Syntheses, structures, and catalytic reactions of palladium adducts of chiral diphosphines that contain a rhenium stereocenter in the backbone

Article abstract of DOI:10.1002/1522-2675(200206)85:6<1778::AID-HLCA1778>3.0.CO;2-9

Reactions of the title diphosphines [(η⁵-C₅H₄PPh₂) Re(NO)(PPh₃)((CH₂)_nPPh₂)] (n=0, (R)-1; n=1, racemic or (S)-2) with [PdCl₂(PhCN)₂] give the palladium/rhenium chelate complexes [η⁵-C₅H₄PPh₂) Re(NO)(PPh₃)(C₅H₄PPh₂) Re(NO)(PPh₃)((μ-CH₂)_nPPh₂) PdCl₂] (n=0, (S)-5;n=1, racemic or(S)-6) in 75 - 92% yield. The crystal structure of racemic 6 shows a twisted-boat conformation of the chelate ring, giving a chiral pocket very different from that in a related rhodium chelate. However, NOE experiments suggest a similar ensemble of conformations in solution. Catalysts are generated from various combinations of a) Pd(OAc)₂ and (R)-1 or (S)-2 (1:2), b) (S)-5 or (S)-6 and (R)-1 or (S)-2 (1:2), or c) (i-Bu)₂AIH with the preceding recipes. These factors effect the Heck arylation of 2,3-dihydrofuran with phenyl trifluoromethylsulfonate. In contrast to analogous reactions with (R)-binap (=(R)-2,2′-bis(diphenylphosphanyl)-1,1′-binaphthalene), the major product 2- phenyl-2,3-dihydrofuran is nearly racemic (≤ 12% ee).

Full text of DOI:10.1002/1522-2675(200206)85:6<1778::AID-HLCA1778>3.0.CO;2-9

1778
Helvetica Chimica Acta Vol. 85 (2002)
Syntheses, Structures, and Catalytic Reactions of Palladium Adducts of
Chiral Diphosphines That Contain a Rhenium Stereocenter in the Backbone
by Klemenz Kromm, Frank Hampel, and J. A. Gladysz*
Institut f¸r Organische Chemie, Friedrich-Alexander-Universit‰t Erlangen-N¸rnberg, Henkestrasse 42,
D-91054 Erlangen
Dedicated to the memory of Professor Luigi M. Venanzi,
a pioneer in the chemistry of group-10 diphosphine chelate complexes
Reactions of the title diphosphines [(h5-C5H4PPh2)Re(NO)(PPh3)((CH2)nPPh2)] (n ˆ 0, (R)-1;
n ˆ 1, racemic or (S)-2) with [PdCl2(PhCN)2] give the palladium/rhenium chelate complexes [(h5-
C5H4PPh2)Re(NO)(PPh3)((m-CH2)nPPh2)PdCl2] (n ˆ 0, (S)-5; n ˆ 1, racemic or (S)-6) in 75 92% yield. The
crystal structure of racemic 6 shows a twisted-boat conformation of the chelate ring, giving a chiral pocket very
different from that in a related rhodium chelate. However, NOE experiments suggest a similar ensemble of
conformations in solution. Catalysts are generated from various combinations of a) Pd(OAc)2 and (R)-1 or (S)-
2 (1:2), b) (S)-5 or (S)-6 and (R)-1 or (S)-2 (1:2), or c) (i-Bu)2AlHwith the preceding recipes. These factors
effect the Heck arylation of 2,3-dihydrofuran with phenyl trifluoromethylsulfonate. In contrast to analogous
reactions with (R)-binap (ˆ(R)-2,2'-bis(diphenylphosphanyl)-1,1'-binaphthalene), the major product 2-
phenyl-2,3-dihydrofuran is nearly racemic (ꢀ12% ee).
Introduction.
The synthesis of chiral ferrocene-based chelating ligands and
applications of their metal complexes as catalysts for enantioselective organic synthesis
has seen rapidly increasing attention over the last decade [1]. Many spectacularly
successful reactions have been developed, and, more recently, efforts have been
extended to nonferrocene metal fragments [2 4]. Besides the direct extrapolation of
ruthenocenes [1a] and our own work detailed below [3][4], examples include [Cr(h6-
arene)(CO)3] [2b,c], [M(h5-C5R5)(CO)3] (MˆMn, Re) [2a,d,e], and [Co(h4-C4R4)(h5-
C5R5)] [2f] derived systems. In accord with the diverse architectural possibilities
inherent in such templates, all three types of stereogenic elements (centers, planes, and
axes) have been employed, often in combination.
Many types of chelating diphosphines have been described that are chiral by virtue
of carbon or phosphorus stereocenters [5]. We reported the first family of chelating
diphosphines that are chiral by virtue of a transition-metal stereocenter, as depicted in
the structures of 1 and 2 [3], which feature a chiral rhenium fragment in the backbone
and are easily obtained in enantiomerically pure form. Rhodium adducts of the
structures 3 and 4 were prepared and were found to be excellent catalyst precursors for
the asymmetric hydrogenation of protected dehydroamino acids. Catalyst activities,
lifetimes, and enantioselectivities compared well with the best systems in the literature
[3].
We sought to extend the applicability of this new family of diphosphines to a wider
set of chelated metal fragments and catalytic reactions. In this paper, we report the

