The hydroxide-promoted catalytic hydrodefluorination of fluorocarbons by ruthenium in aqueous media

Article abstract of DOI:10.1002/adsc.201200870

A ruthenium complex ligated by the non-innocent protic ligand 2,6-imidizoylpyridine (IPI) in aqueous potassium hydroxide media is shown to be a suitable catalyst for the C-F bond activation and subsequent hydrodefluorination (and hydrogenolysis) of a model fluorocarbon. Furthermore, the conversion of C-F bonds was enhanced upon increasing the concentration of potassium hydroxide [KOH]. A simple, non-ligated heterogeneous analogue derived from ruthenium(III) trichloride trihydrate dissolved in aqueous potassium hydroxide is shown to be an even more active catalyst, capable of hydrodefluorination of aryl fluorocarbons at room temperature and alkyl fluorocarbons under more forcing conditions. Copyright

Full text of DOI:10.1002/adsc.201200870

COMMUNICATIONS
DOI: 10.1002/adsc.201200870
The Hydroxide-Promoted Catalytic Hydrodefluorination of
Fluorocarbons by Ruthenium in Aqueous Media
a,
a
a
Michael M. Konnick, * Steven M. Bischof, Roy A. Periana,
a,
and Brian G. Hashiguchi *
Scripps Energy & Materials Center, Department of Chemistry, The Scripps Research Institute, Jupiter, FL 33458, USA
E-mail: mkonnick@scripps.edu or bhashigu@scripps.edu
a
Received: September 26, 2012; Revised: December 7, 2012; Published online: February 22, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200870.
Abstract: A ruthenium complex ligated by the non-
innocent protic ligand 2,6-imidizoylpyridine (IPI) in
aqueous potassium hydroxide media is shown to be
a suitable catalyst for the CÀF bond activation and
heterolytic cleavage; and it is now generally accepted
that kinetic factors (rather than thermodynamic) are
[
4]
the primary origin of its stability. Indeed, most re-
ported systems for catalytic CÀF defluorination ex-
ploit the polarity of the CÀF bond by utilizing either
subsequent hydrodefluorination (and hydrogenoly-
sis) of a model fluorocarbon. Furthermore, the con-
version of CÀF bonds was enhanced upon increas-
[
5]
highly electrophilic or nucleophilic species to induce
reactivity. One area of study to address this challenge,
which has received increasing focus in the past de-
cades, is the use transition metal catalysts to affect CÀ
ing the concentration of potassium hydroxide
[
KOH]. A simple, non-ligated heterogeneous ana-
[
6,7]
logue derived from ruthenium(III) trichloride trihy-
A
H
U
T
E
N
N
F bond conversion.
drate dissolved in aqueous potassium hydroxide is
shown to be an even more active catalyst, capable
of hydrodefluorination of aryl fluorocarbons at
room temperature and alkyl fluorocarbons under
more forcing conditions.
Our group has been exploring the use of reversible
proton transfer reactions between non-innocent li-
gands and strongly acidic or basic solvents to gener-
ate, respectively, highly electrophilic or nucleophilic
[
8]
homogeneous metal complexes. Recently, we have
shown that the Ru(II) complex [produced via the
in situ reduction of (2,6-imidizoylpyridine)rutheni-
Keywords: base-enhanced catalysis; CÀF bond acti-
[
9]
vation; fluorine; hydrodefluorination; ruthenium
um
ligand catalyzes reversible CÀH activation when dis-
solved in aqueous KOH. Surprisingly, rather than
ACHTUNGTRENNGNU( III) trichloride, 1] with a non-innocent protic
[
10]
observing rate inhibition, the rates of CÀH activation
are enhanced with increasing [KOH]. We postulate
The utility of organofluorine compounds across the that reversible deprotonation of the non-innocent
chemical industries of pharmaceuticals, agrochemicals, protic ligands leads to a coordinatively unsaturated,
[
1]
materials, and fine chemicals cannot be understated.
