Atom-transfer radical addition reactions catalyzed by RuCp* complexes: A mechanistic study

Article abstract of DOI:10.1002/chem.200901396

Kinetic and spectroscopic analyses were performed to gain information about the mechanism of atomtransfer radical reactions catalyzed by the complexes [RuCl₂Cp*(PPh₃)] and [RuClCp*(PPh ₃)₂] (Cp* = pentamethylcyc

Full text of DOI:10.1002/chem.200901396

FULL PAPER
DOI: 10.1002/chem.200901396
Atom-Transfer Radical Addition Reactions Catalyzed by RuCp* Complexes:
A Mechanistic Study
Mariano A. Fernꢀndez-Zfflmel,[a] Katrin Thommes,[a] Gregor Kiefer,[a]
Andrzej Sienkiewicz,[b] Katarzyna Pierzchala,[b] and Kay Severin*[a]
Abstract: Kinetic and spectroscopic
analyses were performed to gain infor-
mation about the mechanism of atom-
transfer radical reactions catalyzed by
with ethyl trichloroacetate, ethyl di-
chloroacetate, or dichloroacetonitrile
were used as test reactions. The results
show that for substrates with high in-
trinsic reactivity, such as ethyl tri
ACHTUNGTRENrNUGN oACHTUNGTRENNaNUG cetate, the oxidation state of the
ACHTUNGTRENNUNGchlo-
catalyst in the resting state is +3, and
that the reaction is zero-order with re-
spect to the halogenated compound.
Furthermore, the kinetic data suggest
that the metal catalyst is not directly
involved in the rate-limiting step of the
reaction.
the complexes [RuCl2Cp*CAHTUNGTERNUN(NG PPh3)] and
[RuClCp*(PPh3)2] (Cp*=pentamethyl-
ACHTUNGTRENNUNG
Keywords: atom transfer · homoge-
neous catalysis · kinetics · radical
reactions · ruthenium
cyclopentadienyl), in the presence and
in the absence of the reducing agent
magnesium. The reactions of styrene
Introduction
tions. An additional advantage from a practical point of
view is the fact that complex 1 is not air-sensitive and is
easy to make. Complex 1 has been used in combination with
either AIBN (AIBN=azobisisobutyronitrile)[6] or powdered
magnesium[7] as the co-catalyst (Scheme 1). The co-catalyst
The atom-transfer radical addition (ATRA) of halogenated
ꢀ
compounds to olefins represents a versatile C C coupling
reaction with many applications in organic synthesis.[1] The
first investigations about ATRA reactions were performed
by Kharasch and his group in the 1940s.[2] Three decades
later it was discovered that [RuCl2ACHTNUTRGEN(UNG PPh3)3] is able to cata-
lyze this type of reaction.[3] Meanwhile, many transition-
metal complexes are known to promote inter- and intramo-
lecular ATRA reactions.[1] The highest activities are typical-
ly found for copper[1,4] and ruthenium[1,5–7] catalysts.
One of the best catalysts described so far is the half-sand-
wich complex [RuCl2Cp*ACTHNUTRGNE(NUG PPh3)] (1; Cp*=pentamethylcy-
clopentadienyl).[6,7] Turnover numbers (TONs) of more than
1000 have been achieved for different substrate combina-
Scheme 1. Intra- and intermolecular ATRA reactions of halogenated
compounds and olefins are efficiently catalyzed by the ruthenium com-
plex 1 in combination with AIBN or Mg.
[a] Dr. M. A. Fernꢀndez-Zfflmel, K. Thommes, G. Kiefer,
Prof. K. Severin
is believed to have a twofold role: it activates the RuIII cata-
lyst by reduction to a RuII complex, and it helps to avoid an
accumulation of RuIII complexes due to termination reac-
tions of carbon-centered radicals.
Institut des Sciences et IngØnierie Chimiques
École Polytechnique FØdØrale de Lausanne (EPFL)
1015 Lausanne (Switzerland)
Fax : (+41)21-693-9305
E-mail: kay.severin@epfl.ch
In view of the substantial efforts to optimize the perfor-
mance of Ru-based ATRA catalysts,[5–7] it is surprising that
there are hardly any mechanistic studies on this process
apart from some early kinetic investigations with the first-
[b] Dr. A. Sienkiewicz, K. Pierzchala
Institut de Physique de la Mati›re Complexe (IPMC)
École Polytechnique FØdØrale de Lausanne (EPFL)
1015 Lausanne (Switzerland)
generation catalyst [RuCl2ACTHNUTRGNEUNG
(PPh3)3].[8] The work reported
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901396.
below aims to fill this gap. We will demonstrate that the
Chem. Eur. J. 2009, 15, 11601 – 11607
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11601

highly active [RuCl2Cp*
in a markedly different way from the “classical” catalyst
[RuCl2A(PPh3)3]. Furthermore, we will provide evidence that
ACHTUNGERTN(NUNG PPh3)]–Mg catalyst system behaves
CHTUNGTRENNUNG
the Ru catalyst is not directly involved in the rate-limiting
step of ATRA reactions with highly active substrates such
as ethyl trichloroacetate.