Helvetica Chimica Acta Vol. 85 (2002)
1779
+
Ph2P
Rh
Ph2P
Ph2P
–
PF6
NO
Re
PPh3
Re
Re
PPh3
PPh3
ON
ON
Ph2P
C
H2
PPh2
CH2PPh2
n
(S)-3 (n = 0)
(S)-4 (n = 1)
(S)-2
(R)-1
synthesis of two palladium chelates, the structural characterization of one in solution
and in the solid state, and preliminary screening results that establish their viability as
catalyst precursors for CÀC bond forming reactions. The synthesis and crystal structure
of a palladium chelate of a dirhenium-containing diamine, in which the Re-atoms are
located in the N-alkyl substituents as opposed to the backbone, have been described
elsewhere [4].
Results. As shown in Scheme 1, (R)-1 as well as racemic and (S)-2 were treated
1
each with [PdCl2(PhCN)2] in TH F). Work-up gave the heterobimetallic palladium/
rhenium complexes, (S)-[(h5-C5H4PPh2)Re(NO)(PPh3)(m-PPh2)PdCl2] ((S)-5) and
racemic and (S)-[(h5-C5H4PPh2)Re(NO)(PPh3)(m-CH2PPh2)PdCl2] (racemic and (S)-
6), as air-stable, red or orange powders in 85 75% yield. The compounds were
characterized by IR and NMR (1H, 13C, 31P) spectroscopy, as summarized in the Exper.
Part and Table 1. The racemate gave a correct microanalysis, but the nonracemic
complexes were more difficult to purify, consistent with the lower crystallinity
commonly encountered with such chiral Re compounds.
Scheme 1. Synthesis of Palladium Chelate Complexes1)
The spectroscopic properties fully supported the proposed structures. MS gave
molecular ions with the correct masses and isotope distributions. NMR spectra showed
1
distinct H- and 13C-signals for all cyclopentadienyl H- and C-atoms, and the former
could be assigned as described below. As illustrated in Table 1, the 31P-NMR spectra
1
)
The enantiomers of diphosphines 1 and 2 used to prepare nonracemic chelates have configurations opposite
to those depicted in previous papers in this series. For ease of comparison, the configurations of all
compounds and catalysis products have been inverted in the text, Schemes, and Figures. The true
configurations are given in the Exper. Part. The basis for assignment of (R/S) descriptors has been
summarized earlier [3a].

1780
Helvetica Chimica Acta Vol. 85 (2002)
Table 1. Summaryof 31P{1H}-NMR Dataa)
Complex RePPh3
Re(CH2)nPPh2X
C5H4PPh2X'
1b)c)
20.2 (d, 2J(P,P) ˆ 15)
À 45.2 (d, 2J(P,P) ˆ 15)
À 49.2 (ddd, 1J(P,Rh) ˆ 127,
2J(P,P) ˆ 19, 2J(P,P) ˆ 14)
À 19.1 (d, 2J(P,P) ˆ 14)
À 16.2 (s)
3c)d)e)
9.8 (dd, 2J(P,P) ˆ 14,
3J(P,P) ˆ 5)
50.4 (ddd, 1J(P,Rh) ˆ 183,
2J(P,P) ˆ 19, 3J(P,P) ˆ 5)
53.3 (d, 2J(P,P) ˆ 14)
5f)g)
9.1 (s)
2g)h)
4e)f)g)
26.3 (d, 3J(P,P) ˆ 8)
6.9 (dd, 3J(P,P) ˆ 3, 3J(P,P) ˆ 8) À 17.7 (d, 3J(P,P) ˆ 3)
20.2 (dd, 3J(P,P) ˆ 18, 3J(P,P) ˆ 4) 50.5 (ddd, 1J(P,Rh) ˆ 148,
2J(P,P) ˆ 34, 3J(P,P) ˆ 18)
23.9 (ddd, 1J(P,Rh) ˆ 166,
2J(P,P) ˆ 34, 3J(P,P) ˆ 4)
21.1 (d, 2J(P,P) ˆ 19)
6d)g)
26.2 (s)
62.5 (d, 2J(P,P) ˆ 19)
a) At r.t.; J in Hz. b) In (D8)THF. c) 121 MHz. d) In CD2Cl2. e) À144.0 (sept., J(P,F) ˆ 708). f) In CDCl3.
g) 162 MHz. h) In C6D6.
showed diagnostic chemical-shift and coupling-constant patterns, similar to those of the
cationic rhodium/thenium complexes 3 and 4 [3]. With (S)-5, the signal of C5H4PPh2
was shifted further downfield from that of the uncomplexed diphosphine (d ˆ 53.3 vs.
À16.2) than with racemic or (S)-6 (d ˆ 21.1 vs. À17.7). In contrast, the RePPh2 signal of
(S)-5 was shifted downfield by a smaller value (d ˆ À19.1 vs. À45.2) than the
ReCH2PPh2 signal in racemic and (S)-6 (d ˆ 62.5 vs. 6.9). The cationic complex 3 gives
an opposite chemical-shift trend (d ˆ À49.2 vs. À45.2).
Orange needles of a solvate of racemic 6 were grown from CH2Cl2/hexane. The
crystal structure was determined as described in Table 2 and the Exper. Part. The
structure is given in Fig. 1 and confirms the above assignment. Key bond lengths, bond
angles, and torsion angles are summarized in Table 3, and selected conformational
features are analyzed below.
Palladium chelates of diphosphines are widely used in catalysis [6 8]. In some
protocols, they are first isolated and purified, and, in other cases, they are simply
generated in situ. Both approaches have advantages. For initial screening purposes, the
Table 2. Crystallographic Data of 6 ¥ 3 CH2Cl2
Empirical formula
C51H47Cl8NOP3PdRe D (calc.) [g cmÀ3
1359.1
173(2)
]
1.66
3.082
Mr [g molÀ1
T [K]
]
Absorption coefficient [mmÀ1
Crystal dimensions [mm]
V range [8]
]
0.30 Â 0.25 Â 0.25
Diffractometer
Wavelength [ä]
Nonius KappaCCD
0.71073
ꢀ 25
Range/indices (h,k,l)
À 13 ꢀ h ꢀ 13,
À16 ꢀ k ꢀ 16,
À22 ꢀ l ꢀ 22
17185
Crystal system
Space group
triclinic
P1
Reflections collected
Unique reflections
≈
9289
Unit cell dimensions:
a [ä]
b [ä]
c [ä]
a [8]
b [8]
g [8]
V [ä3]
Z
11.282(2)
Reflections observed (I > 2s(I)) 7729
13.738(3)
18.879(4)
92.48(3)
93.34(3)
111.48(3)
2711.7(9)
2
Refined parameters
Refinement
Rint
594
least squares on F2
0.0358
R indices (I > 2s(I))
R indices (all data)
Goodness-of-fit
Largest diff. peak and
R1 ˆ 0.0723, wR2 ˆ 0.1581
R1 ˆ 0.0871, wR2 ˆ 0.1648
1.121
2.961 and À1.848
hole [e ¥ äÀ3
]

