The AZARYPHOS family of ligands for ambifunctional catalysis: Syntheses and use in ruthenium-catalyzed anti-markovnikov hydration of terminal alkynes

Article abstract of DOI:10.1002/chem.200900563

The family of AZARYPHOS (aza-aryl-phosphane) phosphane ligands, containing a phosphine unit and sterically shielded nitrogen lone pairs in the ligand periphery, is introduced as a tool for developing ambifunctional catalysis by the metal center and nitrogen lone pairs in the ligand sphere. General synthetic strategies have been developed to synthesize over 25 examples of structurally diverse (6-aryl-2pyridyl)phosphanes (ARPYPHOS), (6alkyl-2-pyridyl)phosphanes (ALPY-PHOS), 4,6-disubsituted l,3-diazin-2ylphosphanes or l,3,5-triazin-2- ylphosphanes, quinazolinylphosphanes, quinolinylphosphanes, and others. The scalable syntheses proceed in a few steps. The incorporation of AZARYPHOS ligands (L) into complexes [RuCp(L)₂(MeCN)][PF₆] (Cp = cyclopentadieny1)gives catalysts for the anti-Markovnikov hydration of terminal alkynes of the highest known activities. Electronic and steric ligand effects modulate the reaction kinetics over a range of two orders of magnitude. These results highlight the importance of using structurally diverse ligand families in the process of developing cooperative ambifunctional catalysis by a metal and its ligand.

Full text of DOI:10.1002/chem.200900563

FULL PAPER
DOI: 10.1002/chem.200900563
The AZARYPHOS Family of Ligands for Ambifunctional Catalysis:
Syntheses and Use in Ruthenium-Catalyzed anti-Markovnikov Hydration of
Terminal Alkynes**
Lukas Hintermann,*[a] Tuan Thanh Dang,[a] Aurꢀlie Labonne,[a] Thomas Kribber,[a]
Li Xiao,[a] and Pance Naumov[b]
Abstract: The family of AZARYPHOS
(aza–aryl–phosphane) phosphane li-
gands, containing a phosphine unit and
sterically shielded nitrogen lone pairs
in the ligand periphery, is introduced as
a tool for developing ambifunctional
catalysis by the metal center and nitro-
gen lone pairs in the ligand sphere.
General synthetic strategies have been
developed to synthesize over 25 exam-
ples of structurally diverse (6-aryl-2-
pyridyl)phosphanes (ARPYPHOS), (6-
PHOS), 4,6-disubsituted 1,3-diazin-2-
ylphosphanes or 1,3,5-triazin-2-ylphos-
phanes, quinazolinylphosphanes, quino-
linylphosphanes, and others. The scala-
ble syntheses proceed in a few steps.
The incorporation of AZARYPHOS li-
gands (L) into complexes [RuCp(L)2-
gives catalysts for the anti-Markovni-
kov hydration of terminal alkynes of
the highest known activities. Electronic
and steric ligand effects modulate the
reaction kinetics over a range of two
orders of magnitude. These results
highlight the importance of using struc-
turally diverse ligand families in the
process of developing cooperative am-
bifunctional catalysis by a metal and its
ligand.
A
ACHTUNGTRENNUNG
·
·
phosphane ligands · ruthenium ·
synthetic methods
alkyl-2-pyridyl)phosphanes
(ALPY-
Introduction
means of an appropriately placed functional group within
the ligand structure.[2,3] If this secondary interaction assists
in accelerating the slowest step of a catalytic reaction, the
overall reaction rate may dramatically increase. Several ex-
amples involving pyridylphosphane ligands imply that coop-
erative ambifunctional catalysis may be a generally useful
concept in developing or optimizing catalytic reactions
(Scheme 1): Drent and co-workers found that the palladi-
um-catalyzed methoxycarbonylation of propyne is faster and
more regioselective with pyridylphosphanes[4] instead of tri-
phenylphosphane as the ligand.[5,6] The protonated pyridine
unit may act as a general acid in an assisted hydrodemetala-
tion.[5a,b] Alternatively, a proton transfer from pyridinium to
the metal-bound alkyne could facilitate the generation of a
vinyl-palladium(II) species (Scheme 1a).[7] Notably, the
placement of a methyl group in the 6-position of the pyridyl
unit further increased the regioselectivity of the catalytic re-
action.[5]
Ligand steering effects are crucial for tuning catalyst perfor-
mance, and extended ligand screens are now primary tools
in the search for new catalysts.[1] The concept of (coopera-
tive) ambifunctional catalysis (or bifunctional/multifunction-
al catalysis) proposes that a ligand, modulating the catalyti-
cally active metal center, additionally interacts with metal-
bound substrates in a specific and directed manner by
[a] Dr. L. Hintermann, Dr. T. T. Dang, Dr. A. Labonne,
Dipl.-Gyml. T. Kribber, Dr. L. Xiao
Institute of Organic Chemistry
RWTH Aachen University
Landoltweg 1, 52074 Aachen (Germany)
Fax : (+49)241-8092391
E-mail: lukas.hintermann@oc.rwth-aachen.de
[b] Prof. Dr. P. Naumov
Department of Material and Life Science
Graduate School of Engineering, Osaka University
2-1 Yamadaoka, Suita, Osaka 565-0871 (Japan)
Berke and co-workers synthesized 6-tert-butyl-2-diphenyl-
phosphinopyridine to explore the steric shielding of a nitro-
gen lone pair in pyridylphosphanes as a tool for preventing
[**] AZARYPHOS=aza–aryl–phosphane.
[8]
ꢀ
P N chelation. When studying heterocyclic phosphanes as
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900563.
ligands in bifunctional catalysis,[3] Grotjahn and co-workers
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7167

structural optimization (Scheme 2). Single representatives of
the ligand family have previously been prepared.[17] In the
present work, we describe straightforward synthetic ap-
proaches to AZARYPHOS ligands bearing all kinds of sub-
Scheme 1. Metal-bound pyridylphosphanes involved in (presumed) ambi-
functional interactions with catalysis substrates: a) selectivity- and rate-
enhancement in alkyne methoxycarbonylation, according to Drent[5] and
[10a]
Matteoli;[7] b) hydrogen bridges to acetylene C H bonds;
c) a reaction
ꢀ
intermediate of catalytic anti-Markovnikov alkyne hydration;[10b] d) heter-
olytic splitting of H2 at ruthenium;[13] e) a presumed catalytic intermedi-
ate in a H2/D2 exchange reaction;[13b] f) acceleration of ruthenium-cata-
lyzed nitrile hydration through inner-sphere general base activation of
water.[14]
Scheme 2. General structure definition of AZARYPHOS ligands and
pertinent sub classes, for which syntheses are described in the present
work.
found that the same ligand impressively accelerates rutheni-
um-catalyzed anti-Markovnikov hydration of terminal al-
kynes.[9] The pyridine lone pairs are hydrogen-bond accept-
stituents.[18] An evaluation of the ligand class in ruthenium-
catalyzed anti-Markovnikov hydration of terminal alkynes
reveals that the new AZARYPHOS ligands cover a wider
range of catalytic activities, depending on steric and elec-
tronic ligand properties. This proves that the concept of
fine-tuning of ligand steering effects extends to ambifunc-
tional catalysis.
