Controlling selectivity in intermolecular alkene or aldehyde hydroacylation reactions catalyzed by {Rh(L2)}+ fragments

Article abstract of DOI:10.1021/om9011104

Rhodium(III) dihydrido complexes [Rh(L₂)(H)₂(acetone) ][BAr^F₄] (Ar^F = C₆H ₃(CF₃)₂) containing the potentially hemilabile ligands L₂ = 9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene (Xantphos) and [Ph₂P(CH₂)₂]₂O (POP′) have been prepared from their corresponding norbornadiene rhodium(I) precursors. In solution these complexes are fluxional by proposed acetone dissociation, which can be trapped out by addition of MeCN to form [Rh(L₂)(H)₂(NCMe)][BAr^F₄], which have been crystallographically characterized. Addition of alkene (methyl acrylate) to these complexes results in reduction to a rhodium(I) species and when followed by addition of the aldehyde HCOCH₂CH₂SMe affords the new acyl hydrido complexes [Rh(L₂)(COCH ₂CH₂SMe)H][BAr^F₄] in good yield. The solid-state and solution structures show a tight binding of the POP′ and Xantphos ligands, having a trans-arrangement of the phosphines with the central ether linkage bound. This is similar to the previously reported complex [Rh(DPEphos)(COCH₂CH₂SMe)H][BAr^F₄] (DPEphos = [Ph₂P(C₆H₄)]₂O). Unlike the DPEphos complex, the Xantphos and POP′ ligated complexes are not effective catalysts for the hydroacylation reaction between methyl acrylate and HCOCH₂CH₂SMe. This is traced to their inability to dissociate the central ether link in a hemilabile manner to reveal a vacant site necessary for alkene coordination. Consistent with this lack of availability of the vacant site, these complexes also are stable toward reductive decarbonylation. Complexes [Rh(Ph₂P(CH₂) _nPPh₂)(acetone)₂][BAr^F₄] (n = 2-5) have also been studied as catalysts for the hydroacylation reaction between methyl acrylate and HCOCH₂CH₂SMe at 22 °C. As found previously, for n = 2 this affords the product of alkene hydroacylation, but as the chain length is progressively increased to n = 5, the reaction also progressively changes to favor the product of aldehyde hydroacylation. This is suggested to occur by a decrease in the accessibility of the metal site on increasing the bite angle of the chelate ligand, so that alkene coordination to a putative Rh(III)-acyl hydrido intermediate is progressively disfavored and aldehyde coordination (followed by hydride transfer) is progressively favored. These, and previous, results show that the overall conversion in the hydroacylation reaction can be controlled by the hemilabile nature of the chelating phosphine in the catalyst (e.g., DPEphos versus Xantphos), and the course of the reaction can also be tuned by changing the bite angle of the phosphine, cf. Ph₂P(CH₂)₂PPh₂ and Ph₂P(CH₂)₅PPh₂.

Full text of DOI:10.1021/om9011104

Organometallics 2010, 29, 1717–1728 1717
DOI: 10.1021/om9011104
Controlling Selectivity in Intermolecular Alkene or Aldehyde
Hydroacylation Reactions Catalyzed by {Rh(L2)}þ Fragments
Rebekah J. Pawley,† Gemma L. Moxham,‡ Romaeo Dallanegra,† Adrian B. Chaplin,†
Simon K. Brayshaw,‡ Andrew S. Weller,*,†,‡ and Michael C. Willis†,‡
†Department of Chemistry, Inorganic Chemistry Laboratory, University of Oxford, South Parks Road,
Oxford, OX1 3QR, U.K., and ‡Department of Chemistry, University of Bath, Bath, BA2 7AY, U.K.
Received December 23, 2009
Rhodium(III) dihydrido complexes [Rh(L2)(H)2(acetone)][BArF ] (ArF=C6H3(CF3)2) containing the
4
potentially hemilabile ligands L2 = 9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene (Xantphos) and
[Ph2P(CH2)2]2O (POP0) have been prepared from their corresponding norbornadiene rhodium(I) precur-
sors. In solution these complexes are fluxional by proposed acetone dissociation, which can be trapped out
by addition of MeCN to form [Rh(L2)(H)2(NCMe)][BArF ], which have been crystallographically
4
characterized. Addition of alkene (methyl acrylate) to these complexes results in reduction to a rhodium(I)
species and when followed by addition of the aldehyde HCOCH2CH2SMe affords the new acyl hydrido
complexes [Rh(L2)(COCH2CH2SMe)H][BArF ] in good yield. The solid-state and solution structures
4
show a tight binding of the POP0 and Xantphos ligands, having a trans-arrangement of the phosphines with
the central ether linkage bound. This is similar to the previously reported complex [Rh(DPEphos)-
(COCH2CH2SMe)H][BArF ] (DPEphos=[Ph2P(C6H4)]2O). Unlike the DPEphos complex, the Xantphos
4
and POP0 ligated complexes are not effective catalysts for the hydroacylation reaction between methyl
acrylate and HCOCH2CH2SMe. This is traced to their inability to dissociate the central ether link in a
hemilabile manner to reveal a vacant site necessary for alkene coordination. Consistent with this lack of
availability of the vacant site, these complexes also are stable toward reductive decarbonylation. Complexes
[Rh(Ph2P(CH2)nPPh2)(acetone)2][BArF ] (n = 2-5) have also been studied as catalysts for the hydro-
4
acylation reaction between methyl acrylate and HCOCH2CH2SMe at 22 °C. As found previously, for n=2
this affords the product of alkene hydroacylation, but as the chain length is progressively increased to n=5,
the reaction also progressively changes to favor the product of aldehyde hydroacylation. This is suggested to
occur by a decrease in the accessibility of the metal site on increasing the bite angle of the chelate ligand, so
that alkene coordination to a putative Rh(III)-acyl hydrido intermediate is progressively disfavored and
aldehyde coordination (followed by hydride transfer) is progressively favored. These, and previous, results
show that the overall conversion in the hydroacylation reaction can be controlled by the hemilabile nature of
the chelating phosphine in the catalyst (e.g., DPEphos versus Xantphos), and the course of the reaction can
also be tuned by changing the bite angle of the phosphine, cf. Ph2P(CH2)2PPh2 and Ph2P(CH2)5PPh2.