Helvetica Chimica Acta Vol. 85 (2002)
1781
Fig. 1. Structure of 6 ¥ 3 CH2Cl2 with solvate molecules omitted
test reaction shown in Scheme 2 was selected. This Heck arylation is commonly
conducted with phenyl trifluoromethylsulfonate (PhOTf) as the limiting reactant
[7 9]. Most palladium catalysts based on chelating diphosphine ligands yield mixtures
of dihydrofurans, in which the 2,3-isomer (7) dominates over the 3,4-isomer (8).The
former is obtained in high ee with a catalyst generated in situ from Pd(OAc)2/(R)-binap
(1:2 mol ratio) ((R)-binap ˆ (R)-2,2'-bis(diphenylphospanyl)-1,1'-binaphthalene)
[7a]. The base employed also affects the enantioselectivity. Therefore, we used the
same base (Et3N) in all of our experiments.
We first verified that we could obtain the range of enantioselectivities reported for
the Pd(OAc)2/(R)-binap system (Scheme 2, Entry1 ) [7a]. We then examined
analogous recipes derived from Pd(OAc)2 and the uncomplexed diphosphines (R)-1
and (S)-2 (Entries 2 and 3). All three of these systems exhibited substantial induction
periods, which in the first case has been attributed to the required generation of a Pd(0)
species, with excess diphosphine serving as reductant [10]. The diphosphines (R)-1 and
(S)-2 gave longer induction periods and somewhat slower reactions as reflected by the
time and temperature data in Scheme 2. Additional details are provided elsewhere [11].
However, neither diphosphine gave 7 with a significant enantiomeric excess. The
nonracemic 7 prepared in Entry1 was added to the catalyst system in Entry2 . After
48 h, no racemization was observed.
Experiments with the preformed palladium chelates were investigated next. Not
unsurprisingly, (S)-5 was not very active alone, and (S)-6 gave no reaction whatsoever.
As shown in Entries 4 and 5, combinations of the chelates and free diphosphines (S)-5/

1782
Helvetica Chimica Acta Vol. 85 (2002)
Table 3. KeyDistances [ä] and Angles [8] of 6 ¥ 3 CH2Cl2
ReÀN(1)
1.763(9)
2.248(3)
2.179(10)
1.796(12)
1.947
P(3)ÀC(11)
PdÀP(3)
PdÀP(2)
ReÀCl(1)
ReÀCl(2)
ReÀPd
1.802(10)
2.248(3)
2.286(3)
2.344(3)
2.341(3)
4.332
ReÀP(1)
ReÀC(1)
C(1)ÀP(2)
ReÀCp(centroid)
ReÀC(11)
2.253(7)
N(1)ÀReÀP(1)
N(1)ÀReÀC(1)
C(1)ÀReÀP(1)
ReÀC(1)ÀP(2)
C(24)ÀReÀC(1)
ReÀC(24)ÀP(3)
PdÀP(3)ÀC(24)
P(2)ÀPdÀCl(2)
88.8(3)
99.5(4)
88.0(3)
117.1(5)
85.5(3)
127.6(5)
111.9(4)
89.04(11)
P(2)ÀPdÀP(3)
P(3)ÀPdÀCl(1)
Cl(1)ÀPdÀCl(2)
PdÀP(2)ÀC(50)
PdÀP(2)ÀC(60)
PdÀP(3)ÀC(70)
PdÀP(3)ÀP(80)
92.95(11)
89.78(10)
88.24(11)
113.3(2)
108.5(2)
111.0(2)
115.1(2)
ReÀC(24)ÀP(3)ÀC(70)
ReÀC(24)ÀP(3)ÀC(80)
P(1)ÀReÀC(1)ÀP(2)
N(1)ÀReÀC(1)ÀP(2)
À 173.5(6)
82.4(7)
ReÀC(1)ÀP(2)ÀPd
À 57.6(6)
68.7(6)
177.4(5)
ReÀC(1)ÀP(2)ÀC(50)
ReÀC(1)ÀP(2)ÀC(60)
À 153.6(5)
À 65.2(6)
Scheme 2. Summaryof Enantioselective Arylations
Entry
Catalyst System, Conditions
Conv.
[%]
Ratio (7: 8)
Yield (7 ‡ 8)
ee (R)-7
[%]
[%]
1
2
3
4
5
6
7
Pd(OAc)2/(R)-binap (1 : 2), 408, 8 d
Pd(OAc)2/(R)-1 (1 : 2), 408, 14 d
Pd(OAc)2/(S)-2 (1 : 2), 608, 14 d
(S)-5/(R)-1 (1 : 2), 408, 14 d
(S)-6/(S)-2 (1 : 2), 608, 8 d
(S)-5/(R)-1/(i-Bu)2AlH(1 : 1.5 : 2), 60 8, 9 d
(S)-6/(S)-2/(i-Bu)2AlH(1 : 1.5 : 2), 40 8, 4 d
95
90
80
85
60
35
95
88 : 12
78 : 22
92 : 8
88 : 12
90 : 10
86 : 14
90 : 10
72
75
69
72
74
5
5
10
5
n.d.
12
84
(R)-1 or (S)-6/(S)-2 (1:1 molar ratio) yielded after induction periods deep red
solutions of active catalysts that again afforded low enantioselectivity. The reductant
(i-Bu)2AlHis sometimes employed to generate Pd(0) species [12]. Accordingly,
Entries 6 and 7 are similar to 4 and 5, but with 2.0 equivalents of (i-Bu)2AlH. Entry7
gave the highest rate and enantioselectivity of all, but the latter (12% ee) was far below
useful levels. Similar data were obtained with cyclohexenyl tirfluoromethylsulfonate, as
detailed elsewhere [11].
The preceding data prompted us to test the conformations of 6 and of the related
rhodium/rhenium complex 4 in solution by NMR with the one dimensional DPFGSE-
NOE technique [13][14]. In the first series of experiments, all nonphenyl
H-atom signals were separately irradiated. Each cyclopentadienyl H-atom signal
irradiated gave a 1.5 2.2% enhancement of the neighboring cyclopentadienyl H-atom

