Acceleration of reductive elimination of [Ar-Pd-Csp3] by a phosphine/electron-deficient olefin ligand: A kinetic investigation

Article abstract of DOI:10.1002/chem.200802209

The kinetics of the reductive elimination step of a C_sp3-C _sp2 Negishi cross-coupling catalyzed by a 1:1 complex 2 of palladium and the phosphine/ electron-deficient olefin ligand (E)-3-(2- diphenylphosphanylphenyl)-1 -phenyl-propeno

Full text of DOI:10.1002/chem.200802209

FULL PAPER
DOI: 10.1002/chem.200802209
Acceleration of Reductive Elimination of [Ar-Pd-C ] by a Phosphine/
3
Electron-Deficient Olefin Ligand: A Kinetic sIpnvestigation
Heng Zhang,[a] Xiancai Luo,[a] Kittiya Wongkhan,[b] Hui Duan,[a] Qiang Li,[a]
Lizheng Zhu,[a] Jian Wang,[c] Andrei S. Batsanov,[b] Judith A. K. Howard,[b]
Todd B. Marder,*[b] and Aiwen Lei*[a, d]
Abstract: The kinetics of the reductive
first order in [Pd] is consistent with ho-
mogeneous catalysis by Pd complexes
rather than by Pd nanoparticles (NPs).
Through systematic kinetic investiga-
tions of the Negishi coupling of ethyl
2-iodobenzoate with alkylzinc chlo-
(dppbz=1,2-bis(diphenylphosphino)-
benzene). The use of a 2:1 ratio of
1:Pd resulted in reduced catalytic activ-
ity and selectivity, presumably because
the olefin moiety could no longer assist
in the reductive elimination step. Im-
portantly, hydrogenation of the C=C
double bond in ligand 1 generated a
ꢀ
elimination step of a Csp3 Csp2 Negishi
cross-coupling catalyzed by a 1:1 com-
plex 2 of palladium and the phosphine/
electron-deficient olefin ligand (E)-3-
(2-diphenylphosphanylphenyl)-1-
phenyl-propenone (1) was studied.
Complex 2 is an exceptionally efficient
and highly selective catalyst for Negishi
cross-coupling reactions involving pri-
mary and secondary alkylzinc reagents
bearing b-hydrogen atoms. Turnover
numbers (TONs) as high as 105 and
turnover frequencies (TOFs) as high as
1000 sꢀ1 were observed. The reactions
occurred rapidly and selectively even
at 08C. The fact that the reaction was
ACHTUNGTRENNrUNG ides, the rate constants for reductive
elimination of [Ar-Pd-Csp3] were deter-
mined to be >0.3 sꢀ1, which is about 4
or 5 orders of magnitude greater than
the values previously reported for [Pd-
satACTHUNTRGNENGuU rated ligand (1H2), which was not
only less effective than 1, but also gave
rise to substantial amount of ethylben-
zoate formed by competing b-hydride
elimination. Thus, the p-accepting
olefin moiety in 1 must enhance reduc-
tive elimination rates, and, consequent-
ly, inhibit formation of byproducts re-
sulting from b-hydride elimination.
ACHUTNRGENUN(G dppbz)] and [PdACHTUNTGREN(NGUN PPh3)2] systems
Keywords: cross-coupling · kinetics ·
olefins · palladium · reductive elimi-
nation
Introduction
[a] Dr. H. Zhang, X. Luo, H. Duan, Q. Li, L. Zhu, Prof. Dr. A. Lei
College of Chemistry and Molecular Sciences
Wuhan University, 430072 (P. R. China)
Fax : (+86)27-68754067
Transition-metal-catalyzed cross-coupling reactions are
widely used in organic synthesis.[1,2] However, the transfor-
mations involving alkyl groups remain problematic due to
E-mail: aiwenlei@whu.edu.cn
ꢀ
ꢀ
ꢀ
slow reductive elimination (i.e., Csp2 Csp2 >Csp2 Csp3 >Csp3
Csp3) with deleterious but fast b-H elimination.[3–5] Culkin
and Hartwig investigated the reductive elimination of com-
plex A through stoichimetric reactions (Scheme 1), and
found the half life of the complex at 408C to be 23 min, cor-
[b] K. Wongkhan, Dr. A. S. Batsanov, Prof. Dr. J. A. K. Howard,
Prof. Dr. T. B. Marder
Department of Chemistry
Durham University, South Road
Durham DH1 3 LE (UK)
E-mail: todd.marder@durham.ac.uk
responding to
a
rate constant of about 5ꢀ10ꢀ4 sꢀ1
[c] Dr. J. Wang
(Scheme 1).[6] Espinet et al. reported the rate constant for
reductive elimination of complex B to be 8.9ꢀ10ꢀ6 sꢀ1 at
258C (Scheme 1).[7] Some progress has been made to pro-
mote the rate of oxidative addition by employing sterically
hindered and/or electron-rich ligands.[4,8–17] Although the
steric hindrance assists the reductive elimination, the elec-
tronic factors aimed at facilitating the oxidative addition[18]
inhibit the reductive elimination process.[4] Effective means
Real-Time Analytics, Mettler Toledo Auto Chem
7075 Samuel Morse Dr., Columbia, MD 21046 (USA)
[d] Prof. Dr. A. Lei
State Key Laboratory of Organometallic Chemistry
Shanghai Institute of Organic Chemistry
Chinese Acadamy of Sciences, 354 Fenglin Lu
Shanghai, 200032 (China)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802209.