Introduction
years, and a variety of;mainly rhodium-based;catalyst sys-
tems have been developed.3-15 Similar carbonyl hydroacyla-
tion systems have also been described.16-19 The key steps in
The transition-metal-catalyzed hydroacylation reaction that
couples alkenes, or alkynes, with aldehydes is an attractive
atom-efficient method for the formation of unsymmetrical
ketones, eq 1.1,2 It has attracted significant attention in recent
(10) Shibata, Y.; Tanaka, K. J. Am. Chem. Soc. 2009, 131, 12552–
12553.
(11) Coulter, M. M.; Dornan, P. K.; Dong, V. M. J. Am. Chem. Soc.
2009, 131, 6932–6933.
*Corresponding author. E-mail: andrew.weller@chem.ox.ac.uk.
(1) Suggs, J. W. J. Am. Chem. Soc. 1978, 100, 640–641.
(2) Willis, M. C. Chem. Rev. 2009, 110, 725-748.
(3) Fairlie, D. P.; Bosnich, B. Organometallics 1988, 7, 946–954.
(4) Fairlie, D. P.; Bosnich, B. Organometallics 1988, 7, 936–945.
(5) Marder, T. B.; Roe, D. C.; Milstein, D. Organometallics 1988, 7,
1451–1453.
(12) Osborne, J. D.; Randell-Sly, H. E.; Currie, G. S.; Cowley, A. R.;
Willis, M. C. J. Am. Chem. Soc. 2008, 130, 17232–17233.
(13) Osborne, J. D.; Willis, M. C. Chem. Commun. 2008, 5025–5027.
(14) Willis, M. C.; Randell-Sly, H. E.; Woodward, R. L.; McNally, S.
J.; Currie, G. S. J. Org. Chem. 2006, 71, 5291–5297.
(15) Willis, M. C.; McNally, S. J.; Beswick, P. J. Angew. Chem., Int.
Ed. 2004, 43, 340–343.
(6) Kundu, K.; McCullagh, J. V.; Morehead, A. T., Jr. J. Am. Chem.
Soc. 2005, 127, 16042–16043.
(7) Jun, C. H.; Jo, E. A.; Park, J. W. Eur. J. Org. Chem. 2007, 1869–
1881.
(16) Bergens, S. H.; Fairlie, D. P.; Bosnich, B. Organometallics 2002,
9, 566–571.
(17) Fuji, K.; Morimoto, T.; Tsutsumi, K.; Kakiuchi, K. Chem.
Commun. 2005, 3295–3297.
(8) Roy, A. H.; Lenges, C. P.; Brookhart, M. J. Am. Chem. Soc. 2007,
129, 2082–2093.
(9) Lenges, C. P.; White, P. S.; Brookhart, M. J. Am. Chem. Soc.
(18) Phan, D. H. T.; Kim, B.; Dong, V. M. J. Am. Chem. Soc. 2009,
131, 15608–15609.
(19) Shen, Z.; Dornan, P. K.; Khan, H. A.; Woo, T. K.; Dong, V. M.
1998, 120, 6965–6979.
J. Am. Chem. Soc. 2009, 131, 1077–1091.
r
2010 American Chemical Society
Published on Web 03/01/2010
pubs.acs.org/Organometallics

1718 Organometallics, Vol. 29, No. 7, 2010
Scheme 1. Hemilabile DPEphos, Protecting the Vacant Sitea
Pawley et al.
a Anions are omitted for clarity.
Chart 1. The Bidentate Ligands Used in This Study
the mechanism have been identified and involve oxidative
addition of aldehyde to a Rh(I) center to give a hydrido-acyl
species, followed by migratory insertion of an η-bound
alkene (or heteroalkene), and reductive elimination of the
C-C coupled product.3,19-22 A major obstacle in the devel-
opment of this potentially powerful disconnection in organic
synthetic methodology is competing reductive decarbonyla-
tion, which results in an inactive metal carbonyl. This occurs
from a low-coordinate intermediate in the catalytic cycle,
which can undergo migratory deinsertion of a carbonyl from
an acyl group to a cis vacant site, followed by reductive
elimination of a saturated product (e.g., R-H).
catalytic reactivity is significantly attenuated, while for the
chelating, but nonhemilabile ligands Ph2P(CH2)nPPh2 (n =
2-5, Chart 1) the unexpected result of switching the reaction
from almost exclusively alkene hydroacylation (n = 2) to
almost exclusively aldehyde hydroacylation (Tishchenko-
type27 product, n = 5) is revealed. One of us has recently
reported a related switch in reactivity between hydroacylation
and reductive aldol on moving between DPEphos and dppe
ligands in rhodium-based catalyst systems.13
Various strategies have been adopted to attenuate this
detrimental decarbonylation pathway, one being chelation
control from the substrate.1,7,15,23 We have recently reported
an alternative metal-ligand-based approach to this problem
by using the hemilabile24 phosphine-ether ligand DPEphos
[bis(2-diphenylphosphinophenyl) ether25] to protect the va-
cant site, thus stabilizing the catalyst toward decarbonyla-
tion, while also being able to move away to allow approach
of substrate (e.g., Scheme 1).21,26 This produces an efficient
intermolecular catalyst that can couple β-S-substituted alde-
hydes and unactivated alkenes such as 1-octene.
Results
A study of the hydroacylation reaction using rhodium
catalysts requires a suitable Rh(I) precursor entry point. A
library of these are readily generated by the synthesis of the
corresponding cationic norbornadiene adducts partnered
with the weakly coordinating anion [BArF4]- (ArF = C6H3-
(CF3)2). Chart 1 shows the phosphines used, and Scheme 2
summarizes the complexes made. These are all isolated as
crystalline materials in good yield. Complexes 2 and 4-7
have previously been reported as different salts.4,28 Care
has to be taken in the synthesis of [Rh(L4)(nbd)][BArF4] (4,
nbd = norbornadiene) to avoid the contaminant [Rh(L4)2]-
[BArF4], and a revised procedure allows for this (Experi-
mental Section). 31P{1H} NMR, 1H NMR, and ESI-MS
spectra of these norbornadiene complexes are in full accord
with their formulation as Rh(I) pseudo-square-planar com-
plexes. Notably single 31P environments showing coupling
to 103Rh consistent with the Rh(I) center [e.g., 3 δ 9.74,
J(RhP) 142 Hz] are observed for all species. Active precata-
lysts can be accessed via addition of H2 to these norborna-
diene adducts in acetone solution.29 We discuss first those com-
plexes formed with the POP0 (L2) and Xantphos (L3) ligands.