Helvetica Chimica Acta Vol. 85 (2002)
1783
signals, establishing connectivity2). Both 6 and 4 exhibited a high-field cyclopenta-
dienyl H-atom signal (d ˆ 3.19 and 4.34; designated as Ha) with only a single neighbor.
Irradiation of these signals also gave enhancements of the high-field ReCHH' H-atom
signal (1.2%) but not the low-field signal (<0.1%). The reverse irradiations gave
enhancements of 2.3% and 2.5%. All other NOE relationships were also very similar.
Representative spectra are illustrated in Fig. 2. These data are analyzed (Discussion) in
relation to the partial structures in Fig. 3.
Fig. 2. 1H-NMR Spectra of 6 (CD2Cl2; see Fig. 3 for definitions of the indices a À g, R and S). a) Standard 500-
MHz spectrum (solvent impurities are designated by *); b) DPFGSE-NOE spectrum from the irradiation of
signal HS, with enhancements relative to the irradiated signal ( 100%); c) as in b) but irradiation of signal HR;
d) as in b but irradiation of signal Ha; e) as in b but irradiation of signal Hg.
Discussion. The preceding data establish that the rhenium-containing diphos-
phines 1 and 2 readily form PdCl2 chelates. Together with the previously reported
rhodium chelates 3 and 4, it can be assumed that such diphosphines will chelate a
variety of metal fragments. The rotational degree of freedom intrinsic to the
2
)
Moderate enhancements of some aromatic H-atom resonances were also observed. With 4, irradiation of
the trinorbornadiene ˆ CHH-atoms gave either enhancements of or excitation transfer to all other
trinorbornadiene signals, indicating an exchange process that is further described in the Exper. Part.

1784
Helvetica Chimica Acta Vol. 85 (2002)
Fig. 3. Selected conformational features of chelate complexes. a) 6 ¥ 3 CH2Cl2 with the palladium chloride ligands,
one PPh3 phenyl ring, and solvate omitted; b) as in a but with only the cyclopentadienyl C-atoms and H-atoms,
(calc. positions), and immediately bonded atoms of the chelate ring; c) as in b but for rhodium complex 4.
cyclopentadienyl-based PPh2 group enables a wide range of PP distances and bite
angles. As analyzed earlier, the Re(CH2)nPPh2 moieties are much more basic and
nucleophilic than model organic phosphines that lack the rhenium substituent [3].
Enhanced phosphine basicity, as well as steric bulk, increases the rates of many metal-
catalyzed reactions [15].
As noted above, 3 and 4 were highly effective catalyst precursors for the
enantioselective hydrogenation of protected dehydroamino acids. However, the
palladium chelates 5 and 6, or related species generated in situ with Pd(OAc)2, are
active but are poorly enantioselective catalyst precursors with respect to the Heck
arylation in Scheme 2. No correlation would necessarily be expected, as these
transformations involve much different types of configuration-determining steps, as
well as catalyst electronic configuration (Rh(I)/(III) or d8/d6 vs. Pd(0)/Pd(II) or d10/d8).

Helvetica Chimica Acta Vol. 85 (2002)
1785
In any event, this inauspicious beginning should not diminish enthusiasm for palladium
catalysts derived from diphosphines 1 or 2. The enantioselectivities of Heck arylations
are known to be highly dependent upon a number of factors. For example, when aryl
trifluoromethylsulfonates are replaced by aryl iodides, racemic or essentially racemic
products are obtained [7a].
The crystal structure of 6 (Fig. 1) shows the normal types of bond lengths and
angles about the formally octahedral Re-atom (Table 3) [3][4]. The P(2)ÀC(1) linkage
bends into the interstice on the Re-atom between the least bulky nitrosyl and
cyclopentadienyl ligands, as indicated by the N(1)ÀReÀC(1)ÀP(2) and
P(1)ÀReÀC(1)ÀP(2) torsion angles (À65.28 and À153.68). Fig. 3 shows two views
of the −chiral pocket× generated by removing the ClÀ ligands from the Pd-atom (a, b).
View a highlights the face/edge dispositions of the four phenyl groups, which have been
carefully analyzed by Brunner et al. [5a] and which transmit chirality to the reaction
site. View b highlights the conformation of the six-membered chelate ring and
cyclopentadienyl ligand. For comparison, the analogous segment of rhodium/rhenium
complex 4 is shown in view c.
In contrast to the chair-like conformation of the rhodium chelate in c, the palladium
chelate in b adopts a folded, twist-boat-like conformation. The torsion angles about the
C(1)ÀP(1) bonds are quite similar in both structures. Hence, the major difference
stems from the rotational degree of freedom involving the substituted cyclopentadienyl
ligand. In this regard, note that the C(24)ÀReÀC1 angle in b is much smaller than in c
(85.5(3)8 vs. 99.4(3)8). These differences lead to different categories of chiral pockets.
That in 6 (see a) exhibits a P3M configuration, whereas 4 shows a P2M2 configuration. P
and M are helical chirality descriptors relating the four phenyl groups, derived by
Brunner×s protocol [5a]3).
The PdÀP bond lengths in b are close to the RhÀP bond lengths in c (2.248(3) À
2.286(3) vs. 2.295(2) À 2.360(2) ä), and the remaining bond distances are nearly
identical. Therefore, the differing conformations must either represent responses to the
other metal ligands (ClÀ on Pd-atom vs. diene on Rh-atom), or a crystal-packing effect.
The close similarity of the NOE data for 6 and 4 suggests a similar ensemble of
conformations in solution, supporting a packing effect. Although the palladium/
rhenium chelate 5 was not structurally investigated, the diminished degrees of freedom
resulting from the smaller ring size renders a close correspondence to the rhodium/
rhenium analog 3 even more likely.
View c of 4 shows the only possible assignment of Ha that would give an
enhancement of a ReCHH' H-atom, which occupies an axial position and, when
rhenium has the absolute configuration depicted in c, corresponds to HS. The
conformation in view b of 4 features a greater distance between Ha and the
corresponding ReCHH' H-atom (3.02 vs. 2.54 ä) and would give a much lower NOE
enhancement due to the l/r6 dependence. However, since there is no way to calibrate
the exact value expected for each conformation, we can only conclude that the relative
conformer populations are similar. The upfield chemical shifts of the Ha signals are
logically attributed to the shielding effect of a PPh3 phenyl ring.
3
)
A table of structural data relating to the chiral pocket defined by the PPh2 group of 6 and 4 can be obtained
from the author.

