Operando high-pressure NMR and IR study of the hydroformylation of 1-hexene by 1,1′-bis(diarylphosphino)metallocene-modified rhodium(I) catalysts

Article abstract of DOI:10.1021/om050241l

High-pressure nuclear magnetic resonance and infrared study of the hydroformylation of 1-hexene in the presence of rhodium(1) complexes acting as catalysts was presented. The catalysts exhibited good activity and moderate regioselectivity. It was found th

Full text of DOI:10.1021/om050241l

3692
Organometallics 2005, 24, 3692-3702
Operando High-Pressure NMR and IR Study of the
Hydroformylation of 1-Hexene by
1,1′-Bis(Diarylphosphino)metallocene-Modified
Rhodium(I) Catalysts
Claudio Bianchini,* Werner Oberhauser, and Annabella Orlandini
Istituto di Chimica dei Composti Organometallici (ICCOM-CNR), Area di Ricerca CNR di
Firenze, via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
Carlo Giannelli and Piero Frediani
Dipartimento di Chimica Organica, Universita` degli Studi di Firenze, Polo Scientifico,
via della Lastruccia 13, 50019 Sesto Fiorentino, Italy
Received March 31, 2005
Some rhodium(I) complexes of the general formula [Rh(P-P)(COD)]X were synthesized
and characterized by multinuclear NMR spectroscopy (COD ) cyclocta-1,5-diene; P-P )
1,1′-bis(diphenylphosphino)ferrocene, dppf, X ) BPh4, PF6; P-P ) 1,1′-bis(diphenylphos-
phino)ruthenocene, dppr, X ) BPh4; P-P ) 1,1′-bis(diphenylphosphino)osmocene, dppo,
X ) BPh4, PF6; P-P ) 1,1′-bis(diphenylphosphino)octamethylferrocene, dppomf, X ) BAr′4;
P-P ) (1,1′-bis(di(o-isopropylphenyl)phosphino)ferrocene, o-iPr-dppf, X ) BAr′4). These
complexes were employed as catalyst precursors for the hydroformylation of 1-hexene in
THF either in standard autoclaves or in high-pressure (HP) NMR tubes and IR cells. All
catalysts exhibited good activity (TOFs ranging from 700 to 1000 mol aldehyde (mol cat)-1
h-1) and moderate regioselectivity in n-heptanal (67-74%). Irrespective of the rhodium
precursor, the HP-NMR experiments under catalytic conditions showed the formation of
kinetic dicarbonyl products at room temperature, which were independently prepared by
reaction of the COD precursors with 1 bar CO in THF. Square-planar dicarbonyl complexes
containing two cis carbonyl groups were obtained with the dppf and dppomf ligands, while
the precursors with the dppr, dppo, and o-iPr-dppf ligands gave trigonal-bipyramidal
dicarbonyl complexes with the equatorial positions occupied by two carbonyl groups and by
the metallocene metal atom. The complexes [Rh(CO)2(dppf)]PF6 and [Rh(CO)2(dppo)]PF6 were
isolated in the solid state and characterized by single-crystal X-ray analysis. On increasing
gradually the temperature of the HP-NMR hydroformylation experiments, the dppf, dppr,
and dppo dicarbonyl complexes disappeared. Formed in their place were neutral five-
coordinate hydride(dicarbonyl) complexes RhH(CO)2(P-P) that exist in solution as two
rapidly equilibrating geometric isomers. The reaction of the o-iPr-dppf precursor with syngas
at 60 °C gave a trigonal-bipyramidal dicarbonyl complex with a dative Fe-Rh bond, while
the dppomf complex decomposed to various CO-containing rhodium complexes. Unlike HP-
NMR spectroscopy, HP-IR spectroscopy showed no kinetic product at any stage of the catalytic
reactions. Also, HP-IR spectroscopy allowed us to distinguish the ee and ea geometric isomers
of the hydride(dicarbonyl) resting states with dppf, dppr, and dppo. Irrespective of the
temperature, o-iPr-dppf formed a stable dicarbonyl complex as a result of the catalytic
reaction, while the dppomf dicarbonyl was unstable under hydroformylation conditions,
converting into phosphine-free carbonyl Rh compounds.
Introduction
to design molecular structures with different sand-
wiched metals and substituents either on the cyclopen-
tadienyl ligands (Cp) or on the phosphorus atoms. Most
of the research work in catalysis has been carried out
with palladium(II) precursors of the prototypical ligand
1,1′-bis(diphenylphosphino)ferrocene (dppf): represen-
tative catalytic reactions include cross-coupling reac-
tions,2 Heck reaction,3 carbonylation of chloroarenes,4
aryl halide amination,5 hydroamination of alkynes,6
glioxylate-ene reaction with chiral controllers,7 alternat-
ing copolymerization of CO and ethylene,8 and meth-
oxycarbonylation of alkenes.9
1,1′-Bis(phosphino)metallocenes are potentially chelat-
ing tri- and bidentate ligands with many applications
in homogeneous reactions catalyzed by transition metal
compounds.1 The success of these ligands in catalysis
is largely due to their unique flexibility that allows one
* To whom correspondence should be addressed.
(1) (a) Colacot, T. J. Platinum Met. Rev. 2001, 45, 22. (b) Gan, K.-
S.; Hor, T. S. A. In Ferrocenes: Homogeneous Catalysis, Organic
Synthesis, Materials Science; Togni, A., Hayashi, T., Eds.; VCH-
Wiley: Weinheim, 1995; Chapter 1.
10.1021/om050241l CCC: $30.25 © 2005 American Chemical Society
Publication on Web 06/16/2005

Hydroformylation of 1-Hexene by Rh(I) Catalysts
Organometallics, Vol. 24, No. 15, 2005 3693
Chart 1
To acquire as much information as possible, standard
methods, such as characterization of precursors and
intermediates, reactions with isolated compounds, and
batch catalytic reactions, have been backed up by
operando high-pressure NMR and IR experiments.
Experimental Section
General Comments. All manipulations were routinely
performed under an atmosphere of nitrogen, using Schlenk
tubes. [RhCl(COD)]2 and NaBAr′4 (Ar′4 ) m-(CF3)2C6H3) were
synthesized following reported procedures.12,13 Likewise, the
1,1′-bis(diorganylphosphino)metallocene ligands were prepared
as reported in the literature.8c,14 1-Hexene was purified by
elution through a neutral Al2O3 (70-230 mesh) chromato-
graphic column, followed by distillation and storage under
nitrogen. Commercial trans hex-2-ene (purity > 99%) was used
without further purification. Elemental analyses were per-
The hydroformylation of alkenes by 1,1′-bis(diarylphos-
phino)ferrocene rhodium(I) catalysts has been exclu-
sively investigated by Unruh10 and van Leeuwen.11 Both
authors have reported that the introduction of electron-
withdrawing groups in the para position of the P-aryl
rings increases the catalytic activity, while the regiose-
lectivity is controlled by several factors, including the
ligand basicity, the alkene, and the ligand-to-rhodium
ratio.10,11 Despite these excellent studies, much remains
to understand on the structure-activity relationships
regarding the hydroformylation of alkenes by 1,1′-bis-
(diarylphosphino)metallocene rhodium catalysis. In par-
ticular, there is no information on hydroformylation
catalysts containing a different metallocene metal than
iron or bearing alkyl substituents either on the Cp’s or
in the ortho position of the P-aryl rings. Indeed, similar
structural variations have been shown to affect remark-
ably both the activity and the selectivity of alkene
alkoxycarbonylation reactions.8a The lack of such spe-
cific information on a reaction that still raises much
interest both in industry and academically prompted us
to study the hydroformylation of 1-hexene by a family
of 1,1′-bis(diarylphosphino)metallocene rhodium(I) com-
plexes where the chelating diphosphine ligand is dppf,
dppo, dppr, dppomf, or isopropyl-dppf (Chart 1).
1
formed with a Perkin-Elmer Series II CHNS/O analyzer. H
and 31P{1H} NMR spectra were recorded on a Varian VXR300
spectrometer operating at 299.94 and 121.42 MHz, respec-
tively. 13C{1H} NMR spectra were recorded on a Varian
Mercury 400 spectrometer operating at 100.57 MHz. High-
pressure NMR experiments were performed on a Bruker ACP
200 spectrometer operating at 200.13 MHz for 1H spectra and
81.01 MHz for 31P{1H} spectra. The 10 mm-o.d. sapphire tube
was purchased from Saphikon, Milford, NH, while the tita-
nium high-pressure charging head was constructed at ICCOM-
CNR (Firenze, Italy).15 Caution: Since high gas pressures are
involved, safety precautions must be taken at all stages of
studies involving high-pressure NMR tubes. IR spectra at
atmospheric pressure as well as at high pressure were recorded
with a FT-IR Perkin-Elmer BX spectrometer. The high-
pressure IR cell, constructed at ICCOM-CNR, is constituted
by a 75 mL autoclave equipped with ZnS windows (4 mm
thickness, 8 mm diameter, optical path length 0.2 mm).16 The
hydroformylation reactions were performed in a 125 mL
stainless steel autoclave, equipped with a magnetic drive
stirrer and a Parr 4842 temperature and pressure controller.
