New rhodium pyridylphosphine complexes and their application in hydrogenation reactions

Article abstract of DOI:10.1021/om970010h

New phosphine ligands containing substituted pyridyl rings, namely tris[3-(2,6-dimethoxypyridyl)]phosphine (1), bis[3-(2,6-dimethoxypyridyl)]phenylphosphine (2), and 3-(2,6-dimethoxypyridyl)diphenylphosphine (3), have been prepared. The purpose was to overcome previous problems associated with the use of pyridylphosphine ligands in catalytic hydrogenation reactions and to develop a new class of catalysts that can be separated from reaction mediums via phase separation. To prevent the formation of polymeric transition-metal complexes of pyridylphosphines (which took place via the coordination of the P and N atoms of the ligands to separate metal centers), the ortho-hydrogen atoms of the pyridyl rings were replaced with more bulky substituents. The use of methoxy substituents at the ortho positions of the pyridyl ring in this new class of pyridylphosphine ligands successfully prevented the coordination of the pyridyl group to the metal center and allowed the formation of discrete rhodium pyridylphosphine complexes, which were found to be active in the catalytic hydrogenation of olefins, aldehydes, and imines. The solubility of these rhodium pyridylphosphine complexes in organic solvent or in aqueous solution can be altered by adjusting the acidity of the solution. The separation of the catalyst from the reaction product in a water-immiscible organic solvent was achieved by extracting the catalyst with hydrochloric acid. Upon neutralization of the aqueous extract with sodium carbonate, the catalyst could be re-extracted into an organic layer and could be reused in a new batch of hydrogenation reaction with similar reactivity as the freshly prepared complexes.

Full text of DOI:10.1021/om970010h

Organometallics 1997, 16, 3469-3473
3469
New Rh od iu m P yr id ylp h osp h in e Com p lexes a n d Th eir
Ap p lica tion in Hyd r ogen a tion Rea ction s
Albert S. C. Chan,* Chih-Chiang Chen, and Rong Cao
Union Laboratory of Asymmetric Synthesis and Department of Applied Biology and Chemical
Technology, The Hong Kong Polytechnic University, Hong Kong
Maw-Rong Lee
Department of Chemistry, National Chung Hsing University, Taiwan
Shie-Ming Peng and Gene Hsiang Lee
Department of Chemistry, National Taiwan University, Taiwan
Received J anuary 8, 1997X
New phosphine ligands containing substituted pyridyl rings, namely tris[3-(2,6-dimethoxy-
pyridyl)]phosphine (1), bis[3-(2,6-dimethoxypyridyl)]phenylphosphine (2), and 3-(2,6-
dimethoxypyridyl)diphenylphosphine (3), have been prepared. The purpose was to overcome
previous problems associated with the use of pyridylphosphine ligands in catalytic
hydrogenation reactions and to develop a new class of catalysts that can be separated from
reaction mediums via phase separation. To prevent the formation of polymeric transition-
metal complexes of pyridylphosphines (which took place via the coordination of the P and N
atoms of the ligands to separate metal centers), the ortho-hydrogen atoms of the pyridyl
rings were replaced with more bulky substituents. The use of methoxy substituents at the
ortho positions of the pyridyl ring in this new class of pyridylphosphine ligands successfully
prevented the coordination of the pyridyl group to the metal center and allowed the formation
of discrete rhodium pyridylphosphine complexes, which were found to be active in the
catalytic hydrogenation of olefins, aldehydes, and imines. The solubility of these rhodium
pyridylphosphine complexes in organic solvent or in aqueous solution can be altered by
adjusting the acidity of the solution. The separation of the catalyst from the reaction product
in a water-immiscible organic solvent was achieved by extracting the catalyst with
hydrochloric acid. Upon neutralization of the aqueous extract with sodium carbonate, the
catalyst could be re-extracted into an organic layer and could be reused in a new batch of
hydrogenation reaction with similar reactivity as the freshly prepared complexes.
In tr od u ction
thousands of publications on the use of transition-metal
phosphine catalysts in the literature in the past three
decades.