In this contribution we build upon these results by describ-
ing the effect that changing the hemilabile ligand has on the
rate and selectivity of the hydroacylation reaction between the
β-S-substituted aldehyde, HCOCH2CH2SMe I, and methyl
acrylate II to afford the coupled product III (eq 2). We find
that when using the ligands Xantphos (9,9-dimethyl-4,5-bis-
(diphenylphosphino)xanthene)25 or (POP0) ([Ph2P(CH2)2]2O),
(20) Benhamou, L.; Cesar, V.; Lugan, N.; Lavigne, G. Organometal-
lics 2007, 26, 4673–4676.
(21) Moxham, G. L.; Randell-Sly, H.; Brayshaw, S. K.; Weller, A. S.;
Willis, M. C. Chem.;Eur. J. 2008, 14, 8383–8397.
(22) Hyatt, I. F. D.; Anderson, H. K.; Morehead, A. T.; Sargent, A.
L. Organometallics 2008, 27, 135–147.
(23) Tanaka, K.; Shibata, Y.; Suda, T.; Hagiwara, Y.; Hirano, M.
Org. Lett. 2007, 9, 1215–1218.
(24) Braunstein, P.; Naud, F. Angew. Chem., Int. Ed. 2001, 40, 680–
699.
(25) Kranenburg, M.; Vanderburgt, Y. E. M.; Kamer, P. C. J.;
Vanleeuwen, P.; Goubitz, K.; Fraanje, J. Organometallics 1995, 14,
3081–3089.
(26) Moxham, G. L.; Randell-Sly, H. E.; Brayshaw, S. K.; Wood-
ward, R. L.; Weller, A. S.; Willis, M. C. Angew. Chem., Int. Ed. 2006, 45,
7618–7622.
(27) Crimmin, M. R.; Barrett, A. G. M.; Hill, M. S.; Procopiou, P. A.
Org. Lett. 2007, 9, 331–333.
(28) Brown, J. M.; Chaloner, P. A.; Kent, A. G.; Murrer, B. A.;
Nicholson, P. N.; Parker, D.; Sidebottom, P. J. J. Organomet. Chem.
1981, 216, 263–276.
(29) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98, 2134–
2143.

Article
Organometallics, Vol. 29, No. 7, 2010 1719
Scheme 2. Synthesis of the Rh(I) Norbornadiene Precursors
Scheme 3. Sequential Reaction of 2 with H2 (8), Alkene (10), and Aldehyde (11) or MeCN (9)a
a [BArF - anions are not shown.
]
4
Addition of H2 (1-4 atm) to an acetone solution of 2
results in the hydrogenation of the strained norbornadiene
ligand and the formation of the Rh(III) hydrido complex
[Rh(L2)(H)2(acetone)][BArF4] (8) (Scheme 3), which was
characterized by NMR spectroscopy and ESI-MS. The
associated with hydride ligands trans to a vacant site.31 These
chemical shifts are similar to those reported for related six-
coordinate pincer systems, such as [(PCP-iPr)Rh(H)(ace-
tone)2][BF4], δ -21.29 [PCP-iPr = 1,3-bis(di-isopropylphos-
phinomethyl)benzene].32 We also discount an alternative
structure for 8 that invokes trans-hydrides (with the POP0 ligand
either cis or trans spanning), as such species might be expected to
show chemical shifts for the hydrides that are much less
shielded,33 e.g., trans-[(PNP)Ir(CO)(H)2][PF6] (δ -6.63)34 and
trans-[iPr-PONOP)Ru(H)2(PPh3] (δ -6.09)35 [PNP = 2,6-bis-
1
room-temperature H NMR spectrum of 8 shows a broad,
relative integral 2H, peak at δ -19.84 assigned to the hydride
ligands; while there is a single 31P environment (δ 48.4)
showing coupling to 103Rh [J(RhP) 121 Hz] in the 31P{1H}
NMR spectrum. On cooling (225 K), the hydride peak splits
into two multiplets in a 1:1 ratio, indicating inequivalent
hydride environments, viz., δ -19.25 and -20.27. The coupl-
ing in these signals is best described as an overlapping
doublet of triplets of doublets and has been simulated
i
(di-tert-butylphosphinomethyl)pyridine, Pr-PONOP = C5H3-
N-1,3-(OPiPr2)2]. The observed fluxionality at room tempe-
rature is likely due to a mechanism that passes through a
five-coordinate intermediate, via loss of acetone, that makes
the hydride ligands equivalent on the NMR time scale. ESI-MS
(electrospray ionization mass spectrometry) shows a peak at
m/z = 605.1 (calcd 605.1), which corresponds to the suggested
formulation, and a major peak at m/z = 547.1 identified as [M]þ
- acetone (calcd 547.1). The methanol adduct analogous to 8,
[Rh(L2)(H)2(MeOH)][BF4], has been reported by Brown36 and
has very similar spectroscopic characteristics. The isolation of a
1
successfully and also probed using H{31P} decoupling ex-
periments, in which the hydride signals collapse into two
doublets of doublets, demonstrating coupling to two equiva-
lent 31P nuclei. The size of the coupling to the phosphines and
the other hydride places them all mutually cis [J(PH) ∼14,
J(HH) 11 Hz]. The 31P{1H} NMR spectrum at this tempera-
ture shows one phosphorus environment, δ 49.0, showing
coupling to 103Rh [121 Hz]. These data suggest a likely
structure for 8 that has a trans-spanning POP0 ligand,30 cis
hydrides, and the sixth coordination site occupied by acet-
one. The low-temperature data discount a five-coordinate
pseudo-trigonal-bipyramidal structure with equivalent hy-
dride ligands. It also argues against a five-coordinate square-
based pyramid with a vacant site trans to the hydride, as the
hydride resonances do not show distinctive high-field shifts
(31) Cooper, A. C.; Clot, E.; Huffman, J. C.; Streib, W. E.; Maseras,
F.; Eisenstein, O.; Caulton, K. G. J. Am. Chem. Soc. 1999, 121, 97–106.
(32) Montag, M.; Schwartsburd, L.; Cohen, R.; Leitus, G.; Ben-
David, Y.; Martin, J. M. L.; Milstein, D. Angew. Chem., Int. Ed. 2007,
46, 1901–1904.