1786
Helvetica Chimica Acta Vol. 85 (2002)
In summary, this work has extended the synthetic, structural, and catalytic
chemistry of architecturally novel metal complexes derived from chiral-chelating
diphosphines that contain a transition-metal stereocenter in the backbone. Although
the enantioselectivities in the reaction initially explored with the new palladium/
rhenium precatalyst 6 are modest, NOE experiments indicate an ensemble of
conformations similar to those in the highly effective rhodium/rhenium precatalyst 4.
Chelate ligands that incorporate other types of chiral, non-ferrocene metal fragments
are seeing rapid development in other groups [2]. Additional examples and
applications of metal-containing donor ligands in catalysis will be reported in the near
future [16].
Experimental Part
General. General procedures, chemical purifications, and instrumentation were identical with those in two
recent full papers [3][4]. Chemicals new to this work were obtained and treated as follows: di(benzonitrilo)di-
chloropalladium([PdCl2(PhCN)2]), prepared from PdCl2 (Johnson Matthey, 99.9%) [17]; phenyl trifluorometh-
ylsulfonate, (PhOTf), prepared from phenol, pyridine, and trifluoromethylsulfonic acid anhydride (Fluka) [18];
(i-Bu)2AlH( ꢁ1.2m in toluene, Fluka), Et3N (Fluka, 99.99%), and (‡)-tris{3-[(heptafluoro)hydroxymethyle-
ne]camphorato}europium(III) ((‡)-[Eu(hfc)3]; Aldrich) were used as received.
(R)-[(h5-C5H4PPh2)Re(NO)(PPh3)(m-PPh2)PdCl2] ((R)-5)1). A Schlenk flask was charged with (S)-[(h5-
C5H4PPh2)Re(NO)(PPh3)(PPh2)] ((S)-1; 1.150 g, 1.260 mmol) [3a]1) and THF (30 ml).
A soln. of
[PdCl2(PhCN)2] (0.4833 g, 1.260 mmol) in THF (20 ml) was added with stirring. After 30 min, pentane
(30 ml) was added to the purple-red heterogeneous mixture. After 60 min, the solid was collected on a sintered
glass filter, washed with THF (2 Â 3 ml) and pentane (2 Â 10 ml), and dried in vacuo to give (R)-5 (1.025 g,
0.9401 mmol, 75%)1) as a carmine powder that was >95% pure by 31P-NMR. M.p. 230 2358 (dec). IR (powder
film, cmÀ1): 1687s (NO). 1H-NMR (400 MHz, CDCl3): 8.18 8.13, 7.97 7.92, 7.79 7.68, 7.64 7.25, 7.15 6.96,
6.95 6.84, 6.64 6.59 (7m, 7 C6H5); 5.71, 5.63, 5.57, 4.72 (4 br. s, C5H4). 13C{1H}-NMR (100.5 MHz, CDCl3): 142.1
(d, J(C,P) ˆ 30, Ci (PPh2)); 137.8 (d, J(C,P) ˆ 30, Ci (PPh2)); 136.1 (d, J(C,P) ˆ 14, Co (PPh2)); 134.2
(d, J(C,P) ˆ 11, C'o (PPh2)); 133.7 (s, Cp (PPh2)); 133.7 (d, J(C,P) ˆ 13, C'' (PPh2)); 133.4 (d, J(C,P) ˆ 10.2,
o
Co''' (PPh2)); 132.5 (s, C' (PPh2)); 130.8 (s, Cp (PPh3)); 130.5 (s, C'' (PPh2)); 129.0 (s, C''' (PPh2)); 128.8
p
p
p
(d, J(C,P) ˆ 12, Cm (PPh2)); 128.5 (d, J(C,P) ˆ 10, Cm (PPh3)); 127.9 (d, J(C,P) ˆ 11, C' (PPh2)); 127.3
m
(d, J(C,P) ˆ 10, C'm' (PPh2)); 127.1 (d, J(C,P) ˆ 10, C''' (PPh2)); 112.9 (br. s, C5H4); 100.6 (br. s, C5H4); 98.3 (s,
m
‡
‡
C5H4); 94.9 (s, C5H4); 84.2 (s, C5H4). 31P{1H}-NMR: see Table 1. FAB-MS4): 1090 (5, M ), 1055 (9, [M À Cl] ),
‡
‡
‡
1018 (10, [M À 2Cl] ), 914 (9, [M À PdCl2] ), 834 (100, [M À PdCl2 À C6H5] ), no other significant peaks above
>700. Anal. calc. for C47H39Cl2NOP3PdRe (1090.3): C 51.77, H3.60, N 1.28; found: C 50.24, H3.84, N 1.18.
[(h5-C5H4PPh2)Re(NO)(PPh3)(m-CH2PPh2)PdCl2] (6). Procedure A. A Schlenk tube was charged with
[Re(h5-C5H4PPh2)(CH2PPh2)(NO)(PPh3)] (2; 0.0818 g, 0.0882 mmol) [3a] and THF (4 ml). A soln. of
[PdCl2(PhCN)2] (0.0338 g, 0.0882 mmol) in benzene (4 ml) was added with stirring. After 30 min, pentane
(10 ml) was added to the beige heterogeneous mixture. The solid was collected on a sintered glass filter, washed
with pentane (2 Â 5 ml), and dried in vacuo (0.0789 g, 0.0714 mmol, 81%). The crude product was filtered (SiO2,
10 Â 1 cm, MeOH/CH2Cl2), and the filtrate was concentrated in vacuo to 5 ml. The precipitate was collected on
a sintered glass filter, washed with pentane (5 ml), and dried in vacuo to give 6 as a rusty-brown powder
(0.0731 g, 0.0662mmol, 75%) that was >99% pure by 31P-NMR. M.p. 215 2208 (dec.). IR (powder film, cmÀ1):
1661s, (NO). 1H-NMR (400 MHz, CDCl3)5): 7.94 7.31 (m, 7 C6H5): 6.22 (br. s, Hd (C5H4)); 5.57 (br. s, Hg
(C5H4)); 5.27 (br. s, Hb (C5H4)); 3.19 (br. s, Ha (C5H4)); 2.26 2.22 (m, CHRHS): 2.05 2.00 (m, CHRHS). 13C{1H}-
NMR (100.5 MHz, CD2Cl2): 138.8 (d, J(C,P) ˆ 39, Ci (PPh2)); 136.7 (d, J(C,P) ˆ 12, Co (PPh2)); 135.5
(d, J(C,P) ˆ 37, C'i (PPh2)); 135.3 (d, J(C,P) ˆ 10, Co' (PPh2)); 134.1 (d, J(C,P) ˆ 54, Ci (PPh3)); 133.8
4
)
)
m/z Values correspond to the most intense peak of the isotope envelope; relative intensities are for the
specified mass range.
The H-atoms designated a, b, g, d, HS, and HR were assigned from the NOE experiments, and the positions
are defined in Fig. 3.
5