Chem. Eur. J. 2009, 15, 3823 – 3829
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3823

Scheme 1. Selected rate constants for reductive elimination.
to accelerate the reductive elimination are still scarce for
cross-coupling reactions involving alkyl groups.
Additives with p-acceptor properties (e.g., maleic anhy-
dride, fumaronitrile, p-fluorostyrene, and other olefins) are
known to accelerate reductive elimination in stoichiometric
reactions.[19–23] They have seldom been used as ligands in cat-
Figure 1. X-ray molecular structure of 3·2CH2Cl2. The Pd atom lies at an
inversion center and has square-planar coordination. Bond lengths (ꢂ):
ꢀ
ꢀ
Pd Cl(1) 2.2973(5), Pd P 2.3425(6). Thermal ellipsoids are shown at the
50% probability level.
alytic processes.[24–33] Recently, we reported an efficient Csp3
-
involved Negishi-coupling catalyzed by a complex of palladi-
um and the phosphine/electron-deficient olefin ligand 1.[34]
that we formulate as a monophosphine, Cl-bridged dimer or
oligomer on the basis of solubility, elemental analysis, solid-
state 31P NMR and IR spectroscopy, the last of which shows
no sign of alkene bonding to Pd. In contrast, reaction of
[PdACHTNUTGRNEUNG
(dba)2] with two equivalents of 1 gives a novel Pd0 spe-
cies 4 (Figure 2), in which each of two molecules of 1 are
both P- and p-olefin-bound to Pd. Interestingly, in the solid-
state structure of 4, the metal coordination by two P atoms
The reaction proceeds exceptionally cleanly, providing a
good model for kinetic study. In an attempt to gain a better
understanding of the unique role of the ligand, we conduct-
ed a series of quantitative kinetic investigations. Herein, we
report the finding that ligand 1 dramatically enhances the
rate of reductive elimination of a [Ar-Pd-Csp3] intermediate,
thus making alkyl–aryl coupling reactions possible with high
efficiency and selectivity, that is, without competing b-hy-
dride elimination.
Results and Discussion
Structures of Pd/phosphine/electron-deficient olefin ligand
complexes: Reaction of [PdCl2ACHTUNRTGNEUNG(PhCN)2] with two equiva-
lents of 1 gives 3 in excellent yield (Figure 1, for details see
Supporting Information); however, 1:1 reaction gives 2 in
solution, which rapidly forms a much less soluble complex
Figure 2. X-ray molecular structure of 4 in 4·0.5CH2Cl2. Bond lengths in
ꢀ
ꢀ
the two independent molecules (ꢂ): Pd P(1) 2.3776(9) [2.3673(8)], Pd
ꢀ
ꢀ
P(2) 2.3707(8) [2.3545(9)], Pd C(7) 2.173(3) [2.203(3)], Pd C(8) 2.289(3)
ꢀ
ꢀ
[2.265(3)],
Pd C(34)
2.161(3)
[2.144(3)],
Pd C(35)
2.277(3)
[2.276(3)].Thermal ellipsoids are shown at the 50% probability level.