ꢀ
ors towards C H bonds of coordinated acetylene
(Scheme 1b).[10a] In a reaction intermediate of catalytic
alkyne hydration, a protonated pyridine unit acts as hydro-
gen-bond donor to a metal acyl species (Scheme 1c).[10b]
Considering these findings, an active involvement of the pyr-
idyl group as a proton shuttle in catalysis is very probable.
Grotjahn also found that heterocyclic phosphane ligands
speed up the prototropic equilibrium between p-alkyne and
metal vinylidene complexes of rhodium[11] or ruthenium,[10a]
and observed an impressive acceleration of ruthenium-cata-
lyzed olefin isomerization by means of ionic 1,3-hydrogen
shifts.[12] Catalytic H2/D2-exchange reactions may benefit
from ambifunctional activation of the heterolytic split of co-
ordinated molecular hydrogen by a pyridylphosphane ligand
(Scheme 1d).[13] It was proposed that the resulting hydride
complex stores a bridging proton between two pyridyl
groups (Scheme 1e).[13b] Finally, the acceleration of rutheni-
um-catalyzed hydration of nitriles by 2-diphenylphosphino-
pyridine has been ascribed to general base activation of
water by a pyridine lone pair.[14] Other ligand classes have
been applied in ambifunctional catalysis,[2,3,15] but many ex-
amples and much substantial mechanistic evidence involve
the pyridylphosphane fragment. Although the parent 2-di-
phenylphosphinopyridine (L1) has found several applica-
tions in catalysis,[4b,16] the above precedence implies that
shielded pyridylphosphanes are particularly valuable ligands.
Of those, only 6-methyl-2-diphenylphosphinopyridine (L2)
is commercially available. In the absence of general synthet-
ic routes to 6-substituted pyridylphosphanes,[17] the ligand
screening approach in bifunctional catalysis development is
limited. We now define the family of AZARYPHOS (aza–
aryl–phosphane) ligands for application in ambifunctional
catalysis, which lend themselves to ligand screening and
Results and Discussion
Definition of the AZARYPHOS ligand structure: In a most
convincing demonstration of ambifunctional acceleration in
catalysis, Grotjahn and co-workers[3,9] included tBuPyPPh2
(L7; Py=pyridine-2,6-yl)[8] into the cationic fragment
[RuCp(L7)2]+ (Cp=cyclopentadienyl) to obtain a catalyst
for anti-Markovnikov hydration of terminal alkynes to alde-
hydes[9,19–24] with a roughly 1000 times higher activity than
[RuClCp
G
(dppm=1,1-bis(diphenylphosphino)me-
thane).[9] Through working on applications of alkyne hydra-
tion in synthetic sequences,[24] we were immediately drawn
to the Grotjahn catalyst. However, the involved synthesis of
L7[8] presented an obstacle in the way of further studies of
the new catalyst. Furthermore, ligand L7 was singular in its
peculiar ability to accelerate catalysis and did not lend itself
easily to ligand variation studies. We wished to develop new
tunable and readily available classes of ligands for bifunc-
tional catalysis. In 2006, we introduced 6-aryl-2-diphenyl-
phosphinopyridines as highly efficient ligands for rutheni-
um-catalyzed anti-Markovnikov hydration of terminal al-
kynes.[23] Later, our discovery of the catalytic cross-coupling
of tertiary alkyl nucleophiles opened a straightforward syn-
thetic route to 6-tert-alkylpyridylphosphanes including L7.[25]
Now, we generalize and extend our studies by defining the
class of AZARYPHOS (aza–aryl–phosphane) ligands
7168
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
(Scheme 2), which share the properties of a 1,2,3-arrange-
ment in a 6-membered heterocyclic core of a phosphorus
donor, a nitrogen lone pair, and an R substituent of varying
size (Scheme 2). In terms of applications in ambifunctional
catalysis, the phosphorus donor will bind to a metal, the ni-
trogen lone pair will undergo secondary interactions with
substrates, and the additional substituent R will serve as a
steric shield of the nitrogen lone pair (see below).
In the following we present general synthetic routes to (6-
alkyl-2-pyridyl)phosphanes (ALPYPHOS) bearing primary,
secondary, or tertiary alkyl groups, to (6-aryl-2-pyridyl)phos-
phanes (ARPYPHOS), and to other aza–aryl phosphanes
(AZARYPHOS) including pyrimidine, 1,3,5-triazine, quino-
line, quinazoline, and 6-alkoxypyridine derivatives. A key
structural element is the steric shielding of the nitrogen lone
pair by an ortho substituent; the size of this group R
(Scheme 3) will determine the stability difference between
Scheme 4. Retrosynthetic disconnection strategies for some AZARY-
PHOS ligands.
carbonyl precursors such as pyrimidones or pyridones
(Scheme 4).
Synthesis of ALPYPHOS ligands from halogenated precur-
sors: All aza–aryl–phosphanes in this study have been ob-
tained by means of nucleophilic phosphination from halo-
genated precursors. ALPYPHOS (alkyl–pyridyl–phosphane)
ligands with either primary, secondary, or tertiary alkyl
groups in the 6’-position of the pyridine nucleus have been
prepared (Table 1).
Scheme 3. Role of the shielding group R in suppressing chelation and
opening up of a coordination site for the catalysis substrate in proximity
to the ligand nitrogen lone pair.
the chelate and the open coordination mode through desta-
bilizing steric interactions with additional steering ligands
(L) at the metal (Scheme 3). Larger R substituents will help
to liberate a free site of coordination for a substrate at the
metal center.
The commercially available compounds 2-diphenylphos-
phinopyridine (L1), also readily prepared from 2-chloropyri-
dine (entry 1), and 6-methylpyrid-2-yl-diphenylphosphane
(L2) from 4 (entries 2 and 3), are included for completeness.
The phosphinations proceeded uneventfully. The synthesis
of 1’-hydroxy-1,1-diphenylmethylpyridylphosphane L10 de-
serves special mention, since this ligand contains a hydroxy
group as added functionality (Scheme 5).
Syntheses of AZARYPHOS ligands
Retrosynthetic analysis: The synthesis of pyridylphosphanes
has been reviewed.[4a] Initial preparations relied on the reac-
tion of metalated pyridines (M=Li, Mg, Zn[26]) with chloro-
phosphanes.[4a] Substitutions (SNAr) of halopyridines with
alkali phosphanides[27,28] (LiPPh2, NaPPh2, or KPPh2) have
since emerged as the more general method. The problem of
accessing a certain aza–aryl phosphane is usually reduced to
accessing the corresponding halogenated heterocycle. Only
simple starting materials like 2-chloropyridine, 2-bromopyri-
dine, or 6-methyl-2-chloropyridine are commercially avail-
able. There is no single general route to 6-substituted 2-hal-
opyridines.[17] Our retrosynthetic analysis of the AZARY-
Scheme 5. Synthesis of a hydroxyalkyl pyridylphosphane by means of
temporary silyl protection.