(33) Grellier, M.; Vendier, L.; Sabo-Etienne, S. Angew. Chem., Int.
Ed. 2007, 46, 2613-2615, and references therein.
(34) Kloek, S. M.; Heinekey, D. M.; Goldberg, K. I. Organometallics
2006, 25, 3007–3011.
(35) Salem, H.; Shimon, L. J. W.; Diskin-Posner, Y.; Leitus, G.; Ben-
David, Y.; Milstein, D. Organometallics 2009, 28, 4791–4806.
(36) Brown, J. M.; Cook, S. J.; Kent, A. G. Tetrahedron 1986, 42,
5097–5104.
(30) Bessel, C. A.; Aggarwal, P.; Marschilok, A. C.; Takeuchi, K. J.
Chem. Rev. 2001, 101, 1031–1066.

1720 Organometallics, Vol. 29, No. 7, 2010
Pawley et al.
Figure 2. Molecular structure of the cationic portion of [Rh(L2)-
(COCH2CH2SMe)H][BArF4], 11. The anion and hydrogen atoms,
apart from H10, are omitted for clarity. Ellipsoids are drawn at
Figure 1. Molecular structure of the cationic portion of [Rh-
(L2)(H)2(NCMe)][BArF4], 9. Ellipsoids are drawn at the 50% pro-
bability level. The anion and hydrogen atoms, apart from H0A and
˚
the 50% probability level. Selected bond lengths (A): Rh1-H10,
1.51(2); Rh1-C1, 1.969(2); Rh1-P2, 2.2938(4); Rh1-P1,
2.2971(5); Rh1-O1, 2.302(1); Rh1-S1, 2.4369(4). Selected bond
angles (deg): P2-Rh1-P1, 156.21(2); C1-Rh1-S1, 87.57(6);
H10-Rh1-S1, 174.5(9); C1-Rh1-O1, 177.17(6). C1-S1-
Rh1-O1 torsion, 88.4(2)°.
˚
H0B, are omitted for clarity. Selected bond lengths (A): Rh1-P1,
2.2765(10); Rh1-P2, 2.2778(10); Rh1-O1, 2.232(3); Rh1-N1,
2.134(4). Selected bond angles (deg): P1-Rh1-P2, 164.95(4);
N1-Rh1-O1, 89.97(12).
Rh(III) dihydride species is in contrast to the Rh(I) bis-acetone
adduct that is formed with the DPEphos ligand.21 Addition of
MeCN to acetone solutions of 8 afforded the acetonitrile adduct
Chart 2
[Rh(L2)(H)2(NCMe)][BArF ], 9, which was characterized by
4
NMR spectroscopy and ESI-MS. 1Hand31P{1H} NMR spectra
for 9 are very similar to those of 8 observed at 225 K, showing
that the fluxional process observed in 8 is acetone dissociation:
the more strongly bound MeCN in 9 is reluctant to dissociate.
The solid-state structure of 9 has also been established by X-ray
crystallography, Figure 1, and this shows it to have a meridinal
POP0 ligand, cis hydrides, and a coordinated MeCN ligand, as
suggested by the NMR data for the solution structure of the
acetone adduct 8.
Addition of aldehyde I to an acetone solution of 10,
generated in situ, affords the product of aldehyde oxidative
addition, [Rh(L2)(COCH2CH2SMe)H][BArF4], 11, in essen-
tially quantitative yield by NMR spectroscopy. The solid-
state structure of 11 is shown in Figure 2, which shows a
Rh(III) species with a trans-spanning POP0 ligand and the
hydride and thioether ligands mutually trans-orientated. It is
very closely related to that of 1b, [Rh(DPEphos)(COC6-
H4SMe)H][CB11H6Cl6] (Chart 2).21 The phosphorus atoms
sit approximately trans to one another [P1-Rh1-P1
156.21(2)°], similar to 1b [155.93(5)°]. The Rh-O distance
A suitable Rh(I) precursor that can enter the catalytic cycle
comes from adding excess methyl acrylate, II, to an acetone
solution of 8. This results in the transfer of the hydrides to the
unsaturated substrate, coordination of another equivalent of
alkene, and the generation in situ of a complex spectroscopically
characterized by 1H and 31P{1H} NMR spectroscopy and ESI-
MS as the Rh(I) alkene adduct [Rh(L2)(η2-H2CdCHCO2-
Me)][BArF ], 10. The31P{1H} NMR spectrum shows a doublet
4
1
at δ 39.1 [J(RhP) 129 Hz], while in the H NMR spectrum
˚
broad signals attributable to bound olefin are observed at δ
3.60-2.64, which are similar to those reported for (triphos)Rh-
(H)(η2-H2CdCHCO2Me).37 Excess methyl acrylate is obser-
ved as broadened signals (δ 6.41-5.82 and 3.70), this suggest-
ing an exchange between both bound and free alkene, although
we have not attempted to determine a rate constant for this.
ESI-MS shows a parent ion at m/z = 631.1 (calcd 631.1), con-
firming the empirical formula. MS/MS interrogation of this peak
forces the loss of methyl acrylate, giving an ion at m/z = 545.1
(calcd 545.1), further supporting the suggested formulation.
in 11 is longer than that in 1b [2.3017(12) versus 2.248(3) A,
respectively], although, as noted later, the ether ligand in 11
is not displaced by incoming ligands (e.g., MeCN), unlike 1a/
b, which readily form the corresponding MeCN adducts. The
reasons for this difference in reactivity are not clear, given the
trend in the bond lengths, but they might reflect the differ-
ence in backbone ligand architecture between POP0 and
DPEphos. The SMe group is canted to one side, making
the two phosphorus atoms chemically inequivalent. This is as
observed for 1b in the solid state and for 1a (Scheme 1) from
low-temperature solution NMR data.
The room-temperature 31P{1H} NMR spectrum of 11
shows a broad signal at δ 31.1. A single hydride environment
(37) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Bachechi, F.
Organometallics 1991, 10, 820–823.

Article
Organometallics, Vol. 29, No. 7, 2010 1721
Scheme 4. Addition of H2 to 3a
a [BArF - anions are not shown.