powerful p-donor that can interact more readily with
This unique position in modern chemistry is derived the anti-bonding orbital (LUMO) of a CÀH bond;
from both the incredible thermodynamic (>130 kcal leading to MÀC bond formation (Scheme 1). We con-
À1
mol in certain cases) and kinetic stability of the CÀ sidered that this approach of base-accelerated cataly-
[2]
F bond. However, these same features have histori- sis could be useful for activating other kinetically
cally made the degradation of organofluorine com- inert bonds that operate via a nucleophilic mecha-
pounds challenging with perfluorocarbon manipula- nism, such as the activation and reduction of CÀF
tion being especially difficult, as individual CÀF bond bonds. We envision that the formal activation of CÀF
strengths typically increase as additional fluorine bonds would operate via a similar donation of elec-
[
3]
atoms are added to a molecule. This inert nature trons from the metal d-orbitals into the anti-bonding
allows fluorocarbons to persist in the environment for orbital of the CÀF bond (Scheme 1). Indeed, it
extended periods of time; several of which are both seemed plausible that the much higher overall CÀF
bio-accumulative and toxic (such as perfluorooctanoic bond polarity and stronger RuÀF bond strength in
acid) or highly potent greenhouse gases (such as a putative Ru(R)(F) intermediate would likely make
CH F , CF ). Despite the inherent strength of the CÀ our catalyst even more active for CÀF activation than
2
2
4
F bond, high bond polarity makes it susceptible to CÀH activation (Scheme 1). The goal was to utilize
632
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 632 – 636

The Hydroxide-Promoted Catalytic Hydrodefluorination of Fluorocarbons
Table 1. HDF and hydrogenolysis of p-F-C H COOH, 2, cat-
6
4
[a]
alyzed by 1 at varied time, [KOH] and H pressure.
2
[b]
[c]
[d]
Entry
t
A
H
U
G
R
N
U
G
PH2
Conv.
%]
Product
[
[Yield (%)]
1
1 h
1M
400
400
400
400
400
55
89
95
22
86
3 (24)
4
5
(24)
(6)
2
3
4
5
3 h
7 h
1 h
1 h
1M
3 (18)
4
5
(66)
(4)
1M
3 (12)
4
5
(80)
(1)
0.25M
3M
3 (11)
4
5
(10)
(1)
Scheme 1. Model for base-accelerated CÀH and CÀF activa-
tion by strong p-donor catalysts with non-innocent ligands
and proposed net HDF reaction.
3 (35)
4
5
(40)
(8)
6
7
1 h
1 h
1M
1M
50
150
12
19
3 (12)
3 (13)
this CÀF cleavage to carry out the overall hydrode-
fluorination (HDF) reaction shown in Scheme 1. It is
important to highlight the three advantages of using
basic solvents (e.g., KOH): (i) catalyst activation, (ii)
4
5
(4)
(2)
[a]
General conditions: 50 mM 2, 1 mM (2 mol%) 1, 2 mL
KOH(aq) solvent, 1008C.
formation of H O helps drive the reaction, and (iii)
2
À
innocuous F is generated rather than HF. Herein, we
[
[
b]
c]
In psi.
present our work on Ru-catalyzed hydrodefluorina-
tion (HDF) of fluorocarbons in aqueous hydroxide
media.
Conversions were determined via comparison to an inter-
nal standard (MeOH) of known concentration, which
was added post reaction.
Product identities and yields were determined via com-
parison to internal standard (MeOH).
We initiated our investigations into CÀF activation
[d]
by examining the pre-catalyst (IPI)Ru ACHTUNGTERNNUN(G III)Cl , 1,
3
which we previously identified for the CÀH activation
[
10]
reaction in aqueous KOH media. To minimize
issues related to mass transfer, we focused our efforts
on para-fluorobenzoic acid, 2, as a base-soluble fluo-
rocarbon. As can be seen in Table 1, treatment of 2 in
1
4
M KOH with 2 mol% 1 at 1008C for 1 h under
00 psi H resulted in the partial conversion of 2
2
(
55%) to a mixture of BzOK, 3, and the fully hydro-
genated product Cy-COOK, 4, as the major products
24% and 24%, respectively); along with minor
amounts of the partially hydrogenated product 5
(
[
11]
(
6%) (Scheme 2; Table 1, entry 1). At longer reac-
Scheme 2. HDF of 2 catalyzed by (IPI)RuCl , 1; with subse-
tion times, 4 is the major product, with 3 and 5 as
minor products (Table 1, entries 2 and 3). Control ex-
periments show that no reaction is observed in the ab-
3
quent hydrogenation of BzOK, 3, to form CyCOOK, 4, and
C H COOK, 5.