PHOS ligand structure (Scheme 4) starts with the disconnec-
[29]
tion of C R/R’ (path A) or C P bonds (path B) to halo-
genated precursors. By experiment, we found that path A is
viable for SNAr reactions with, for example, alkoxides, but
not successful for catalytic cross-coupling reactions of phos-
phorus-bearing precursors. Path B, which leads to substitut-
ed halogenated azacycles, is more general. The latter are fur-
Selective lithiation of 2,6-dibromopyridine (2)[30,31] and re-
action with benzophenone[32] gave hydroxyalkylpyridine 12
(Scheme 5). Reaction of the latter with NaPPh2 gave deha-
logenated alcohol 13 and presumably a phosphine coupling
product 14 instead of the desired L10 (Scheme 5). Appa-
rently, the presence of a proton donor induces this redox re-
action. Thus, alcohol 12 was first silylated and the trimethyl-
silane (TMS)-protected phosphine L9 obtained successfully
ꢀ
ꢀ
ꢀ
ther disconnected by C C cleavage to give multiply halo-
ꢀ
genated precursors, or by C X cleavage to give heterocyclic
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7169

L. Hintermann et al.
Table 1. Synthesis of ALPYPHOS ligands from halogenated precur-
sors.[a]
matography on silica gel under argon, or by
recrystallization, which was more convenient on a large
scale. Selective crystallization by seeding of cooled (ꢀ408C)
supersaturated methanol solutions was advantageous.
Entry
1
Substrate
M
K
Product
Yield
[%]
Synthesis of other AZARYPHOS ligands: Heterocyclic
phosphanes were also obtained from chlorinated diazine
and triazine precursors, chloroquinolines, or chloroquinazo-
lines (Table 3). These phosphinations proceeded uneventful-
ly. Both mono-[34] and bis-phosphination[35] of dihalopyri-
dines have been reported. We achieved monosubstitution of
2,6-dichloropyridine (1) with NaPPh2. An attempted alkyla-
tion of L25 with 4 equiv of a tBuMgCl/CuCN (2:1) re-
agent[36] gave a reaction mixture from which the desired L7
could not be isolated. On the other hand, the chlorine in
L25 is readily substituted with sodium alkoxides in dimethyl
formamide to give alkoxypyridylphosphanes L26 and L27
(Table 4).
69
3
L1
2
Li
77[b]
4
4
L2
L2
3
4
Li
K
89[c]
61
5
6
L3
L4
5
6
7
8
K
70
57
65
73
Properties of AZARYPHOS ligands: Aza–aryl–phosphanes
are colorless crystalline solids, which crystallize from metha-
nol (exceptions: L5 is low-melting, L3 and L4 were obtained
as oils). They are stable to air as solids, but oxidize to phos-
phane oxides in solution. The recording of NMR spectra in
CDCl3 occasionally gave erratic results, presumably due to
reactions with solvent impurities. Consistent spectral data
were obtained for samples in degassed deuterobenzene.
Large colorless crystals of L8 crystallized from methanol at
ꢀ208C. The structure as determined by X-ray crystallogra-
phy is shown in Figure 1.[37] The arrangement of the diphe-
nylphosphino group (with one N-C-P-Caryl dihedral angle at
roughly 1808) has also been found in other pyridylphos-
phanes and is often retained in metal complexes.[38]
K
7
8
L5
L6
Na
Na
9
L7
9
Na
69
10
10
L8
L8
10
11
Li
79[d]
39
Na
11
L9
12
–
26[e]
L9
L10
[a] See the Supporting Information for experimental details. [b] Ref. [8],
compare with ref. [28]. [c] Ref. [9]. [d] Large-scale reaction (0.18 mol).
[e] 26% in two steps from 12 without isolation of L9.
in a nonoptimized yield of 39%.[33] Trifunctional ligand L10
was then liberated by acidic hydrolysis of L9 in a nonopti-
mized yield of 26% over 3 steps.
Figure 1. X-ray crystal-structure determination of L8;[37] ORTEP view.
Synthesis of ARPYPHOS ligands from halogenated precur-
sors: The nucleophilic phosphination of 2-aryl-6-halopyri-
dines in THF or in solvent mixtures composed of either di-
ethyl ether or THF and toluene (added for solubility rea-
sons) gave the target (6-aryl-2-pyridyl)phosphanes (ARPY-
PHOS) (Table 2). They were either purified by flash chro-
Synthesis of halogenated precursors
6-Alkyl-2-halopyridines by means of a new Wittig synthesis
of pyridones: Berke et al. had obtained 6-tert-butyl-2-chloro-
pyridine (9) via pyridone 30 from pinacoline.[8] The lengthy
procedure gave an overall yield of 6% over 5 steps,[39] limit-
7170
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
Table 2. Synthesis of ARPYPHOS ligands.[a]
hyde and ylide 31, the latter
derived from phosphonium
salt 32 and base (Scheme 6a).
Alternatively, the mixing of a
sodium b-ketoaldehyde eno-
late (33: R=tBu) and 32
should give an ion pair 34 in
an acid/base equilibrium with
b-ketoaldehyde and ylide.
Entry
1
Substrate
M
Product
Yield
[%]
Na
86
Eventually,
should lead to pyridones
(Scheme 6b).
Indeed, mixing of enolate
33 and 32[40] gave Wittig cou-
pling products that cyclized
to 30 after acidification. Pyri-
done 30 was produced in
79% yield (>30 g scale) in a
both
routes
15
16
17
18
19
20
L11
L12
L13
L14
2
3
4
5
6
Na
Na
Na
Na
K
79
75
71
78
69
one-pot
reaction
(Scheme 7a). This new pyri-
done synthesis is more gener-
al, but not always high yield-
ing. In another application
example,
6-cyclohexylpyri-
din-2(1H)-one (35) was ob-
tained from ketoaldehyde 36,
phosphonium salt 32, and a
base in moderate overall
yield (Scheme 7b). Side prod-
ucts included triketone 37
(by trimerization of 33) and
2,6-dicyclohexylpyridine (38).
The intermediacy of open-
chain unsaturated amides in
this synthesis was evident in
the reaction of hydroxy-
methylene camphor (39)[41]
with excess 32 and triethyla-
mine in ethanol (Scheme 8).
The resulting amide mixture
(E/Z)-40 cyclized to pyridone
41 under forcing conditions
in an ammonium acetate/
acetic acid melt (Scheme 8).
The intermediates (E)-40
(major product) and (Z)-40
(minor product) could also
be isolated. Pure (E)-40 in
EtOH with acid (p-toluene
L15
L15
7
8
Na
Na
76
69
21
22
L16
L17
9
K
58
70
23
24
L18
L19
10
Na
sulfonic
acid
(pTsOH))
under reflux conditions gave
an equilibrium mixture of
(E)-40 and (Z)-40. The E/Z
isomerization is a prerequi-
site to cyclization, however,
pyridone 41 was not formed
[a] Typical conditions: 1 equiv of MPPh2 in THF, reaction in Et2O/toluene or THF at 08C to RT. Reaction
time: 1 to 8 h. See the Supporting Information for details.
ing access to the material. We devised a shorter synthesis of
6-pyridones by retrosynthetic disconnection to a b-ketoalde-
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7171

L. Hintermann et al.
Table 3. Synthesis of pyrimidyl, triazinyl, and annelated ligands.[a]
Entry
1
Substrate
Product
Yield
[%]
87
25
26
L20
L21
2
69
Scheme 6. Projected one-pot synthesis of 2-(1H)pyridones by means of
Wittig condensation.