Results and Discussion
Our interest in exploring the mechanism of RuCp*-cata-
lyzed ATRA reactions was triggered by the following unusu-
al observation: ethyl dichloroacetate and ethyl trichloroace-
tate both underwent ATRA reactions with styrene, with the
former reacting faster under similar conditions (Scheme 2a
and b). However, if both substrates were used in parallel,
only the ATRA product of ethyl trichloroacetate was
formed, whereas ethyl dichloroacetate was not consumed at
all (Scheme 2c).
Scheme 3. Reactions of ethyl trichloroacetate and ethyl dichloroacetate
with styrene catalyzed by a) the RuIII complex 1 with Mg, b) the RuII
complex 2 with Mg, c) the RuII complex 2 without Mg, and d) the RuIII
complex 1 without Mg.
Figure 1. Reactions of ethyl dichloroacetate with styrene catalyzed by the
II
RuIII complex 1 with Mg ( ), by the Ru complex 2 with Mg ( ), by the
~
*
RuII complex 2 without Mg ( ), and by the Ru complex 1 without Mg
III
&
^
(
). Reaction conditions: [ethyl dichloroacetate]=100 mm, [styrene]=
100 mm, [Ru]=0.5 mm, [Mg]=100 mg (4.1 mmol), toluene (1000 mL),
358C. The data represent averaged values of two independent experi-
ments. The yields were determined by GC using mesitylene as an internal
standard.
Scheme 2. Ru-catalyzed ATRA reactions of ethyl trichloroacetate and
ethyl dichloroacetate with styrene. The selectivity observed in the compe-
tition reaction c) is in contrast to the reactivity of the substrates a) and
b).
with ethyl dichloroacetate (Figure 1) were consistently faster
than the reactions with ethyl trichloroacetate (Figure 2).
This reactivity order is surprising in view of the fact that ho-
To explain these results, we first investigated the time
course of the reaction of ethyl trichloroacetate and ethyl di-
chloroacetate with styrene in more detail. The reactions
were performed with four different catalyst systems: the
RuIII complex 1 (0.5 mol%) in the absence and in the pres-
ence of an excess of Mg, and the RuII complex [RuClCp*-
ACHTUNGTRENNUNG(PPh3)2] (2, 0.5 mol%) in the absence and in the presence of
an excess of Mg (Scheme 3).
The time courses for the 2”4 reactions are shown in Fig-
ures 1 and 2. Depicted are the yields of the reactions. The
values for the conversion (styrene consumption) are slightly
higher, but the differences are below 3%. The reactions
ꢀ
molytic C Cl bond cleavage is easier for the trichloroester.
There was no significant difference between reactions cata-
lyzed by 1+Mg or by 2+Mg. Apparently, the catalyst pre-
cursors 1 and 2 were rapidly converted to the same active
species (see below). Reactions with the RuII catalyst 2 with-
out the reducing agent Mg were considerably slower. The
beneficial effect of Mg was in line with what was observed
previously.[7] When the RuIII complex 1 was used without
Mg, no product was detected for reactions with ethyl di-
chloroacetate. The ATRA of ethyl trichloroacetate, how-
11602
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11601 – 11607

Reactions Catalyzed by RuCp* Complexes
FULL PAPER
periments of di- and trichloroesters with dichloroacetoni-
trile. By itself, dichloroacetonitrile was found to be less reac-
tive than ethyl dichloroacetate, but more reactive than ethyl
trichloroacetate (see Supporting Information). In competi-
tion experiments, dichloroacetonitrile blocked the reaction
of ethyl dichloroacetate with styrene completely
(Scheme 4a). However, when equimolar amounts of ethyl
trichloroacetate and dichloroacetonitrile were used, the
ATRA reaction of the nitrile was slowed down but not
blocked completely (Scheme 4b; for details see Supporting
Information).
Figure 2. Reactions of ethyl trichloroacetate with styrene catalyzed by
II
the RuIII complex 1 with Mg ( ), by the Ru complex 2 with Mg ( ), by
~
*
the RuII complex 2 without Mg ( ), and by the Ru complex 1 without
III
&
^
Mg
(
). Reaction conditions: [ethyl trichloroacetate]=100 mm, [sty-
rene]=100 mm, [Ru]=0.5 mm, [Mg]=100 mg (4.1 mmol), toluene
(1000 mL), 358C. The data represent averaged values of two independent
experiments. The yields were determined by GC using mesitylene as an
internal standard.
ever, proceeded slowly to give a yield of 1.2% after 40 min.