One of the most important breakthroughs in the study
of homogeneous catalysis in the past three decades is
the development of rhodium organophosphine catalysts
for hydrogenation and hydroformylation reactions. In
1965, Wilkinson et al. and Coffey independently devel-
oped chlorotris(triphenylphosphine)rhodium as a facile
catalyst for the catalytic hydrogenation of olefins.1-3
Further development of this chemistry by Wilkinson et
al. and Pruett et al. resulted in the successful com-
mercial application of HRh(CO)(PPh3)3 as a catalyst
precursor in the hydroformylation of propene.4 The
development of rhodium chiral phosphine complexes
created an exciting, new field of research: homogeneous
catalytic asymmetric hydrogenation. The Monsanto
L-Dopa process based on the asymmetric hydrogenation
of an enamide catalyzed by Rh(dipamp)+ was success-
fully commercialized in the mid-1970’s.5 The substan-
tial interest in this area has been reflected in the
As an effort to further develop the arylphosphine
ligands and their application in homogeneous catalysis,
transition-metal complexes containing pyridylphosphine
ligands have been synthesized and tested in catalytic
hydrogenation,6 hydroformylation,6-8 and carbonylation
reactions.9-11 The potential advantages of using a
pyridyl ring to replace a phenyl ring in the arylphos-
phine ligands are that (1) the added functionality of the
pyridyl group in the ligand may introduce more inter-
esting chemistry and (2) the solubility of the new
ligands, either in aqueous or in organic solvents, may
be easily controlled simply by adjusting the acidity of
the solution. (The latter property may be useful for the
separation of the catalyst from the products via phase
(5) Knowles, W. S. Acc. Chem. Res. 1983, 16, 106.
(6) Kurtev, K.; Ribola, D.; J ones, R. A.; Cole-Hamilton, D. J .;
Wilkinson, G. J . Chem. Soc., Dalton Trans. 1980, 55.
(7) Fell, B.; Papadogianakis, G. J . Mol. Catal. 1991, 66, 143.
(8) Gladiali, S.; Pinna, L.; Arena, C. G.; Rotondo, E., Faraone, F. J .
Mol. Catal. 1991, 66, 183.
(9) Drent, E.; Budzelaar, P. H. M.; J ager, W. W. Eur. Pat. Appl.
EP386,833; Chem. Abstr. 1991, 114, 142,679z.
(10) Drent, E.; Budzelaar, P. H. M. Eur. Pat. Appl. EP386,834;
Chem. Abstr. 1991, 114, 142,680t.
X Abstract published in Advance ACS Abstracts, J uly 1, 1997.
(1) Young, J . F.; Osborn, J . A.; J ardine, F. H.; Wilkinson, G. J . Chem.
Soc. Chem. Commun. 1965, 131.
(2) Osborn, J . A.; J ardine, F. H.; Young, J . F.; Wilkinson, G. J . Chem.
Soc. A 1966, 1711.
(3) Coffey, R. S. Brit. Pat. 1,121,642 (assigned to ICI), filed in 1965.
(4) Pruett, R. L. Adv. Organomet. Chem. 1979, 17, 1.
(11) Zao, W.-J .; Fang,T.-Q.; Zhang, S.-Q.; Zhang, Z.-Z.; Yang, L.-M.
J . Chem. Eng. Natural Gas 1992, 17, 3.
S0276-7333(97)00010-1 CCC: $14.00 © 1997 American Chemical Society

3470 Organometallics, Vol. 16, No. 15, 1997
Chan et al.
separation. The use of sulfonated triphenylphosphine
as a water-soluble organophosphine ligand has found
successful commercial application in the Rh-catalyzed
hydroformylation of olefin.12) Unfortunately, in the
previous studies the metal complexes containing py-
ridylphosphine ligands were found to be inactive in the
hydrogenation of olefins.6 This was thought to be due
to the coordination of the nitrogen atom of the pyridyl
ring to the metal center and saturated the coordination
sites. In this paper, we wish to report the results of a
new approach to the pyridylphosphine ligands and their
application in homogeneous hydrogenation reactions.
Resu lts a n d Discu ssion
The synthesis of pyridylphosphines has recently been
reviewed by Newkome.13 Previous studies mostly fo-
cused on the added binding capability of the pyridyl
group in the organophosphine species.13-15 In this
study, we intentionally avoided the coordination of the
pyridyl ring by limiting the access of the nitrogen atom
to the metal center via the use of more hindered groups
to replace the ortho hydrogen atoms. An initial attempt
with methoxy substituents at the ortho positions has
been found to be successful. When 2,6-dimethoxypyri-
dine was lithiated with n-butyllithium in THF at -40
°C followed by the slow addition of phosphorus trichlo-
ride at the same temperature, tris[3-(2,6-dimethoxypy-
ridyl)]phosphine (1) was obtained. The recrystallization
F igu r e 1. ORTEP drawing of [RhCl(C8H12)‚2] with 30%
ellipsoid.