3824
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 3823 – 3829

Reaction Mechanisms
FULL PAPER
and two h2-olefinic bonds is intermediate between square-
planar and tetrahedral, which would be expected for Pd0
complexes. The plane defined by the Pd atom and the mid-
points of both h2-bonds forms a dihedral angle of 51.8 or
53.18 with the PdP2 plane (cf. 0 for planar and 908 for tetra-
hedral coordination). The olefinic C7=C8 bond lengthens
from 1.330(3) ꢂ in 3 to the average of 1.406(4) ꢂ in 4 due
ꢀ
to metal coordination. The adjacent C2 C7 bond lengthens
from 1.467(3) to 1.499(4) ꢂ as the change of the C1-C2-C7-
C8 torsion angle from 174.2(3)8 in 3 to 57.4(4)–78.1(4)8 in 4
breaks the p-conjugation. The angle between aryl planes i
and ii increases from 248 in 3 to 67–808 in 4.[35] For details of
the crystal structure and spectroscopic characterization, see
Supporting Information.
High-turnover-number (TON) and low-temperature experi-
ments: Figure 3 shows the typical concentration profiles
versus time for the reaction [Eq. (1)] of ethyl 2-iodoben-
Figure 4. Reaction of ArI 6 and CyZnCl (7) catalyzed by 2 (0.001m) at
08C and 158C monitored by ReactIRTM. A) [6] versus time. Linear fit of
[6] versus time: between 400–600 s (08C); between 130–200 s (158C).
B) Linear fit of ln[6] versus time: between 400–900 s (08C); between
130–330 s (158C).
ployed in enabling the oxidative addition, transmetallation
and Csp3-involved reductive elimination.
Figure 3. Kinetic profile of reaction of ArI 6 and CyZnCl (7) catalyzed
by 2 (0.001 mol%) at 258C monitored by ReactIRTM
.
The effectiveness of the olefin moiety in ligand 1: The major
difference between our catalyst system and the systems re-
ported previously (Scheme 1) was the ligand on the palladi-
um center. Both of the reported cases employed two phos-
phine ligands, that is, the bidentate chelate, bis(diphenyl-
phosphino)benzene[6] or PPh3 (2 equiv).[7] Our ligand con-
sists of an electron-deficient olefin in addition to a phos-
phine moiety.
zoate (6; 100 mmol) with cyclohexylzinc chloride (7;
500 mmol) at 258C, producing 8 at a final yield of 99% (de-
termined by GC). The reaction was catalyzed by
2
(0.001 mol%). The turnover frequencies (TOFs) were as
high as 103 sꢀ1. Encouraged by the large TON and TOF
values, further kinetic studies were carried out on this Ne-
gishi coupling.
To elucidate the role of the electron-deficient olefin
moiety in the reaction, we examined the coupling of 6 with
9 catalyzed by 1 mol% of complex 3 generated in situ from
[PdCl2ACHTNUGTRNEUNG(CH2CN)2] and ligand 1 (1:2). As expected, and in
contrast to catalysis by 2 [Eq. (2)], formation of the desired
cross-coupling product 10 was hampered by the presence of
the second equivalent of ligand 1, with only 33% conversion
of 6 after 80 min and a molar ratio of ethyl benzoate to 10
of 1.4:1. In addition, ligand 1H2 was prepared by reduction
of the olefin group of ligand 1, and was utilized in the cata-
lytic coupling of 6 with 9. The reaction finished after
100 min, but showed poor selectivity, with a molar ratio of
ethyl benzoate to 10 of 2:1, indicating that b-hydride elimi-
nation is a significant competing process when 1H2 is em-
At 08C, with catalyst 2 (0.5 mol%), the coupling of 6
(1.0 mmol) and 7 (2.25 mmol) finished in 17 min (Figure 4A,
squares). The same reaction at 158C completed within 5 min
(Figure 4 A, circles). The high reactivity and selectivity of
this reaction demonstrated the capability of the catalyst em-
Chem. Eur. J. 2009, 15, 3823 – 3829
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3825

T. B. Marder, A. Lei et al.
ployed, again, in contrast to ligand 1, for which the high se-
lectivity for 10 clearly indicates that reductive elimination
must be much faster than b-hydride elimination.