]
4
at δ -9.58 is observed as a broadened doublet [J(RhH) 24
Hz], while the 1H-1H COSY NMR spectrum shows a
correlation between the hydride and the SMe group (δ
1.79); the latter appears as a doublet. The single resonance
in the 31P{1H} NMR spectrum splits into a tightly coupled
ABX spectrum on cooling to 225 K, with two resonances
observed at δ 34.6 and 28.5 that show trans 31P-31P cou-
pling, J(PP) 309 Hz, as well as coupling to 103Rh, consistent
with a Rh(III) center, viz., 128, 133 Hz. We have previously
observed similar fluxional behavior in solution for 1a and 1b,
which was suggested to be due to an inversion of stereo-
chemistry at the thioether, resulting in a flip of the methyl
group from side to side via a sp2-hybridized at sulfur inter-
mediate.21 The coupling between the methyl and hydride
groups observed at room temperature shows that the Rh-S
bond is not broken in this fluxional process.
Complex 11 is stable toward decarbonylation, remaining
unchanged in d6-acetone solution over a period of 1 week as
monitored by NMR spectroscopy and ESI-MS. This is in
contrast to 1a, which reductively decarbonylates over the
same time scale to give [Rh(L1)(CO)(EtSMe)][CB11H6Cl6],
while [Rh(L4)(COCH2CH2SMe2)(H)]þ cannot be isolated,
as decarbonylation is rapid.21 The stability of 11 is no doubt
related to the tight binding of the POP0 ligand, which does
not decoordinate the ether linkage and thus does not reveal
the cis vacant site necessary for migratory decarbonylation. This
tight binding is further established by the fact that addition of
excess MeCN (100 equiv) to an acetone solution of 11 does not
displace the ether ligand, to the detection limits of a 31P{1H}
NMR experiment. This is in contrast to 1a/b, which immediately
form the MeCN adducts, alongside concomitant decoordination
of the hemilabile ether. Similarly, addition of methyl acrylate to
11 did not result in reaction, again in contrast to 1a/b, which
proceed to give the product of hydroacylation.21
Figure 3. Molecular structure of the cationic portion of
[Rh(L3)(H)2(NCMe)][BArF4], 13. Ellipsoids are drawn at the
50% probability level. The anion and hydrogen atoms, apart
from H0A and H0B, are omitted for clarity. Selected bond
lengths (A): Rh1-P1, 2.2748(8); Rh1-P2, 2.2584(8); Rh1-O1,
2.233(2); Rh1-N1, 2.128(3). Selected bond angles (deg):
P1-Rh1-P2, 164.31(3); N1-Rh1-O1, 95.98(9).
˚
temperature for 12, in contrast to the one seen for 8, suggests a
larger barrier to the fluxional process that makes the hydrides
equivalent for the latter. Similar to 8 addition of MeCN to
acetone solutions of 12 results in the formation of an acetonitrile
adduct, [Rh(L3)(H)2(NCMe)][BArF ] (13), and a sharpening of
4
the hydride resonances, consistent with the cessation of the
1
fluxional processes. 13 has been characterized by H and 31P
NMR spectroscopy and ESI-MS and shows similar solution data
to 9. The solid-state structure of 13 has also been established by
X-ray crystallography (Figure 3), and this shows it to have a
trans-spanning38-40 Xantphos ligand, cis hydrides, and a coordi-
nated MeCN ligand. This geometry is also as suggested by NMR
data for the solution structures of the acetone adducts 8 and 12.
Addition of excess methyl acrylate II to acetone solutions
of 12 gave a material in solution that was tentatively char-
acterized by 1H NMR, 31P{1H} NMR, and ESI-MS as [Rh-
(L3)(η2-H2CdCHCOOMe)][BArF4], 14, by comparison
with the solution analytical data for 10. Addition of aldehyde
I to acetone solutions of 14 formed in this way afforded the
product of oxidative addition of aldehyde to a Rh(I) center,
Similar to the formation of 8, hydrogenation of the Xant-
phos (L3) precursor 3 in acetone results in the formation of
dihydrido species [Rh(L3)(H)2(acetone)][BArF ], 12 (Scheme 4).
4
The NMR data for 12 are similar to 8, and we thus assign a
similar structure in which the ligand is proposed to be trans-
spanning coordinating through the central oxygen with cis
hydrides. A single 31P environment is observed in the room-
temperature 31P{1H} NMR spectrum, δ 45.4 [J(RhP) 121 Hz],
while in the 1H NMR spectrum two broad, relative integral 1
H, hydride environments, δ -18.81 and -19.00, are observed.
One broad methyl signal is also observed for the Xantphos
ligand. Cooling to 225 K resolves the hydride signals into two
sharp multiplets at approximately the same chemical shift as at
room temperature, which show mutually cis coupling to the
other hydride and 31P as well as coupling to 103Rh [doublet
of triplets of doublets]. The 31P{1H} NMR spectrum remains
unchanged at this temperature. ESI-MS confirms the empiri-
cal formulation (m/z = 741.15, calcd 741.15). The observa-
tion of two, albeit broad, hydride environments at room
(38) 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, 872–883.
(39) Sandee, A. J.; van der Veen, L. A.; Reek, J. N. H.; Kamer, P. C.
J.; Lutz, M.; Spek, A. L.; van Leeuwen, P. Angew. Chem., Int. Ed. 1999,
38, 3231–3235.
(40) 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. J.
Chem. Soc., Dalton Trans. 2002, 2308–2317.

1722 Organometallics, Vol. 29, No. 7, 2010
Pawley et al.
Scheme 5. Synthesis of Complex 15a
a [BArF - anion is not shown.
]
4
broad, integral 2 H signals at δ 2.62 and 2.04, and the Xantphos
methyl groups are observed as two sharp, integral 3 H, singlets
at δ 1.97 and 1.51. The S-Me group is observed at δ 1.47 as an
unresolved, integral 3 H, multiplet that also shows a weak
correlation to the hydride signal at δ -8.72 in a 1H/1H COSY
experiment. The 31P{1H} NMR spectrum shows a broad signal
at δ 32.8. These data indicate a fluxional process that makes
equivalent the phosphorus atoms but not the Xantphos methyl
groups. As for 11 and 1a/b, a process that involves inversion at
sulfur is most likely. Cooling to 257 K slows this process so that
two sharp phosphorus environments are now observed in the
31P{1H} NMR spectrum that show an ABX coupling pattern
with trans 31P-31P coupling [J(PP) 300 Hz]. At low temperature
the hydride resonance is essentially unchanged although slightly
better resolved than at room temperature. Four separate reso-
nances for the methylene protons of the aldehyde backbone are
also observed, consistent with a static structure. As found for 11,
complex 15 is stable toward decarbonylation over 1 week (by
ESI-MS and NMR spectroscopy), while addition of MeCN
does not displace the ether group, both observations demon-
strating a tight binding of the trans-spanning ether ligand.