6
9
sence of 1 or H . The observation that 3 is a major
product (1 h) that is further hydrogenated with in- solubilized F . The solution visually appears to
2
À
creasing [4] at longer times and the presence of minor remain homogeneous throughout the reaction, and
1
amounts of 5 is consistent with a fast initial HDF fol-
H NMR spectroscopic analysis indicates no resonan-
lowed by hydrogenation of the aromatic ring to ulti- ces associated with either free or hydrogenated/de-
mately generate 4 (Scheme 2). Additionally, the hy- composed IPI ligand (such as would be expected if
drogenated analogue of 2 with the CÀF bond intact is a dissociated IPI ligand was hydrogenated during the
1
never observed. Quantitative H NMR spectroscopic reaction). This suggests that 1 is stable to decomposi-
analysis demonstrated >98% mass balance based on tion to heterogeneous Ru species. Importantly, exami-
19
starting [2] for all reactions and F NMR comparison nation of the effect of hydroxide on defluorination re-
of reaction solutions to an authentic sample (KF) vealed that reactivity increases upon increasing
verified that the lost F from 2 is converted solely to [KOH]. Thus, analogous to that observed for the CÀ
Adv. Synth. Catal. 2013, 355, 632 – 636
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
633

Michael M. Konnick et al.
COMMUNICATIONS
H activation reaction with 1, reducing [KOH] to
0.25M resulted in only 22% conversion (Table 2,
entry 4) while increasing [KOH] to 3M gave 86%
conversion of 2 (Table 2, entry 5). The reaction also
shows a dependence on H pressure and reducing H
2
2
pressure led to lower conversion of 2. Consistent with
HDF as the initial reaction, lower H2 pressures
showed less hydrogenation and greater selectivity for
the formation of 3 (entries 6 and 7).
While the role of the IPI ligand is essential for
maintaining the homogeneity of 1, we envision that
the increased reactivity upon increasing [KOH] does
not necessarily need to originate from IPI deprotona-
tion. The reversible deprotonation of a coordinated
Scheme 3. HDF of 2 at 258C and 508C catalyzed by 6.
OH could also impart the same p-nucleophilic en- starting 2 for all reactions and verified that the lost F
[12]
À
hancement. Also, it is reasonable to predict (based is converted solely to solubilized F .
on the relative pK ) that reversible deprotonation of
After observing the efficiency and mild tempera-
a
a bound OH (vs. IPI) would generate an even more tures of the catalysis described above, we investigated
[
13]
p-nucleophilic and active species. One possible im- ability of pre-catalyst 6 to affect HDF on a wider vari-
plication of this is that an ancillary organic ligand ety of substrates (Table 2). Remarkably, complete
such as IPI may not be necessary for efficient catalyt- conversion of perfluorobenzoic acid was observed
ic turnover. While precipitation to RuO from a non- when exposed to conditions identical to those utilized
x
ligated Ru is a significant concern in this medium, for 2 (Table 2, entry 1). Although significantly more
there are several examples of successful heterogene- forcing conditions are required (2008C, 5 h, 20 mol%
[
14]
ous transition metal HDF catalysts, and the ability Ru), partial conversion of the fluoroalkyl substrates
of reduced transtion metal colloids to affect reduction F C(CH ) COOK and CH F was also seen (Table 2,
AHCTUNGTRENNUNG
3
2
2
2 2
[15]
chemistry is known.
present study, Young and Grushin have developed a het- conversion were observed for the perfluorinated alkyl
erogeneous Rh catalyst that operates in the presence of substrates F C(CF ) COOK and CF (Table 2, en-
In particular relevance to this entries 2 and 3). Unfortunately, only trace levels of
AHCTUNGTRENNUNG
3
2
2
4
1
9
4
0% NaOH/H which was determined to be more active tries 4 and 5); however, F NMR analysis indicated
than its homogeneous analogue. To the best of our the presence of F dissolved in the aqueous KOH sol-
knowledge however, the ability of simple Ru salts to vent, which was not observed during control reactions
2
[16]
À
catalyze HDF in basic media has not been examined.