Catalysis by Pd0 complexes or Pd nanoparticles (NPs): Lit-
erature reports indicate that an inverse order in [Pd] will be
observed if a reaction involves the formation of Pd NPs.[36–40]
To clarify whether the above reactions were catalyzed by a
Pd0 complex or Pd NPs, we carried out kinetic experiments
using different Pd loadings (from 0.25 mol% to 2 mol%),
with all other factors held constant. The results are shown
on Figure 5. Clearly, we can see that the reaction is first-
Figure 6. Reactions of ArI 6 with RalkylZnCl 9 catalyzed by 2 or Pd
ACHTUNGTRENNUNG(OAc)2
at 258C monitored by ReactIRTM. For both experiments before ArI 6
was added to initiate reaction, 0.3 equiv PPh3 (vs. 2 or PdACHTNUTRGNENG(U OAc)2) was
added and mixing for 5 min.
transitioning into a zero-order kinetic regime. Such an in-
duction period may be explained by slow generation of the
active catalytic species. The plot of concentration versus
time for the 08C experiment (Figure 4A) indicates clearly
that the rate continued to increase sharply until about 33%
conversion, then the rate was almost unchanged until about
79% conversion. The same trend was found for the 158C
experiment.
Due to the long induction period of the reaction between
6 and 7, the global kinetics was difficult to study. The reac-
tion of 6 with 9 was investigated [Eq. (2)]. Because of the
conformational restriction of a cyclohexyl group, the reduc-
tion of PdII to Pd0 by 7 might be a slower process compared
to that by alkylzinc 9. The reaction progress was monitored
in situ by ReactIRTM. Pleasantly, the induction period disap-
peared as expected (Figure 7A). This result further support-
ed the hypothesis that the observed induction period was
due to the slow formation of an active catalyst, through in-
teraction between 2 and the alkylzinc reagent.
Figure 5. The kobs of reactions between ArI 6 and RalkylZnCl 9 catalyzed
by 2 (0.0005, 0.001, 0.002, 0.004m) at 158C monitored by ReactIRTM
.
A distinct kinetic feature of this catalytic reaction is the
existence of a zero-order kinetic regime, with or without the
induction period (e.g., Figure 4A (after the induction
period) and Figure 7A). As shown in Scheme 2, the Negishi
reaction follows a general pathway involving sequential oxi-
dative addition (kOA), transmetallation (kTM), and reductive
elimination (kRE). Within the zero-order kinetic regime, the
reaction rate is independent of the concentrations of both
starting materials (i.e., ArI and RZnX). The fact that the
rate appears to be zero-order in both [ArI] and [RZnX] be-
tween certain conversions, indicates that neither the oxida-
tive addition nor the transmetallation was the rate-determin-
ing step. Therefore, within this zero-order kinetic regime,
the reaction rate could be limited by the reductive elimina-
tion step (kRE).[43,44] The rate constants for reductive elimina-
tion kRE were then obtained from fitting the experimental
data, and the results are listed in Tables 1 and 2. The rate
constants for reductive elimination in the reaction of 6 with
7 were 0.472, and 1.46 sꢀ1 at 08C and 158C, respectively. The
rate constant for reductive elimination in the reaction 6 with
9 was about 0.3 sꢀ1 at 158C (Table 2). These rate constants
order in [Pd]. In addition, the intercept is almost zero,
which strongly indicates that the reaction is taking place
mainly through a single pathway. This is consistent with ho-
mogeneous catalysis by Pd complexes rather than by Pd
NPs. Ligand inhibition experiments are another way to iden-
tify catalytic processes involving NPs, in which less than one
equivalent of ligand (vs. catalyst) can completely shut down
the reactivity. Thus, we examined the reaction in the pres-
ence of catalyst 2 (1 mol%) and PPh3 (0.3 mol%), and
found that the reaction was somewhat slower (Figure 6). In
comparison, when PdACHTNUGTRNEUNG(OAc)2 (1 mol%) was used as the cata-
lyst precursor, the reaction of ArI 6 and alkylzinc 9, which
normally is complete within scores of seconds, was extreme-
ly slow when 0.3 mol% PPh3 was added (Figure 6). Again,
this supports our proposal that catalyst 2 is homogeneous,
whereas PdACHTUNGTRENNUNG
(OAc)2 generates Pd NPs.[41,42]
The rates of reductive elimination determined from the
zero-order kinetic regime: Figures 3 and 4A both revealed
an induction period, during which the rate accelerated until
3826
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 3823 – 3829

Reaction Mechanisms
FULL PAPER
Table 2. Rate constants for the reaction [Eq. (2)] involving n-C12H25ZnCl
at 158C.
[a]
[ArI]0 [m] Conversion kobs
kOA [mꢀ1 sꢀ1
]
kRE [sꢀ1
]
zero
order 0.10
0.050
first
order 0.10
0.050
0.20
0–67%
0–49%
0–50%
0–95%
0–88%
0–89%
3.07ꢀ10ꢀ4 N. A.