Helvetica Chimica Acta Vol. 85 (2002)
1787
(d, J(C,P) ˆ 8, C'' (PPh2)); 133.6 (d, J(C,P) ˆ 10, Co (PPh3)); 133.2 (d, J(C,P) ˆ 10, C''' (PPh2)); 132.8 (s, Cp
o
o
(PPh2)); 131.0 (s, Cp (PPh3)); 130.7 (s, C' (PPh2)); 130.5 (s, C'' (PPh2)); 130.3 (s, C''' (PPh2)); 129.2 (d, J(C,P) ˆ
p
p
p
11, Cm (PPh2)); 129.1 (d, J(C,P) ˆ 11, Cm (PPh3)); 128.1 (d, J(C,P) ˆ 11, C' (PPh2)); 128.1 (d, J(C,P) ˆ 10, C''
m
m
(PPh2)); 127.9 (d, J(C,P) ˆ 10, C''' (PPh2)); 92.3 (s, C5H4); 91.7 (br. s, C5H4); 89.0 (s, C5H4); 87.6 (s, C5H4); À9.6
m
‡
‡
(br. s, CH2); 31P{1H}-NMR: see Table 1. FAB-MS4): 1103 (2, M ), 1068 (82, [M À Cl] ), 834 (42, [M À 2Cl À
‡
‡
CH2PPh2] ), 771 (100, [M À 2Cl À PPh3] ), no other significant peaks above >700. Anal. calc. for
C48H41Cl2NOP3PdRe (1104.3): C 52.21, H3.74, N 1.27; found: C 51.82, H3.60, N 1.31.
Procedure B. A Schlenk tube was charged with (R)-2 (0.546 g, 0.589 mmol) [3a]1) and THF (35 ml). A soln.
of [PdCl2(PhCN)2] (0.2258 g, 0.5886 mmol) in THF (10 ml) was added with stirring. After 30 min, pentane
(50 ml) was added to the beige heterogeneous mixture. The solid was collected on a sintered glass filter, washed
with pentane (2 Â 5 ml), and dried in vacuo (0.5980 g, 0.5415 mmol, 92%). The crude product was dissolved in
CH2Cl2 (10 ml) and layered with hexane (40 ml). After 24 h, the orange needles were collected by filtration,
washed with pentane (2 Â 4 ml) and dried (10À3 bar, 2 h) to give (R)-6 (0.5525 g, 0.5003 mmol, 85%)1) that was
589
>99% pure by 31P-NMR. M.p. 212 2148 (dec.) [a] ˆ À127 Æ 3 (c ˆ 1.20, THF). The NMR spectra (1H, 13C,
26
31P) were similar to those of the racemate. Anal. calc. for C48H41Cl2NOP3PdRe (1104.3): C 52.21, H3.74, N 1.27;
found: C 51.61, H4.00, N 1.20.
Representative Catalytic Reactions. Reaction A. A 5-ml vial was charged under Ar with Pd(OAc)2 (0.007 g,
0.031 mmol), (S)-1 or (R)-2 (0.060 mmol)1), benzene (2 ml), and undecane standard (typically 0.078 g,
0.500 mmol), capped with a septum, and heated with stirring in a sand bath (Scheme 2). A brown precipitate
formed. After 20 min, PhOTf (0.226 g, 1.00 mmol), Et3N (0.202 g, 2.00 mmol), and 2,3-dihydrofuran (0.351 g,
5.00 mmol) were added. The vial was sealed under Ar, and the heterogeneous mixture was stirred until reaction
was complete (GLC, Optima-5, Macherey-Nagel, 25 mm, 25 m  0.32 mm; 100 1808, 208/min). tr: 3.9 min
(PhOTf), 5.7 min (undecane), 7.8 min (2,3-dihydro-2-phenylfuran (7)), 8.3 min (2,5-dihydro-2-phenylfuran
(8)). The mixture was poured into pentane (10 ml) and filtered (SiO2, 1 Â 5 cm, pentane). The combined
eluates were concentrated, and the yellow oil was purified by FC (SiO2, 2 Â 15 cm, pentane/ether 1:1) to afford
a mixture of 7 and 8 as a colorless oil. The enantiomeric purity of 7 was elucidated with the shift reagent (‡)-
[Eu(hfc)3] (1H-NMR, CDCl3: olefinic and aliphatic signals of (S)-7 downfield from (R)-7) [7a].
Reaction B. A 5-ml vial was charged under Ar with (R)-5 or (R)-6 (0.030 mmol)1), (S)-1 or (R)-2
(0.060 mmol)1), and benzene (2 ml), capped with a septum, and heated with stirring in a sand bath (Scheme 2).
After 20 min, PhOTf (0.226 g, 1.00 mmol), Et3N (0.202 g, 2.00 mmol), and 2,3-dihydrofuran (0.351 g,
5.00 mmol) were added. The reaction was monitored and worked up as described for Reaction A.
Reaction C. A 5-ml vial was charged under Ar with (R)-6 (0.0430 g, 0.0389 mmol)1), (R)-2 (0.0541 g,
0.0584 mmol)1) and benzene (3 ml), capped with a septum, and heated with stirring in a sand bath. After 5 min,
(i-Bu)2AlH(0.065 ml, 0.0778 mmol; 1.2 m in toluene) was added dropwise with stirring. Gas bubbles formed and
the mixture immediately turned deep red. Then, PhOTf (0.294 g, 1.30 mmol), Et3N (0.263 g, 2.6 mmol), and 2,3-
dihydrofuran (0.456 g, 6.50 mmol) were added, and the homogeneous mixture was stirred at 408. The reaction
was monitored and worked up as described for Reaction A.
Crystallography. A CH2Cl2 soln. of 6 was layered with hexanes. After 1 d at r.t., orange needles of 6 ¥ 3
CH2Cl2 had formed. Data were collected as outlined in Table 2. Cell parameters were obtained from 10 frames
with a 108 scan and refined with 53094 reflections. Lorentz, polarization, and absorption corrections [19] were
applied. The space group was determined from systematic absences and subsequent least-squares refinement.
The structure was solved by direct methods. The parameters were refined with all data by full-matrix least-
squares on F2 by means of SHELXL-97 [20]. Non-H-atoms were refined with anisotropic thermal parameters.
The H-atoms were fixed in idealized positions by means of a riding model. Scattering factors were taken from
the literature [21]. The asymmetric unit contained 3 molecules of CH2Cl2. Two were disordered and refined to
occupancy ratios of 51:49 (C91/C91', Cl3/Cl3', Cl4/Cl4') and 60 :40 (Cl5/Cl5', Cl6/Cl6').
Crystallographic data (excluding structure factors) have been deposited with the Cambridge Crystallo-
graphic Data Centre, No. CCDC 176280. Copies of this information can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2 1EZ UK (fax: ‡ 44(1223)336033; e-mail:
deposit@ccdc.ca.ac.uk).
NOE Experiments. A 5-mm NMR tube was placed in a Schlenk tube, evacuated, heated, purged with N2,
and charged with a soln. of 4 ¥ CH2Cl2 or 6 (0.0158 mmol) in dry CDCl3 or CD2Cl2 (0.7 ml; 0.022m). The soln. was
purged with a gentle N2 stream (10 min). The tube was sealed with a Teflon septum and transferred to a JEOL
Alpha-500 spectrometer for DPFGSE-NOE measurements [13][14]. Chemical shifts were referenced to
residual solvent 1H-signals (d ˆ 7.27, CDCl3; 5.32, CH2Cl2). The data allowed additional assignments for the
previously reported 4 ¥ CH2Cl2 [3] (CDCl3, 248)5): 5.55 (br. s, Hd (C5H4)); 5.51 (br. s, Hg (C5H4)); 4.92 (br. s,