This unexpected catalytic activity, although not very high,
continues for long periods of time, affording a yield of 69%
after a reaction time of eleven days. Both substrates, that is,
styrene and ethyl trichloroacetate, could potentially form
radicals by thermal or photochemical activation. These radi-
cals would be able to reduce the RuIII complex 1 and start
the ATRA reaction. The following observations suggest that
a substrate-induced activation of 1 is not an important con-
tribution under our reaction conditions: a reduction of com-
plex 1 with styryl radicals can be excluded because we
would observe ATRA adducts for ethyl trichloroacetate and
for ethyl dichloroacetate, which is not the case. The reduc-
Scheme 4. Ru-catalyzed ATRA competition reactions of ethyl tri
AHCTUNGTERGaNNUN cetate or ethyl dichloroacetate and dichloroacetonitrile with styrene.
ACHTUNGTRENNUNGchloACHTUNGTRENNUNGro-
tion of complex 1 by radicals derived from ethyl tri
chlo
E
The competition experiments show that ethyl tri
ACHTUNGTRENNUNGchloACHTUNGTRENNUNGro-
A
ACHTUNGTRENaNUGN cetate and dichloroacetonitrile are both able to block the
ethyl trichloroacetate and styrene without Mg is not very ef-
ficient, and only very low turnover numbers were achieved
(Figure 2). The observed yield of 69% (TON=138) with
complex 1 could only be explained if we assume a repetitive
activation of RuIII complexes by the halogenated substrate.
However, we were not able to detect any decomposition of
ethyl trichloroacetate in reactions with styrene without Ru
ATRA reaction of ethyl dichloroacetate and styrene. This
can be explained by assuming the following scenario: ethyl
trichloroacetate and dichloroacetonitrile react rapidly with
the catalytically active RuII complex to give an intermediate,
which is then slowly transformed into the product. The cata-
lyst in the resting state, however, is not able to react with
ethyl dichloroacetate, prohibiting its conversion.
ꢀ
(analysis by GC after six days). The homolytic C Cl bond
cleavage of ethyl trichloroacetate thus seems to be negligi-
ble under our reaction conditions.
If the reaction of the Ru catalyst with ethyl trichloroace-
tate is fast in comparison to the rate-limiting step, then the
reaction order with respect to this substrate should be zero.
This was confirmed by kinetic experiments with 0.25, 0.50,
1.00, 1.50, or 2.00 equivalents of ethyl trichloroacetate with
respect to a fixed amount of styrene, using 0.5 mol% of
complex 1 in the presence of Mg. The rates of product for-
mation (and the rates of styrene consumption) were very
similar for the five reactions (see Supporting Information).
A related behavior was observed for dichloroacetonitrile:
when ATRA reactions were performed with variable nitrile
concentrations, a constant reaction rate was obtained for di-
chloroacetonitrile concentrations of 0.1m or higher (see Sup-
porting Information). However, decreasing the substrate
concentration below 0.1m resulted in a slightly decreased re-
Subsequently, we examined the time course of the reac-
tion of styrene with an equimolar amount of ethyl tri
ACHTUNGTRENNUNGacetate and ethyl dichloroacetate. For this experiment we
ACHUTGTNRNEUGcN hloACHTUNGTERNrNUGN o-
used the RuIII catalyst 1 (0.5 mol%) in conjunction with Mg.
There was zero conversion of the dichloroester as long as
there was trichloroester in the reaction mixture (see Sup-
porting Information). However, once the trichloroester was
completely consumed, the reaction of the dichloroester with
styrene started. These results suggested that the trichloroes-
ACHTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 11601 – 11607
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11603

K. Severin et al.
action rate. The zero-order rate dependence with respect to
the chlorinated substrates is in contrast to what was found
for the first-generation catalyst [RuCl2ACHTNUTRGENG(UN PPh3)3]: for addition
reactions of CCl4 to 1-octene, a first-order dependence of
the rate on the CCl4 concentration was observed.[8]
Reactions with a variable concentration of styrene (12.5,
25, 37.5, 50, or 100 mm) and a fixed amount of ethyl trichlo-
ACHTUNGTRENNUNGroACHTUNGTRENNUNGacetate (100 mm) revealed a linear dependence of the ini-
tial reaction rate on the olefin concentration (see Supporting
Information). However, for higher styrene concentrations
(150 or 200 mm) a deviation from a linear relationship was
observed. This saturation behavior could be explained by
the increased formation of polymeric side products at high
styrene concentrations. Indeed, the plot for the rate of sty-
rene consumption versus initial styrene concentration is
linear even at higher styrene concentrations (see Supporting
Information).
Figure 3. UV/Vis spectra of three different solutions: a toluene solution
containing ethyl trichloroacetate, styrene, and complex 2 recorded 2 min
after mixing (dotted line); a toluene solution containing styrene, ethyl tri-
chloroacetate, and complex 1 (dashed line); a toluene solution containing
styrene and complex 2 (solid line). Concentrations: [ethyl trichloroace-
tate]=100 mm, [styrene]=100 mm, [Ru]=0.5 mm.