3
4
5
82
80
75
27
28
L22
L23
dures. Therefore, for synthesizing L6 and L7 we now prefer
the cross-coupling approaches described below for 6-ar/
alkyl-2-chloropyridines and the copper-catalyzed tert-alkyla-
tion of multiply chlorinated azacycles. The new pyridone
synthesis will still be valuable for accessing targets that
cannot be obtained by cross-coupling approaches, like tricy-
clic pyridone 41 (Scheme 8).
29
L24
2-Bromo-6-arylpyridines by monoarylation of 2,6-dibromo-
pyridine (2): Cross-coupling of 2,6-dihalopyridines (1: Cl; 2:
Br) are usually performed as
[a] See the Supporting Information for experimental details.
Table 4. Synthesis of ALPYPHOS ligands from halogenated precursors.[a]
double substitutions. Prece-
Entry
1
Substrate
Reagent[b]
T
t
Product
Yield
[%]
dence implies that monosubsti-
[8C] [h]
tutions,[42] though seldom per-
formed with
CACTHNGUTERNNUG
(sp3) nucleo-
NaPPh2 (1.0), THF
0
1.0
62
philes,[43] should proceed selec-
tively with 2,6-dibromopyridine
(2) and hindered aryl metals.[44]
Bulky aryl Grignard reagents
1
1
L25
L25
2
3
NaPPh2 (1.0), tBuOMe
0
1.0
1.5
91
60
tAmOH (1.5), NaH (1.5), DMF
90
were coupled with
[NiBr2A(DME)]/PCy3 (DME=
dimethoxyethane, Cy=cyclo-
2 and a
L25
L25
L26
CTHUNGTRENNUNG
hexyl) catalyst in dioxane at ex-
tended reaction times of 72 h
by Kempe and co-workers.[45]
We used a mixed THF/toluene
solvent and air-stable complex
(1R)-menthol (1.5), NaH (1.5), DMF 90
1.5
95
L27
[a] See the Supporting Information for experimental details. [b] tAmOH=tert-amyl alcohol.
[NiCl2ACHTUNGTRNEUNG
(PCy3)2][46] as the catalyst
under these milder conditions, possibly reflecting ring strain
in the tricyclic product.
instead, and reliably obtained 2-bromo-6-arylpyridines in
only a few hours (Table 5).
Pyridones 30 and 35 were chlorinated either by POCl3
alone, PCl5 in POCl3, or POCl3 in DMF. The mixed reagents
are more reactive. Pyridone 30 reacted only partially in
POCl3, but was fully consumed by POCl3/PCl5
(Scheme 9).[8,9] The chloropyridines were then phosphinated
as described in Table 1. The new Wittig synthesis of pyri-
dones provides scalable access to ligands L7 (33% over 4
steps from pinacolone) and L6 (12% over 4 steps), but the
pyridone syntheses require laborious purification proce-
This approach was less successful for small aryl groups,
where the resulting mixtures of mono- and diarylation prod-
ucts with starting material were difficult to separate (en-
tries 1–4). The solubility of the reaction products dropped
with increasing size of the C-6 aryl substituent. Steric limits
of the reaction were evident in the case of the 2,4,6-triphe-
nylphenyl derivative (entry 8). A reaction time of several
days and a higher catalyst loading was required. Remarka-
bly, the coupling catalyst developed by Herrmann and co-
7172
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
Table 5. Synthesis of ARPYPHOS ligands from halogenated precursor-
s.[a]
Entry
1
ArMgBr
[equiv]
[Ni]
[%]
t
Product
Yield
G
[h]
[b]
1.1
0.3
3
–
42
16
2
3
1.1
0.3
7
<70[c]
[b]
1.1
1.1
0.3
0.3
12
–
Scheme 7. Application examples of the new 2-pyridone synthesis.
21-Br
4
16
<61[c]
22
24
18
17
19
5
6
7
8
1.2
1.1
1.1
1.5
0.6
0.3
0.3
1.0
15
15
83
67
30
83
Scheme 8. Open-chain intermediates in the synthesis of a camphor-de-
rived pyridone.
192
57[d]
[a] See the Supporting Information for experimental details. [b] The
product could not be separated from impurities. [c] The products con-
tained a little diarylpyridine impurity. [d] Modification: heating for 4 d at
758C and for 4 d at 858C.
Scheme 9. Conversion of 2-pyridones to 2-chloropyridines and ALPY-
PHOS ligands.
these conditions; its amount depended on the excess of
Grignard reagent used (Table 6).
A general two-step synthesis of 6-ar/alkyl-2-chloropyridines
from 2-tert-butoxy-6-chloropyridine (45): Since the direct ar-
ylation approach described above failed for small aryl
Grignard reagents, an indirect synthesis of 2-halo-6-arylpyri-
dines was needed. Starting from cheap 2,6-dichloropyridine,
workers,[47] consisting of [Ni
ACTHNUTRGNEN(UG acac)2] (acac=acetylacetonate)
and the imidazolium salt IPr·HCl (43),[48] precursor to a N-
heterocyclic carbene ligand,[49] gave 19 within a few hours at
858C (Table 6)![50] Even diarylpyridine 44 formed under
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7173

L. Hintermann et al.
Table 6. Large scale cross-coupling synthesis of 19.[a]
a strategy was devised, in which a bulky “dummy” nucleo-
phile desymmetrizes the starting material to leave a single
halogen for cross-coupling.[51] In a final step, the “dummy”
group is transformed back to a halogen. According to Breit
and Seiche, 2,6-dichloropyridine (1) reacts with potassium
tert-butoxide to give monoether 45 in high yield.[52] The
latter is readily cross-coupled according to Kumada et al.[53]
with a range of Grignard nucleophiles (Table 7).
The nickel/carbene catalyst reported by Herrmann and
co-workers gave best results with small aryl nucleophiles,[47]
whereas primary and secondary alkyl-Grignard reagents
were coupled under the conditions reported by Kumada
et al.;[53] 1,3-bis(diphenylphosphino)propane (dppp) was the
preferred ligand for couplings of primary, and 1,2-bis(diphe-
Entry Scale
[mmol]
ArlMgBr
[equiv]
[Ni][b]
[mol%]
Yield of 19[c] Yield of 44
[%]
[%]
1
2
3
125
62
254
1.50
1.33
1.08
2.0
2.7
2.0
50
72
74
28
13
6
[a] See the Supporting Information for experimental details. [b] [Ni-
ACHTUNGTRENNUNG(acac)2] was combined with 1.1 equiv of IPr·HCl (43) as ligand precursor.