1788
Helvetica Chimica Acta Vol. 85 (2002)
ˆCH(trinorbornadiene)); 4.89 (br. s, Hb (C5H4)); 4.63, 4.54 (2 br. s, 2 ˆ CH(trinorbornadiene)); 4.10 (br. s,
ˆCH(trinorbornadiene)); 4.02, 3.85 (2 br. s, 2 bridgehead CH); 1.52 (br. s, CHH'); 2.38 2.28 (m, CHRHS);
1
2.23 2.18 (m, CHRHS). A H-NOESY/EXSY experiment showed cross peaks for three pairs of signals (4.92/
4.10, 4.63/4.54, 4.02/3.85), consistent with a well-precedented exchange involving a formal 1808 rotation about
the rhodium-trinorbonadiene bond axis6).
We thank the Deutsche Forschungsgemeinschaft (DFG, GL 300/4-1), Dr. Walter Bauer for assistance with
the NOE measurement, and Johnson MattheyPMC (palladium loan) for support.
REFERENCES
[1] a) A. Togni, in −Metallocenes×, Eds. A. Togni, R. L. Halterman, Wiley-VCH, Weinheim, 1998, Vol. 2,
Chapt. 10; b) C. J. Richards, A. J. Locke, Tetrahedron: Asymmetry 1998, 9, 2377; c) A. Togni, N. Bieler, U.
Burckhardt, C. Kˆllner, G. Pioda, R. Schneider, A. Schnyder, Pure Appl. Chem. 1999, 71, 1531; d) D. A.
√
Dobbs, K. P. M. Vanhessche, E. Brazi, V. Rautenstrauch, J.-Y. Lenoir, J.-P. Genet, J. Wiles, S. H. Bergens,
Angew. Chem. 2000, 112, 2080; Angew. Chem., Int. Ed. 2000, 39, 1992.
[2] a) S. Kudis, G. Helmchen, Angew. Chem. 1998, 110, 3210; Angew. Chem., Int. Ed. 1998, 37, 3047; b) C.
¬
Bolm, K. Munƒiz, Chem. Soc. Rev. 1999, 28, 51; c) C. Pasquier, L. Pelinski, J. Brocard, A. Mortreux, F.
Agbossou-Niedercorn, Tetrahedron Lett. 2001, 42, 2809; d) S. U. Son, K. H. Park, S. J. Lee, Y. K. Chung,
D. A. Sweigart, Chem. Commun. 2001, 1290; e) C. Bolm, M. Kesselgruber, N. Hermanns, J. P. Hildebrand,
G. Raabe, Angew. Chem. 2001, 113, 1536; Angew. Chem., Int. Ed. 2001, 40, 1488; f) G. Jones, C. J. Richards,
Organometallics 2001, 20, 1251.
[3] a) K. Kromm, B. D. Zwick, O. Meyer, F. Hampel, J. A. Gladysz, Chem. Eur. J. 2001, 7, 2015; b) B. D. Zwick,
A. M. Arif, A. T. Patton, J. A. Gladysz, Angew. Chem. 1987, 99, 921; Angew. Chem., Int. Ed. 1987, 26, 910.
[4] L. J. Alvey, O. Delacroix, C. Wallner, O. Meyer, F. Hampel, S. Szafert, T. Lis, J. A. Gladysz, Organometallics
2001, 20, 3087.
[5] a) H. Brunner, A. Winter, J. Breu, J. Organomet. Chem. 1998, 553, 285; b) T. Ohkuma, M. Kitamura, R.
Noyori, in −Catalytic Asymmetric Synthesis× 2nd edn., Ed. I. Ojima, Wiley-VCH, Weinheim, 2000. Chapt. 1.
[6] −Transition Metals for Organic Synthesis×, Eds. M. Beller, C. Bolm, Wiley-VCH, Weinheim, 1998, Vol. 1 2;
A. Tenaglia, A. Heumann, Angew. Chem. 1999, 111, 2316; Angew. Chem., Int. Ed. 1999, 38, 2180; S. A.
Raynor, J. M. Thomas, R. Raja, B. F. G. Johnson, R. G. Bell, M. D. Mantle, Chem. Commun. 2000, 1925.