Unlike the regular pyridylphosphinerhodium com-
plexes that are inactive in hydrogenation reactions,
these new types of pyridylphosphine complexes are good
catalysts for the hydrogenation of olefins, aldehydes,
and imines. At ambient temperature and under 200
psig of H2, the catalysts are not active in the hydrogena-
tion of ketones. When a conjugated eno ketone is
subjected to hydrogenation with these catalysts, only
the CdC double bond is hydrogenated. Several typical
examples of the catalytic hydrogenation reactions are
listed in Table 4.
of the crude product from ethanol gave a 45% yield of
pure tris[3-(2,6-dimethoxypyridyl)]phosphine.
Similarly, pure bis[3-(2,6-dimethoxypyridyl)]phenyl-
phosphine (2) and 3-(2,6-dimethoxypyridyl)diphenylphos-
phine (3) were obtained in 70% and 45% yields, respec-
tively, by reacting the lithiated 2,6-dimethoxypyridine
with dichlorophenylphosphine and chlorodiphenylphos-
phine, respectively.
An obvious added dimension of the new pyridylphos-
phine ligands (versus triphenylphosphine) is the con-
trollable solubility, either in organic solvent or in water,
via the control of the acidity of the solution. This
property offers an opportunity for phase separation of
the transition-metal catalysts containing these ligands
from the water-immiscible organic medium and, there-
fore, may facilitate the recycling of the transition-
metal-phosphine catalysts. In this regard, we tested
the rhodium catalyst containing 1 in hydrogenation
reactions in benzene followed by the extraction of the
catalyst with 12 N hydrochloric acid. Atomic absorption
studies of the aqueous and the organic layers indicated
that 96% of the catalyst was extracted into the aqueous
layer. A 31P NMR study indicated that the catalyst was
intact during the extraction process and the acidified
catalyst could be re-extracted from the aqueous solution
with a water-immiscible organic solvent (e.g., benzene
or toluene) after the neutralization of the HCl with a
base. These results clearly shed light on the potential
benefits for the further development of this type of new
ligands and catalysts.
Like triphenylphosphine, all three of these substi-
tuted pyridylphosphine ligands can be used to make
discrete transition-metal complexes without the com-
plication of the coordination of the pyridyl ring. For
example, when [Rh(COD)Cl]2 (COD ) 1,5-cyclooctadi-
ene) was allowed to react with 2 equiv of these ligands,
discrete, monomeric [RhCl(COD)‚L (L) 1, 2, or 3) was
obtained. The molecular structure of [RhCl(COD)‚2]
was unambiguously characterized by single-crystal X-
ray diffraction as a square planar complex. A perspec-
tive view of this complex is shown in Figure 1. The
relevant crystal data, bond lengths, and bond angles are
summarized in Tables 1-3. It was noted that C(3) and
C(8) of the COD ligand had higher thermal parameters
in the X-ray study. This is not uncommon in ethylene
groups and may be due to some conformational disorder.
(12) Kuntz, E. G. Chemtech 1987, Sep, 570.
(13) Newkome, G. R. Chem. Rev. 1993, 93, 2067.
(14) Yang, H.; Alvarez, M.; Lugan, N.; Mathieu, R. J . Chem. Soc.,
Chem. Commun. 1995, 1721.
(15) Brunner, H.; Bublak, P. Synthesis 1995, 36.