GC-MS analyses were carried out using a Shimadzu GCMS-
QP5050A with a SP capillary column (length, 30 m; diameter,
0.25 mm; 0.1 µm film thickness). The hydroformylation
products of 1-hexene were analyzed with a Shimadzu GC14
gas chromatograph coupled with a Shimadzu C-R4A computer
equipped with a packed column (length, 2 m, diameter, 3.17
mm) of the type PPG LB-550-X 15% (polypropylene glycol
(15%) supported on Chromosorb W) and a flame ionization
detector. The residual amount of 1-hexene, the isomerization
products of 1-hexene, and hexane were analyzed using a
Perkin-Elmer 8320 gas chromatograph equipped with a capil-
lary Chromopack column of Al2O3/Na2SO4 PLOT (length, 50
(2) (a) Ogasawara, M.; Yoshida, K.; Hayashi, T. Organometallics
2000, 19, 1567. (b) Bosch, B. E.; Bru¨mmer, I.; Kunz, K.; Erker, G.;
Fro¨lich, R.; Kotila, S. Organometallics 2000, 19, 1255. (c) Xie, Y.; Tan,
G. K.; Yan, Y. K.; Vittal, J. J.; Ng, S. C.; Hor, T. S. A. J. Chem. Soc.,
Dalton Trans. 1999, 773. (d) Fong, S.-W. A.; Hor, T. S. A. J. Cluster
Sci. 1998, 9, 351. (e) Hayashi, T.; Konishi, M.; Kobori, Y.; Kumada,
M.; Higuchi, T.; Hirotsu, K. J. Am. Chem. Soc. 1984, 106, 158.
(3) (a) Jutand, A.; Hii, K.-K. (Mimi); Thornton-Pett, M.; Brown, J.
M. Organometallics 1999, 18, 5367. (b) Boyes, A. L.; Butler, I. R.;
Quayle, S. C. Tetrahedron Lett. 1998, 39, 7763. (c) Brown, J. M.; Hii,
K.-K. (Mimi) Angew. Chem., Int. Ed. 1996, 35, 657.
(4) Ma¨gerlein, W.; Indolese, A. F.; Beller, M. Angew. Chem., Int. Ed.
2001, 40, 2856.
(5) (a) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2090. (b)
Hartwig, J. F. Acc. Chem. Res. 1998, 31, 852. (c) Driver, M. S.; Hartwig,
J. F. J. Am. Chem. Soc. 1996, 118, 7217.
m; diameter, 0.45 mm) and a flame ionization detector.
-
Synthesis of [Rh(dppf)(COD)]X (X ) BPh4-, 1a; PF6
,
1b). [RhCl(COD)]2 (10 mg, 0.0203 mmol) was dissolved in 5
mL of methanol, solid dppf (22.50 mg, 0.0406 mmol) was
added, and the resulting suspension was stirred for 1 h at room
temperature to obtain a clear solution. A dark yellow crystal-
line precipitate of 1a was obtained by adding NaBPh4 (15.28
mg, 0.0447 mmol), which was filtered off, washed three times
(6) (a) Li, K.; Horton, P. N.; Hursthouse, M. B.; Hii, K.-K. (Mimi) J.
Organomet. Chem. 2003, 665, 250. (b) Mu¨ller, T. E.; Berger, M.;
Grosche, M.; Herdtweck, E.; Schmidtchen, F. P. Organometallics 2001,
20, 4384.
(7) Mikami, K.; Aikawa, K. Org. Lett. 2002, 4, 99.
(8) (a) Bianchini, C.; Meli, A.; Oberhauser, W.; van Leeuwen, P. W.
N. M.; Zuideveld, M. A.; Freixa, Z.; Kamer, P. C. J.; Spek, A. L.; Gusev,
O. V.; Kal’sin, A. M. Organometallics 2003, 22, 2409. (b) Gusev, O. V.;
Kal’sin, A. M.; Peterleitner, M. G.; Petrovskii, P. V.; Lyssenko, K. A.;
Akhmedov, N. G.; Bianchini, C.; Meli, A.; Oberhauser, W. Organome-
tallics 2002, 21, 3637. (c) Gusev, O. V.; Kalsin, A. M.; Petrovskii, P.
V.; Lyssenko, K. A.; Oprunenko, Y. F.; Bianchini, C.; Meli, A.;
Oberhauser, W. Organometallics 2003, 22, 913.
(12) Giordano, G.; Crabtree, R. H. Inorg. Synth. 1979, 218.
(13) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992,
11, 3920.
(14) (a) Li, S.; Wei, B.; Low, P. M. N.; Lee, H. K.; Hor, T. S. A.; Xueb,
F.; Mak, T. C. W. J. Chem. Soc., Dalton Trans. 1997, 1289. (b) Hamann,
B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1998, 120, 3694 (c) Gusev, O.
V.; Peterleitner, M. G.; Kal’sin, A. M.; Vologdin, N. V. Russ. J.
Electrochem. 2003, 39, 1293. (d) Szymoniak, J.; Besanqon, J.; Dormond,
A.; Moise, C. J. Org. Chem. 1990, 55, 1429. (e) Trouve, G.; Broussier,
R.; Gautheron, B.; Kubicki, M. M. Acta Crystallogr. 1991, C47, 1966.
(15) Bianchini, C.; Meli, A.; Traversi, A. Ital. Pat. FI A000025, 1997.
(16) Heaton, B. Mechanisms in Homogeneous Catalysis, A Spectro-
scopic Approach; Wiley-VCH: Weinheim, 2004.
(9) Bianchini, C.; Meli, A.; Oberhauser, W.; Parisel, S.; Gusev, O.
V.; Kal’sin, A. M.; Vologdin, N. V.; Dolgushin, F. M. J. Mol. Catal. A
2004, 224, 35.
(10) Unruh, J. D.; Christenson J. R. J. Mol. Catal. 1982, 14, 19.
(11) Nettekoven, U.; Kamer, C. C. J.; Widhalm, M.; van Leeuwen,
P. W. N. M. Organometallics 2000, 19, 4596.

3694 Organometallics, Vol. 24, No. 15, 2005
Bianchini et al.
with MeOH (2 mL), and then dried under vacuum. Yield:
41.39 mg (94%). Anal. Calcd for C66H60BFeP2Rh: C, 73.08; H,
5.58. Found: C, 73.29; H, 5.75. Compound 1b was synthesized
by an identical procedure, using NH4(PF6) (7.25 mg, 0.0445
mmol) instead of NaBPh4. Yield: 33.27 mg (90%). Anal. Calcd
for C42H40FeF6P3Rh: C, 55.43; H, 4.39. Found: C, 55.29; H,
4.20.
NMR data for 1a: 1H NMR (299.94 MHz, CDCl3) δ 7.85-
7.77 (m, 8H, P(C6H5)2); 7.63-7.52 (m, 12H, P(C6H5)2); 7.50-
7.37 (m, 8H, o-B(C6H5)4); 7.10-6.90 (m, 8H, m-B(C6H5)4); 6.90-
6.80 (m, 4H, p-B(C6H5)4); 4.34 (m, 4H, CH(COD)); 4.29 (s, 4H,
R-H(Cp)); 4.19 (s, 4H, â-H(Cp)); 2.26-2.04 (s, 8H, CH2(COD));
31P{1H} NMR (121.42 MHz, CDCl3) δ 23.12 (d, 1J(RhP) ) 148.5
Hz).
1H NMR (299.94 MHz, CDCl3) δ 7.81-7.38 (m, 32H,
P(C6H5)2 + B(C6H3(m-CF3)2)4); 4.56 (m, 4H, CH(COD)), 2.40-
2.04 (m, 8H, CH2(COD), 1.63 (s, 12H, R-(CH3)Cp); 1.58 (s, 12H,
â-(CH3)Cp); 31P{1H} NMR (121.42 MHz, CDCl3) δ 23.22 (d,
1J(RhP) ) 145.8 Hz).
Synthesis of [Rh(o-iPr-dppf)(COD)]BAr′4 (5). [RhCl-
(COD)]2 (14.13 mg, 0.0287 mmol) was dissolved in 5 mL of
oxygen-free dichloromethane, solid o-iPr-dppf (41.37 mg, 0.0573
mmol) was added, and the solution was then stirred for 1 h at
room temperature. NaBAr′4 (50.79 mg, 0.0573 mmol) was
added, and stirring was continued for 1 h. The NaCl precipitate
was removed by filtration of the solution through Celite, and
the resulting solution was concentrated to dryness under
vacuum to give analytically pure 5. Yield: 64.40 mg (62%).
Anal. Calcd for C86H76F24FeP2Rh: C, 57.48; H, 4.26. Found:
C, 57.23; H, 4.12.
NMR data for 1b: 1H NMR (299.94 MHz, CDCl3) δ 7.83-
7.75 (m, 8H, P(C6H5)2); 7.60-7.51 (m, 12H, P(C6H5)2); 4.33 (m,
4H, CH(COD)); 4.30 (s, 4H, R-H(Cp)); 4.20 (s, 4H, â-H(Cp));
2.28-2.10 (s, 8H, CH2(COD)); 31P{1H} NMR (121.42 MHz,
1H NMR (299.94 MHz, CDCl3) δ 7.66-6.83 (m, 32H,
P(C6H5)2 + B(C6H3(m-CF3)2)4); 4.33 (s, 4H, R-H(Cp)) 4.20 (m,
4H, CH(COD)), 4.07 (s, 4H, â-H(Cp)), 3.86 (m, 4H, CH(CH3)2),
1
CDCl3) δ 23.22 (d, J(RhP) ) 148.0 Hz).
3
1.48-1.81 (m, 8H, CH2(COD), 1.29 (d, J(HH) ) 6.4 Hz, 6H,
Synthesis of [Rh(dppr)(COD)]BPh4 (2). [RhCl(COD)]2
(10 mg, 0.0203 mmol) was dissolved in 5 mL of methanol, dppr
(24.34 mg, 0.0406 mmol) was added, and the resulting mixture
was stirred for 1 h at room temperature. The addition of
NaBPh4 (15.28 mg, 0.0447 mmol) caused the precipitation of
a dark yellow precipitate, which was filtered off, washed
several times with 2 mL portions of MeOH, and dried under
vacuum. Yield: 43.40 mg (94%). Anal. Calcd for C66H60BP2-
RhRu: C, 70.16; H, 5.35. Found: C, 69.80; H, 5.20.
