A readily prepared neutral heterobimetallic titanium(IV)-rhodium(I) catalyst for intramolecular hydroacylation

Article abstract of DOI:10.1039/b504195c

The combination of HOCMe₂CH₂PPh₂, Ti(OzPr)₄, and [Rh(cod)Cl]₂ (3:1:1) in either benzene or dichloromethane produces a discrete species (tentatively formulated as complex 14) that is an active catalyst for intramolecular hydroacylation reactions of 3-substituted pentenals. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b504195c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A readily prepared neutral heterobimetallic titanium(IV)–rhodium(I)
catalyst for intramolecular hydroacylation{
John P. Morgan,* Kousik Kundu* and Michael P. Doyle
Received (in Berkeley, CA, USA) 24th March 2005, Accepted 3rd May 2005
First published as an Advance Article on the web 1st June 2005
DOI: 10.1039/b504195c
The leading results of Wolczanski, et al. prompted us to focus
on titanium(IV) as the second metal center in the heterobimetallic
complex.11 This previous report revealed that complex 13 (Fig. 2)
could be readily synthesized from TiCl4, but the product eluded
full characterization due to the facile formation of aggregates. We
were surprised to find that switching to Ti(OiPr)4 resulted in the
formation of a discrete Rh–phosphine complex without the
concomitant formation of by-products. Notably, the reaction is
carried out in two steps: first, Ti(OiPr)4 is added to a solution of
3 eq. ligand HOCMe2CH2PPh2 10 (Fig. 2(a)) in dry dichlor-
omethane.12 Within 0.5 h the entirety of the ligand has complexed
the titanium, evidenced by a small (1.5 ppm) upfield shift in the
31P{1H} NMR spectrum (Fig. 2(b)) and the formation of a
persistent light yellow color. Subsequently [RhCl(cod)]2 is
added and the formation of a sharp doublet at 18.19 ppm occurs
The combination of HOCMe2CH2PPh2, Ti(OiPr)4, and
[Rh(cod)Cl]2 (3:1:1) in either benzene or dichloromethane
produces a discrete species (tentatively formulated as complex
14) that is an active catalyst for intramolecular hydroacylation
reactions of 3-substituted pentenals.
Catalysts with multiple metal centers can show a breadth of
reactivity that often exceeds their monometallic counterparts.1
Each metal center may play a unique catalytic or structural role in
the overall ‘‘multimetallic catalyst system.’’ A necessary condition
for the use of two (or more) metals is their compatibility, a
situation that has been addressed in ‘‘early–late’’ bimetallic
transition metal complexes.2–4 Ultimately both the identity of the
metals and their proximity to each other may lead to ‘‘synergistic’’
effects, in which one metal perturbs the reactivity of the other in an
advantageous manner.5 This metal–metal cooperativity has
recently been suggested in hydrocarbonylation chemistry, one of
our continuing interests.6
A specific hydrocarbonylation process (hydroacylation) was
targeted for initial investigation with a proposed heterobimetallic
catalyst involving rhodium(I) (Scheme 1).7 The most active Rh(I)
catalysts (developed by Bosnich and coworkers) were BINAP-
derived cationic rhodium(I) systems which demonstrated remark-
able reactivity for the cyclization of 4-substituted pentenals.8
Recently however, Peters has shown that zwitterionic metal–
metalloid rhodium–boronate complexes demonstrate the highest
turnover frequency of all cationic rhodium systems (Fig. 1).9 In
each case the cis configuration of phosphines, locked by chelation,
prevents significant sequestering of catalyst as the inactive form D
(in Scheme 1).
The high activity of these boronate complexes indicates that
electronic perturbation of the cationic rhodium(I) center by a
proximally located charged heteroatom can dramatically increase
hydroacylation reactivity. Rather than work with complexes
containing a cationic rhodium atom, we chose instead to focus
on neutral rhodium, due to its vastly improved stability relative to
the cationic systems.7f Additionally there is significant room for
improvement in neutral rhodium(I) hydroacylation catalysts; a
general, highly active neutral rhodium system is currently
unknown.10 In this Communication, we report the construction
and hydroacylation catalysis of a unique neutral rhodium(I)
bimetallic complex.
Scheme 1 Mechanism for Rh(I)-catalyzed intramolecular hydroacyla-
tion using a Wilkinson’s-type complex. For neutral complexes, X is
typically
a halide; for cationic complexes X represents an empty
coordination site.