Complexes 11 and 15 do not react with methyl acrylate, II,
remaining unchanged over 24 h on addition of a 10-fold-
excess, as measured by NMR spectroscopy and ESI-MS.
This is in direct contrast to 1a/b, which react rapidly to give
the product of hydroacylation.21 This lack of reactivity is in
line with the unavailability of the vacant site due to the strong
binding of the ether linkage in the POP0 or Xantphos ligands.
The norbornadiene precursor materials (2 and 3) were also
screened in the catalysis of the hydroacylation reaction
between aldehyde I and methyl acrylate II, after activation
with H2 (i.e., to form complexes 8 and 12, respectively, in
situ). As Table 1 shows (entries 2 and 3), these species are very
poor catalysts for the hydroacylation reaction (product III),
and over an extended time (∼150 h) the product of aldehyde
hydroacylation, IV (formally a Tishchenko reaction27,42), is
also formed, interestingly in a greater proportion than the
alkene hydroacylation product. This overall low reactivity is
consistent the attenuated-hemilabile nature of the phosphine
ligands L2 and L3 and a resulting reactivity impasse at the
acyl-hydride product.
Figure 4. Molecular structure of the cationic portion of [Rh(L3)-
(COCH2CH2SMe)H][BArF4], 15. Major disordered component
shown. The anion and hydrogen atoms, apart from H0, are omit-
ted for clarity. Ellipsoids are drawn at the 50% probability level.
˚
Selected bond lengths (A): Rh1-H0 1.58(2), Rh1-C1, 1.981(3);
Rh1-P2, 2.2841(8); Rh1-P1, 2.2896(8); Rh1-O1, 2.308(2);
Rh1-S1, 2.4431(9). Selected bond angles (deg): P2-Rh1-P1,
157.60(3); C1-Rh1-S1, 86.60(11); H0-Rh1-S1, 180(2);
C1-Rh1-O1, 176.31(12). C4-S-Rh1-O1 torsion, 121.3(8)°.
[Rh(L3)(COCH2CH2SMe)H][BArF ], 15, in essentially quan-
4
titative yield, which is directly comparable to 11 and 1a/b
(Scheme 5). The solid-state structure of 15 is shown in Figure
4, and this shows a trans-spanning Xantphos ligand coordi-
nated to a Rh(III) acyl hydride. The thio-ether group is
disordered over two sites, related by a small change in the twist
of the S-Me group and a slight conformational change in the
ethyl backbone. This was modeled successfully, and only the
major component is shown. A related complex, Ir(L3)(CO)-
(H)2(COEt), with a cis-coordinated Xantphos ligand has re-
cently been reported.41 The stereochemistry around the metal
center in 15 is approximately the same as observed for 1b and 11
Given that ligands L2 and L3 did not readily reveal a vacant
site, we have explored the reactivity of a ligand that would do
this but give similar electronic and steric profiles otherwise:
Ph2P(CH2)5PPh2 (L7). As a further point of reference, we also
prepared the complexes with the common bidentate phosphines
Ph2P(CH2)nPPh2 (n = 2-4, L4-L6, Chart 1) and [Rh(Ph2P-
˚
in the solid state, the Rh-O distance being 2.308(2) A and the
P-Rh-P angle measured as 157.60(3)°. It is thus no surprise to
find that the solution NMR data are very similar to these two
1
complexes. In particular in the room-temperature H NMR
(CH2)nPPh2)(nbd)][BArF ], 4-7 (Scheme 2). These complexes
4
spectrum a broad doublet is observed at δ -8.72 [J(RhH) 23
Hz], the methylene protons of the aldehyde are observed as two,
have been prepared previously as alternate salts.4,28 Addition of
H2 (d6-acetone solution) afforded the previously reported bis-
acetone adducts [Rh(Ph2P(CH2)nPPh2)(acetone)2][BArF4],
(41) Fox, D. J.; Duckett, S. B.; Flaschenriem, C.; Brennessel, W. W.;
Schneider, J.; Gunay, A.; Eisenberg, R. Inorg. Chem. 2006, 45, 7197–
7209.
(42) Seki, T.; Nakajo, T.; Onaka, M. Chem. Lett. 2006, 35, 824–829.

Article
Organometallics, Vol. 29, No. 7, 2010 1723
Table 1. Comparison of the Catalysts Reported in This Study in the Hydroacylation Reaction of I and IIa
initial rate
for the
initial rate
for the
% alkene % aldehyde
hydroacylation hydroacylation total conversion
formation of III ꢁ10-3 formation of IV ꢁ10-3
entry ligand temp (°C)
III
IV
of aldehyde
100
time (h)
(mol/L/h)c
(mol/L/h)c
1
2
3
4
5
6
7
8
L1
L2
L3
L4
L5
L6
L7
L7
L7
L7
22
22
22
22
22
22
22
40
55
18
93
6
7
44
20
8
20
165
145
60
100
8
96(6)
10(7)
2(2)
30(2)
21(1)
3(1)
38(12)
2.0(4)
1.0(3)
25.0(3)
136(9)
50d
30d
99d
100
100
100
100
100
76
10
91
38
50
13
17
19
30
62
50
87
83
81
46
134(10)
14(7)
6
130(2)
e
e
0.5
0.1
15
e
e
9
10b
76(8)
140(20)
a Conditions: all manipulations under an argon atmosphere, 0.043 M solution of precatalyst in d6-acetone, hydrogenated using standard procedures,
catalyst:aldehyde ratio = 1:10, substrate ratio = 1:1 aldehyde:alkene, conversion determined by 1H NMR spectroscopy. b Substrate ratio 1:5 aldehyde:
alkene. c Initial rate determined from an exponential fit to the data at t = 0. d Total conversion at given time. e Reaction almost complete before a reliable
first data point can be taken to determine an initial rate.
Scheme 6. Synthesis of Complexes 14 to 17a
a [BArF - anions are not shown.