(CH)CH3)2); 0.91 (d, 3J(HH) ) 6.4 Hz, 6H, (CH)CH3)2); 31P-
1
{1H} NMR (121.42 MHz, CDCl3) δ 29.97 (d, J(RhP) ) 150.1
Hz).
Synthesis of [Rh(CO)2(dppf)]PF6 (6) and [Rh(CO)2-
(dppo)]PF6 (8). Both these complexes were obtained through
an identical procedure; therefore only one reaction is described
here. In a Schlenk tube containing 5 mL of oxygen-free CH2-
Cl2 was dissolved 1b (3b) (0.0333 mmol). Then CO was bubbled
through the solution for 5 min. The orange-red solution of 6
(or the yellow solution of 8) was concentrated to dryness under
vacuum, and the solid was suspended in n-pentane and stirred
at room temperature for 10 min. Analytically pure 6 (8) was
filtered off and dried under a slight stream of nitrogen. Yield
of 6: 24.06 mg (85%). Anal. Calcd for C36H28F6FeO2P3Rh: C,
50.38; H, 3.29. Found: C, 50.60; H, 3.60. Yield of 8: 26.6 mg
(80%). Anal. Calcd for C36H28F6O2OsP3Rh: C, 43.57; H, 2.82.
Found: C, 43.86; H, 2.95.
1H NMR (299.94 MHz, CDCl3) δ 7.83-7.75 (m, 8H, P(C6H5)2);
7.61-7.49 (m, 12H, P(C6H5)2); 7.45-7.37 (m, 8H, o-B(C6H5)4);
7.05-6.98 (m, 8H, m-B(C6H5)4); 6.90-6.82 (m, 4H, p-B(C6H5)4);
4.72 (s, 4H, R-H(Cp)); 4.51 (s, 4H, â-H(Cp)); 4.24 (m, 4H,
CH(COD)); 2.37-1.96 (s, 8H, CH2(COD)); 31P{1H} NMR (121.42
1
MHz, CDCl3) δ 20.48 (d, J(RhP) ) 149.8 Hz).
Synthesis of [Rh(dppo)(COD)]X (X ) BPh4-, 3a; PF6
-
,
3b). [RhCl(COD)]2 (10 mg, 0.0203 mmol) was dissolved in 5
mL of methanol, solid dppo (27.96 mg, 0.0406 mmol) was
added, and the resulting reaction mixture was stirred for 1 h
at room temperature. The addition of NaBPh4 (15.28 mg,
0.0447 mmol) caused the precipitation of a dark yellow
precipitate of 3a, which was filtered off, washed several times
with 2 mL of methanol, and dried under vacuum. Yield: 44.89
mg (90%). Anal. Calcd for C66H60BP2RhOs: C, 65.03; H, 4.96.
Found: C, 64.80; H, 4.83. Compound 3b was synthesized
analogously, using NH4(PF6) instead of NaBPh4. Yield of 3b:
38.60 mg (91%). Anal. Calcd for C42H40F6P3OsRh: C, 48.30;
H, 3.83 Found: C, 48.20; H, 3.73.
NMR data for 3a: 1H NMR (299.94 MHz, CDCl3) δ 8.02-
7.96 (m, 8H, P(C6H5)2); 7.78-7.56 (m, 12H, P(C6H5)2); 7.46-
7.34 (m, 8H, o-B(C6H5)4); 7.25-7.18 (m, 8H, m-B(C6H5)4); 7.08-
6.03 (m, 4H, p-B(C6H5)4); 5.16 (s, 4H, R-H(Cp)); 4.83 (s, 4H,
â-H(Cp)); 4.43 (m, 4H, CH(COD)); 2.37-2.12 (s, 8H, CH2-
(COD)); 31P{1H} NMR (121.42 MHz, CDCl3) δ 24.15 (d,
1J(RhP) ) 149.7 Hz);
NMR and IR data for 6: 1H NMR (299.94 MHz, CD2Cl2) δ
7.80-7.40 (m, 20H, P(C6H5)2); 4.59 (s, 4H, R-H(Cp)), 4.28 (s,
4H, â-H(Cp)); 13C{1H} NMR (100.57 MHz, CD2Cl2) δ 181.90
(br s, CO), 133.57-129.50 (P(C6H5), 76.51 (t, 2J(CP) ) 6.0 Hz,
R-C(Cp)), 74.66 (t, 3J(CP) ) 4.0 Hz, â-C(Cp)), 71.20 (d,
1J(PC) ) 64 Hz, ipso-C(Cp)); 31P{1H} NMR (121.42 MHz, CD2-
1
Cl2) δ 24.97 (d, J(RhP) ) 127.3 Hz); IR (KBr) 2094 (s), 2048
(s) cm-1; IR (THF) 2098 (s), 2053 (s) cm-1
.
NMR and IR data for 8: 1H NMR (299.94 MHz, CD2Cl2) δ
4
7.75-7.50 (m, 20H, P(C6H5)2); 5.51 (t, J(PH) ) 1.7 Hz, 4H,
â-H(Cp)), 4.33 (t, 3J(PH) ) 1.8 Hz, 4H, R-H(Cp)); 13C{1H} NMR
(100.57 MHz, CD2Cl2) δ 189.00 (br, m, CO), 134.36-127.98
(m, P(C6H5)2, 73.45 (s, â-C(Cp)), 68.09 (t, 2J(CP) ) 5.2 Hz,
R-C(Cp)), 68.0 (d, 1J(CP) ) 64.9 Hz, ipso-C(Cp); 31P{1H} NMR
(121.42 MHz, CD2Cl2) δ 5.17 (d, 1J(RhP) ) 90.1 Hz); IR (KBr)
2023 (s), 1971 (s) cm-1; IR (THF) 2027 (s), 1982 (s) cm-1
.
In Situ Synthesis of [Rh(CO)2(dppr)]BPh4 (7), [Rh-
(CO) (dppomf)]BAr′4 (9), and [Rh(CO)2(o-iPr-dppf)]BAr′4
(10). 2In a Schlenk tube containing 0.9 mL of oxygen-free CD2-
Cl2 was dissolved 2, 4, or 5 (0.03 mmol). After the tube was
cooled to 0 °C, CO was bubbled into the CD2Cl2 solution for 5
min. The resulting solution was transferred into a 5 mm NMR
tube for product characterization. All our attempts to isolate
7, 9, or 10 in the solid state were unsuccessful due to their
fast decomposition in the absence of a protective CO atmo-
sphere.
NMR and IR data for 7: 1H NMR (299.94 MHz, CD2Cl2) δ
7.83-7.40 (m, 20H, P(C6H5)2), 7.40-6.80 (m, 20H, B(C6H5)4);
5.26 (s, 4H, â-H(Cp)), 4.06 (s, 4H, R-H(Cp)); 13C{1H} NMR
(100.57 MHz, CD2Cl2) δ 188.00 (br, m, CO), 164.03 (q, 1J(CB)
NMR data for 3b: 1H NMR (299.94 MHz, CDCl3) δ 8.02-
7.96 (m, 8H, P(C6H5)2); 7.78-7.56 (m, 12H, P(C6H5)2); 5.16 (s,
4H, R-H(Cp)); 4.83 (s, 4H, â-H(Cp)); 4.43 (m, 4H, CH(COD));
2.37-2.12 (s, 8H, CH2(COD)); 31P{1H} NMR (121.42 MHz,
1
CDCl3) δ 24.23 (d, J(RhP) ) 149.1 Hz).
Synthesis of [Rh(dppomf)(COD)]BAr′4 (4). [RhCl(COD)]2
(20 mg, 0.0406 mmol) was dissolved in 5 mL of degassed
dichloromethane, solid dppomf (54.0 mg, 0.081 mmol) was
added, and the solution was stirred for 1 h at room tempera-
ture. Solid NaBAr′4 (71.75 mg, 0.081 mmol) was added, and
stirring was continued for another hour. The NaCl precipitate
was removed by filtration through Celite. The dark yellow
solution was concentrated to dryness under vacuum to give
analytically pure 4. Yield: 104 mg (74%). Anal. Calcd for
C82H68BF24FeP2Rh: C, 56.57; H, 3.94. Found: C, 56.33; H,
3.85.
) 49.2 Hz, ipso-C(B(C6H5)4), 135.39-121.70 (P(C6H5)2
+
B(C6H5)4), 79.66 (s, â-C(Cp)), 75.44 (t, 2J(CP) ) 5.3 Hz,
R-C(Cp)), 68.0 (d, 3J(CP) ) 64.9 Hz, ipso-C(Cp); 31P{1H} NMR

Hydroformylation of 1-Hexene by Rh(I) Catalysts
Organometallics, Vol. 24, No. 15, 2005 3695
(121.42 MHz, CD2Cl2) δ 11.31 (d, 1J(RhP) ) 90.3 Hz); IR (THF)
of 6 and 15 could be detected. Heating the TDF solution to
higher temperature until 70 °C for 1 h gave the conversion of
15 into the known compound 11. As compound 15 was unstable
in the absence of syngas, only the significant chemical shifts
are reported for this compound. An analogous HP-NMR
experiment in CD2Cl2 exhibited only compound 6. Neither 15
nor 11 was observed, even after prolonged heating at 70 °C.
An analogous HP-IR experiment in THF as well as in CH2Cl2
gave no hint for the formation of 15, as in THF only 11 was
observed pressurizing the THF solution of 6 with CO/H2 (15/
30 bar) at room temperature, while in CH2Cl2 only compound
6 was observed, even at 70 °C. Due to the overlapping of the
1H NMR signals in the phenyl region of compounds 6, 11, and
15, no reliable phenyl signals for 15 can be reported.
2040 (m), 1991 (s) cm-1
.