The influence of the catalyst loading was studied as well.
Reactions with 0.125, 0.25, 0.50, 0.75, or 1.00 mol% of com-
plex 1 showed that the reaction rate increased with an ap-
parent reaction order of 1/3 with respect to the Ru concen-
tration. It should be noted that the kinetic analysis is com-
plicated by the fact that the catalyst was a mixture of a solu-
ble Ru complex and the insoluble magnesium powder. A
large excess of Mg was used with respect to the Ru catalyst,
but the total amount of Mg was still crucial. This was evi-
denced by control experiments with Ru complex 1 (1 mmol,
1 mol% with respect to styrene) and Mg (0.8 or 4.1 mmol).
The latter reaction was approximately twice as fast as the
former. This indicated that the reaction rate was controlled
by the amount of the homogeneous catalyst and the amount
of the heterogeneous co-catalyst.
To learn more about the catalyst in the resting state, we
performed in situ UV/Vis measurements of reactions cata-
lyzed by the RuII complex 2 in the absence of Mg. The only
colored species in solution were the Ru complexes. Impor-
tantly, the spectrum of the RuIII complex 1 was clearly dis-
tinguishable from that of the RuII complex 2. An oxidation
of 2 into 1 can therefore be followed by UV/Vis spectrosco-
py.[9] The results of these experiments are shown in Figures 3
and 4.
For reactions between ethyl trichloroacetate, styrene, and
0.5 mol% of complex 2 in toluene, the UV/Vis spectrum re-
corded after 2 min at room temperature was nearly identical
to that of a solution of the pure complex 1 (Figure 3). This
suggested that the dominant Ru species in reactions with
ethyl trichloroacetate was the RuIII complex 1. This was fur-
ther substantiated by stoichoimetric reactions of complex 2:
upon addition of only ten equivalents of ethyl trichloroace-
tate, an instantaneous and quantitative oxidation to the
RuIII complex 1 occurred, as shown by UV/Vis spectroscopy.
In this context, a recent study by Matyjaszewski et al. is of
interest. They have shown that complex 2 in conjunction
with AIBN is a potent catalyst for atom transfer radical
polymerization reactions.[10] It was postulated that complex
2 is reversibly oxidized to give a 19eꢀ complex of the formu-
Figure 4. UV/Vis spectra of three different solutions: a toluene solution
containing ethyl dichloroacetate, styrene, and complex 2 recorded 2 min
after mixing (dotted line); a toluene solution containing styrene, ethyl di-
chloroacetate, and complex 1 (dashed line); a toluene solution containing
styrene and complex 2 (solid line). Concentrations: [ethyl dichloroace-
tate]=100 mm, [styrene]=100 mm, [Ru]=0.5 mm.
gest that the oxidation product of 2 is more likely to be the
17eꢀ complex [RuXClCp*
ACTHNUTRGEN(NUG PPh3)].
The UV/Vis spectrum of a solution containing ethyl di-
chloroacetate instead of ethyl trichloroacetate was in agree-
ment with the presence of a mixture of complexes 1 and 2
(Figure 4). Further evidence for a partial oxidation of 2 to 1
was obtained by recording spectra of solutions containing
complex 2 and different concentrations of ethyl dichloroace-
tate. The spectra gradually changed from that of pure 2 to
that of complex 1, with a clear isosbestic point at 411 nm
(see Supporting Information). From the UV/Vis spectra it
can be concluded that the resting state for Mg-free reactions
with ethyl trichloroacetate is the RuIII complex 1. For ethyl
dichloroacetate, on the other hand, we observe a mixture of
the RuII complex 2 and the RuIII complex 1.
Additional information about the nature of the Ru com-
plexes and possible radical intermediates formed under cata-
lytic conditions was obtained by electron spin resonance
(ESR) spectroscopy. For this study only ethyl trichloroace-
tate was used, as the data suggested that this substrate react-
ed to a greater extent with complex 2. Preliminary studies
la [RuXClCp*
ACHTUNGTRENNUNG(PPh3)2] (X=Br, Cl). Our UV/Vis studies sug-
11604
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11601 – 11607

Reactions Catalyzed by RuCp* Complexes
FULL PAPER
showed that the room-temperature ESR spectra displayed a
broad, non-resolved signal centered at g=2.175, with a line
width of ꢁ280 G. In contrast, much better-resolved ESR
spectra were observed for frozen solutions. Therefore, we
focused on low-temperature
RuCp*-catalyzed ATRA of ethyl trichloroacetate to styrene
(Scheme 5). In the first step (shown in purple) the RuIII
complex 1 is reduced by Mg to a RuII complex. We have no
experimental data for the structure of this complex, but the
ESR studies performed at
20 K. For comparison, we first
recorded the ESR spectrum of
the RuIII complex 1. Ethyl tri-
chloroacetate and styrene were
added to the solution to imi-
tate the reaction conditions
(the substrates were not ex-
pected to react with complex 1
to a significant extent). The re-
corded trace showed a well-re-
solved three-line spectrum
characteristic of a paramagnet-
ic ruthenium d5 complex of
low spin S=1/2 with distorted
octahedral
symmetry
(Figure 5, top).[11] The corre-
sponding g-factor values for
the observed peaks were: gx =
2.501, gy =2.067, and gz =1.946.