New Rhodium Pyridylphosphine Complexes
Organometallics, Vol. 16, No. 15, 1997 3471
Ta ble 3. Bon d An gles of [Rh Cl(C8H12)‚2]
Ta ble 1. Cr ysta l Da ta a n d Con d ition s for
Cr ysta llogr a p h ic Da ta Collection a n d Str u ctu r e
Refin em en ta
Cl-Rh-P
87.45(8)
89.39(24)
90.99(24)
159.06(25)
162.7(3)
162.4(3)
161.8(3)
93.94(24)
96.91(23)
35.5(4)
P-C9-C14
116.3(6)
118.8(7)
121.8(7)
120.6(8)
118.8(8)
120.3(8)
119.6(8)
120.5(6)
123.3(6)
115.8(7)
120.4(8)
117.9(8)
123.9(7)
116.6(8)
119.5(7)
125.1(7)
117.0(6)
117.9(7)
122.2(5)
121.9(5)
115.4(6)
120.3(7)
117.7(7)
125.1(7)
116.8(7)
118.1(7)
125.8(6)
116.7(6)
117.5(6)
116.9(7)
115.6(6)
117.2(7)
118.8(6)
117.6(6)
117.9(6)
83.7(15)
95.0(16)
113.4(15)
Cl-Rh-C1
Cl-Rh-C2
Cl-Rh-C5
Cl-Rh-C6
P-Rh-C1
P-Rh-C2
P-Rh-C5
P-Rh-C6
C1-Rh-C2
C1-Rh-C5
C1-Rh-C6
C2-Rh-C5
C2-Rh-C6
C5-Rh-C6
Rh-P-C9
Rh-P-C15
Rh-P-C22
C9-P-C15
C9-P-C22
C15-P-C22
Rh-C1-C2
Rh-C1-C8
C2-C1-C8
Rh-C2-C1
Rh-C2-C3
C1-C2-C3
C2-C3-C4
C3-C4-C5
Rh-C5-C4
Rh-C5-C6
C4-C5-C6
Rh-C6-C5
Rh-C6-C7
C5-C6-C7
C6-C7-C8
C1-C8-C7
P-C9-C10
C10-C9-C14
C9-C10-C11
C10-C11-C12
C11-C12-C13
C12-C13-C14
C9-C14-C13
P-C15-C16
title
[RhCl(COD)‚2]
formula
fw
RhPO5N2C31H39Cl
688.98
diffractometer used
Nonius
space group
a (Å)
monoclinic P21/c
8.219(3)
P-C15-C19
b (Å)
18.312(5)
C16-C15-C19
C15-C16-C17
C16-C17-C18
C17-C18-N1
C17-C18-O1
N1-C18-O1
C15-C19-N1
C15-C19-O2
N1-C19-O2
P-C22-C23
c (Å)
21.639(8)
95.2(3)
81.4(3)
81.2(3)
89.9(3)
â (deg)
V (Å3)
Z
98.98(3)
3217.2(19)
4
1.422
0.7107
1424
Dcalcd (g‚cm-3
λ (Å)
)
37.7(3)
118.70(24)
113.8(3)
110.19(23)
101.2(3)
106.3(3)
105.4(3)
72.9(5)
107.5(6)
125.4(9)
71.6(5)
109.1(6)
125.7(9)
116.8(8)
115.1(8)
111.3(6)
72.1(5)
F(000)
unit cell detn: no. (2 θ range)
25; (19.12-26.84 deg)
scan type
θ/2θ
scan width (deg)
scan speed (deg/min)
2θmax
2(0.65 + 0.35 tan(θ))
2.06-8.24
45.0
(-8; 8)(0; 19)(0; 23)
6.929
P-C22-C26
C23-C22-C26
C22-C23-C24
C23-C24-C25
C24-C25-N2
C24-C25-O3
N2-C25-O3
C22-C26-N2
C22-C26-O4
N2-C26-O4
C18-N1-C19
C25-N2-C26
C18-O1-C20
C19-O2-C21
C25-O3-C27
C26-O4-C28
O5-C29-C30
O5-C29-C31
C30-C29-C31
hkl ranges
µ (cm-1
)
cryst size (mm)
transmission
0.10 × 0.30 × 0.35
0.957; 1.000
T (K)
298
no. of measd reflns
no. of obsd reflns (I > 2.0σ(I))
no. of unique reflns
Rf; Rw
4199
2983
4199
0.045; 0.047
GoF
2.79
126.0(8)
70.3(5)
refinement program
no. of atoms
NRCVAX
80
112.0(6)
125.8(8)
115.8(8)
116.9(9)
124.9(6)
no. of refined params
minimize function
wt scheme
335 (2983 out of 4199 reflns)
2
∑(w|Fo - Fc| )
1/[σ2(Fo) + KFo
0.000 010
0.41(5)
]
2
2
wt modifier K in KFo
g (2nd ext coeff) × 104
(δ/σ)max
0.0055
-0.430; 2.310
Ta ble 4. Exa m p les of th e Hom ogen eou s
Hyd r ogen a tion Rea ction s Ca ta lyzed by
[Rh Cl(COD)‚1]
residual in final D-map (e/A3)
a
Rf ) ∑|Fo - Fc|/∑|Fo|; Rw ) ([∑(w(Fo - Fc)2)/∑(wFo2)])0.5. GoF
) ([∑(w(Fo - Fc)2)/(no. of reflns - no. of params)]0.5
solvated: [RhCl(C8H12)‚2](Me2CO).