NMR and IR data for 9: 1H NMR (299.94 MHz, CD2Cl2) δ
7.90-7.40 (m, 32H, P(C6H5)2 + BAr′4); 1.63 (s, 12H, R-CH3-
(Cp)), 1.37 (s, 12H, â-CH3(Cp)); 13C{1H} NMR (100.57 MHz,
CD2Cl2) δ 182.0 (br m, CO), 162.10 (m, ipso-C(B(Ar′)4, 135.37,
132.28, 129.0 (P(C6H5)), 134.80 (s, o-C(Ar′)), 124.58 (q,
1J(CF) ) 272.4 Hz, CF3), 117.46 (s, p-C(Ar′)), 86.74 (s, R-C(Cp)),
86.11 (s, â-C(Cp)), 72.24 (d, 1J(PC) ) 57.8 Hz, ipso-C(Cp)),
12.30 (s, R-(CH3)Cp), 8.32 (s, â-(CH3)Cp); 31P{1H} NMR (121.42
MHz, CD2Cl2) δ 30.53 (d, 1J(RhP) ) 132.1 Hz); IR (THF) 2094
(s), 2044 (s) cm-1
.
NMR and IR data for 10: 1H NMR (299.94 MHz, CD2Cl2) δ
8.10-7.40 (m, 32H, P(C6H5)2 + BAr′4); 5.26 (br s, 4H, â-H(Cp)),
4.04 (br s, 4H, R-H(Cp)), 3.38 (m, 4H, CH(CH3)2), 0.88 (br s,
24H, (CH)CH3)2); 13C{1H} NMR (100.57 MHz, CD2Cl2) δ 190.20
(br m, CO), 162.20 (m, ipso-C(B(Ar′)4, 153.09 (s, C(CH(CH3)),
134.80 (s, o-C(Ar′)), 133.27, 128.36, 126.83 (P(C6H5)), 124.58
NMR data for 15: 1H NMR (200.13 MHz, TDF) δ 5.10 (br
s, â-H(Cp)), 4.25 (br s, R-H(Cp)); 31P{1H} NMR (81.01 MHz,
1
TDF) δ 24.72 (d, J(RhP) ) 91.6 Hz).
Synthesis of RhH(CO)2(dppomf) (14). Compound 4 (20
mg, 0.011 mmol) and NEt3 (3.2 µL, 0.021 mmol) were dissolved
in 1.8 mL of degassed THF-d8. This mixture was transferred
under nitrogen into a 10 mm-o.d. sapphire tube. The tube was
pressurized with CO/H2 (1:2) up to a total pressure of 45 bar
and then introduced into the NMR probe at room temperature.
1
(q, J(CF) ) 272.4 Hz, CF3), 117.46 (s, p-C(Ar′)), 77.82 (br s,
â-C(Cp)), 72.56 (br s, R-C(Cp)) + ipso-C(Cp), 32.49 (s, CH-
(CH3)2), 22.92 (s, CH(CH3)2), 22.57 (s, CH(CH3)2); 31P{1H} NMR
(121.42 MHz, CD2Cl2) δ 7.14 (d, 1J(RhP) ) 89.1 Hz); IR (THF)
2025 (m), 1980 (s) cm-1
.
1
The conversion of 4 into 14 was followed by 31P{1H} and H
In Situ Synthesis of RhH(CO)2(dppf) (11), RhH(CO)2-
(dppr) (12), and RhH(CO)2(dppo) (13). Compound 1b, 2,
or 3b (0.018 mmol) was dissolved in 1.8 mL of degassed THF-
d8, and the solution was transferred under nitrogen into a 10
mm-o.d. sapphire tube. The tube was pressurized with syngas
(CO/H2, 1:2) to a total pressure of 45 bar and then introduced
into the NMR probe at room temperature. 31P{1H} NMR
spectra acquired at different temperatures in the interval from
20 to 60 °C showed 1b and 2 to convert completely into 11
and 12, respectively, within 1 h, while 3b converted into 13
only after heating at 60 °C for 3 h. NMR data for 11,11 12,
and 13 are given at room temperature. Our attempts to isolate
11, 12, and 13 in the solid state were unsuccessful due to their
instability in the absence of CO and H2. The IR absorption
spectra for 11, 12, and 13 were acquired at room temperature
using a high-pressure IR cell pressurized at 45 bar of CO/H2
(1:2).
NMR at room temperature. The conversion was complete after
15 min. The IR spectrum of 14 was acquired at room temper-
ature after pressurizing the IR cell, containing a solution of 4
(20 mg, 0.011 mmol) and 3.2 µL (0.021 mmol) of NEt3 in 24
mL of THF, up to 45 bar of CO/H2 (1:2).
1H NMR (200.13 MHz, TDF) δ 1.77 (s, 12H, R-CH3(Cp)), 1.47
1
2
(s, 12H, â-CH3(Cp)), -10.09 (td, J(RhH) ) 9.6 Hz, J(PH) )
42.9 Hz, 1H, Rh-H); 31P{1H} NMR (81.01 MHz, TDF) δ 30.02
(d, J(RhP) ) 125.0 Hz); IR (THF) 2063 (w, ee), 1984 (m, ee
1
and ea), 1941 (s, ea) cm-1
.
Batch Hydroformylation Reactions. Typically, a THF
solution (25 mL) containing 2 µmol of the precatalyst and 2
mmol of 1-hexene was introduced by suction into the autoclave,
previously evacuated by a vacuum pump. The autoclave was
then pressurized at room temperature with CO and H2 in a
1:2 ratio to a total pressure of 45 bar, stirred at 300 rpm, and
then heated to the desired temperature. After the desired
reaction time, the autoclave was cooled to room temperature
and the gases were vented. The composition of the solution
was analyzed by GC and GC-MS.
NMR and IR data for 11: 1H NMR (200.13 MHz, TDF) δ
8.11-7.20 (m, 20H, P(C6H5)2), 4.43 (s, 4H, R-H(Cp)), 4.16 (s,
4H, â-H(Cp)), -9.45 (td, 1J(RhH) ) 9.8 Hz, 2J(PH) ) 42.9 Hz,
1H, Rh-H); 31P{1H} NMR (81.01 MHz, TDF) δ 31.10 (d,
1J(RhP) ) 121 Hz); IR (THF): 2036 (w, ee), 1989 (m, ee and
ea), 1945 (s, ea) cm-1 (ee: bis-equatorial coordination mode,
ea: equatorial-axial coordination mode).
High-Pressure NMR Study in THF-d8 of 1a, 2, 3a, 4,
and 5 under Syngas Either in the Presence or Absence
of 1-Hexene. All high-pressure NMR experiments were
performed analogously; therefore only a typical procedure is
described here. A 10 mm-o.d. sapphire tube was charged under
nitrogen with 1.8 mL of a degassed THF-d8 solution of each
rhodium complex (0.0125 mmol). Then 1-hexene (156 µL, 1.25
mmol) was syringed into the NMR tube. 31P{1H} and 1H NMR
spectra were acquired at room temperature (20 °C). Afterward,
the sapphire tube was removed from the NMR probe and
charged with CO/H2 (1:2) to a total pressure of 45 bar. The
sapphire tube was placed again into the NMR probe and
31P{1H} NMR and 1H NMR spectra were recorded every 10
°C from room temperature to 60 °C. After heating the solution
for 1 h at 60 °C, the NMR probe was cooled to room temper-
NMR and IR data for 12: 1H NMR (200.13 MHz, TDF) δ
8.11-7.20 (m, 20H, P(C6H5)2), 4.52 (s, 4H, R-H(Cp)), 4.26 (s,
1
2
4H, â-H(Cp)), -9.60 (td, J(RhH) ) 7.8 Hz, J(PH) ) 30 Hz,
1H, Rh-H); 31P{1H} NMR (81.01 MHz, TDF) δ 30.5 (d,
1J(RhP) ) 128 Hz); IR (THF) 2038 (w, ee), 1990 (s, ee and ea),
1943 (s, ea) cm-1
.
NMR and IR data for 13: 1H NMR (200.13 MHz, TDF)
(phenyl-H difficult to assign, overlapped with BAr′4-H), δ 5.10
1
(s, 4H, R-H(Cp)), 4.90 (s, 4H, â-H(Cp)), -9.72 (td, J(RhH) )
8.7 Hz, 2J(PH) ) 28.2 Hz, 1H, Rh-H); 31P{1H} NMR (81.01
1
MHz, TDF) δ 35.20 (d, J(RhP) ) 128.2 Hz); IR (THF) 2036
(w, ee), 1992 (m, ee and ea), 1942 (s, ea) cm-1
.
1
ature and 31P{1H} and H NMR spectra were recorded.
In Situ Reaction of 6 with Syngas in THF-d8 and in
CD2Cl2. A 10 mm-o.d. sapphire tube was charged first with
10.7 mg (0.0125 mmol) of compound 6 and dissolved in 1.8
mL of degassed TDF and then with 15 bar of CO. Afterward
it was placed into the NMR probe at room temperature, and
31P{1H} NMR and 1H NMR spectra were recorded. Only
compound 6 was observed by 31P{1H} NMR spectroscopy. The
NMR tube was removed from the probe and pressurized with
Analogous experiments were carried out in the absence of
1-hexene, and no substantial difference was noticed in the
1
31P{1H} and H NMR spectra.
High-Pressure IR Study in THF of 1a, 2, 3a, 4, and 5
under Syngas Either in the Presence or Absence of
1-Hexene. All high-pressure IR measurements were per-
formed analogously; therefore only a typical procedure is
described here. Compound 1a, 2, 3a, 4, or 5 (0.0315 mmol)
was dissolved in 24 mL of oxygen-free THF. Then 1-hexene
(390 µL, 3.15 mmol) was added, and the solution introduced
into the high-pressure IR cell under nitrogen. The IR cell was
1
30 bar of hydrogen. Immediately after, 31P{1H} NMR and H
NMR spectra at room temperature were acquired. The conver-
sion of 6 into the as yet unidentified compound 15 started,
and after 1 h at room temperature nearly equal concentrations

3696 Organometallics, Vol. 24, No. 15, 2005
Bianchini et al.