Table 7. Nickel-catalyzed cross-coupling of 2-tert-butoxy-6-chloropyridine (45) with Grignard reagents, and chlorodealkoxylation to 2-chloro-6-ar/alkyl-
pyridines.[a]
Entry
1
RMgX[b]
G
Ni[c]
[%]
Ligand[d]
IPr
T
[8C]
t
Product A
Yield
A [%]
Product B
Yield
B [%]
[h]
1
RT
RT
15
5
98
90
69
60
46
15
2
1
IPr
47
48
49
21
5
3
4
1.5
1.4
0.3
0.3
dppp
dppp
RT
RT
4
4
89
97
75
79
6
5
6
1.5
1.3
0.3
0.3
dppe
dppe
RT
RT
4
4
92
96
77
94
50
51
7
8
7
8
1.5
3
2
PBu3
60
60
10
8
53
67
66
89
52
53
23
20
1.25
IPr
[a] See the Supporting Information for experimental details. [b] Grignard reagents as solutions in THF. [c] The nickel precursor in combination with IPr
is [Ni(acac)2]; in all other cases, the complexes [NiCl2A(PR3)2] were used; quantity in mol%. [d] Ligands: IPr=1,3-bis(2,6-diisopropylphenyl)imidazolyli-
dene carbene, derived from imidazolium salt 43.
A
CHTUNGTRENNUNG
7174
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
nylphosphino)ethane (dppe) for secondary alkyl groups
(Table 7, 45 ! A). The 9-anthracenyl Grignard reagent
gave unsatisfactory results with carbene ligands, but coupled
Chlorodiazines from 1,3-diketones and urea: Generally, 1,3-
diketones condense with urea in acidic solution,[55,56] but the
reaction turned out to be very slow with the bulky dipiva-
loylmethane (57): hydrolytic decomposition of urea compet-
ed with heterocyclization to 58, and enamine 59 was the
major reaction product (56%) in hot acetic acid (Table 9).
well with [NiCl2ACHTUNGTRENNUNG(PBu3)2] as the catalyst (entry 7). The con-
trary was true for the 2,4,6-triphenylphenyl Grignard re-
agent (entry 8). Reaction of the 2-tert-butoxy-6-ar/alkylpyri-
dines with Vilsmeyer reagent gave chloropyridines in good
yields (Table 7, A!B).[54] This approach of synthesizing 6-
ar/alkyl-2-halopyridines from 2,6-dichloropyridine (1) via 2-
tert-butoxy-6-chloropyridine is very general and extends to
primary or secondary alkyl, and small or hindered aryl
groups, giving pyridylphosphane ligands in only 3 steps from
the storable precursor 45. So far, it only fails for tertiary
alkyl groups, for which catalytic cross-coupling reactions
with chloroarenes are hardly known.
Table 9. 2-Chloropyrimidines from 1,3-diketones via pyrimidones.[a]
Entry
R
Substrate A Product B Yield [%] Product C Yield [%]
1
2
tBu 56
Ph 60
58
61
19
64
25
26
95
75
Copper-catalyzed tert-alkylation of multiply chlorinated aza-
cycles: The difficulties originally encountered in the synthe-
sis of 6-tert-butyl-2-chloropyridine (9) translate to the lack
of suitable cross-coupling methodology for tertiary alkyl nu-
cleophiles and the nonsymmetric nature of the target, which
asked for an lengthy de novo synthesis.[8] We have recently
found a general copper-catalyzed cross-coupling reaction of
tertiary alkyl Grignard reagents with multiply chlorinated
azacycles that solves this problem: dichlorinated azacycles 1
or 54 are selectively monoalkylated, whereas 2,4,6-trichloro-
pyrimidine (55) and trichlorotriazine (56) are selectively dis-
ubstituted (Table 8). Details of this transformation have al-
ready been reported.[25]
[a] See the Supporting Information for experimental details.
The pyrimidones gave 2-
chloropyrimidines by reaction
with POCl3 (Table 9), which
were readily converted to the
symmetric pyrimidylphosphanes
as previously shown in Table 3.
Ligand structure–activity relationships in ruthenium-cata-
lyzed anti-Markovnikov hydration of terminal alkynes: The
anti-Markovnikov hydration of terminal alkynes to alde-
hydes serves as the first test reaction to study ligand struc-
ture–activity relationships. The Grotjahn group found that
complex [RuCp(Ln)2ACTHNUTRGENNUG(MeCN)]ACHTUNTGERN[NGUN PF6] catalyzes the reaction
Table 8. Cu-catalyzed selective monoalkylation of multiply chlorinated
azacycles.[a]
with unprecedented rates when using the bulky ligand L7,
but that incorporation of the smaller L1 or L2 led to either
inactive or very weakly active catalysts.[9] In our communica-
tion[23] we had established that bulky ARPYPHOS ligands
also give highly active in situ catalysts for anti-Markovnikov
hydration of terminal alkynes. Here we generalize our study
of structure–activity relationships of AZARYPHOS ligands.
The hydration catalysts were generated in situ by ligand ex-
Entry
Substrate
RMgX
[equiv]
CuI
[%]
Product
Yield
[%]
AHCTUNGTRENNUNG
1
1.5
5
85
1
9
2
3
1.05
5
92
ACTHNUTRGNEUNG
change from [RuCp(h6-naphthalene)][PF6] (62)[57] with the
AZARYPHOS ligands, followed by evaporation to dryness
54
29
(Scheme 10a).[23]
The catalytic hydration of 1-octyne to octanal was chosen
as the standard reaction for kinetic comparisons
(Scheme 10b). Reaction progress was determined from the
time-dependent octanal concentration by means of gas chro-
matography with tetradecane as the internal standard. Reac-
tion-progress curves were fitted[58] to the rate law shown in
Equation (1):
2.5
10
83
55
25
4
5
2.4
2.0
3.5
5
90
74
56
1
27
10
d½octanalꢁ=dt ¼ kLn½Ruꢁ0½H2Oꢁ½octyneꢁ
ð1Þ
The rate constants for individual ligands (kLn) were divided
by the rate constant of the catalyst derived from L7 (kL7) at
608C to obtain a relative catalytic activity (RCA). As evi-
[a] See the Supporting Information for experimental details.