Considering the catalytic ability of 1 and the points tion will be required, these initial results are sugges-
raised above, we next examined the simple Ru(III) tive that Ru-catalyzed HDF in basic media may be
catalyst RuCl (H O) , 6. Astonishingly, at conditions feasible for aliphatic perfluorinated compounds,
in the absence of 6. While further study and optimiza-
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
3
2
3
otherwise identical to those utilized for 1, 6 catalyzed which is comparatively rare in the literature.
the HDF and hydrogenolysis of 2 at room temperature
Table 2. Hydrodefluorination of selected fluorocarbons cata-
lyzed by 6.
(
258C), in 40% conversion to BzOK, 3, as the major
[a]
product (Scheme 3). This result contrasts with the re-
actions of CÀH bonds, where controls show that 6
[d]
[d]
Entry Substrate
Conv. [%] Product
Yield (%)]
[
10]
shows no activity for H/D exchange in basic media.
[
No reaction was observed in either the absence of H2
[
[
b]
c]
[e]
1
2
3
4
5
C F COOK
>99
Cy-COOK (92)
C H COOK (73)
6
5
or KOH (in H O), further demonstrating the necessi-
2
F C
3
A
H
U
G
R
N
N
(CH ) COOK 85
2
2
3
7
ty of KOH for turnover. Increasing the temperature
to 508C resulted in the complete consumption of 2
with the fully hydrogenated 4 as the sole product
CH F
26
trace
trace
CH (22)
2
2
4
C F COOK
–
3
7
[c]
CF4
CH (<1)
4
(
Scheme 3). While the reaction at 258C resulted in
[a]
General conditions: unless otherwise noted, 50 mM sub-
strate, 20 mmol (20 mol%) 6, 2 mL 1M KOH(aq), 400 psi
the formation of moderate amounts of heterogeneous
Ru, it appears that complete precipitation (of presum-
[
17]
H , 2008C, 5 h.
2
ably RuO ) is observed at 508C.
Control experi-
x
[b]
2
1
mmol (2 mol%) 6, 1 h, 508C.
5 psi (1 atm, 0.54 mn) of substrate (4 mol% 6).
ments where a solution of 6 is preheated to 1008C for
[c]
30 min to induce precipitation prior to the addition of
[d]
Conversions and yields were determined via comparison
2
demonstrated that the heterogeneous Ru is an
to an internal standard (MeOH for entries 1, 2, and 4
1
active catalyst, as complete conversion to 4 is still ob-
served at qualitatively similar rates. NMR spectro-
scopic analysis demonstrated >97% mass balance on
[
H NMR]; neon for entries 3 and 5 [GC-MS]).
[e]
Minor amounts of 3 (2%) and 5 (1%) were also ob-
served.
634
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 632 – 636

The Hydroxide-Promoted Catalytic Hydrodefluorination of Fluorocarbons
Scheme 4. Proposed reduced metal insertion (A) and hydride metathesis (B) mechanisms for HDF catalyzed by 1.
Considering our previous studies concerning the CÀ tion. A more complete mechanistic study including
H activation reaction in aqueous KOH and other lit- these experiments is currently underway.