Table 1. Experimental X-ray Diffraction
Parameters and Crystal Data of 6 and 8
Chart 2
6
8
empirical formula
fw
C
36H28F6FeO2P3Rh C36H28F6O2OsP3Rh
858.25 992.60
cryst size (mm)
cryst syst
space group
a (Å)
0.40 × 0.35 × 0.22 0.35 × 0.10 × 0.05
monoclinic
P21/a
triclinic
P1h
9.662(2)
36.711(10)
9.971(2)
90
92.13(2)
90
15.524(1)
11.609(2)
9.860(6)
87.74(3)
81.50(2)
93.50(1)
1752.1(1)
1.881
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
3534.2(1)
1.613
less favorable reflections to parameters ratio. Moreover in 8
the significant decay of the crystal has to be pointed out, which
is responsible for the somewhat poor data. In both structures
hydrogen atoms were introduced in their calculated positions,
with thermal parameters 20% larger than those of the respec-
tive carbon atoms. The function minimized during the refine-
Dcalcd (Mg/m3)
Z
4
2
µ(Mo KR) (mm-1
F(000)
)
1.077
4.295
1720
960
diffractometer
radiation
temp (K)
θ range for data
collection (deg)
index range
Enraf Nonius CAD4, λ(Mo KR) )
0.71073 Å
ment was ∑w(Fo - Fc2)2, where w is respectively defined as
2
293(2)
2.04-24.97
293(2)
2.64-20.02
1/[σ2(Fo ) + (0.0580P)2 + 3.02P] (6) and 1/[σ2(Fo ) + (0.0218P)2
2
2
+ 82.41P] (8), where P ) (max(Fo ) + 2Fc2)/3. All the calcula-
2
tions were performed on a PC using the WINGX package20a
with SIR-97,20b SHELX-97,20c and ORTEP-3 programs.20d
Crystallographic data for the structures (excluding structure
factors) reported in this paper have been deposited at the
Cambridge Crystallographic Data Center: CCDC-266231 for
6, CCDC-266232 for 8. Copies of the data can be obtained free
of charge on application to The Director, CCDC, 12 Union
Road, Cambridge, CB21EZ, U.K. (fax, internat. +(1223)
336-033; e-mail, deposit@ccdc.cam.ac.uk; web, //www.ccdc.cam.
ac.uk.
-11 e h e 11
0 e k e 43
0 e l e 11
-14 e h e 14
-11 e k e 11
0 e l e 9
3597
no. of reflns collected 6787
no. of indep reflns 6192
3261
no. of refined params 442
322
R1 (2σ(I))
0.0420
0.0673
R1 (all data)
0.0655
0.1086
0.963
0.0993
wR2 (all data)
0.1574
goodness of fit on F2
1.006
largest diff peak and 0.727/-0.577
0.944/-0.968
hole (e/Å3)
Results and Discussion
pressurized at room temperature with CO/H2 (1:2) to a total
pressure of 45 bar and then placed into the spectrometer.
Spectra were acquired at room temperature and at 60 °C under
stirring. After 1 h at 60 °C, the spectra showed no significant
change in the 4000-750 cm-1 region. Afterward the autoclave
was cooled to room temperature and the gases were vented
out. The GC-MS analysis of the solution showed the presence
of COD and the formation of aldehydes with the same
regioselectivity observed in the corresponding batch reactions.
Analogous HP-IR experiments, carried out without added
1-hexene, gave identical results.
X-ray Structure Determination. Single crystals of 6 and
8 were obtained by diffusion of n-hexane in a CH2Cl2 solution
of 6 and 8 at room temperature. Diffraction data of 6 and 8
were collected at room temperature on an Enraf Nonius CAD4
automatic diffractometer with Mo KR radiation (graphite
monochromator). Unit cell parameters of both structures were
determined from a least-squares refinement of the setting
angles of 25 carefully centered reflections. Crystal data and
data collection details are given in Table 1. The intensities
were rescaled, and a standard deviation σ(I) was calculated
using the value of 0.03 for the instability factor k.17 Lorentz-
polarization and absorption corrections were applied.18 Atomic
scattering factors were taken from ref 19a, and an anomalous
dispersion correction, real and imaginary part, was applied.19b
The structures were solved by direct methods and refined by
full-matrix F2 refinement. Whereas in 6 anisotropic thermal
parameters were assigned to all non-hydrogen atoms, in 8 the
phenyl carbon atoms were refined isotropically, owing to the
Synthesis and Characterization of the Rhodium
Complexes Investigated in this Study. The COD
catalyst precursors 1a,b, 2, 3a,b, 4, and 5 (Chart 2) were
synthesized through two slightly different protocols. In
the first procedure, a stoichiometric amount of the
desired 1,1′-bis(diarlphosphino)metallocene ligand was
added to a MeOH solution of [RhCl(COD)]2. The further
addition of NaBPh4 or NH4(PF6) allowed the isolation
-
of the complex cation [Rh(P-P)(COD)]+ as BPh4 or
PF6 salt. This procedure could be applied only to the
-
dppf, dppr, and dppo derivatives, as these diphosphine
ligands show a low solubility in MeOH. In the case of
dppomf and o-iPr-dppf, which are insoluble in methanol,
both [RhCl(COD)]2 and the diphosphine ligand were
dissolved in dichloromethane, and the precipitation of
the cationic metal complexes was achieved by adding
NaBAr′4. Unambiguous characterization of the [Rh(P-
P)(COD)]X products was obtained by multinuclear NMR
spectroscopy, IR spectroscopy, and elemental analysis.
It is worth mentioning that all complexes exhibit a
Rh-P coupling constant of ca. 150 Hz, which is typical
for square-planar rhodium diphosphine complexes with
a P-Rh-P bite angle close to 90°.21
(20) (a) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837. (b)
Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo,
C.; Guagliardi, A.; Moliterni, G. G.; Polidori, G.; Spagna, R. J. Appl.
Crystallogr. 1999, 32, 115. (c) Sheldrick, G. M. SHELX-97; University
of Go¨ttingen: Go¨ttingen, Germany, 1997. (d) Burnett, M. N.; Johnson,
C. K. ORTEP-3; Report ORNL-6895; Oak Ridge National Laboratory:
Oak Ridge, TN, 1996.
(17) Corfield, P. W. R.; Doedens, R. J.; Ibers, J. A. Inorg. Chem. 1967,
6, 197.
(18) Parkin, S.; Moezzi, B.; Hope, H. J. Appl. Crystallogr. 1995, 28,
53.
(19) (a) Wilson, J. C. In International Tables for X-ray Crystal-
lography; Kluwer: Dordrecht, 1992; p 500. (b) Wilson, J. C. In
International Tables for X-ray Crystallography; Kluwer: Dordrecht,
1992; p 219.
(21) (a) Anderson, M. P.; Pignolet, L. H. Inorg. Chem. 1981, 20, 4101.
(b) Bakos,. J.; To´th, I.; Heil, B.; Szalontai, G.; Pa`rka`nyi, L.; Fu¨lo¨p, V.
J. Organomet. Chem. 1989, 370, 263. (c) Kalck, P.; Randrianalimanana,
C.; Ridmy, M.; Thorez, A. New J. Chem. 1988, 12, 679.

Hydroformylation of 1-Hexene by Rh(I) Catalysts
Organometallics, Vol. 24, No. 15, 2005 3697
Table 2. Selected Bond Lengths [Å] and Angles
[deg] for 6 and 8a
Scheme 1
6
8
Rh(1)-P(1)
2.357(1)
2.371(1)
1.888(5)
1.893(5)
2.035(12)
1.639(5)
2.038(7)
1.644(5)
2.286(6)
2.283(6)
1.74(3)
1.81(4)
Rh(1)-P(2)
Rh(1)-C(1)
Rh(1)-C(2)
Fe(1)-CCp1 (av)
Fe(1)-Cp1centroid
Fe(1)-CCp2 (av)
Fe(1)-Cp2centroid
Os(1)-CCp1 (av)
Os(1)-Cp1centroid
Os(1)-CCp2 (av)
Os(1)-Cp2centroid
O(1)-C(1)
2.20(2)
1.82(2)
2.19(2)
1.84(2)
1.20(3)
1.16(3)
1.123(6)
1.134(6)
4.234(1)
O(2)-C(2)
Rh(1)‚‚‚Fe(1)
Since the HP-NMR experiments at room temperature
showed the formation of dicarbonyl rhodium(I) com-
plexes at the early stage of the hydroformylation reac-
tions (vide infra), 1a, 2, 3b, 4, and 5 were independently
reacted with CO in CH2Cl2 to see whether the dicar-
bonyl complexes could be isolated for authentication.
Irrespective of the CO pressure (1-15 bar), mono-
cationic dicarbonyl complexes of the formula [Rh(CO)2-
(P-P)]+ were selectively obtained (Scheme 1).