. Acetone
Ta ble 2. Bon d Dista n ces (Å) of [Rh Cl(C8H12)‚2]
Rh-Cl
2.3617(23)
2.3183(22)
2.204(8)
2.221(8)
2.121(8)
2.144(8)
1.839(8)
1.812(8)
1.823(7)
1.349(14)
1.517(15)
1.498(14)
1.483(16)
1.482(12)
1.377(13)
1.514(13)
1.451(15)
1.361(11)
1.373(11)
1.369(11)
1.369(13)
1.370(14)
1.414(12)
C15-C16
C15-C19
C16-C17
C17-C18
C18-N1
C18-O1
C19-N1
C19-O2
C20-O1
C21-O2
C22-C23
C22-C26
C23-C24
C24-C25
C25-N2
C25-O3
C26-N2
C26-O4
C27-O3
C28-O4
O5-C29
C29-C30
C29-C31
1.390(11)
1.393(11)
1.379(12)
1.389(13)
1.318(11)
1.345(10)
1.336(9)
1.343(10)
1.409(12)
1.405(10)
1.404(10)
1.380(10)
1.365(11)
1.371(11)
1.323(10)
1.364(9)
Rh-P
Rh-C1
Rh-C2
Rh-C5
Rh-C6
P-C9
P-C15
P-C22
C1-C2
C1-C8
C2-C3
C3-C4
C4-C5
C5-C6
C6-C7
C7-C8
C9-C10
C9-C14
C10-C11
C11-C12
C12-C13
C13-C14
1.330(9)
1.355(8)
1.410(11)
1.419(10)
1.26(3)
1.562(24)
1.376(25)
a
Reaction conditions: 25 °C; 200 psig of H2; substrate/cat. (wt)
) 100; quantitative selectivity was obtained in all cases (therefore,
the percentage of conversion was the same as the percentage of
yield).
In conclusion, in the design and synthesis of new
pyridylphosphine ligands, we have been able to prepare
substituted pyridylphosphine ligands that form discrete,
well-defined rhodium complexes that are active in the
catalytic hydrogenation of olefins, aldehydes, and imi-
nes. The phase separation of these catalysts from
water-immiscible organic solvents with hydrochloric
acid also has been achieved. Further development of
this type of substituted pyridylphosphine ligands, par-
ticularly in chiral forms, is in good progress.
Exp er im en ta l Section
Gen er a l Meth od s a n d P r oced u r es. Unless otherwise
indicated, all experiments were carried out under nitrogen

3472 Organometallics, Vol. 16, No. 15, 1997
Chan et al.
atmosphere. All NMR data were recorded on
a
Bruker
added to quench the reaction. The solvent was evaporated in
a rotary evaporator, and the residue was redissolved in 30 mL
of chloroform. The chloroform solution was washed with 30
mL of water and dried over anhydrous sodium sulfate.
Evaporation of the solvent in vacuo gave a crude material that
was recrystallized from ethanol to give 1.45 g of white powdery
product (45% theoretical yield). 1H NMR in CDCl3, pyridyl
ring δ: 3.88 (s, 3H, OCH3); 3.92 (s, 3H, OCH3); 6.22(d, 1H,
J H-H ) 8.2 Hz); 6.92 (dd, 1H, J H-H ) 8.2 Hz, J P-H ) 4.5 Hz);
phenyl rings δ: (m, 10H). 13C[H] NMR in CDCl3, pyridyl ring
δ: 53.59 and 53.71 (OCH3’s); 108.45 (d, J P-C ) 12.5 Hz, C-3);
145.57 (C-4); 101.77 (C-5); 164.07 (C-2); 164.32 (C-6); phenyl
ring δ: 128.45, 133.53, 136.60 (d, J P-C ) 10.6 Hz). 31P[H] NMR
in CDCl3 δ: -21.97 ppm. Molecular weight according to high-
resolution mass spectrometry (EI): 323.1083 (calcd molecular
weight ) 323.1075; error ) 0.8 mmu.) Anal. Calcd for
DPX400 instrument. The high-resolution mass spectrometry
was carried out on a J EOL J MS SX/SX 102A four-sector
tandem mass spectrometer. Purchased reagents were used
as received, and all solvents were dried and distilled before
use.