[
18]
erature precedents
we propose two general path-
In conclusion, we have demonstrated that the base-
[
19]
ways by which HDF could occur with 1:
reduced enhanced nucleophilic Ru catalyst 1, though originally
metal insertion (Scheme 4, cycle A) or CÀF bond designed for the CÀH activation in basic media, is
metathesis (Scheme 4, cycle B). In the former, the also a capable of affecting CÀF activation. Analogous
role of H is as a reducing agent to generate an active to our prior work, CÀF bond conversion is enhanced
2
Ru(II) species, which then inserts into the CÀF bond. upon increasing [KOH]; further supporting our gener-
Subsequent protonolysis of the MÀR intermediate al premise that reversible deprotonation of non-inno-
would then generate RH, with the catalyst then re- cent ligands in KOH generates a more active catalyst
duced by H to regenerate the active catalyst. In the via enhancement of coordinative lability and p-nucle-
2
latter, the role of H is to generate an active RuÀH ophilicity. Additionally, a heterogeneous analogue of
2
species, which then forms the MÀR intermediate via 1, (generated from 6) was shown to be highly active
s-bond metathesis. Subsequent hydrogenolysis (via for HDF of fluoroaryl substrates in aqueous KOH.
metathesis with H or stepwise protonolysis and re- While our initial examinations have focused only
2
duction) again closes the catalytic cycle. We propose upon CÀF bond conversion, it is interesting to specu-
the beneficial role of KOH in the proposed cycles is late if this general class of reactivity could be applied
two-fold: (i) reversible deprotonation of the ancillary to other strong CÀX bonds, such as CÀS and CÀO,
ligand by KOH generates a catalyst [either a Ru(II) which are relevant to the fields of hydrocarbon desul-
or RuÀH species] of higher nucleophilicity and coor- furization and biomass reduction, respectively. Fur-
dinative lability; and (ii) the water base solvent pro- thermore, it is reasonable that this general motif
vides a thermodynamically strong fluoride trap in the could be applied to other metal centers to affect simi-
hydrogen bonding network present in the media itself. lar chemistry, such as Os, or more attractively, Fe. In-
While no formal mechanistic study has been per- vestigations into these concepts are currently under-
formed to date, RuÀH generation from the deproto- way.
[
20]
nation of Ru(H ) complexes is well established, and
2
H in strongly basic media is known to reversibly de-
2
protonate to generate very reactive incipient or sol-
[21]
Experimental Section
vated hydrides. These precidents coupled with the
observation that HDF occurs with subsequent hydro-
genation of 2 may support a Ru hydride-catalyzed
pathway. Additionally, it is important to note that de-
Representative HDF/Hydrogenolysis Reaction of 2
Catalyzed by 1
spite the evidence that 1 does not form heterogeneous In a 15-mL vial, 5.0 mg (12 mmol) of (IPI)RuCl , 1, and
3
Ru during catalysis, it is possible that the origin of the 84.1 mg (0.6 mn) of p-FC H -COOH, 2, were massed. To
6
4
observed reactivity with 1 is from the formation of this solid, 12 mL of 1M KOH(aq) were added, the vial was
sealed, shaken and sonicated to give a light brown colored
undetectable amounts of the highly catalytically
homogeneous solution ([1]=1 mM; [2]=50 mM). A 2.0-mL
aliquot of this stock solution was added to a 4-mL vial, and
active heterogeneous Ru; and that the observed in-
crease in reactivity upon increasing [KOH] is a result
of increased formation of such a species from 1. To di-
rectly address this point, several carefully designed
studies will be needed to determine if indeed this is
the case; including detailed selectivity studies and ex-
2
50 mL of a MeOH/KOD/D O standard solution were added
2
1
19
for analysis by H and F NMR, providing the t point for
0
the reaction. Another 2.0-mL aliquot of this solution was
added to a 10-mL Teflon cup equipped with a micro-cross
stir bar; this cup was then placed inside of a custom built
amining if there is a time-dependence upon reaction 316-SS mini-reactor, and the reactor was sealed. The solu-
rate when pre-heating 1 or 6 before substrate addi- tion was degassed via cycling 10ꢁ with 500 psi argon and
Adv. Synth. Catal. 2013, 355, 632 – 636
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
635

Michael M. Konnick et al.