Rh(1)‚‚‚Os(1)
3.139(2)
155.9(2)
123.8(18)
94.9(8)
92.4(11)
97.8(9)
97.4(10)
P(1)-Rh(1)-P(2)
C(1)-Rh(1)-C(2)
C(1)-Rh(1)-P(1)
C(2)-Rh(1)-P(1)
C(1)-Rh(1)-P(2)
C(2)-Rh(1)-P(2)
Cp1centroid-Fe(1)-Cp2centroid
Cp1centroid-Os(1)-Cp2centroid
O(1)-C(1)-Rh(1)
O(2)-C(2)-Rh(1)
95.96(4)
87.4(2)
88.53(15)
175.49(15)
171.7(2)
88.30(15)
179.1(2)
165.9(9)
167(3)
179(3)
175.3(6)
176.8(4)
However, only two of them, namely, [Rh(CO)2(dppf)]-
PF6 (6) and [Rh(CO)2(dppo)]PF6 (8), could be isolated,
while [Rh(CO)2(dppr)]PF6 (7), [Rh(CO)2(dppomf)]PF6 (9),
and [Rh(CO)2(o-iPr-dppf)]PF6 (10) were stable only in
the presence of a protective CO atmosphere. The
characterization of the latter complexes was therefore
achieved by in situ HP-NMR and HP-IR techniques.
The dicarbonyl complexes exhibit two different coor-
dination geometries: the dppf complex 6 and the dppomf
complex 9 are square planar with cis CO ligands, while
the dppr, dppo, and o-iPr-dppf complexes 7, 8, and 10
exhibit a distorted trigonal-bipyramidal coordination
geometry due to an actual weak bonding interaction
between the sandwiched metal and rhodium (Scheme
1). Distinctive features for a square-planar coordination
about the rhodium center in 6 and 9 are a 31P{1H} NMR
Rh-P coupling constant of 127 and 132 Hz, respectively,
and two strong IR absorption bands, in THF solution,
of almost equal intensity at 2098 and 2053 cm-1 for 6
and 2094 and 2044 cm-1 for 9.22 Moreover, small ∆δ
values separate the Cp R and â hydrogen atoms in 6
(0.42 ppm) as well as the Cp methyl groups in 9 (0.26
ppm), which indicates the absence of an intramolecular
metal-rhodium bond.8b,23
a Cp1: C(51)-C(55); Cp2: C(61)-C(65).
larger dihedral angle between the two Cp planes
increases the difference of the chemical shift of the R
and â hydrogen atoms of the Cp rings.8a,c,23 Since the
trigonal-bipyramidal complexes 7, 8, and 10 are stereo-
chemically rigid in solution, the observed ν(CO) bands
(2020-2040 and 1990-1980 cm-1) may be safely as-
signed to equatorial cis CO ligands.24
The formation of κ3-P,P,M 1,1′-bis(diarylphosphino)-
metallocene complexes, as in 7, 8, and 10, is controlled
by both steric and electronic factors.8a,c,23 In particular,
the κ3-P,P,M bonding mode is promoted by the presence
of alkyl substituents on either Cp or P-aryl groups, while
the strength of the metal-metal dative bond increases
in the order Fe < Ru < Os, due to the increasing
diffusion of the metal-centered e2g electron density down
to the iron triad.23a-c
A square-planar geometry for 6 and a trigonal-
bipyramidal geometry for 8 were unequivocally con-
firmed also in the solid state by single-crystal X-ray
analyses. ORTEP plots of the complex cations [Rh(CO)2-
(dppf)]+ and [Rh(CO)2(dppo)]+ are reported in Figure 1.
Crystallographic data for both compounds and selected
bond lengths and angles are reported in Table 1 and
Table 2, respectively.
In 6, the rhodium center is square-planarly coordi-
nated by a chelating dppf ligand and by two terminal
carbonyl groups. Deviations from the ideal geometry are
put in evidence by the trans angles P(1)-Rh(1)-C(2)
and P(2)-Rh(1)-C(1) of 175.5(2)° and 171.7(2)°, respec-
tively, with the rhodium atom displaced by 0.049 Å from
the least-squares plane passing through the P and the
CCO donors. The ferrocenyl moiety, with the Fe-
Cpcentroid distances averaging 1.642(3) Å and the
Cp1centroid-Fe(1)-Cp2centroid of 179.1(2)°, adopts an ar-
A κ3-P,P,M ligand coordination in 7, 8, and 10 was
readily inferred by a significant decrease of the value
of the Rh-P coupling constant (ca. 90 Hz) as well as by
1H NMR ∆δ values, relative to the R and â hydrogen
atoms of the Cp rings, of ca. 1.2 ppm. Indeed, in a
trigonal-bipyramidal structure with trans P atoms, the
(22) (a) Betley, Th. A.; Peters, J. C. Angew. Chem., Int. Ed. 2003,
42, 2385. (b) Fairle, D. P.; Bosnich, B. Organometallics 1988, 7, 946.
(23) (a) Sato, M.; Shigeta, H.; Sekino, M.; Akabori, S. J. Organomet.
Chem. 1993, 458, 199. (b) Zuideveld, M. A.; Swennenhuis, B. H. G.;
Boele, M. D. K.; Guari, Y.; van Strijdonck, G. P. F.; Reek, J. N. H.;
Kamer, P. C. J.; Goubitz, K.; Fraanje, J.; Lutz, M.; Spek, A. L.; van
Leeuwen, P. W. N. M. J. Chem. Soc., Dalton Trans. 2002, 2308. (c)
Zuideveld, M. A.; Swennenhuis, B. H. G.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M. J. Organomet. Chem. 2001, 637, 805. (d) Akabori, S.;
Kumagai, T.; Shirahige, T.; Sato, S.; Kawazoe, K.; Tamura, C.; Sato,
M. Organometallics 1987, 6, 526. (e) Akabori, S.; Kumagai, T.;
Shirahige, T.; Sato, S.; Kawazoe, K.; Tamura, C.; Sato, M. Organome-
tallics 1987, 6, 2105.
(24) (a) Dixon, F. M.; Eisenberg, A. H.; Farrell, J. R.; Mirkin, C. A.;
Liable-Sands, L. M.; Rheingold, A. L. Inorg. Chem. 2000, 39, 3432. (b)
Siegl, W. O.; Lapporte, S. J.; Collman, J. P. Inorg. Chem. 1971, 10,
2158.

3698 Organometallics, Vol. 24, No. 15, 2005
Bianchini et al.
Figure 1. ORTEP plots of 6 (left) and 8 (right). Thermal ellipsoids are shown at the 30% probability level. The PF6
anions have been omitted for clarity.
Table 3. Hydroformylation of 1-Hexene Using [Rh(P-P)(COD)]X Complexesa (P-P:
1,1′-bis(diarylphosphino)metallocene)
precatalyst/
ligand
time
(h)
conversion
(%)
hydroformylation
products (%)
selectivityc
hexane
(%)
1-heptanol
(%)
isomerization
(%)
run
TOFb
(%)
1
2
1a/dppf
1
2
1
2
1
2
1
1
1
2
68.3
75.1
65.2
93.2
32.8
98.9
98.8
98.9
71.9
76.4
683
376
652
466
328
495
988
988
719
382
52.1
59.5
49.1
71.7
20.7
68.4
67.1
63.4
52.9
56.4
67.1
73.5
69.7
72.0
69.4
72.3
63.5
62.8
67.2
72.5
0
0
16.2
15.0
16.0
20.7
8.9
25.6
28.4
30.2
18.9
17.6
1a/dppf
0.6
0.1
0.8
3.3
3.7
0.8
0.9
0
0
3
2/dppr
0
4
2/dppr
0
5
3a/dppo
0
6
3a/dppo
1.1
2.5
4.3
0
7
4/dppomf
[RhCl(COD)]2
5/o-iPr-dppf
5/o-iPr-dppf
8
9
10
0.7
1.7
a Reaction conditions: precatalyst 2 µmol; substrate 2 mmol; 25 mL THF; 45 bar CO/H2 (1:2) at 20 °C; reaction temperature 60 °C;
stirring rate 300 rpm. b Expressed as (mol of aldehyde) (mol of Rh)-1 h-1 c Selectivity towards the linear aldehyde.
.
rangement of minimum repulsion with respect to the
phenyl rings. Consistently, the Cp groups exhibit a
staggered conformation, with the torsion angle P(1)-
Cp1centroid-Cp2centroid-P(2) of 32.7°. The rhodium-iron
distance of 4.234 Å excludes any interaction between
the two metals, the sum of the covalent radii of Rh and
Fe being 2.69 Å.
The larger atom radius of osmium as compared to iron
(1.35 vs 1.26 Å) and the larger distances of the Cp rings
from the Os center (1.83(2) Å) well account for the short
Rh-Os distance (3.139(2) Å) observed in the structure
of 8, which indicates the occurrence of a weak, yet
actual, metal-metal bonding interaction.25 The two Cp
rings in 8 are not parallel to each other, which is shown
by the Cp1centroid-Os(1)-Cp2centroid angle of 165.9(8)°.
Indeed, the two Cp rings deviate from an eclipsed
conformation by an angle of 27.3°. The carbonyl groups
dispose themselves in the equatorial plane of a distorted
with a Zr-Rh bond length of 2.444(1) Å and a CO-Rh-
CO angle of 107.8°. The values of the Rh-P distances
averaging 2.364(7) and 2.284(2) Å respectively in 6 and
8 are in agreement with the trans influence exerted by
the opposite CO groups. The CO groups in 6 exhibit
comparable linearity and substantially equal Rh-C
(1.89 Å av) and C-O (1.14 Å av) bond distances.22a The
two carbonyl groups in 8 are less defined, due to the
poor quality of the data, and therefore a truly reliable
comparison with compound 6 cannot be made.
Batch Hydroformylation Reactions. The perfor-
mance of 1a, 2, 3a, 4, and 5 as catalyst precursors for
the hydroformylation of 1-hexene was investigated in
THF in a standard, mechanically stirred autoclave.
Following a partial optimization procedure, these condi-
tions were selected to compare the activity and regiose-
lectivity of the catalysts investigated: substrate-to-
catalyst ratio 1000, 60 °C, 45 bar CO/H2 (1:2). The
results are reported in Table 3.