Tr is[3-(2,6-d im eth oxyp yr id yl)]p h osp h in e (1). A solu-
tion of n-butyllithium (6.25 mL of a 1.6 M hexane solution, 10
mmol) was slowly added to a solution of 2,6-dimethoxypyridine
(1.39 g, 10 mmol) in 20 mL of THF at -40 °C under magnetic
stirring over a period of about 1 h. After all the n-butyllithium
was added, the mixture continued to stir for 1 h at -40 °C
and 4 h at ambient temperature. The solution was cooled to
-40 °C again, and 0.28 mL of PCl3 (3.3 mmol) in 10 mL of
THF was added dropwise to it while stirring was continued.
The solution was brought to ambient temperature, and 2 mL
of water was added to quench the reaction. The solvent was
evaporated in a rotary evaporator, and the residue was re-
dissolved in 30 mL of chloroform. The chloroform solution was
washed with 30 mL of water and dried over anhydrous sodium
sulfate. Evaporation of the solvent in vacuo gave a crude
material that was recrystallized from ethanol to give 0.667 g
of a white powdery product (45% theoretical yield). 1H NMR
in CDCl3 δ: 3.84 (s, 9H, OCH3); 3.91(s, 9H, OCH3); 6.22 (d,
3H, J H-H ) 8.1 Hz); 6.95 (dd, 3H, J H-H ) 8.1 Hz, J P-H ) 4.1
Hz). 13C[H] NMR in CDCl3 δ: 53.24 (OCH3); 106.43 (d, J P-C
) 13.5 Hz, C-3); 144.53 (C-4); 101.25 (C-5); 163.81 (C-2 and
C-6; could not be resolved on a DPX400 NMR). 31P[H] NMR
in CDCl3: δ: -47.67 ppm. Molecular weight according to
high-resolution mass spectrometry (EI): 445.1401 (calcd mo-
lecular weight ) 445.1400; error ) 0.1 mmu.) Anal. Calcd for
C
19H18O2NP: C, 70.58; H, 5.61. Found: C, 70.28; H, 5.63.
[Rh Cl(C8H12)‚2]. [Rh(COD)Cl]2 (50 mg, 0.1 mmol) and 2
(77 mg, 0.2 mmol) were stirred in 5 mL of benzene in a 25 mL
round-bottom flask at ambient temperature for 1 h. The
solvent was evaporated in vacuo, and a yellow residue was
collected. NMR studies indicated that the reaction was
essentially quantitative, and the product was used for catalytic
reactions. 1H NMR in CDCl3, coordinated COD δ: 1.95, 2.02
(m, 8H, CH2’s); 5.33 (b, 4H, CH’s of COD); substituted pyridyl
δ: 3.92 (s, 3H, OCH3); 3.94 (s, 3H, OCH3); 6.35 (d, 1H, J H-H
)
8.1 Hz); 6.96 (b); phenyl δ: 7.27 (m, 1H), 7.37 (m, 2H), 8.05
(dd, 2H, J H-H ) 8.5 Hz, J P-H ) 10.7 Hz). 13C[H] NMR in
CDCl3: coordinated COD δ: 29.00, 33.05 (CH2’s); 70.65, 70.78
(CH’s); substituted pyridyl δ: 101.47, 103.02 (d, J P-C ) 48 Hz),
148.86, 163.22, 165.22; phenyl δ: 127.48, 128.87, 132.03,
131.23 (d, J P-C ) 44.5 Hz). 31P[H] NMR in CDCl3 δ: 14.43 (d,
J Rh-P ) 148.5 Hz). Molecular weight according to high-
resolution mass spectrometry (FAB): 630.0928 (calculated
molecular weight ) 630.0921; error ) 0.7 mmu.)
C
21H24O6N3P: C, 56.63; H, 5.43. Found: C, 56.29; H, 5.41.