0.307
0.299
0.327
2.99ꢀ10ꢀ4 N. A.
3.27ꢀ10ꢀ4 N. A.
3.03ꢀ10ꢀ3 3.03
4.20ꢀ10ꢀ3 4.20
6.88ꢀ10ꢀ3 6.88
0.20
@0.6[b]
@0.4[b]
@0.7[b]
[a] The unit of kobs for the zero order reaction is sꢀ1, the unit of kobs for
the first-order reaction is msꢀ1. [b] Under these conditions, OA is deemed
to be the rate-determining step; the rate of RE is much faster than OA,
so kRE is expressed as @kOA ꢀ[ArI]0.
sion of 33–98% for the experimental data of the reaction of
6 with CyZnCl 7 (Figure 4B, and Table 1). Similar kinetic
analysis was obtained for the reaction of 6 with alkylzinc 9
(Figure 7B and Table 2). In this regard, first-order kinetics
was assigned to the decline of ArI 6, which meant that oxi-
dative addition was the rate-determining step, and the rate
constants for reductive elimination should be @[6]ꢀkOA. By
linear fitting of ln[6] versus time (Figures 4B and 7B), two
sets of kRE were obtained, listed in Tables 1 and 2. Although
the exact value of kRE cannot be obtained, they are certainly
much larger than the reported numbers shown in Scheme 1.
The rates of reductive elimination determined from the
more complex kinetic regime: Although each single experi-
ment gave a good first-order fit, surprisingly, the kobs from
three runs of the reaction of 6 with 9 using three different
[6]0, which should be same, are different (Table 2). The ki-
netic data might indicate that the reaction is more complex.
If we consider that reductive elimination is not rate-limiting,
and with the steady state approximation, the rate equation
can be expressed as Equation (3) (see the Supporting Infor-
mation for detailed mathematical deduction), in which one
can see that there is no simple order for electrophile 6 or
nucleophile 9.
Figure 7. Reaction of ArI 6 with RalkylZnCl 9 catalyzed by 2 (0.001m) at
158C monitored by ReactIRTM. A) Kinetic profile of 6 versus time, treat-
ed as a zero order reaction, that is, reductive elimination is the rate-deter-
mining step. B) Treated as a first-order reaction, that is, oxidative addi-
tion is the rate-determining step.
Scheme 2. The proposed catalytic cycle.
Table 1. Rate constants of reaction [Eq. (1)] involving CyZnCl.
k1k2½ArIꢁ½RZnClꢁ½Pdꢁtotal
[a]
T [8C] Conversion [%] kobs
kOA [mꢀ1 sꢀ1
]
kRE [sꢀ1
]
ð3Þ
rate ¼
k1½ArIꢁ þ k2½RZnClꢁ
zero
order 15
first
order 15
0
33–79
35–83
33–98
35–99
4.72ꢀ10ꢀ4 N.A.
0.472
1.46ꢀ10ꢀ3 N.A.
7.08ꢀ10ꢀ3 7.08
2.14ꢀ10ꢀ2 21.4
1.46
0
@1.4[b]
@4.3[b]
To see whether catalyst deactivation was involved or not
in this reaction, according to Blackmondꢃs theory of “reac-
tion progress kinetic analysis”,[43,44] the “same excess” ex-
periments (see Supporting Information for detailed explana-
tion) were investigated. Figure 8 shows the results of these
“same excess” experiments, clearly indicating some degree
of catalyst deactivation.
[a] The unit of kobs for the zero order reaction is sꢀ1, the unit of kobs for
the first-order reaction is msꢀ1. [b] Under these conditions, oxidative ad-
dition is deemed to be the rate-determining step, the rate of reductive
elimination is much faster than oxidative addition, so kRE is expressed as
@kOA ꢀ[ArI]0.