[7] a) F. Ozawa, A. Kubo, Y. Matsumoto, T. Hayashi, E. Nishioka, K. Yanagi, K. Moriguchi, Organometallics
1993, 12, 4188, and earlier refs. cited therein; b) Y. Hashimoto, Y. Horie, M. Hayashi, K. Saigo,
Tetrahedron: Asymmetry 2000, 11, 2205.
[8] O. Loiseleur, P. Meier, A. Pfaltz, Angew. Chem. 1996, 108, 218; Angew. Chem., Int. Ed. 1996, 35, 200; O.
Loiseleur, M. Hayashi, N. Schmees, A. Pfaltz, Synthesis 1997, 1338.
[9] S. R. Gilbertson, Z. Fu, D. Xie, Tetrahedron Lett. 2001, 42, 365, and earlier refs. cited therein.
[10] F. Ozawa, A. Kubo, T. Hayashi, Chem. Lett. 1992, 2177.
[11] K. Kromm, PhD Thesis, Universit‰t Erlangen-N¸rnberg, in preparation.
[12] E. Negishi, R. A. John, J. Org. Chem. 1983, 48, 4098.
[13] K. Stott, J. Keeler, Q. N. Van, A. J. Shaka, J. Mag. Res. 1997, 125, 302.
[14] W. Bauer, A. Soi, A. Hirsch, Magn. Reson. Chem. 2000, 38, 500; C. Gaul, P. I. Arvidsson, W. Bauer, R. E.
Gawley, D. Seebach, Chem. Eur. J. 2001, 7, 4117.
[15] Q. Shelby, N. Kataoka, G. Mann, J. Hartwig, J. Am. Chem. Soc. 2000, 122, 10718; K. E. Torraca, S.-I.
Kuwabe, S. L. Buchwald, J. Am. Chem. Soc. 2000, 122, 12907; C. Dai, G. C. Fu, J. Am. Chem. Soc. 2001, 123,
2719; G. Y. Li, Angew. Chem. 2001, 113, 1561; Angew. Chem., Int. Ed. 2001, 40, 1513.
[16] S. Eichenseher, K. Kromm, O. Delacroix, J. A. Gladysz, Chem. Commun. 2002, 1046; K. Kromm, P. L.
Osburn, J. A. Gladysz, submitted for publication.
[17] −Synthetic Methods of Organometallic and Inorganic Chemistry×, Eds. W. A. Herrmann, A. Salzer, Thieme,
Stuttgart, 1996, Vol. 1, p. 159 160.
[18] M. E. Mowery, P. DeShong, J. Org. Chem. 1999, 64, 3266; L. R. Subramanian, M. Hanack, L. W. K. Chang,
M. A. Imhoff, P. v. R. Schleyer, F. Effenberger, W. Kurtz, P. J. Stang, T. E. Dueber, J. Org. Chem. 1976, 26,
4099; E. Anders, M. Stankowiak, Synthesis 1984, 1039.
6
)
The actual mechanism is known to be more complex [22].

Helvetica Chimica Acta Vol. 85 (2002)
1789
[19] −Collect×. Data collection software, Nonius B. V., 1998; Z. Otwinowski, W. Minor, Methods Enzymol. 1997,
276, 307.
[20] G. M. Sheldrick, SHELX-97, Program for refinement of crystal structures, University of Gˆttingen, 1997.
[21] D. T. Cromer, J. T. Waber, in −International Tables for X-Ray Crystallography×, Eds. J. A. Ibers, W. C.
Hamilton, Kynoch; Birmingham, 1974.
[22] H. Berger, R. Nesper, P. S. Pregosin, H. R¸egger, M. Wˆrle, Helv. Chim. Acta 1993, 76, 1520; M. Valentini,
K. Selvakumar, M. Wˆrle, P. S. Pregosin, J. Organomet. Chem. 1999, 587, 244; B. Crociani, S. Antonaroli,
M. L. Di Vona, S. Licoccia, J. Organomet. Chem. 2001, 631, 117.
Received December 21, 2001