The average g-factor value for
Scheme 5. Proposed mechanism for the RuCp*-catalyzed ATRA of ethyl trichloroacetate to styrene.
these three peaks, hgi=[g2x/3+gy2/3+gz2/3]1/2, is 2.184, which is
also close to the g-factor value observed at room tempera-
ture for a non-resolved ESR spectrum.
16eꢀ complex [RuClCp*
ACTHNGUTRE(NNUG PPh3)] (3) seems to be a plausible
candidate. It should be noted that previous attempts to
make complex 3 on a preparative scale have failed.[12] How-
ever, a structurally related complex with a sterically de-
manding cyclopentadienyl ligand has recently been charac-
terized crystallographically.[13] For reactions with the catalyst
precursor 2, the active intermediate 3 can be accessed by
loss of one phosphine ligand. The RuII complex 3 starts the
main catalytic cycle (Scheme 5, blue) by rapidly reacting
with ethyl trichloroacetate to give complex 1 along with the
Subsequently, we recorded the ESR spectra of reaction
mixtures containing styrene, ethyl trichloroacetate, and the
RuII complex 2. Two experimental procedures were used: in
one case, the sample was allowed to react for 40 min at
room temperature, and then the sample was frozen rapidly.
Alternatively, the sample was frozen directly after mixing
the reagents. In both cases the ESR spectra were very simi-
lar to that obtained for complex 1, indicating that rapid and
quantitative oxidation of the RuII complex 2 had occurred
(Figure 5, bottom).
C
radical CCl2CO2Et. Chloro-atom transfer reactions of this
kind are known to be reversible, but the equilibrium in this
case is completely on the side of the RuIII complex (k2 @
C
Based on all of the above-mentioned experiments we
would like to propose the following mechanism for the
kꢀ2). The subsequent coupling between CCl2CO2Et and sty-
rene (k3) is the rate-limiting step of the reaction. This as-
sumption is supported by the UV/Vis and ESR experiments,
which show that complex 1 is the resting state of the reac-
tion. It also explains the observation that ethyl trichloroace-
tate is able to monopolize the catalyst in competition experi-
ments with ethyl dichloroacetate. The latter substrate is ex-
pected to react more slowly with the active RuII complex 3
because of its lower intrinsic reactivity. In the resting state,
on the other hand, the Ru complex 1 is not able to activate
ethyl dichloroacetate. The main product of the reaction is
formed in a chloro-atom transfer between complex 1 and
the benzyl radical. The equilibrium of this reversible reac-
tion is expected to be on the side of RuII complex (k4 >
kꢀ4).[9] The following control experiment provided evidence
that the last chloro-atom transfer step is not rate-determin-
ing: reactions between ethyl dichloroacetate[14] and styrene
were performed with 0.5 mol% of the RuII complex 2 with-
Figure 5. X-band ESR spectra of a frozen toluene solution (20 K) of com-
plex 1 (top) and of a rapidly frozen mixture of complex 2 (1 mm), styrene
(200 mm), and ethyl trichloroacetate (200 mm) in toluene (bottom).
Chem. Eur. J. 2009, 15, 11601 – 11607
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11605

K. Severin et al.
out Mg. The reaction rate would be expected to increase
with addition of the RuIII complex 1 if step 4 was crucial for
the overall rate, but no such increase was observed.
In this case, the nature of the catalyst will be important. Sec-
ondly, a better catalyst could also be a more stable catalyst,
which gives rise to higher turnover numbers, or a more se-
lective catalyst, which gives low amounts of side products.
However, the fact that the rate-limiting step of ATRA reac-
tions can be metal-independent should be considered for
future studies, even more so as it is common practice to test
new catalysts with intrinsically active substrates such as
CCl4.[5]
Another interesting finding of our study is the observation
that a RuIII complex such as 1 can be catalytically active
without prior reduction. In this case, the ATRA reactions
could be mediated by a RuIII/RuIV redox couple, but further
investigations are needed to verify this point. Under normal
reaction conditions, the contribution of this second catalytic
cycle is expected to be small. However, it is conceivable that
other RuIII or OsIII complexes will display a higher propensi-
ty to catalyze ATRA reactions through this alternative path-
way. Investigations along these lines are currently being per-
formed in our laboratory.
For reactions between ethyl trichloroacetate and styrene,
a second catalytic pathway (Scheme 5, red) seems to be op-
erational next to the main catalytic cycle (Scheme 5, blue).