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7175

L. Hintermann et al.
Steric ligand effects on catalyst performance are crucial in
the ARPYPHOS series. The activity of the catalysts in-
creased in the order R (R=aryl group at pyridine C-6)=Ph
(L11)
< mesityl (L12) < 2,4-iPr2C6H3 (L14) < 2,4,6-
iPr3C6H2 (L13) < 2,4,6-Ph3C6H2 (L15). This roughly corre-
lates with the size of the aryl group. Substitutions in the aryl
group were most effective in the ortho positions. Exception-
ally, the diisopropoxy derivative L19 gave a less-active cata-
lyst than L11 (phenyl) or L12 (mesityl), which indicates that
the oxygen functionality actively interferes with the cataly-
sis. Substitution in the meta position had little effect, since
the catalyst derived from L17 was only slightly more active
than that of L11 (entry 9 vs. 10). In the ALPYPHOS series,
the inactivity of 2-diphenylphosphinopyridine (L1) and the
low catalytic activity of L2 (RCA ca. 0.01) had earlier been
noted.[9] We find that replacing methyl (in L2) by a secon-
dary alkyl group (L6) increases catalytic activity (entry 15),
but only substitution by a tertiary alkyl group (L7; entry 5)
causes a large jump of catalyst activity (0.02!1.0). Within a
series of tertiary ALPYPHOS ligands, L8 is slightly more ef-
fective than L7 (entries 4 vs. 5),[59] but L9 and L10 with func-
tionalized 1,1-diphenylmethyl groups gave less-active cata-
lysts (entries 8 vs. 11). Even though L10 is less bulky than
Scheme 10. a) General scheme of ligand exchange of 62 with AZARY-
PHOS (Ln) in acetonitrile. b) Model reaction for the kinetic comparison
of ligand performance.
dent from Table 10, many in situ complexes were catalytical-
ly active, but their efficiency varied over two orders of mag-
nitude. The regioisomeric byproduct 2-octanone was not de-
tected in any of these reactions, implying a mechanism-
based regiospecificity of the reaction.[3,20c]
Table 10. Relative catalytic activity induced by ligands in the hydration of 1-octyne.[a]
Entry
Ligand
Ln
RCA[b]
[458C]
RCA[b]
[608C]
Entry
Ligand
Ln
RCA[b]
[458C]
RCA[b]
[608C]
G
E
N
ACHTUNGTRENNUNG
[d]
1
L15
0.73
0.41
4.2
1.8
9
L17
–
0.48
0.41
2
L13
10
L11
0.09
[d]
3
4
5
6
L14
L8
0.32
0.35
0.34
0.25
1.3
1.1
1[c]
11
12
13
14
L9
–
0.28
0.16
0.14
0.04
[d]
L21
L22
L19
–
[d]
L7
–
[d]
L12
0.59
–
[d]
[d]
7
8
L20
L10
–
0.56
0.56
15
16
L6
–
0.02
0[e]
[d]
[d]
–
L26/L27
–
[a] See Scheme 10b for the test reaction. Conditions: [1-octyne]=0.25m, [Ru]=5 mol%, 5 equiv H2O. [b] Relative catalytic activities in the test reaction,
see text. [c] Reference value. [d] Not determined. [e] No catalytic activity observed.
7176
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
L9, it generated a slightly more active catalyst. It is tempting
to assume involvement of the hydroxy group of L10 in the
reaction, which could be considered a case of “trifunctional”
catalysis. The effect is quite small and far-reaching conclu-
sions are premature, but future investigations of even more
functionalized ligands are certainly justified. Compared with
the reference hydration catalyst derived from L7, in situ cat-
alysts derived from pyrimidine L20 and triazine L22 were
less active (Scheme 11). The additional 4-tert-butyl groups in
L20 and L22 are presumably not sterically active, because
they will point away from the active site if phosphorus binds
to ruthenium.
class in a few steps with a scalable methodology. Syntheses
often require as little as two steps from readily available
starting materials. General routes to 6-aryl-2-phosphinopyri-
dines, 6-alkyl-2-phosphinopyridines (alkyl=primary, secon-
dary, and tertiary), or other AZARYPHOS ligands like dia-
zinylphosphanes, triazinylphosphanes, or 6-alkoxy-2-phos-
phinopyridines have been elaborated. ARPYPHOS (6-aryl-
2-phosphinopyridines) ligands have been applied in catalysis
for the first time and were found to be highly efficient in
the ruthenium-catalyzed anti-Markovnikov hydration of ter-
minal alkynes. Ambifunctional catalysis by a metal and a
functional group within the steering ligand is a promising
concept for developing new reaction chemistry. Our study of
the ruthenium-catalyzed anti-Markovnikov hydration of ter-
minal alkynes implies that steric and electronic ligand varia-
tion is an essential element also in ambifunctional catalysis
reaction development. The readily available AZARYPHOS
ligands open a door to systematically explore the potential
of ambifunctional catalysis chemistry by means of ligand
screening approaches.
Scheme 11. Relative catalytic activities (RCA) of pyridyl, pyrimidyl, and
triazinyl ligands.
Experimental Section
All experimental data and copies of 1H and 13C NMR spectra of new
compounds have been given in the Supporting Information.
The lower activity of catalysts derived from the nitrogen-
rich ligands is probably a consequence of the lower basicity
of these heterocycles,[60] with a concomitant weaker interac-
tion of their nitrogen lone pairs with the substrate. Addi-
tionally, the s-acceptor effect of the heterocycles will reduce
electron density at ruthenium, with an as yet unclear influ-
ence on the alkyne/vinylidene interconversion[61] or other
steps within the overall mechanism of the reaction. Future
studies should address this issue by modulating the electron
density through variation of the non-heterocyclic aryl
groups at phosphorus. For some AZARYPHOS ligands, the
derived species [RuCp(L)2]+ was not catalytically active:
these include alkoxy-substituted pyridyl ligands L26 and
L27, and the quinoline-derived ligand L23. For the latter, a
Acknowledgements
This work was supported by the DFG (Emmy Noether Programm) and
the Fonds der Chemischen Industrie. We thank Prof. Carsten Bolm,
RWTH Aachen University, for continuing support.
[1] N. G. Anderson, Practical Process Research & Development, Aca-
demic Press, San Diego, 2000, Chapter 9.
[2] a) G. J. Rowlands, Tetrahedron 2001, 57, 1865; b) D. Natale, J. C.
Mareque-Rivas, Chem. Commun. 2008, 425; c) Y. J. Park, J.-W.
Park, C.-H. Jun, Acc. Chem. Res. 2008, 41, 222.
[3] a) D. B. Grotjahn, Dalton Trans. 2008, 6497; b) D. B. Grotjahn,
Chem. Eur. J. 2005, 11, 7146.
[4] a) G. R. Newkome, Chem. Rev. 1993, 93, 2067; b) Z.-Z. Zhang, H.
Cheng, Coord. Chem. Rev. 1996, 147, 1.
[5] a) E. Drent, P. Arnoldy, P. H. M. Budzelaar, J. Organomet. Chem.
1994, 475, 57; b) E. Drent, P. Arnoldy, P. H. M. Budzelaar, J. Orga-
nomet. Chem. 1993, 455, 247; c) J. Keijsper, P. Arnoldy, M. J. Doyle,
E. Drent, Recl. Trav. Chim. Pays-Bas 1996, 115, 248; d) E. Drent,
EP0441446A1, 1991; e) E. Drent, P. H. M. Budzelaar, W. W. Jager,
EP0386833A1; f) G. Kiss, Chem. Rev. 2001, 101, 3435.
chelating species [RuCp
in [D3]MeCN (doublets at d
(P,P)=37.7 Hz). As Grotjahn had previously found, chela-
(h2-L23)(h1-L23)]
ACHTUNGTERNGNUN ACHUTNGTRENNGNU ACHTNGURTEN[NUNG PF6] was formed
AHCTUNGTRENNUNG
JACHTUNGTRENNUNG
tion is a cause of catalytic inactivity with sterically insuffi-
ciently demanding ligands.[3b] In agreement with earlier
work, we found no catalytic activity for the complexes
(PPh3)2].[9,16,62]
[RuCpACHTUNGTRENNUNG(MeCN)ACHTUNGTRENNUNG(PPh3)2]ACHTNUREGTG[NUNN PF6] or [RuClCpCAHTNUGTRNENGUN
[6] a) S. Gladiali, L. Pinna, C. G. Arena, E. Rotondo, F. Faraone, J.