COMMUNICATIONS
then 10ꢁ with 400 psi H . The reactor was pressurized to the
desired H2 pressure after the last gas cycle (typically
117, 8674; c) J. Vela, J. M. Smith, Y. Yu, N. A. Ketterer,
C. J. Flaschenriem, R. J. Lachicotte, P. L. Holland, J.
Am. Chem. Soc. 2005, 127, 7857; d) U. Jꢆger-Fiedler,
M. Klahn, P. Arndt, W. Baumanna, A. Spannenberg,
V. V. Burlakov, U. Rosenthal, J. Mol. Catal. A 2007,
261, 184; e) K. Fuchibe, Y. Ohshima, K. Mitomi, T.
Akiyama, J. Fluorine Chem. 2007, 128, 1158; f) D.
Breyer, T. Braun, A. Penner, Dalton Trans. 2010, 39,
7513; g) M. F. Kꢃhnel, D. Lentz, Angew. Chem. 2010,
122, 2995; Angew. Chem. Int. Ed. 2010, 49, 2933;
h) J. A. Panetier, S. A. Macgregor, M. K. Whittlesey,
Angew. Chem. 2011, 123, 2835; Angew. Chem. Int. Ed.
2011, 50, 2783; i) H. Yang, H. Gao, R. Angelici, Orga-
nometallics 1999, 18, 2285, and references cited therin.
[8] B. G. Hashiguchi, S. M. Bischof, M. M. Konnick, R. A.
Periana, Acc. Chem. Res. 2012, 45, 885.
2
400 psi). The reactor was then heated to the desired temper-
ature (1008C) for the desired length of time (typically 1 h).
The reactor was then cooled in an ice-water bath, depressur-
ized and the headspace was cycled 5ꢁ with argon to ensure
removal of all H . The reactor was then opened, and 250 mL
2
of the MeOH/KOD/D O standard solution were added to
2
the solution in the reactor. The solution was stirred for 5–
1
19
1
0 min, and an aliquot was taken for H and F NMR analy-
sis. A procedure identical to that described above is fol-
lowed when RuCl (H O), 6, is utilized.
ACHTUNGTRENNUNG
3
2
See the Supporting Information for additional experimen-
tal details, reaction considerations, MeOH/KOD/D O stock
2
1
19
solution preparation, representative H and F NMR spec-
troscopic analyses, and pictorial examples of solutions of 6
in 1M KOH(aq).
[9] While in the original report we generated the active
Ru(II) species via reduction of 1 with Zn, we have re-
cently determined that H can also serve as a suitable
reductant to initiate this chemistry.
10] B. G. Hashiguchi, K. J. H. Young, M. Yousufuddin,
W. A. Goddard III, R. A. Periana, J. Am. Chem. Soc.
2
Acknowledgements
[
[
[
We gratefully acknowledge The Scripps Research Institute for
financial support.
2
010, 132, 12542.
11] The ability of late-transition metals to catalyze HDF
with subsequent arene hydrogenolysis has long been
[6a]
References
known; see ref. and references cited therin.
12] Initial studies suggested that the metal species present
in solution before treatment with either Zn(0) or H2
[
1] a) D. OꢂHagan, J. Fluorine Chem. 2010, 131, 1071;
b) W. K. Hagmann, J. Med. Chem. 2008, 51, 4359;
c) K. L. Kirk, Org. Process Res. Dev. 2008, 12, 305;
d) A. M. Thayer, Chem. Eng. News 2006, 84, 15;
e) S. M. Ametamey, M. Honer, P. A. Schubiger, Chem.
Rev. 2008, 108, 1501; f) S. J. Ainsworth, Chem. Eng.
News 2008, 86, 15; g) J. H. Clark, D. Wails, T. W. Ba-
stock, Aromatic Fluorination, CRC, Boca Raton, 1996.
2] a) K. Uneyama, Organofluorine Chemistry, Blackwell,
Oxford, 2006; b) D. F. McMillen, D. M. Golden, Annu.