{ Electronic supplementary information (ESI) available: experimental
procedures and catalyst characterization. See http://www.rsc.org/suppdata/
cc/b5/b504195c/
*jpmorgan@umd.edu (John P. Morgan)
kousik@glue.umd.edu. (Kousik Kundu)
Fig. 1 Rhodium(I) catalyst with a proximally located charged hetero-
atom shows enhanced reactivity.9
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 3307–3309 | 3307

View Article Online
Scheme 2 Hydroacylation of styrene 2-carboxaldehyde (8).8
occurs. Similarly compound 8 is known to produce significant
amounts of polymeric side product when exposed to cationic
rhodium catalysts (Scheme 2).8 With 14, the desired hydroacyl-
ation product is cleanly produced and no polymer side products
are formed.
Importantly, none of the catalyst components ([RhCl(cod)]2,
Ti(OiPr)4, or ligand) can individually (or in pairwise combination)
catalyze the hydroacylation of 1, suggesting that the active species
is indeed a unique three-component complex. Compound 12 (i.e.
without Rh) is ineffective at hydroacylation, as is the combination
of [RhCl(cod)]2 + Ti(OiPr)4 (without added ligand). Most
significantly, the mixture of [RhCl(cod)]2 and 3 equivalents of
ligand 10 (i.e. without Ti) does not produce a viable catalyst. A
‘‘carbon analogue’’ of the catalyst, (triphos)RhCl (triphos 5 tris-
(tris-(diphenylphosphinomethyl)ethane), in which the titanium
center is substituted by a quaternary carbon atom, also does not
catalyze the hydroacylation of 1 either at room temperature or at
45 uC.15 These control experiments indicate that the Ti(IV) center
must not only organize the ligands but must also play a unique
role in catalysis.
Fig. 2 In situ preparation of (iPrO)Ti(m:g1,g1-OCMe2CH2PPh2)3RhCl
(14) with associated 31P spectra. Complete spectra are supplied in
Supporting Information.
over 0.5 h (JRhP 5 147 Hz, Fig. 2(c)). The resulting complex,
tentatively formulated as (iPrO)Ti(m:g1,g1-OCMe2CH2PPh2)3-
RhCl, 14, decomposes to an oily solid under vacuum,13 but it is
stable in solution for over 36 hours. Catalyst synthesis (and the
hydroacylation reactions) can be readily carried out on the
benchtop in dry glassware under a nitrogen atmosphere; a
glovebox is not required. Several attempts to crystallize the
complex were unsuccessful.
In Table 2, the catalytic turnover of the bimetallic system is
compared to other monometallic neutral Rh(I) systems formed
in situ. For 14, the TOF is an order of magnitude larger than other
monometallic species. Although the exact reason for this higher
activity remains unclear, it is feasible that the coordination of the
aldehyde to the Lewis acidic titanium center (of one catalyst
complex) may activate the aldehydic C–H bond for Rh insertion
(by a second catalyst complex). This activation may then lead to a
faster rate of catalysis by shifting the insertion equilibrium toward
the hydrido-acyl-Rh(III) species A (Scheme 1). However, addition
of Ti(OiPr)4 to the reaction between RhCl(PPh3)3 and 1 did not
change reaction rate or yield, indicating that this Lewis acid-
catalyzed pathway should not be relevant.
Complex 14 can perform the clean intramolecular hydroacyla-
tion ring-closure reaction of 3-phenyl-4-pentenal 1 and styrene
2-carboxaldehyde 8 (Table 1 and Scheme 2). In particular,
compound 1 can lead to 6 different possible products (2–7 in
Scheme 1) and three have been observed using cationic catalysts.14
When compound 14 is utilized, however, the cyclopentanone is the
only observable product – no rearrangement or alkene migration
Table 1 Hydroacylation of 3-substituted pentenal 1 at room temper-
aturea
Product (% yield)
An alternative explanation is ‘‘bimetallic cooperativity,’’ in
which the cationic Ti center directly activates the Rh center by
charge-charge repulsion (Scheme 3). As the Rh center becomes
more electron deficient (Rh(III)), reductive elimination to Rh(I)
(and therefore catalytic turnover) should be favored.