Based on the proposed catalytic pathway in Scheme 2, as-
suming that reductive elimination is not the rate-determin-
ing step, and that there is catalyst deactivation (we regarded
this deactivation as a first-order process, with rate constant,
kdecay),[46] kinetic simulation was used to obtain the rate con-
stant.[47,48] The fitting of the reaction mechanism to the ex-
perimental curves was performed with the kinetic simulator
software Dynafit.[48] The results are listed in Figure 9 and
Table 3. The rate constant for oxidative addition (kOA) is
are far greater than those reported by both Hartwig (5ꢀ
10ꢀ4 sꢀ1)[6] and Espinet (8.9ꢀ10ꢀ6 sꢀ1),[7] which might reveal
the effects of designed ligand 1 on the reductive elimination
step by comparison with dppbz and (PPh3)2.[45]
The rates of reductive elimination determined from the
first-order kinetic regime: As Figure 4 showed, ln[6] plotted
versus time, gave a better linear fit in the range of conver-
Chem. Eur. J. 2009, 15, 3823 – 3829
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3827

T. B. Marder, A. Lei et al.
Conclusion
The results from the experiments above manifested the ef-
ꢀ
fects of the olefin part of ligand 1 in enhancing the Csp2 Csp3
cross-coupling reaction, particularly in terms of the reduc-
tive elimination rate. It is reasonable to conclude that the
olefin moiety in 1, with good p-acceptor properties, acceler-
ated the reductive elimination of an [Ar-Pd-Csp3] species,
thus resulting in the highly selective and efficient coupling
reaction. These results provide new insight into ligand
ꢀ
design for C C coupling reactions. For example, 1 is a
ꢀ
highly effective ligand for Pd-catalyzed Csp Csp cross-cou-
pling reactions.[49] Further applications of ligand 1 are under
Figure 8. The “same excess” experiments for reaction between ArI 6 and
RalkylZnCl 9 catalyzed by 2 (0.001m) at 158C.
active investigation in our laboratories.
Experimental Section
Typical procedure for ReactIRTM experiments: Compound
2
(0.001 mmol) and CyZnCl (7; 1m, 500 mL) were placed in a 3-necked
vessel, and then ethyl 2-iodobenzoate (6; 28 g, 0.1 mol) was added to ini-
tiate the reaction. The mixture was monitored by ReactIRTM at 258C.
During the reaction, aliquots were removed and analyzed by GC, and the
results were used to calibrate the ReactIRTM spectra and to obtain the
conversion and yield. At the end of the reaction, no ArI 6 was detected,
and a 99% GC yield was obtained for the cross-coupling product 8.
Acknowledgement
Figure 9. The simulated curve and the experimental data of reaction be-
This work was support by National Natural Science Foundation of China
(20502020, 20702040, 20832003) and a startup fund from Wuhan Univer-
sity. K.W. thanks the Royal Thai Government for a postgraduate scholar-
ship. Q.L. thanks the China Scholarship Council for a Fellowship to work
at Durham University. T.B.M. thanks the Royal Society of Chemistry for
a Journals Grant for International Authors and Durham University for
an HEIF Grant.
tween ArI 6 and RalkylZnCl 9 catalyzed by 2 (0.001m) at 158C.
Table 3. Rate constants determined by fitting the experimental data of
Figure 7A with Dynafit.
kOA [mꢀ1 sꢀ1
]
kTM [mꢀ1 sꢀ1
]
Kdecay [sꢀ1
]
kRE [sꢀ1
@0.5[a]
]
constant
error
10.8
0.51
2.50
0.57
2.55ꢀ10ꢀ3
2.1ꢀ10ꢀ4
[1] F. Diederich, P. J. Stang, Metal-Catalyzed Cross-Coupling Reactions,
2nd ed., Wiley-VCH, Weinheim, 2004.
[2] E.-i. Negishi, Handbook of Organopalladium Chemistry for Organic
Synthesis, Wiley, New York, 2002.
[a] Under these conditions, the rate of reductive elimination is deemed to
be much faster than oxidative addition or transmetallation, so kRE is ex-
pressed as @kOA ꢀ[ArI]0 or kTM ꢀ[RZnCl]0.
[3] D. J. Cꢄrdenas, Angew. Chem. 1999, 111, 3201–3203; Angew. Chem.
Int. Ed. 1999, 38, 3018–3020.
[4] D. J. Cꢄrdenas, Angew. Chem. 2003, 115, 398–401; Angew. Chem.
Int. Ed. 2003, 42, 384–387.
10.8 msꢀ1 and kTM is 2.5 msꢀ1. The exact value of rate con-
stant for reductive elimination was not obtained from this
[5] T.-Y. Luh, M.-k. Leung, K.-T. Wong, Chem. Rev. 2000, 100, 3187–
3204.
[6] D. A. Culkin, J. F. Hartwig, Organometallics 2004, 23, 3398–3416.