Rev. Phys. Chem. 1982, 33, 493.
are the mixed aquo/hydroxo species [(H IPI)Ru-
m
3
Àn (1+mÀn)
A
H
U
G
R
N
U
G
nA
H
N
T
E
N
N
]
, which are rapidly formed
upon loss of the bound Cl from 1 to solution in aque-
2
À
[10]
ous KOH; see ref.
[13] The authors realize that such an argument assumes
À
a mononuclear Ru species, ignoring the effects of O
stabilization via the formation of multinuclear oxo-
[
bridged complexes.
[6a]
[14] See ref. and references cited therein.
[15] A. Roucoux, J. Schulz, H. Patin, Chem. Rev. 2002, 102,
3757.
[3] a) S. J. Blanksby, G. B. Ellison, Acc. Chem. Res. 2003,
3
6, 255; b) D. OꢂHagan, Chem. Soc. Rev. 2008, 37, 308.
[16] R. J. Young Jr, V. V. Grushin, Organometallics 1999, 18,
294.
[
4] R. Bosque, E. Clot, S. Fantacci, F. Maseras, O. Eisen-
stein, R. N. Perutz, K. B. Renkema, K. G. Caulton, J.
Am. Chem. Soc. 1998, 120, 12634.
[17] It is reasonable to suggest that under these reducing
condtions that 6 is readily reduced to Ru metal. How-
ever, control experiments utilizing Ru powder [Ru(0)
metal] as the ruthenium source under identical condi-
tions showed no conversion of 2; suggesting that such
species are not responsible for catalytic turnover.
[
5] a) C. Douvris, O. V. Ozerov, Science 2008, 321, 1188;
b) V. J. Scott, R. C¸ elenligil- C¸ etin, O. V. Ozerov, J. Am.
Chem. Soc. 2005, 127, 2852; c) R. Panisch, M. Bolte, T.
Mꢃller, J. Am. Chem. Soc. 2006, 128, 9676; d) C. Douv-
ris, E. S. Stoyanov, F. S. Tham, C. A. Reed, Chem.
Commun. 2007, 1145, and references cited therein.
6] For excellent relevant reviews concerning this topic,
please see: a) H. Amii, K. Uneyama, Chem. Rev. 2009,
[6b]
[18] See ref. and references cited therein.
[19] The mechanistic proposals described for 1 below can
be seen as theoretically valid for heterogeneous surface
chemistry as well (utilizing the precatalyst 6); however,
the heterogeneous nature would likely complicate
a more detailed mechanistic analysis.
[20] a) G. J. Kubas, in: Metal Dihydrogen and Sigma-bond
Complexes-Structure, Theory and Reactivity, (Ed.:
J. J. P. Fackler), Kluwer Academic/Plenum, New York,
2001; b) C. P. Lau, S. M. Ng, G. Jia, Z. Lin, Coord.
Chem. Rev. 2007, 251, 2223, and references cited there-
in.
[
1
09, 2119–2183; b) A. Nova, R. Mas-Ballestꢄ, A.
Lledꢅs, Organometallics 2012, 31, 1245–1256; c) R. P.
Hughes, Eur. J. Inorg. Chem. 2009, 4591; d) T. Braun, F.
Wehmeier, Eur. J. Inorg. Chem. 2011, 613–625; e) E.
Clot, O. Eisenstein, N. Jasim, S. A. Macgregor, J. E.
McGrady, R. N. Perutz, Acc. Chem. Res. 2011, 44, 333–
3
48; f) V. V. Grushin, Acc. Chem. Res. 2010, 43, 160;
g) W. D. Jones, Dalton Trans. 2003, 3991; h) H. Torrens,
Coord. Chem. Rev. 2005, 249, 1957.
[21] a) W. K. Wilmarth, J. C. Dayton, J. M. Flournoy, J. Am.
Chem. Soc. 1953, 75, 4549; b) J. M. Flournoy, W. K.
Wilmarth, J. Am. Chem. Soc. 1961, 83, 2257.
[
7] a) M. Aizenberg, D. Milstein, Science 1994, 265, 359;
b) M. Aizenberg, D. Milstein, J. Am. Chem. Soc. 1995,
636
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 632 – 636