sample with [1a ]0 ) [3b] ) (1.00 ( 0.03)
× 10-2 mol L-1 and [tht] ) 6 × 10-2 mol L-1, prepared as
described above, was allowed to react at 322.6 K. The reaction
was followed by 19F NMR, calculating the concentrations of
the products (Figure 4). 19F NMR (282 MHz, CDCl3) data for
heteroaryl products are given.9 1c: δ -92.67 (s, o-C6Cl2F2F),
-117.10 (s, p-C6Cl2F2F), -119.34 (m, o-C6F2F3), -158.90 (m,
p-C6F4F), -161.63 (m, m-C6F2F3). 2c: δ -89.83 (m, o-C6-
Cl2F2F), -116.34 (m, o-C6F2F3), -118.20 (s, p-C6Cl2F2F),
-160.18 (m, p-C6F4F), -162.84 (m, m-C6F2F3).
Exp er im en ta l Section
General methods were as reported elsewhere.9 The com-
plexes [AuCl(tht)],24 trans-[Pd(C6Cl2F3)2(tht)2] (1a ),25 trans-[Pd-
(C6F5)2(tht)2] (1b),26 cis-[Pd(C6Cl2F3)2(tht)2] (2a ),25 cis-[Pd-
(C6F5)2(tht)2] (2b ),26 [Au(C6F5)(tht)] (3b ),27 (NBu4)2[Pd(C6-
Cl2F3)4],25 [Au(tht)2](ClO4) (5),28 and [Li(C6Cl2F3)],25 were pre-
pared as reported in the literature.
[Au (C6Cl2F 3)(th t)] (3a ). To a solution of [Li(C6Cl2F3)] (5.15
mmol) in diethyl ether (80 mL) at -78 °C was added finely
ground [AuCl(tht)] (1.50 g, 4.68 mmol). The mixture was
stirred for 2 h, allowing the temperature to increase slowly.
Stirring was continued at room temperature for an additional
15 min. The resulting white suspension was treated with
water (60 mL), and the organic layer was separated, washed
with water (2 × 20 mL), dried over anhydrous MgSO4, and
concentrated to ca. 5 mL. Upon cooling at -28 °C, white
needles of 3a separated, which were filtered, washed with cold
diethyl ether (3 × 1 mL), and air-dried (yield 1.98 g, 87%): IR
(KBr) 1435 (vs), 1403 (vs), 1055 (s), 1037 (s), 773 (vs), 713 (m),
704 (m); 1H NMR (300 MHz, CDCl3) δ 3.44 (s, br, SCH2), 2.23
(s, br, CCH2); 19F NMR (282 MHz, CDCl3) δ -90.28 (s, o-CF),
-116.62 (s, p-CF). Anal. Calcd for C10H8AuCl2F3S: C, 24.76;
H, 1.66. Found: C, 24.84; H, 1.73.
Rea ction s of [Au (C6Cl2F 3)(th t)] (3a ) w ith th t. A (10.0
( 0.03) × 10-3 mol L-1 sample in 3a and (5.0 ( 0.2) × 10-3
mol L-1 in tht, prepared as described above, was examined by
NMR. The 19F NMR spectrum showed only signals from 3a
without changes in δ. The 1H NMR (300 MHz, CDCl3)
spectrum showed very broad (fluxional) signals at δ 2.83
(SCH2), 1.95 (CCH2).
Er r or An a lysis. Errors were estimated as reported before.9
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.
(NBu 4)[P d (C6Cl2F 3)3(t h t )] (NBu 4‚4).
A solution of
(NBu4)2[Pd(C6Cl2F3)4] (204 mg, 0.147 mmol) and 1a (110, 0.161
mmol) in THF (6 mL) was stirred at 50 °C for 5 h. The yellow
solution was filtered to remove traces of black palladium, and
evaporated to dryness. The residue was treated with diethyl
ether (5 mL) to give a white solid (NBu4).4 which was
separated, washed with diethyl ether (2 × 1 mL) and air-dried
(yield 0.204 g, 67%): IR (KBr) 2965 (s), 1425 (vs), 1396 (vs),
1040 (vs), 770 (vs); 1H NMR (300 MHz, CDCl3) δ 3.25 (m, CH2),
2.64 (m, SCH2), 1.82 (m, SCCH2), 1.70 (m, CH2), 1.45 (m, CH2),
Ap p en d ix
Der iva tion of Ra te Eq 10. The steady-state con-
centration of intermediates A and C in Scheme 3 are
k1[3a ][1a ] + k-2[C]
[A] )
[C] )
(13)
(14)
k-1[tht] + k2
(23) Other observations further support the proposed mechanism.
For example, isomerizations on similar Pt(II) complexes are much
slower than for Pd, in agreement with the different behavior of these
two metals in associative substitutions (see ref 20). This and other
results will be reported in a forthcoming paper.
k2[A] + k-3[2a ][3a ]
k3[tht] + k-2
(24) Uso´n, R.; Laguna, A.; Laguna, M. Inorg. Synth. 1989, 26, 86.
(25) Espinet, P.; Mart´ınez-Ilarduya, J . M.; Pe´rez-Briso, C.; Casado,
A. L.; Alonso, M. A. J . Organomet. Chem. 1998, 551, 9-20.
(26) Uso´n, R.; Fornie´s, J .; Mart´ınez, F.; Toma´s, M. J . Chem. Soc.,
Dalton Trans. 1980, 888-893.
k1(k-2 + k3[tht])[1a ][3a ] + k-2k-3[2a ][3a ]
{k1(k-2 + k3[tht])}[tht]
[A] )
(27) Uso´n, R.; Laguna, A.; Vicente, J . J . Organomet. Chem. 1977,
131, 471.
(15)
(28) Uso´n, R.; Laguna, A.; Laguna, M.; J ime´nez, J .; Go´mez, M. P.;
Sainz, A.; J ones, P. G. J . Chem. Soc., Dalton Trans. 1990, 3457-3463.
The reaction rate is

Isomerization of trans-[PdR2L2] Complexes
Organometallics, Vol. 17, No. 17, 1998 3683
our measurements have been made at the early stages
of the isomerization of 1a to 2a . This means that [2a ]
is small and k-1k-2 , k2k3, giving
∂[1a ]
∂t
riso,cat ) -
) k1[1a ][3a ] - k-1[A]
(16)
(17)
k1k2k3[1a ][3a ] - k-1k-2k-3[2a ][3a ]
k-1k-2 + k2k3 + k-1k3[tht]
riso,cat
)
k1k2[3a ]
riso,cat ) kcat[1a ] )
[1a ]
(10)
k2 + k-1[tht]
Equation 17 holds for the reversible reaction. How-
ever, it can be simplified to an irreversible isomerization
of 1a since eq 3 is shifted far toward the cis isomer and
OM980283S