[7] J. A. Casares, P. Espinet, B. Fuentes, G. Salas, J. Am. Chem. Soc.
2007, 129, 3508–3509.
kinetic analysis, while it can be expressed as kRE @kOA
ꢀ
[ArI]0 or kRE @kTM ꢀ[RZnCl]0. It is kRE @0.5 sꢀ1, which of
course, is far greater than the reported values in
Scheme 1.[6,7]
In other words, no matter how we treated the experimen-
tal data as zero-order (the rate-determining step was reduc-
tive elimination) or first-order, or even with a more compli-
cated kinetic profile in which reductive elimination is not
rate-determining, in the two last cases, the rate constants for
reductive elimination are about four orders of magnitude
greater than that reported by Hartwig[6] and about five
orders of magnitude greater than that reported by Espinet.[7]
[8] N. Kataoka, Q. Shelby, J. P. Stambuli, J. F. Hartwig, J. Org. Chem.
2002, 67, 5553–5566.
[9] D. W. Old, J. P. Wolfe, S. L. Buchwald, J. Am. Chem. Soc. 1998, 120,
9722–9723.
[10] G. Altenhoff, R. Goddard, C. W. Lehmann, F. Glorius, Angew.
Chem. 2003, 115, 3818–3821; Angew. Chem. Int. Ed. 2003, 42, 3690–
3693.
[11] O. Navarro, R. A. Kelly III, S. P. Nolan, J. Am. Chem. Soc. 2003,
125, 16194–16195.
[12] M. Eckhardt, G. C. Fu, J. Am. Chem. Soc. 2003, 125, 13642–13643.
[13] J. E. Milne, S. L. Buchwald, J. Am. Chem. Soc. 2004, 126, 13028–
13032.
3828
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 3823 – 3829

Reaction Mechanisms
FULL PAPER
[14] B. C. Hamann, J. F. Hartwig, J. Am. Chem. Soc. 1998, 120, 7369–
7370.
[15] A. F. Littke, G. C. Fu, J. Org. Chem. 1999, 64, 10–11.
[16] C. Dai, G. C. Fu, J. Am. Chem. Soc. 2001, 123, 2719–2724.
[17] N. Hadei, E. A. B. Kantchev, C. J. OꢃBrien, M. G. Organ, Org. Lett.
2005, 7, 3805–3807.
[18] A. F. Littke, G. C. Fu, Angew. Chem. 2002, 114, 4350–4386; Angew.
Chem. Int. Ed. 2002, 41, 4176–4211.
[19] A. Yamamoto, S. Ikeda, J. Am. Chem. Soc. 1967, 89, 5989–5990.
[20] T. Yamamoto, A. Yamamoto, S. Ikeda, J. Am. Chem. Soc. 1971, 93,
3350–3359.
[21] R. Sustmann, J. Lau, Chem. Ber. 1986, 119, 2531–2541.
[22] A. E. Jensen, P. Knochel, J. Org. Chem. 2002, 67, 79–85.
[23] T. Yamamoto, M. Abla, Y. Murakami, Bull. Chem. Soc. Jpn. 2002,
75, 1997–2009.
[24] R. Giovannini, T. Studemann, G. Dussin, P. Knochel, Angew. Chem.
1998, 110, 2512–2515; Angew. Chem. Int. Ed. 1998, 37, 2387–2390.
[25] R. Giovannini, T. Stuedemann, A. Devasagayaraj, G. Dussin, P.
Knochel, J. Org. Chem. 1999, 64, 3544–3553.
R(F)=0.040 [6286 data with I=2s(I)], wR(F2)=0.113 (all data).
Crystal data for 4: C54H42O2P2Pd·0.5CH2Cl2, Mr =933.68, ortho-
rhombic, space group Pca21, a=38.814(6), b=10.691(2), c=
20.880(3) ꢂ, V=8664(2) ꢂ3; Z=8, 1calcd =1.432 gcmꢀ3
,
m=
0.61 mmꢀ1, 73864 reflections with 2q=568, 17706 unique, Rint
=
0.047, R(F)=0.031 [15582 data with I=2s(I)], wR(F2)=0.065 (all
data).
[36] M. T. Reetz, J. G. de Vries, Chem. Commun. 2004, 1559–1563.
[37] I. J. S. Fairlamb, S. Grant, P. McCormack, J. Whittall, Dalton Trans.
2007, 859–865.