Computational analysis of the complex suggests that the Rh and
Ti centers lie too far apart for direct orbital overlap (greater than
Substrate Catalyst
[Rh(S-BINAP)]+b
Time
45 min
51
42
79
0
7
[Rh(S,S-chiraphos)]+b 45 min 21
0
0
16
14c
8 h
98
3.1 A), consistent with this cooperativity model as a through-
˚
space charge–charge effect. Also proximity of the dissociated
phosphine may play an important role in the reductive elimination
step.
a
See Supporting Information for full experimental details. All
In conclusion, the 1:1:3 Rh:Ti:ligand complex 14 is an easily
constructed neutral Rh(I) catalyst which exhibits hydroacylation
reactivity. The catalyst components are readily available and the
catalyst preparation can be easily monitored by 31P NMR. The
titanium center is uniquely important to the overall catalytic
reactions were performed in CD2Cl2 at room temperature. Yields
using catalyst 14 were measured by NMR integration or gas
b
chromatography.
c
Catalyst loading is 10 mol%, prepared in situ.
Reference 14. Catalyst loading is 1 mol%.
3308 | Chem. Commun., 2005, 3307–3309
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Table 2 Turnover frequencies for the intramolecular hydroacylation of 1 using various neutral hydroacylation catalysts prepared in situa
Catalyst
Solvent
Time to completion
TOF (h21
)
(iPrO)Ti(m:g1,g1-OCMe2CH2PPh2)3RhCl, 14
RhCl(PPh3)3
C6D6
C6D6
C6D6
CD2Cl2
CD2Cl2
CD2Cl2
8 h
Slow (23% in 16 h)
Slow
72 h
.72 h
80 h
1.2
0.14
—
0.14
,0.12
0.13
[Rh(cod)Cl]2 + R-(+)-BINAP
[Rh(cod)Cl]2 + R-(+)-BINAP
[Rh(cod)Cl]2 + S,S-(+)-DIOP
[Rh(cod)Cl]2 + R,R-(+)-BDPP
a
Conditions: [Rh(cod)Cl]2 (3 mmol), ligand (3.5 mmol) in solvent (1 mL) at room temperature.7f Catalyst loading is 10 mol% for all cases.
BINAP 5 2,29-bis(diphenylphosphino)-1,19-binaphthyl; DIOP 5 (4S,5S)-O-isopropylidene-2,3-dihydroxy-1,4-bis(diphenylphosphino)butane;
BDPP 5 (2R,4R)-2,4-bis(diphenylphosphino)pentane; Triphos 5 tris(diphenylphosphinomethyl)ethane.
4 For reviews on ligands in bimetallic or polymetallic metal systems, see:
(a) A. D. Garnovskii, B. I. Kharisov, L. M. Blanco, A. P. Sadimenko,
A. I. Uraev, I. S. Vasilchenko and D. A. Garnovskii, J. Coord. Chem.,
2002, 55, 1119–1141; (b) S. Lotz, P. H. van Rooyen and R. Meyer, Adv.
Organomet. Chem., 1995, 37, 219–320.
5 (a) T. G. Gray, A. S. Veige and D. G. Nocera, J. Am. Chem. Soc., 2004,
126, 9760–9768; (b) A. Ceccon, S. Santi, L. Orian and A. Bisello, Coord.
Chem. Rev., 2004, 248, 683–724; (c) E. N. Jacobsen, Acc. Chem. Res.,
2000, 33, 421–431.
6 (a) M. S. Shanklin, M. V. Zlokazov and M. P. Doyle, Synlett, 1994,
615–616; (b) M. S. Shanklin and M. P. Doyle, Organometallics, 1993,
Scheme 3 Acceleration of rate-limiting reductive elimination step by
charge-charge repulsion and proximity of coordinating phosphine.