This was concluded from the results of reactions with the
RuIII catalyst 1 without Mg, which showed a slow but steady
formation of the ATRA product (see above). It seems plau-
sible to invoke a RuIV complex as a catalytic intermediate,
but attempts to characterize a catalytically relevant RuIV
species have so far not been successful.
Along with the productive catalytic pathways, there are
two reactions which give low but detectable amounts of side
C
products. First, the carbon radical CCl2CO2Et can undergo
a homocoupling to give the dimer (CCl2CO2Et)2 (small
amounts were detected by GC–MS). This termination reac-
tion is not as likely for the less stable benzyl radical, but the
latter can react with styrene to give oligomers and polymers.
It is expected that a mechanism similar to that described
in Scheme 5 is operational for the ATRA reaction of ethyl
dichloroacetate and styrene. One important difference is
that the equilibrium for the first chloro-atom transfer reac-
tion is not completely on the side of RuIII complex 1
(k2ffikꢀ2). Furthermore, the catalytic RuIII/IV pathway
(Scheme 5, shown in red) is not relevant for ethyl dichloro-
Experimental Section
General: The complexes [RuCl2Cp*ACTHNUTRGENNUG ACHTUNGTRENNUNG(PPh3)2]
(PPh3)] (1)[15] and [RuClCp*
(2)[16] were prepared according to literature procedures. Mg powder
(>99%) was purchased from Fluka, and was agitated by means of a stir-
ring bar under an atmosphere of dry dinitrogen for 10 days before use.
All ATRA reactions were performed in a glove box under an atmos-
phere of dinitrogen. The solvents and the commercially available sub-
strates were distilled from appropriate drying agents and stored under ni-
trogen. GC measurements were made on a Varian Chrompack CP-3380
apparatus (Chrompack CP-SIL8CB column; 30 m; 250 mm) coupled to a
FID detector. UV/Vis measurements were made on a Perkin–Elmer
Lambda 40 UV/Vis spectrometer.
ACHTUNGTRENNUNGaceACHTUNGTRENNUNGtate, presumably because of the lower oxidizing power of
this substrate compared to ethyl trichloroacetate.
Conclusion
Kinetic and spectroscopic analysis have been performed to
gain information about the mechanism of atom-transfer rad-
ical reactions (ATRA) catalyzed by the complexes
General procedure for the ATRA of chlorinated esters or of dichloroace-
tonitrile to styrene: Complex 1 or 2 (400 mL of a 1.25 mm stock solution
in toluene) was added to a 1.5 mL vial that contained Mg (100 mg; if Mg
was used for the reaction). The total volume was increased to 900 mL
with toluene and the resulting mixture was stirred at 358C for 10 min.
The reaction was then initiated by addition of a freshly prepared stock
solution containing styrene, the chlorinated compound, and mesitylene as
an internal standard (100 mL; final concentrations: [styrene]=100 mm;
[chlorinated compound]=100 mm; [internal standard]=10 mm). The so-
lution was stirred at 358C and samples (25 mL) were removed at given
times from the reaction mixtures, diluted with non-deoxygenated acetone
(500 mL), and analyzed by GC.
[RuCl2Cp*ACHTUNGTRENNUNG(PPh3)] and [RuClCp*AHCTUNGTRENN(GUN PPh3)2] in the presence
and in the absence of the reducing agent magnesium. The
ATRA reactions of styrene with ethyl trichloroacetate, ethyl
dichloroacetate, or dichloroacetonitrile were used as test re-
actions. The results show that for substrates with high intrin-
sic reactivity such as ethyl trichloroacetate or dichloroaceto-
nitrile, the oxidation state of the catalyst in the resting state
is +3, and that the reaction is zero-order with respect to the
halogenated compound. This is in contrast to what is ob-
UV/Vis measurements: Samples (450 mL) of the corresponding reaction
mixtures were removed two minutes after adding the substrate and were
transferred to a quartz cuvette of 0.5 mm thickness, which was tightly
closed. The UV/Vis spectra of the reaction mixtures were recorded and
compared with standards of complexes 1 and 2 of the same Ru concen-
tration.
served for less active catalysts such as [RuCl2ACHTUNTRGENUNG(PPh3)3], which
show a first-order dependence of the reaction rate on the
chlorinated substrate. Furthermore, the kinetic data suggest
that the metal catalyst is not directly involved in the rate-
limiting step of the reaction. An important consequence of
this finding is that it will not be possible to make a faster
catalyst for the ATRA reaction of ethyl trichloroacetate or
dichloroacetonitrile and styrene, at least under reaction con-
ditions similar to those described above. This should not
imply that it is not worthwhile to search for better ATRA
catalysts. First of all, for substrates with a lower intrinsic re-
activity (more stable carbon-halogen bonds), the first
chloro-atom transfer step may easily become rate-limiting.