Bis[3-(2,6-d im eth oxyp yr id yl)]p h en ylp h osp h in e (2). A
solution of n-butyllithium (6.25 mL of a 1.6 M hexane solution,
10 mmol) was slowly added to a solution of 2,6-dimethoxypy-
ridine (1.39 g, 10 mmol) in 20 mL of THF at -40 °C under
magnetic stirring over a period of about 1 h. After all the
n-butyllithium was added, the mixture continued to stir for 1
h at -40 °C and 4 h at ambient temperature. The solution
was cooled to -40 °C again, and 0.68 mL of dichlorophe-
nylphosphine (5 mmol) in 10 mL of THF was added dropwise
to it while the stirring was continued. The solution was
brought to ambient temperature, and 2 mL of water was added
to quench the reaction. The solvent was evaporated in a rotary
evaporator, and the residue was redissolved in 30 mL of
chloroform. The chloroform solution was washed with 30 mL
of water and dried over anhydrous sodium sulfate. Evapora-
tion of the solvent in vacuo gave a crude material that was
recrystallized from ethanol to give 1.34 g of white powdery
product (70% theoretical yield). 1H NMR in CDCl3, pyridyl
rings δ 3.84 (s, 6H, OCH3); 3.91 (s, 6H, OCH3); 6.22 (d, 2H,
J H-H ) 8.2 Hz), 6.94 (dd, 2H, J H-H ) 8.1 Hz, J P-H ) 4.5 Hz);
phenyl ring δ: 7.23-7.30 (m, 5H). 13C[H] NMR in CDCl3,
pyridyl rings δ: 53.49 (OCH3); 107.62 (d, J P-C ) 12.7 Hz, C-3);
145.40 (C-4); 101.58 (C-5); 162.26 (C-2 and C-6; could not be
resolved on a DPX400 NMR); phenyl ring δ: 127.91, 128.22,
128.37, 128.48, 133.37 (d, J P-C ) 21.1 Hz). 31P[H] NMR in
CDCl3 δ: -35.47 ppm. Molecular weight according to high-
resolution mass spectrometry (EI) 384.1234 (calcd molecular
weight ) 384.1239; error ) 0.5 mmu.) Anal. Calcd for
[RhCl(COD)‚1] and [RhCl(COD)‚3] were prepared similarly.
Ca ta lytic Hyd r ogen a tion . A typical procedure for the
catalytic hydrogenation of unsaturated substrates is as follows.
A 50 mL stainless steel autoclave equipped with a magnetic
stirring bar was charged with 1.0 mg of [RhCl(C8H12)‚1], 100
mg of substrate, and 5 mL of methanol under 1 atm of nitrogen
gas. The autoclave was then pressurized with 200 psig of H2,
and the solution was stirred at ambient temperature for 16 h.
The H2 was released at the end of the reaction, and the solution
was transferred to a 50 mL round-bottom flask. The solvent
was evaporated in vacuo, and the residue was analyzed by 1H
NMR and/or HPLC.
Test of th e P h a se Sep a r a tion of Ca ta lyst fr om th e
Or ga n ic P r od u ct. In a typical experiment, a 50 mL auto-
clave with a glass liner and a magnetic stirring bar was
charged with 0.4 mg of [RhCl(COD)‚1] in 2 mL of toluene and
20 mg of benzylideneacetophenone. The autoclave was pres-
surized with 200 psig of H2, and the solution was stirred under
H2 pressure at ambient temperature for 15 h. The H2 gas was
released, and the organic solution was extracted with 2 mL of
12 N HCl solution twice. The aqueous layer was diluted to
50 mL with distilled water and was analyzed for rhodium
content with a Varian Model 1200 atomic absorption spec-
trometer using the rhodium atomic spectral line at 343.5 nm.
A total of 96.2% of the rhodium complex was found in the
aqueous layer.
C
20H21O4N2P: C, 62.50; H, 5.51. Found: C, 62.50; H, 5.50.
3-(2,6-Dim et h oxyp yr id yl)d ip h en ylp h osp h in e (3).
A
Test of t h e In t egr it y of t h e Ca t a lyst d u r in g P h a se
Sep a r a tion . [RhCl(COD)‚1] (2.0 mg) was dissolved in 0.4 mL
of deuterated benzene, and the 31P NMR of the species was
recorded. (31P[H] NMR in C6D6 δ: δ8.48, d, J Rh-P ) 151.7 Hz.)