]
4
16-19,4 Scheme 6. Surprisingly, to our knowledge, the
adduct with Ph2P(CH2)5PPh2 has not been reported, and we
find it exists as a mixture of the Rh(I) bis-acetone complex
of 19b, perhaps suggesting a slow equilibrium between the three.
Similar complexes to 19 have been reported previously as the
methanol adducts and show very similar spectroscopic data.28
These complexes are also in equilibrium with one another. The
observation of dihydrido species on hydrogenation of rhodium-
(I) chelating phosphines with increasing bite angle of the phos-
phine (cf. L4 and L7) has also been noted previously.43,44
Addition of aldehyde I to 16-19 did not afford species that
were readily characterized, with multiple species observed by
31P{1H} NMR spectroscopy that turned into uncharacterized
decomposition products over time. This is as previously repor-
ted by Bosnich for some of these systems with 4-pentenals.3
However for 19 (L7) ESI-MS showed a relatively clean spec-
trum with a strong signal corresponding to the molecular ion
[Rh(L7)(HCOCH2CH2SMe)2]þ (m/z = 751.15, calcd 751.15).
Efforts to isolate this species lead to intractable material. These
observations are in contrast to 1, 11, and 15 and reflect the
controlling influence of ligands L1, L2, and L3, respectively,
on both the stereochemical rigidity and overall stability of the
Rh(III) acyl hydride complexes. Importantly, under these
conditions 19b/c were not observed by ESI-MS or NMR
[Rh(L7)(acetone)2][BArF ] (19a) and two Rh(III) hydrido spe-
4
cies, which are tentatively characterized as the C-H activated
complex [Rh(PPh2CH2CH2CHCH2CH2PPh2)(H)(acetone)2]-
[BArF ] (19b) and the dihydride acetone adduct [Rh-
4
(L7)(H)2(acetone)2][BArF ] (19c) (Scheme 6). Complex 19b
4
shows a hydride resonance in the 1H NMR spectrum at
δ -21.68 [J(PH) 16, J(RhH) 27 Hz] as a doublet of triplets that
collapses into a doublet on decoupling 31P, while 19c shows a
doublet of triplets at δ -22.30 [J(PH) 11, J(RhH) 35 Hz] that
collapses into broad doublet on decoupling 31P. The relative
integral of these signals compared to the aromatic region (32 H)
is 0.20 H (19b) and 0.14 H (19c). Three environments are also
observed in the 31P{1H} NMR spectrum, δ37.3 [d, J(RhP) 201],
35.8 [d, J(RhP) 115 Hz], and 55.3 [d, J(RhP) 133 Hz] in a
10:2:0.8 ratio (19a, 19b, and 19c, respectively, as determined by
1H{31P-selective} experiments). ESI-MS shows a signal due
to the [Rh(L7)(acetone)2][BArF ] (m/z = 659.2, calcd 659.2)
4
fragment. 19a and 19b have the same empirical formula (m/z
659.2), while the signals for the minor component 19c would be
masked by the isotope pattern for 19a/b. Cooling a d6-acetone
solution of this mixture results in no significant change in
chemical shifts, although the ratios do change slightly in favor
(43) DuBois, D. L.; Blake, D. M.; Miedaner, A.; Curtis, C. J.;
DuBois, M. R.; Franz, J. A.; Linehan, J. C. Organometallics 2006, 25,
4414-4419, and references therein.
(44) James, B. R.; Mahajan, D. Can. J. Chem.: Rev. Can. Chim. 1980,
58, 996–1004.

1724 Organometallics, Vol. 29, No. 7, 2010
Pawley et al.
spectroscopy, suggesting that they react with aldehyde to give
the same product as 19a. This would be consistent with an
equilibrium between the three, as suggested.
ring size, and in particular the bite angle, of the ligand can
affect the rate of reaction and selectivity in many catalytic
processes. Efforts to put these influences of the ligand on a
quantitative footing have led to a number of closely related
concepts. A large natural bite angle of a ligand can lead to
enhanced rates of reductive elimination or selectivity, as
exemplified in hydroformylation or hydrocyanation re-
ations.25,38,47-49 Barron and co-workers introduced the con-
cept of pocket angle,50 which relates the interior cone angle of
a ligand (as influenced by the bite angle) to relative catalytic
activity, with an optimum size of the pocket angle deter-
mined by a balance of accessibility of the metal center by the
substrate with catalyst decomposition. The concept of ac-
cessible molecular surface (AMS), as developed by Leitner
and co-workers,51 provides a further refinement on the steric
control of a chelating phosphine on catalysis in {Rh(L2)}þ
fragments (L = chelating phosphine), highlighting the in-
creased rate of reductive elimination of product associated
with bulky ligands with large bite angles. In many of these
studies the ligands L2 to L6 are compared, providing useful
data for our study reported here. When bound to a metal
center, ligand L4 has a smaller bite angle (84°), larger pocket
angle (132°), and larger accessible molecular surface (7.578
A ) than those estimated for L6 (99°, 94°, ∼4.6 A ). Ligand
L7 would be expected to show a similar steric profile to L6.52
Consideration of these factors in concert with a likely3
hydroacylation catalytic cycle that starts with addition of
aldehyde I to precatalysts 16 to 19 would initially give the
acyl-hydrido species A (Scheme 7), as observed for 1a/b, 11,
and 15 and calculated by Morehead and Sargent for the
intramolecular hydroacylation reaction of 4-pentenal by
{Rh(L4)}þ fragments.22 Bifurcation of the pathway can
now potentially occur: coordination of alkene results in
alkene hydroacylation via intermediate B, while coordina-
tion of aldehyde via D gives aldehyde hydroacylation (the
Tishchenko reaction27). A ligand that reveals an open metal
site (e.g., L4, large AMS, large pocket angle, small bite angle)
would allow η2-alkene coordination (B). Importantly it also
allows for a necessary coplanar orientation of alkene and
hydride required for insertion. Aldehyde can also coordi-
nate, but clearly alkene is preferred in this instance. With a
bulkier ligand (e.g., L7, small AMS, small pocket angle, large
bite angle) we suggest that alkene coordination is disfavored,
as is the required orientation for insertion.50 By contrast,
aldehydes can coordinate η1, which is sterically less demand-
ing. Moreover, this more flexible coordination mode might
be expected to facilitate insertion of the hydride, a process
closely related to that suggested to occur in carbonyl hydro-
genation by late-transition-metal hydrides (Chart 3).53
Complexes 16 to 19 (generated in situ by addition of H2 to