ESR measurements: The ESR spectra were recorded at low temperature
(20 K) using an ESR spectrometer, Model E540 EleXys (Bruker BioSpin
GmbH, Karlsruhe, Germany), operating in the microwave X-band,
equipped with a cylindrical TE011 high-Q cavity, model ER 4122SHQE,
and a continuous-flow helium cryostat, Model ESR900, from Oxford In-
struments (Abington, UK). The Instrumental settings of Bruker EleXys
spectrometer were: microwave frequency: ꢁ9.402 GHz; microwave
power: 0.2 mW; scan time: 336 s; modulation frequency: 100 kHz; modu-
lation amplitude: 5 G; time constant: 20.5 ms; gain: 50 dB; single scan of
2000 G. Samples (800 mL) of the corresponding reaction mixtures were
11606
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11601 – 11607

Reactions Catalyzed by RuCp* Complexes
FULL PAPER
removed immediately after thorough mixing of the reagents, and the sol-
utions were transferred into 4 mm OD/3.0 mm ID quartz ESR tubes,
from Wilmad–Labglass (Buena, NJ, USA). The samples were quenched
in liquid nitrogen and stored at 77 K. To acquire the ESR spectra, the
samples were transferred quickly to the precooled cavity of the ESR
spectrometer (ꢁ50 K). Alternatively, the solutions were transferred into
2.4 mm OD/2.0 mm ID quartz capillary tubes, which were subsequently
put into the 4 mm OD/3.0 mm ID quartz ESR tube before freezing.
Lundgren, M. A. Rankin, R. McDonald, M. Stradiotto, Organome-
tallics 2008, 27, 254–258; c) B. Dutta, E. Solari, R. Scopelliti, K. Se-
verin, Organometallics 2008, 27, 423–429; d) Y. Borguet, A. Richel,
S. Delfosse, A. Leclerc, L. Delaude, A. Demonceau, Tetrahedron
Lett. 2007, 48, 6334–6338; e) Y. Motoyama, S. Hanada, K. Shima-
moto, H. Nagashima, Tetrahedron 2006, 62, 2779–2788; f) L. Que-
batte, E. Solari, R. Scopelliti, K. Severin, Organometallics 2005, 24,
1404–1406; g) Y. Motoyama, S. Hanada, S. Niibayashi, K. Shimamo-
to, N. Takaoka, H. Nagashima, Tetrahedron 2005, 61, 10216–10226;
h) L. Quebatte, M. Haas, E. Solari, R. Scopelliti, Q. T. Nguyen, K.
Severin, Angew. Chem. 2005, 117, 1108–1112; Angew. Chem. Int.
Ed. 2005, 44, 1084–1088; i) L. Quebatte, R. Scopelliti, K. Severin,
Eur. J. Inorg. Chem. 2005, 3353–3358; j) L. Quebatte, R. Scopelliti,
K. Severin, Angew. Chem. 2004, 116, 1546–1550; Angew. Chem. Int.
Ed. 2004, 43, 1520–1524; k) B. T. Lee, T. O. Schrader, B. Martín-
Matute, C. R. Kauffman, P. Zhang, M. L. Snapper, Tetrahedron
2004, 60, 7391–7396; l) O. Tutusaus, S. Delfosse, A. Demonceau,
A. F. Noels, C. ViÇas, F. Teixidor, Tetrahedron Lett. 2003, 44, 8421–
8425; m) O. Tutusaus, C. ViÇas, R. NfflÇez, F. Teixidor, A. Demon-
ceau, S. Delfosse, A. F. Noels, I. Mata, E. Molins, J. Am. Chem. Soc.
2003, 125, 11830–11831; n) B. de Clercq, F. Verpoort, Tetrahedron
Lett. 2002, 43, 4687–4690; o) F. Simal, L. Wlodarczak, A. Demon-
ceau, A. F. Noels, Eur. J. Org. Chem. 2001, 2689–2695; p) F. Simal,
L. Wlodarczak, A. Demonceau, A. F. Noels, Tetrahedron Lett. 2000,
41, 6071–6074.
Acknowledgements
This work was supported by the Swiss National Science Foundation and
by the EPFL. We thank Prof. L. Forro, EPFL, for supporting the ESR
measurements.
[1] For reviews, see: a) T. Pintauer, K. Matyjaszewski, Chem. Soc. Rev.
2008, 37, 1087–1097; b) K. Severin, Curr. Org. Chem. 2006, 10, 217–
224; c) L. Delaude, A. Demonceau, A. F. Noels, Top. Organomet.
Chem. 2004, 11, 155–171; d) H. Nagashima in Ruthenium in Organic
Synthesis (Ed.: S.-I. Murahashi), Wiley-VCH, Weinheim, 2004,
pp. 333–343; e) A. J. Clark, Chem. Soc. Rev. 2002, 31, 1–11; f) K. I.