The solution was extracted with 0.4 mL of 12 N hydrochloric
acid, the aqueous layer was collected and a 31P NMR of the
aqueous solution was recorded. (31P[H] NMR in H2O δ: 10.65,
d, J Rh-P ) 150.8 Hz.) The acidic aqueous solution was neutral-
ized with aqueous sodium carbonate solution, and the solution
was extracted with 5 mL of toluene. The organic layer was
separated, and the toluene solvent was evaporated in vacuo.
solution of n-butyllithium (6.25 mL of a 1.6 M hexane solution,
10 mmol) was slowly added to a solution of 2,6-dimethoxypy-
ridine (1.39 g, 10 mmol) in 20 mL of THF at -40 °C under
magnetic stirring over a period of about 1 h. After all the
n-butyllithium was added, the mixture was continued to stir
for 1 h at -40 °C and 4 h at ambient temperature. The
solution was cooled to -40 °C again, and 1.8 mL of chlo-
rodiphenylphosphine (10 mmol) in 10 mL of THF was added
dropwise to it while the stirring was continued. The solution
was brought to ambient temperature, and 2 mL of water was

New Rhodium Pyridylphosphine Complexes
Organometallics, Vol. 16, No. 15, 1997 3473
The residue was redissolved in 0.4 mL of deuterated benzene,
and a final 31P NMR was recorded. (31P[H] NMR in C6D6 δ:
8.45, d, J Rh-P ) 151.4 Hz.)
Cr ysta llogr a p h y. A crystal of dimensions 0.1 × 0.30 ×
0.35 mm was used for data collection on an Enraf-Nonius
diffractometer (graphite monochromatized Mo KR radiation,
λ ) 0.7107) using the θ - 2θ scan method (2θmax ) 45°) at 298
K. Intensity data were corrected for Lorentz and polarization
Test of Ca ta lyst Recycle. In a typical experiment, a 50
mL stainless steel autoclave equipped with a magnetic stirring
bar was charged with 2 mL of toluene, 5 mg of 2-(6-methoxy-
2-naphthyl)acrylic acid (substrate), and 0.05 mg of [RhCl-
(COD)‚(1) (catalyst precursor). The autoclave was pressurized
with 200 psig H2, and the solution was allowed to stir at 25
°C under 200 psig of of H2 for 4 h. The hydrogen was released
under an inert atmosphere (N2). A small sample of the
solution was taken for HPLC analysis, and 84% of the
substrate was found to be hydrogenated to give 2-(6-methoxy-
2-naphthyl)propanoic acid. The solution was extracted with
4 mL of 12 N HCl solution. After phase separation, the
aqueous layer was titrated with 1 N Na2CO3 solution until
the solution just turned basic (according to a litmus paper test).
The catalyst was extracted from the aqueous layer with 10
mL of toluene that, after phase separation, was concentrated
to 2 mL by evaporating off the excess solvent in vacuo. The
resulting solution and 5 mg of 2-(6-methoxy-2-naphthyl)acrylic
acid (substrate) was charged to the autoclave, and the hydro-
genation reaction was carried out again under identical
conditions for 4 h. Product analysis after releasing the H2
indicated 80% of the 2-(6-methoxy-2-naphthyl)acrylic acid was
hydrogenated to give the 2-(6-methoxy-2-naphthyl)propanoic
acid product.
effects; empirical absorptions were based on
scans. A total
of 4199 unique reflections were obtained, 2983 of which were
considered observed (I > 2σ (I)) and used in structure refine-
ment. The structure was solved by Patterson and refined by
full-matrix least-squares methods using the NRCVAX pro-
gram.16 The final least-squares refinement was calculated
with 335 parameters and converged to RF ) 4.5%, RWF ) 4.7%,
and S ) 2.79 with W-1 ) σ2(F) + 0.00001F2. The hydrogen
atoms were placed in idealized positions and were not refined.
The bond lengths and bond angles are listed in Tables 2 and
3, respectively.
Ack n ow led gm en t. We thank the Research Grant
Council of Hong Kong for financial support of this study.
Su p p or tin g In for m a tion Ava ila ble: Tables of atomic
parameters and U values (4 pages). Ordering information is
given on any current masthead page.
OM970010H
(16) Gabe, E. J .; Le Page, Y.; Charland, J . P.; Lee, F. L.; White, P.
S. J . Appl. Crystallogr. 1989, 22, 384.