Mol. Catal. 1991, 66, 183; b) G. Franciꢁ, R. Scopelliti, C. G. Arena,
G. Bruno, D. Drommi, F. Faraone, Organometallics 1998, 17, 338.
[7] a) A. Scrivanti, V. Beghetto, E. Campagna, M. Zanato, U. Matteoli,
Organometallics 1998, 17, 630; b) A. Scrivanti, V. Beghetto, U. Mat-
teoli, Adv. Synth. Catal. 2002, 344, 543; c) A. Scrivanti, U. Matteoli,
Tetrahedron Lett. 1995, 36, 9015.
[8] J. Baur, H. Jacobsen, P. Burger, G. Artus, H. Berke, L. Dahlenburg,
Eur. J. Inorg. Chem. 2000, 1411.
[9] D. B. Grotjahn, D. A. Lev, J. Am. Chem. Soc. 2004, 126, 12232.
[10] a) D. B. Grotjahn, V. Miranda-Soto, E. J. Kragulj, D. A. Lev, G. Er-
dogan, X. Zeng, A. L. Cooksy, J. Am. Chem. Soc. 2008, 130, 20;
b) D. B. Grotjahn, E. J. Kragulj, C. D. Zeinalipour-Yazdi, V. Miran-
Conclusion
The AZARYPHOS class of heterocyclic phosphanes has
been defined as a useful and promising ligand family for ap-
plication in ambifunctional catalysis. Prior to our work, only
a few members of this ligand family had been described in
the literature and no general synthetic methodology was
available. We have developed synthetic strategies to effi-
ciently access almost any desired member of this ligand
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7177

L. Hintermann et al.
da-Soto, D. A. Lev, A. L. Cooksy, J. Am. Chem. Soc. 2008, 130,
10860.
[11] a) D. B. Grotjahn, X. Zeng, A. L. Cooksy, J. Am. Chem. Soc. 2006,
128, 2798; b) D. B. Grotjahn, X. Zeng, A. L. Cooksy, W. S. Kassel,
A. G. DiPasquale, L. N. Zakharov, A. L. Rheingold, Organometallics
2007, 26, 3385.
[35] a) O. Herd, A. Hessler, M. Hingst, M. Tepper, O. Stelzer, J. Organo-
met. Chem. 1996, 522, 69; b) S. Al-Benna, M. J. Sarsfield, M. Thorn-
ton-Pett, D. L. Ormsby, P. J. Maddox, P. Bres, M. Bochmann, J.
Chem. Soc. Dalton Trans. 2000, 4247.
[36] T. W. Bell, L. Hu, S. V. Patel, J. Org. Chem. 1987, 52, 3847.
[37] CCDC 713997 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[38] For this comparison, several X-ray crystal structures from the CSD
database have been considered: ALATUC, PODWUA, WILJOQ,
and WILPIQ.
[39] Prof. D. B. Grotjahn, San Diego State University, has informed us
that this synthesis has been optimized in his group to give higher
overall yields. Nevertheless, the lengthy route and overall yield are
still unsatisfactory. Personal communication by e-mail, 2006.
[40] Salt 32 is obtained from chloroacetamide and PPh3 in hot acetoni-
trile; see the Supporting Information.
[41] M. Gianini, A. von Zelewsky, Synthesis 1996, 702.
[42] Review on selectivity in cross-coupling of heterocyclic substrates
with multiple electrophilic sites: S. Schrçter, C. Stock, T. Bach, Tet-
rahedron 2005, 61, 2245.
[43] For monosubstitutions of dihalopyridines, see: a) H. Ohmiya, H.
Yorimitsu, K. Oshima, Chem. Lett. 2004, 33, 1240; b) A. Minato, K.
Tamao, T. Hayashi, K. Suzuki, M. Kumada, Tetrahedron Lett. 1980,
21, 845; c) A. Minato, K. Suzuki, K. Tamao, M. Kumada, J. Chem.
Soc. Chem. Commun. 1984, 511; d) X. L. Cui, R. S. Brown, J. Org.
Chem. 2000, 65, 5653.
[44] For monosubstitutions with aryl metals, see: a) C. R. Woods, M. Be-
naglia, S. Toyota, K Hardcastle, J. S. Siegel, Angew. Chem. 2001,
113, 771; Angew. Chem. Int. Ed. 2001, 40, 749; b) J. C. Loren, J. S.
Siegel, Angew. Chem. 2001, 113, 776; Angew. Chem. Int. Ed. 2001,
40, 754; c) G. Hungerford, M. Van der Auweraer, J.-C. Chambron,
V. Heitz, J.-P. Sauvage, J.-L. Pierre, D. Zurita, Chem. Eur. J. 1999, 5,
2089; d) M. Alami, J.-F. Peyrat, L. Belachmi, J.-D. Brion, Eur. J.
Org. Chem. 2001, 4207; e) A. Puglisi, M. Benaglia, G. Roncan, Eur.
J. Org. Chem. 2003, 1552; f) R.-Z. Jin, Z. Bian, C.-Q. Kang, H.-Q.
Guo, L.-X. Gao, Synth. Commun. 2005, 35, 1897.
[12] D. B. Grotjahn, C. R. Larsen, J. L. Gustafson, R. Nair, A. Sharma, J.
Am. Chem. Soc. 2007, 129, 9592.
[13] a) A. Caballero, F. A. Jalꢂn, B. R. Manzano, Chem. Commun. 1998,
1879; b) F. A. Jalꢂn, B. R. Manzano, A. Caballero, M. C. Carriꢂn, L.
Santos, G. Espino, M. Moreno, J. Am. Chem. Soc. 2005, 127, 15364.
[14] T. Oshiki, H. Yamashita, K. Sawada, M. Utsunomiya, K. Takahashi,
K. Takai, Organometallics 2005, 24, 6287.
[15] For applications in cross-coupling, see: a) S. Ishikawa, K. Manabe,
Org. Lett. 2007, 9, 5593; b) N. Yoshikai, H. Mashima, E. Nakamura,
J. Am. Chem. Soc. 2005, 127, 17978.
[16] a) H. Yoshida, S. Nakano, Y. Yamaryo, J. Ohshita, A. Kunai, Org.
Lett. 2006, 8, 4157; b) M. T. Reetz, R. Demuth, R. Goddard, Tetra-
hedron Lett. 1998, 39, 7089.
[17] While 6-substituted 2-pyridylphosphanes are usually synthesized by
phosphination of the 2-halo precursors (see ref. [4a]), there is no
single obvious synthetic route for the latter. For example, a recent
paper on catalysis described syntheses of four 6-substituted 2-diphe-
nylphosphinopyridines, in which each of the halogenated precursors
was obtained by a different approach: L. P. Spencer, R. Altwer, P.