Kobrakov, A. V. Ivanov, J. Heterocycl. Chem. 2001, 37, 529–539;
g) R. A. Gossage, L. A. van de Kuil, G. van Koten, Acc. Chem. Res.
1998, 31, 423–431; h) J. Iqbal, B. Bhatia, N. K. Nayyar, Chem. Rev.
1994, 94, 519–564; i) F. Minisci, Acc. Chem. Res. 1975, 8, 165–171.
[2] a) M. S. Kharasch, E. V. Jensen, W. H. Urry, Science 1945, 102, 128–
128; b) M. S. Kharasch, W. H. Urry, E. V. Jensen, J. Am. Chem. Soc.
1945, 67, 1626–1626.
[6] L. Quebatte, K. Thommes, K. Severin, J. Am. Chem. Soc. 2006, 128,
7440–7441.
[7] K. Thommes, B. IÅli, R. Scopelliti, K. Severin, Chem. Eur. J. 2007,
13, 6899–6907.
[8] a) W. J. Bland, R. Davis, J. L. A. Durrant, J. Organomet. Chem.
1985, 280, 397–406; b) W. J. Bland, R. Davis, J. L. A. Durrant, J. Or-
ganomet. Chem. 1984, 267, c45–c48.
[9] W. A. Braunecker, W. C. Brown, B. C. Morelli, W. Tang, R. Poli, K.
Matyjaszewski, Macromolecules 2007, 40, 8576–8585.
[3] H. Matsumoto, T. Nakano, Y. Nagai, Tetrahedron Lett. 1973, 14,
5147–5150.
[4] For selected examples, see a) C. Ricardo, T. Pintauer, Chem.
Commun. 2009, 3029–3031; b) T. Pintauer, W. T. Eckenhoff, C. Ri-
cardo, M. N. C. Balili, A. B. Biernesser, S. J. Noonan, M. J. W.
Taylor, Chem. Eur. J. 2009, 15, 38–41; c) J. M. MuÇoz-Molina, T. R.
Belderraín, P. J. PØrez, Adv. Synth. Catal. 2008, 350, 2365–2372;
d) J. A. Bull, M. G. Hutchings, C. Lujꢀn, P. Quayle, Tetrahedron
Lett. 2008, 49, 1352–1356; e) W. T. Eckenhoff, S. T. Garrity, T. Pinta-
uer, Eur. J. Inorg. Chem. 2008, 563–571; f) R. N. Ram, N. Kumar,
Tetrahedron Lett. 2008, 49, 799–802; g) J. A. Bull, M. G. Hutchings,
P. Quayle, Angew. Chem. 2007, 119, 1901–1904; Angew. Chem. Int.
Ed. 2007, 46, 1869–1872; h) W. T. Eckenhoff, T. Pintauer, Inorg.
Chem. 2007, 46, 5844–5846; i) J. M. MuÇoz-Molina, A. Caballero,
M. M. Díaz-Requejo, S. Trofimenko, T. R. Belderraín, P. J. PØrez,
Inorg. Chem. 2007, 46, 6858–6863; j) C. V. Stevens, E. Van Meenen,
K. G. R. Masschelein, Y. Eeckhout, W. Hooghe, B. Dꢂhondt, V. N.
Nemykin, V. V. Zhdankin, Tetrahedron Lett. 2007, 48, 7108–7111;
k) D. Yang, Y.-L. Yan, B.-F. Zheng, Q. Gao, N.-Y. Zhu, Org. Lett.
2006, 8, 5757–5760.
[10] A. Plichta, W. Li, K. Matyjaszewski, Macromolecules 2009, 42,
2330–2332.
[11] P. Munshi, R. Samanta, G. K. Lahiri, J. Organomet. Chem. 1999,
586, 176–183.
[12] T. Braun, G. Münch, B. Windmüller, O. Gevert, M. Laubender, H.
Werner, Chem. Eur. J. 2003, 9, 2516–2530.
[13] B. Dutta, E. Solari, S. Gauthier, R. Scopelliti, K. Severin, Organo-
metallics 2007, 26, 4791–4799.
[14] This control experiment was performed with ethyl dichloroacetate
and not with ethyl trichloroacetate because the RuIII complex 1 by
itself shows a low catalytic activity for the latter substrate.
[15] T. Arliguie, C. Border, B. Chaudret, J. Devillers, R. Poilblanc, Orga-
nometallics 1989, 8, 1308–1314.
[16] M. S. Chinn, D. M. Heinekey, J. Am. Chem. Soc. 1990, 112, 5166–
5175.
[5] For selected examples, see: a) J. Wolf, K. Thommes, O. Briel, R.
Scopelliti, K. Severin, Organometallics 2008, 27, 4464–4474; b) R. J.
Received: May 25, 2009
Published online: September 11, 2009
Chem. Eur. J. 2009, 15, 11601 – 11607
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11607