From a perusal of the data reported in Table 3, one
may readily infer that the dppf, dppr, and dppo precur-
sors display a similar catalytic behavior, especially in
terms of regio- and chemoselectivity. The only noticeable
difference is provided by the conversions at 1 and 2 h:
a longer induction period seems to affect the dppo
catalyst, as it is much less productive than the dppf one
within 1 h, but it gives almost complete conversion of
the substrate after 2 h (runs 1/2 vs 5/6). The dppr
catalyst 2 lies midway between the dppf and dppo ones
(runs 3/4).
trigonal-bipyramid, and P(1) and P(2) occupy the pseudo-
axial position with an P(1)-Rh(1)-P(2) angle of 155.9-
(2)°. A similar cis disposition of the CO ligands has been
observed by Erker in a 1,1′-bis(diphenylphosphino)-
(methyl)zirconocene dicarbonyl rhodium complex.26 This
exhibits a very distorted trigonal-bipyramidal structure
(25) (a) Trepanier, S. J.; McDonald, R.; Cowie, M. Organometallics
2003, 22, 2638. (b) Trepanier, S. J.; Sterenberg, B. T.; McDonald, R.;
Cowie, M. J. Am. Chem. Soc. 1999, 121, 2613. (c) Sterenberg, B. T.;
McDonald, R.; Cowie, M. Organometallics 1997, 16, 2297. (d) Steren-
berg, B. T.; Hilts, R. W.; Moro, G.; McDonald, R.; Cowie, M. J. Am.
Chem. Soc. 1995, 117, 245. (e) Hilts, R. W.; Franchuk, R. A.; Cowie,
M. Organometallics 1991, 10, 304.
(26) (a) Cornelissen, C.; Erker, G.; Kehr, G.; Fro¨hlich, R. Organo-
metallics 2005, 24, 214. (b) Cornelissen, C.; Erker, G.; Kehr, G.;
Fro¨hlich, R. Dalton Trans. 2004, 4059.
At first glance, the dppomf catalyst appears as the
most active with an almost complete transformation of

Hydroformylation of 1-Hexene by Rh(I) Catalysts
Organometallics, Vol. 24, No. 15, 2005 3699
1-hexene in 1 h and a selectivity in aldehydes of 67%
(run 7). However, a perusal of the product distribution
and a comparison with the results obtained with the
unmodified catalyst precursor [RhCl(COD)]2 (run 8)
raised a doubt about the true nature of the catalyst
generated by 4. This doubt became a certainty after the
reaction was studied by HP-NMR and HP-IR under
hydroformylation conditions that clearly showed a
significant degradation of 4 to various compounds,
including phosphine-free rhodium carbonyls (see below).
The slight increase in 1-hexene conversion exhibited
by the o-iPr-dppf catalyst 5 in the second hour (see runs
9 and 10) was initially taken as an indication of effective
catalyst degradation within the first 60 min. On the
other hand, the enhanced production of linear aldehyde
in the second hour was still consistent with a phosphine-
modified catalyst. A clue to rationalize the behavior of
5 was later provided by operando HP-NMR and HP-IR
experiments that showed 5 to be unable to generate a
hydride(dicarbonyl) resting state (see below), which may
indicate a different hydroformylation mechanism as
compared to 1a, 2, and 3a.
Figure 2. 31P{1H} NMR spectra acquired in the course of
the hydroformylation of 1-hexene in THF-d8 catalyzed by
1b: (a) 1b in the presence of 1-hexene at room temperature
under nitrogen; (b) after the tube was pressurized with CO/
H2 (1:2) to a total pressure of 45 bar at room temperature;
(c) 40 °C, (d) 50 °C, (e) 60 °C; (f) after the tube was cooled
to room temperature.
Finally, all the catalysts promote significant isomer-
ization of 1-hexene, which is consistent with the experi-
mental conditions.
Operando NMR and IR Studies. The hydrofor-
mylation of 1-hexene assisted by 1a, 2, 3b, 4, and 5 was
investigated by HP-NMR and HP-IR spectroscopy under
experimental conditions comparable to those used in the
batch reactions. An almost perfect coincidence of ex-
perimental conditions could be adopted only for the HP-
IR experiments, as the IR cell was a real autoclave. In
contrast, the HP-NMR experiments were carried out in
10 mm-o.d. sapphire tubes that, although spinning
regularly, cannot prevent the occurrence of mass-
transfer limitation due to the lack of effective stirring.
Moreover, a higher concentration of catalyst precursors
was actually used in the HP-NMR experiments to have
a better signal-to-noise ratio. On the other hand, the
combination of HP-IR and HP-NMR experiments al-
lowed us to look at catalytic reactions on different time
scales as well as take advantage of the inherent mass-
transfer limitation of the HP-NMR technique to inter-
cept kinetic products.
Scheme 2
Typically, THF-d8 solutions of 1b, 2, 3b, 4, and 5 were
pressurized in a 10 mL sapphire tube with CO/H2 (1:2)
to a total pressure of 45 bar at room temperature. After
1
acquiring 31P{1H} and H NMR spectra, the solutions
were gradually heated to 60 °C. For every thermal step
(ca. 10 °C), a residence time of 10 min was chosen before
acquiring the next NMR spectrum. Irrespective of the
rhodium precursor, either square-planar or trigonal-
bipyramidal dicarbonyl RhI complexes were formed at
room temperature and COD was released in solution.
A slight anomalous behavior was exhibited by the dppf
complex. For this reason a sequence of 31P{1H} NMR
spectra relative to the reaction catalyzed by 1b is shown
in Figure 2, while Scheme 2 illustrates the results for
all the complexes investigated.
NMR data for the R and â hydrogen atoms of the Cp’s
that are indicative of a transoid disposition of the P
donor atoms (∆δ ) 0.85 ppm).8c,23
As the temperature was increased to 40 °C, 6 rapidly
disappeared while the other signal broadened (Figure
2c). Further increasing the temperature to 50 °C and
then to 60 °C caused both resonances to disappear with
a concomitant increase of the noise-to-signal ratio, which
is typical of NMR spectra taken during a catalytic
reaction where multiple equilibra are at work (Figure
2d,e).16,27
The dppf precursor completely converted to two
species already at room temperature: the square-planar
dicarbonyl 6 and another compound, hereafter quoted
as 15 (Figure 2b). Complex 15 is featured by a 31P{1H}
1
doublet at 24.92 ppm with J(PRh) ) 91.6 Hz and H

3700 Organometallics, Vol. 24, No. 15, 2005
Bianchini et al.
Chart 3
When the temperature of the probe-head was cooled
to room temperature, two well-resolved resonances
reappeared in the spectrum: one was due to 15, the
other to a new species that was safely identified as the
known trigonal-bipramidal hydride(dicarbonyl) RhI com-
plex RhH(CO)2(dppf) (11) (Figure 2f).11 The formation
of the latter complex apparently requires that H2 is
heterolytically split at rhodium, a process that has
several precedents, especially for hydrogenation reac-
tions carried out in a nucleophilic solvent such as THF.28
As a matter of fact, 11 was not obtained by treatment
of 1b with syngas in a nonbasic solvent such as CH2-
Cl2. The identification of 11 was straightforward on the
basis of previous studies by van Leeuwen and co-
workers.11 The NMR parameters (31P{1H} doublet and
1H triplet of doublets for the terminal hydride ligand)
observed for 11 are averaged values due to a fast
interconversion between equatorial-equatorial (ee) and
equatorial-axial (ea) geometric isomers (Scheme 2),
likely proceeding via turnstyle or Berry-type (pseudo)-
rotation.29 Interestingly, the two isomers, which inter-
convert into each other faster than the NMR time scale,
could be distinguished by HP-IR spectroscopy (see
below). At this point it may be worthwhile to recall that
the hydride(dicarbonyl) compounds are commonly re-
garded as true precatalysts for the hydroformylation of
alkenes by rhodium complexes with chelating diphos-
phines. From hydride(dicarbonyl) compounds, CO de-
coordination occurs, followed by alkene complexation,
to regenerate five-coordinate intermediates.30
All our attempts to unambiguously characterize 15
were unsuccessful. This compound is actually un-
stable: it slowly converted to 11 in the NMR tube even
at room temperature and disappeared completely by
venting the NMR tube with N2. A number of indepen-
dent experiments were carried out in an attempt to
identify 15: (i) Isolated 6 was dissolved in THF-d8 and
the solution was introduced into a HP-NMR tube. Upon
pressurization with CO (up to 30 bar), no other com-
pound than 6 was seen by NMR spectroscopy from room
temperature to 60 °C. Only when the tube was pres-
surized with 45 bar CO/H2 (1:2) did 6 start to convert
into 15. (ii) The same experiment as in (i) was performed
in CD2Cl2. No conversion into 15 was observed even in
the presence of syngas, irrespective of the pressure and
temperature. (iii) When the hydroformylation of 1-hex-
ene catalyzed by 1b was studied by operando HP-IR
spectroscopy under fast and effective mixing of the
reactants (see below), no trace of 15 was detected. (iv)
Isolated 6 was dissolved in THF-d8 and the solution was
introduced into an HP-NMR tube. Upon pressurization
with H2 (up to 30 bar) at room temperature, some
hydride(dicarbonyl) 11 was formed, but together with
several unknown, probably reduced, species.
Incorporation of all of these results with the NMR
data for 15 led us to conclude that (i) the P atoms in 15
are likely transoid to each other. (ii) 15 requires the
concomitant presence of H2 and CO to form, yet these
two reagents must be in low concentration as occurs in
the mass-transfer limitation conditions of the HP-NMR
experiment. Indeed, under vigorous stirring as occurs
in the HP-IR cell, neither 6 nor 15 was seen and 1b
was straightforwardly converted into 11 even at room
temperature (see below). (iii) The formation of 15
requires a basic solvent to be accomplished. A number
of possible structures for dppf metal complexes are fully
or partially consistent with the experimental observ-
ables listed above (Chart 3), yet none of them have been
unequivocally proved.