Wei, L. Gelmini, J. Gauld, D. W. Stephan, Organometallics 2003, 22,
3841.
[18] Based on the synthetic methods reported here, the compounds L8
(“ALPYPHOS“) and L15 (“ARPYPHOS“) have been prepared on
a large scale and are currently on offer from the Aldrich chemical
company.
[19] Review on catalytic hydration of alkynes: L. Hintermann, A. Lab-
onne, Synthesis 2007, 1121.
[20] Initial work on anti-Markovnikov hydration of terminal alkynes:
a) M. Tokunaga, Y. Wakatsuki, Angew. Chem. 1998, 110, 3024;
Angew. Chem. Int. Ed. 1998, 37, 2867; b) T. Suzuki, M. Tokunaga, Y.
Wakatsuki, Org. Lett. 2001, 3, 735; c) M. Tokunaga, T. Suzuki, N.
Koga, T. Fukushima, A. Horiuchi, Y. Wakatsuki, J. Am. Chem. Soc.
2001, 123, 11917; d) T. Suzuki, M. Tokunaga, Y. Wakatsuki, Tetrahe-
dron Lett. 2002, 43, 7531.
[45] N. M. Scott, T. Schareina, O. Tok, R. Kempe, Eur. J. Inorg. Chem.
2004, 3297.
[21] D. B. Grotjahn, C. D. Incarvito, A. L. Rheingold, Angew. Chem.
2001, 113, 4002; Angew. Chem. Int. Ed. 2001, 40, 3884.
[22] F. Chevallier, B. Breit, Angew. Chem. 2006, 118, 1629; Angew.
Chem. Int. Ed. 2006, 45, 1599.
[23] A. Labonne, T. Kribber, L. Hintermann, Org. Lett. 2006, 8, 5853.
[24] a) A. Labonne, L. Zani, L. Hintermann, C. Bolm, J. Org. Chem.
2007, 72, 5704; b) T. Kribber, A. Labonne, L. Hintermann, Synthesis
2007, 2809.
[46] B. Heyn, B. Hipler, G. Kreisel, H. Schreer, D. Walther, Anorgani-
sche Synthesechemie, 2nd ed., Springer, Berlin, 1990, p. 64.
[47] V. P. W. Bçhm, T. Weskamp, C. K. W. Gstçttmayr, W. A. Herrmann,
Angew. Chem. 2000, 112, 1672; Angew. Chem. Int. Ed. 2000, 39,
1602.
[48] IPr·HCl=N,N’-bis(2,6-diisopropylphenyl)imidazolium chloride; this
material was prepared according to our improved procedure: L.
Hintermann, Beilstein J. Org. Chem. 2007, 3, 22.
[25] L. Hintermann, L. Xiao, A. Labonne, Angew. Chem. 2008, 120,
8370; Angew. Chem. Int. Ed. 2008, 47, 8246.
[26] P. H. M. Budzelaar, J. H. G. Frijns, A. G. Orpen, Organometallics
1990, 9, 1222, and references therein.
[27] G. R. Newkome, D. C. Hager, J. Org. Chem. 1978, 43, 947.
[28] P. A. A. Klusener, J. C. L. J. Suykerbuyk, P. A. Verbrugge, EP
0499328A2, 1992.
[29] So far, we have restrained ourselves to using the PPh2 fragment.
[30] J. E. Parks, B. E. Wagner, R. H. Holm, J. Organomet. Chem. 1973,
56, 53.
[49] “N-Heterocyclic Carbenes in Transition Metal Catalysis”: Topics in
Organometallic Chemistry, Vol. 21 (Ed.: F. Glorius), Springer, Hei-
delberg, 2007.
[50] Incidentally, this appears to be the first reported case of a catalytic
cross-coupling reaction of the 2,4,6-triphenylphenyl group.
[51] For use of thiophenol as a “dummy” group to desymmetrize 2,6-di-
bromopyridine, see: V. Bonnet, F. Mongin, F. Trꢄcourt, G. Quꢄguin-
er, P. Knochel, Tetrahedron 2002, 58, 4429.
[52] B. Breit, W. Seiche, J. Am. Chem. Soc. 2003, 125, 6608.
[53] a) M. Kumada, K. Tamao, K. Sumitani, Org. Synth. Coll. Vol. 1988,
6, 407; b) K. Tamao, K. Sumitani, M. Kumada, J. Am. Chem. Soc.
1972, 94, 4374.
[31] D. Cai, D. L. Hughes, T. R. Verhoeven, Tetrahedron Lett. 1996, 37,
2537.
[32] Synthesis of 12 via metalation of 2 in CH2Cl2 solution: M. A. Peter-
son, J. R. Mitchell, J. Org. Chem. 1997, 62, 8237.
[33] Catalytic amounts of CuI were added, which may not be necessary.
For copper-mediated phosphinations, see: C. Meyer, H. Grꢃtzmach-
er, H. Pritzkow, Angew. Chem. 1997, 109, 2576; Angew. Chem. Int.
Ed. Engl. 1997, 36, 2471.
[34] a) S. Kuang, Z. Zhang, Q. Wang, T. C. W. Mak, J. Chem. Soc. Dalton
Trans. 1998, 2927; b) T. Zhang, Y. Qin, D. Wu, R. Zhou, X. Yi, C.
Liu, Synth. Commun. 2005, 35, 1889.
[54] The direct conversion of alkoxypyridines to halopyridines is fairly
well documented for methoxy and a few other alkoxy groups: a) M.
Shiao, K. Tarng, Heterocycles 1990, 31, 637; b) T. R. Kasturi,
H. R. Y. Jois, L. Mathew, Synthesis 1984, 743; c) M. Shiao, L. Shyu,
K. Tarng, Y. Tsun, Synth. Commun. 1990, 20, 2971.
[55] a) A. Combes, C. Combes, Bull. Soc. Chim. Fr. 1893, 7, 791; b) P. N.
Evans, J. Prakt. Chem. 1893, 48, 489.
[56] J. Nasielski, A. Standeart, R. Nasielski-Hinkens, Synth. Commun.
1991, 21, 901.
7178
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7167 – 7179

Ambifunctional Catalysis
FULL PAPER
[57] E. P. Kꢃndig, F. R. Monnier, Adv. Synth. Catal. 2004, 346, 901.
[58] DYNAFIT program: P. Kuzmic, Anal. Biochem. 1996, 237, 260.
[59] Due to experimental errors, deviations of ꢂ10% in RCA are not
necessarily significant. However, the slightly higher activity induced
by L8 over L7 has been reproduced in several independent experi-
ments.
[60] Compare the pKa values of pyridine (5.2), pyrimidine (1.3), and
1,3,5-triazine (<0): A. Albert, J. N. Phillips, J. Chem. Soc. 1956,
1294.
[61] Y. Wakatsuki, J. Organomet. Chem. 2004, 689, 4092.
[62] C. Bianchini, J. A. Casares, M. Peruzzini, A. Romerosa, F. Zanobini,
J. Am. Chem. Soc. 1996, 118, 4585.
Received: March 2, 2009
Published online: June 19, 2009
Chem. Eur. J. 2009, 15, 7167 – 7179
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7179