[38] G. P. F. Van Strijdonck, M. D. K. Boele, P. C. J. Kamer, J. G. de Vries,
P. W. N. M. Van Leeuwen, Eur. J. Inorg. Chem. 1999, 1073–1076.
[39] I. J. S. Fairlamb, R. J. K. Taylor, J. L. Serrano, G. Sanchez, New J.
Chem. 2006, 30, 1695–1704.
[40] A. H. M. de Vries, J. M. C. A. Mulders, J. H. M. Mommers, H. J. W.
Henderickx, J. G. de Vries, Org. Lett. 2003, 5, 3285–3288.
[41] While Pd NPs are also known to be effective catalysts for this trans-
formation,[42] both the clear first-order dependence on [Pd] and the
results of PPh3 inhibition experiments (see text and Supporting In-
formation for details) ruled them out as being active species in the
present case.
[26] M. A. Grundl, J. J. Kennedy-Smith, D. Trauner, Organometallics
2005, 24, 2831–2833.
[27] A. Scrivanti, V. Beghetto, U. Matteoli, S. Antonaroli, A. Marini, B.
Crociani, Tetrahedron 2005, 61, 9752–9758.
[28] D. B. G. Williams, M. L. Shaw, Tetrahedron 2007, 63, 1624–1629.
[29] I. J. S. Fairlamb, A. R. Kapdi, A. F. Lee, G. P. McGlacken, F. Weiss-
burger, A. H. M. de Vries, L. Schmieder-van de Vondervoort, Chem.
Eur. J. 2006, 12, 8750–8761.
[30] I. J. S. Fairlamb, A. R. Kapdi, A. F. Lee, Org. Lett. 2004, 6, 4435–
4438.
[31] Y. Mace, A. R. Kapdi, I. J. S. Fairlamb, A. Jutand, Organometallics
2006, 25, 1795–1800.
[42] J. Liu, Y. Deng, H. Wang, H. Zhang, G. Yu, H. Zhang, Q. Li, T. B.
Marder, Z. Yang, A. Lei, Org. Lett. 2008, 10, 2661–2664.
[43] J. S. Mathew, M. Klussmann, H. Iwamura, F. Valera, A. Futran,
E. A. C. Emanuelsson, D. G. Blackmond, J. Org. Chem. 2006, 71,
4711–4722.
[44] D. G. Blackmond, Angew. Chem. 2005, 117, 4374–4393; Angew.
Chem. Int. Ed. 2005, 44, 4302–4320.
[45] Hartwig reported reductive elimination involving an electronic-rich
aryl group, and Espinet employed an electron-poor aryl group. Al-
though the substrates in our system, which involved an electron-
poor aryl group, were different from theirs, the data still indicate the
dramatic accelerating effect on reductive elimination.
[46] G. Rothenberg, S. C. Cruz, G. P. F. Van Strijdonck, H. C. J. Hoef-
sloot, Adv. Synth. Catal. 2004, 346, 467–473.
[32] I. J. S. Fairlamb, Org. Biomol. Chem. 2008, 6, 3645–3656.
[33] P. S. Pregosin, Chem. Commun. 2008, 4875–4884.
[34] X. Luo, H. Zhang, H. Duan, Q. Liu, L. Zhu, T. Zhang, A. Lei, Org.
Lett. 2007, 9, 4571–4574.
[35] CCDC 701728 (3) and 701729 (4) contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
[47] C. S. Consorti, F. R. Flores, J. Dupont, J. Am. Chem. Soc. 2005, 127,
12054–12065.
[48] P. Kuzmic, Anal. Biochem. 1996, 237, 260–273.
[49] W. Shi, Y. Luo, X. Luo, L. Chao, H. Zhang, J. Wang, A. Lei, J. Am.
Chem. Soc. 2008, 130, 14713–14720.
www.ccdc.cam.ac.uk/data_request/cif;
Crystal
data
for
3:
C54H42Cl2O2P2Pd·2CH2Cl2, Mr =1131.97, T=120 K, monoclinic,
space group P21/n, a=9.0882(8), b=25.761(2), c=11.379(1) ꢂ, b=
106.49(1)8, V=2554.4(4) ꢂ3, Z=2, 1calcd =1.472 gcmꢀ3
,
m=
Received: October 25, 2008
Published online: February 19, 2009
0.78 mmꢀ1, 34743 reflections with 2q=608, 7423 unique, Rint =0.033,
Chem. Eur. J. 2009, 15, 3823 – 3829
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3829