A much simpler HP-31P{1H} NMR picture was ob-
served for the dppr and dppo catalysts under hydro-
formylation conditions. In either case, the dicarbonyl
intermediates were initially formed and the neutral
hydride(dicarbonyl) compounds were the only detectable
Rh compounds present in the reaction mixtures in the
course (see HP-IR study below) and at the end of the
hydroformylation catalysis (Scheme 2). The conversion
of the dppo derivative 8 into RhH(CO)2(dppo) (13) was
significantly slower (3 h at 60 °C) than that of the dppr
analogue RhH(CO)2(dppr) (12) (<1 h). Again, averaged
values for H-P and Rh-P coupling constants were
observed in the NMR spectra.
In contrast to 1b, 2, and 3b, neither 4 nor 5 generated
hydride(dicarbonyl) compounds when reacted with syn-
gas in the HP-NMR tube with or without added 1-hex-
ene. The dppomf derivative, after forming a square-
planar dicarbonyl at room temperature, underwent
extensive decomposition to binary CO rhodium com-
pounds as the temperature was increased to 60 °C (see
HP-IR experiment below), while the o-iPr-dppf dicar-
bonyl complex 10 remained unchanged even after 2 h
at 60 °C. In separate experiments, the dicarbonyl
complexes 9 and 10 were reacted with syngas in the
presence of NEt3, which is known to promote the
heterolytic splitting of H2.28 Under these conditions, the
hydride(dicarbonyl) RhH(CO)2(dppomf) (14) was selec-
tively obtained even at room temperature, while 10
decomposed to give rhodium-carbonyl compounds and
free metallocenylphosphine ligand.
(27) Bianchini, C.; Lee, H. M.; Barbaro, P.; Meli, A.; Monetti, S.;
Vizza, F. New J. Chem. 1999, 23, 929.
(28) Bianchini, C.; Moneti, S.; Peruzzini, M.; Vizza, F. Inorg. Chem.
1997, 36, 5818.
(29) (a) Berry, R. S. J. Chem. Phys. 1960, 32, 933. (b) Meakin, P.;
Muetterties, E. L.; Jesson, J. P. J. Am. Chem. Soc. 1972, 94, 5271. (c)
Buisman, G. J. H.; van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M. Organometallics 1997, 16, 5681. (d) Casey, C. P.; Paulsen,
E. L. Beuttenmueller, E. W.; Proft, B. R.; Matter, B. A.; Powell, D. R.
J. Am. Chem. Soc. 1999, 121, 63. (e) Castellanos-Pa´ez, A.; Castillon,
S.; Claver, C.; van Leeuwen, P. W. N. M.; de Lange, W. G. J.
Organometallics 1998, 17, 2543.
(30) (a) Van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Reek, J. N. H.;
Dierkes, P. Chem. Rev. 2000, 100, 2741. (b) Van der Veen, L. A.;
Keeven, P. H.; Schoemaker, G. C.; Reek, J. N. H.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M.; Lutz, M.; Spek, A. L. Organometallics 2000,
19, 872. (c) Casey, C. P.; Whiteker, G. T.; Melville, M. G.; Petrovich,
L. M.; Gavney, J. A.; Powell, D. R. J. Am. Chem. Soc. 1992, 114, 5535.
Parallel to the HP-NMR investigation, an HP-IR
study of the hydroformylation reactions catalyzed by 1b,
2, 3b, 4, and 5 was carried out under conditions
comparable to those of the batch reactions, except for a
substrate-to-catalyst ratio of 100. IR traces in the CO
stretching region, recorded at 60 °C, are reported in
Figure 3 for all reactions.

Hydroformylation of 1-Hexene by Rh(I) Catalysts
Organometallics, Vol. 24, No. 15, 2005 3701
relative ν(CO) bands showed that the metallocenyl-
diphosphines investigated prefer to bind rhodium in ea
fashion, which accounts for the regioselectivity observed
in the batch reactions (67-74% in linear aldehyde).11,30b,32
Conclusions
The hydroformylation of 1-hexene catalyzed by 1,1′-
bis(diarylphosphino)metallocene rhodium(I) complexes
has been investigated, for the first time, by both HP-
NMR and HP-IR spectroscopy under catalytic condi-
tions. In the absence of this study, the hydroformylation
results, especially as regards the chemo- and regiose-
lectivity (Table 3), would have misled one to conclude
that all diphosphine rhodium precursors generate simi-
lar catalytic systems, where unmodified catalysts, formed
upon degradation of the modified catalysts, contribute
to the overall catalytic outcome. Just the operando
studies have made clear that this is not true, unam-
biguously showing that both the mechanism and the
catalytic activity are actually controlled by the structure
of the diphosphine ligand. On the other hand, it is also
apparent that the structural variations of the 1,1′-bis-
(diarylphosphino)metallocenes investigated in this study
affect significantly neither the chemoselectivity nor the
regioselectivity.
The main conclusions provided by the present study
may be summarized as follows.
Figure 3. HP-IR spectra acquired after 1 h of the
hydroformylation of 1-hexene catalyzed by [Rh(P-P)-
(COD)]X (THF, 60 °C, 45 bar 1:2 CO/H2).
Irrespective of the catalyst precursor, all the steps
involved in the hydroformylation reaction at 60 °C are
faster than the NMR time scale, including the genera-
tion of the catalyst from the hydride(dicarbonyl) resting
state. Noteworthy, the latter process seems to be rate
limiting, as the hydride(dicarbonyl) compounds are the
only species visible on the IR time scale during the
catalytic reactions.
The activation of H2 in THF is heterolytic in nature
for the dppf, dppr, dppo, and dppomf precursors. More-
over, the activity of the 1,1′-bis(diarylphosphino)metal-
locene rhodium(I) catalysts is significantly influenced
by the formation (and strength) of dative metal-
rhodium bonds in the RhI dicarbonyl intermediates
which are precursors to the RhI hydride(dicarbonyl)
resting states. The low productivity of the dppo precur-
sor in the first hour is therefore attributable to the
remarkable strength of the dative Os-Rh bonding
interaction in the corresponding dicarbonyl complex.
However, the dppo catalyst is also more efficient than
the dppf and dppr catalysts in the second hour, which
suggests that a strong intramolecular Os-Rh bond may
be important in stabilizing other five-coordinate cata-
lytic intermediates toward their degradation to phos-
phine-free species. Severe degradation has been proved
for the catalyst with dppomf, which, in fact, does not
form a dative Fe-Rh bond.
Irrespective of the temperature, pressurization of the
autoclave with 45 bar syngas caused the COD precur-
sors 1b and 2 to transform rapidly and selectively into
the hydride(dicarbonyl) compounds 11 and 12, respec-
tively (Figure 3a,b). In agreement with the NMR study,
the dppo precursor 3b converted to the hydride(dicar-
bonyl) 13 more slowly than the dppf and dppr precursors
(Figure 3c). Indeed, the complete transformation re-
quired 2 h at 60 °C. No trace of hydride(dicarbonyl)
complex was observed starting from the o-iPr-dppf
precursor, whose dicarbonyl product 10 was stable for
2 h at 60 °C (Figure 3d). A dicarbonyl product was also
obtained with the dppomf precursor; however soon after
its formation, the dicarbonyl 9 started decomposing to
give, inter alia, phosphine-free rhodium compounds
(Figure 3e) that were also obtained by the reaction of
[RhCl(COD)]2 with syngas (Figure 3f).31
Besides confirming the transient nature of the dicar-
bonyl complexes and the resting-state nature of the
hydride(dicarbonyl) complexes with dppf, dppr, and
dppo, the HP-IR experiments provided further valuable
information: (i) The dppomf dicarbonyl intermediate
undergoes decomposition to several CO rhodium species,
which include compounds with no coordinated phos-
phine. (ii) The dynamic equilibrium that averages the
NMR parameters of the ee and ea stereoisomers of the
hydride(dicarbonyl) complexes 11, 12, 13, and 14 (this
compound was independently prepared by adding NEt3,
vide infra) is slow on the IR time scale. Accordingly, both
conformers were detected and the intensity ratio of the
Another case is represented by the o-iPr-dppf ligand:
it promotes the formation of a κ3-P,P,M dicarbonyl
complex, but not of a hydride(dicarbonyl) compound
even in THF. The failure of the o-iPr-dppf dicarbonyl
complex to activate H2 in a heterolytic fashion may be
due to the four bulky o-iPr groups. These may either
(31) (a) Buisman, G. J. H.; Vos, E. J.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M. J. Chem. Soc., Dalton Trans. 1995, 409. (b) Garland, M.;
Pino, P. Organometallics 1991, 10, 1693.
(32) Van der Veen, L. A.; Boele, M. D. K.; Bregman, F. R.; Kamer,
P. C. J.; van Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J.; Schenk,
H.; Bo, C. J. Am. Chem. Soc. 1998, 120, 11616.

3702 Organometallics, Vol. 24, No. 15, 2005
Bianchini et al.
reinforce the dative Fe-Rh bond by favoring a trans
arrangement of the P atoms8a,b,23 or disfavor the exter-
nal nucleophilic attack by THF or NEt3, which is
required for the heterolytic splitting of activated H2.28
Traversi and Carlo Bartoli for the construction of the
high-pressure NMR and IR instruments.
Supporting Information Available: Crystallographic
tables and tables of atomic coordinates, bond distances and
angles, anisotropic thermal parameters, and hydrogen coor-
dinates for compounds 6 and 8 are available free of charge
via the Internet at http://pubs.acs.org.
Acknowledgment. This work has been supported
by the CNR Department “Molecular Design” within the
project “Design of molecules and nanostructures with
catalytic properties”. Thanks are also due to Mr. Aldo
OM050241L