12, 11–12; (c) M. E. Broussard, B. Juma, S. G. Train, W. J. Peng,
S. A. Laneman and G. G. Stanley, Science, 1993, 260, 1784–1788.
7 (a) V. Ritleng, C. Sirlin and M. Pfeffer, Chem. Rev., 2002, 102,
1731–1769; (b) C. H. Jun, J. B. Hong and D. Y. Lee, Synlett, 1999, 1,
1–12 and references therein; (c) R. C. Larock, K. Oertle and G. F. Potter,
J. Am. Chem. Soc., 1980, 102, 190–197; (d) J. W. Suggs, J. Am. Chem.
Soc., 1979, 101, 489; (e) M. Tanaka and K. Sakai, Curr. Org. Chem.,
2003, 7, 353–367; (f) Y. Taura, M. Tanaka, X. M. Wu, K. Funakoshi
and K. Sakai, Tetrahedron, 1991, 47, 4879–4888.
turnover of this bimetallic system. The ready ability to change both
ligands and metals in the catalyst synthesis can rapidly provide a
variety of structurally complex and catalytically active early-late
transition metal complexes of interest.
We thank the National Science Foundation for supporting this
work. We also thank Jason M. Nichols for assistance with
computational work.
8 (a) B. Bosnich, Acc. Chem. Res., 1998, 31, 667–674; (b) D. P. Fairlie and
B. Bosnich, Organometallics, 1988, 7, 936–945; (c) D. P. Fairlie and
B. Bosnich, Organometallics, 1988, 7, 946–954.
9 T. A. Betley and J. C. Peters, Angew. Chem. Int. Ed., 2003, 42,
2385–2389.
John P. Morgan,* Kousik Kundu* and Michael P. Doyle
10 M. Tanaka, M. Imai, M. Fujio, E. Sakamoto, M. Takahashi, Y. Eto-
Kato, X. M. Wu, K. Funakoshi, K. Sakai and H. Suemune, J. Org.
Chem., 2000, 65, 5806–5816 and references therein.
Department of Chemistry and Biochemistry, University of Maryland,
College Park, Maryland 20742, USA. E-mail: jpmorgan@umd.edu;
kousik@glue.umd.edu.; Fax: 301.314.9121; Tel: 301-405-1327
11 L. M. Slaughter and P. T. Wolczanski, Chem. Commun., 1997,
2109–2110.
12 Variation of ligand stoichiometry indicates that 3:1:1 ligand:Rh:Ti is
optimal: lower amounts of added ligand (2 eq.) lead to the formation of
multiple metal species (detected by 31P{1H} NMR), while higher
stoichiometry (4.5 eq.) leads to rapid precipitation of catalytically
inactive, Rh-containing solids.
13 When the oily solid is redissolved in CD2Cl2, the doublet at 18.19 ppm
in the 31P{1H} NMR spectrum is replaced by a singlet at 30.72 ppm
(Supporting Information). The resulting yellowish solution does not
show hydroacylation reactivity with 1.
14 R. W. Barnhart and B. Bosnich, Organometallics, 1995, 14, 4343–4348.
15 (a) C. Bianchini, A. Meli, M. Peruzzini and F. Vizza, Organometallics,
1990, 9, 226–240; (b) To our knowledge (triphos)Rh(I) complexes have
not been previously examined for catalytic hydroacylation.
16 Modeling was performed on an IBM PC with Pentium 4 processor
running Spartan ’02 molecular mechanics (Wavefunction, Inc.).
Notes and references
1 (a) S. Kamijo and Y. Yamamoto, J. Org. Chem., 2003, 68, 4764–4771;
(b) S. Matsunaga, T. Ohshima and M. Shibasaki, Adv. Synth. Catal.,
2002, 344, 3–15; (c) B. Coq and F. Figueras, J. Mol. Cat. A Chem., 2001,
173, 135–184; (d) M. Shibasaki and M. Kanai, Chem. Pharm. Bull.,
2001, 49, 511–524; (e) Multimetallic Catalysts in Organic Synthesis;
M. Shibasaki and Y. Yamamoto, Eds.; Wiley: Weinheim, 2004.
2 (a) J. P. Collman and R. Boulatov, Chem. Rev., 2004, 104, 561–588; (b)
J. W. Park, P. B. Mackenzie, W. P. Schaefer and R. H. Grubbs, J. Am.
Chem. Soc., 1986, 108, 6402–6404; (c) S. Kuwata, S. Kabashima,
N. Sugiyama, Y. Ishii and M. Hidai, Inorg. Chem., 2001, 40, 2034–2040.
3 (a) G. S. Ferguson and P. T. Wolczanski, J. Am. Chem. Soc., 1986, 108,
8293–8295; (b) S. M. Baxter and P. T. Wolczanski, Organometallics,
1990, 9, 2498–2509.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 3307–3309 | 3309