the nbd precursors) were screened as catalysts in the hydro-
acylation reaction between aldehyde I and methyl acrylate II
at 10 mol % loading. Willis and co-workers have successfully
used {Rh(L4)}þ, generated in situ, at higher temperature
(60 °C) with aldehyde I and methyl acrylate II in intermole-
cular hydroacylation,15,21,45 building upon Bosnich’s early
report of its use in the intramolecular hydroacylation of
4-pentenal.4 Fu has reported on {Rh(L4)}þ in the intramole-
cular hydroacylation of alkynes.46 Ligand L5 has also been
used by a number of research groups in intramolecular
hydroacylation reactions that invoke {Rh(L5)}þ fragments
as active catalysts.19,23
Table 1 lists the results of this catalyst screening and also
compares these results with those from use of ligands L1, L2,
and L3. Ligand L4 under these conditions promotes com-
plete conversion of the aldehyde over 60 h, but, interestingly,
an approximate ratio of 9:1 of alkene hydroacylation, III, to
aldehyde hydroacylation, IV, is observed by 1H NMR
spectroscopy (entry 4). Ligand L4 has previously been
reported to be effective for selective alkene hydroacylation
at 55 °C over 2 h using these substrates, and the longer times
for complete conversion under these conditions reported
here are consistent with the lower temperature.26 These
higher temperatures also favor the formation of III. Repeat-
ing the reaction at 55 °C resulted in a shorter reaction time
(2 h) and complete conversion to III. These conditions reported
previously also use an excess of alkene, which would be
expected to promote alkene hydroacylation. Unexpectedly,
use of ligand L7 results in a switch in product distribution,
giving the aldehyde hydroacylation product IV as the major
product in 87% conversion (entry 7). Heating to 40 and 55 °C
using the catalyst derived from L7 results in a gradual
increase in alkene hydroacylation product III (entries 8
and 9), while use of a 5-fold excess of alkene also results in
an increase in the yield of the alkene hydroacylation product
(entry 10), although the rate of catalysis drops. Bosnich has
previously reported substrate inhibition in intermolecular
alkene hydoacylation,3 which in our case presumably arises
from competition for a vacant site prior to oxidative addition
of the aldehyde. When using ligands L5 and L6, products III
and IV are observed in more equal proportions (entries 5
and 6). These data show that gradually increasing the length
of the ligand backbone results in the products of aldehyde
hydroacylation being favored. Figure 5 shows time/conver-
sion plots of alkene versus aldehyde hydroacylation for these
systems. For the systems and conditions reported here ligand
L1 performs the best in terms of selectivity and rate for
alkene hydroacylation. This mirrors our previously reported
results for this system at higher temperature (55 °C).21
Why does the reactivity change as the ligand backbone
length (L4 to L7) is changed? As noted, ligands L2 and L3
(entries 2 and 3) in complexes 11 and 15 do not move aside on
approach of a relatively strong ligand such as acetonitrile or
methyl acrylate, and thus the modest reactivity with alkene II
is perhaps no surprise. It is also well established that chelate
2
2
˚
˚
(47) Casey, C. P.; Whiteker, G. T. Isr. J. Chem. 1990, 30, 299–304.
(48) van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Reek, J. N. H.;
Dierkes, P. Chem. Rev. 2000, 100, 2741–2770.
(49) Birkholz, M. N.; Freixa, Z.; van Leeuwen, P. Chem. Soc. Rev.
2009, 38, 1099–1118.
(50) Koide, Y.; Bott, S. G.; Barron, A. R. Organometallics 1996, 15,
2213–2226.
(51) Angermund, K.; Baumann, W.; Dinjus, E.; Fornika, R.; Gorls,
H.; Kessler, M.; Kruger, C.; Leitner, W.; Lutz, F. Chem.;Eur. J. 1997,
3, 755–764.
(52) The AMS for L7 (Ph2P(CH2)5PPh2) has not been reported.
However those for Ph2P(CH2)4PPh2 and Ph2P(CH2)6PPh2 are both
2
4.66 A (ref 51), suggesting a similar AMS value for L7. Similarly bite
˚
(45) Willis, M. C.; Randell-Sly, H. E.; Woodward, R. L.; Currie, G. S.
Org. Lett. 2005, 7, 2249–2251.
(46) Tanaka, K.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 11492–
11493.
angles for L7 and pocket angles have not been reported, but might be
expected to be the same as for L6.
(53) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201–2237.

Article
Organometallics, Vol. 29, No. 7, 2010 1725
Figure 5. Relative yields of alkene hydroacylation (III, circles) and aldehyde hydroacylation (IV, squares) products using various
catalyst precursors. See Table 1 for more details. Reactions carried out at room temperature, 0.043 mmol/mL of catalyst in d6-acetone,
10 mol % catalyst loading, conversion of aldehyde I to III and IV products followed by 1H NMR spectroscopy. Plotted lines are for
comparison only.
Indeed, Dong and co-workers have recently used DFT
calculations to show that a carbonyl η1-intermediate is
formed prior to hydride insertion in intramolecular ketone
hydroacylation.19 Illustrating this point further, considera-
tion of the space-filling diagrams of the previously reported
structures RhI2(L4)(COMe)54 and RhI2(L6)(COMe),55
which have an acyl group trans to the vacant site and are
thus reasonable models for intermediate A, shows that the
metal surface is indeed less accessible to incoming ligands
for L6 than L4 (Figure 6). That there is an equilibrium
established between D, A and B is shown by the effects
of temperature (increased temperature favors B, entries 8
and 9, Table 1) and alkene concentration (increased con-
centrations favors B, entry 10, Table 1). Closely related
explanations for the selectivity are that alkene binding is fast
and irreversible;leading to alkene hydroacylation;for
small bite angle ligands and reversible for large bite angle
ligands with the reductive elimination favoring aldehyde
hydroacylation. Alternatively large bite angle ligands favor
reductive elimination, and thus insertion becomes rate-de-
termining, favoring aldehyde hydroacylation. For small bite
angle ligands reductive elimination is rate-determining, fa-
voring alkene hydroacylation. Finally, the tentatively as-
signed CH activation product 19b probably plays no part
in the catalysis, as there is a gradual change in catalytic
outcome on changing the ligand backbone length and
(54) Gonsalvi, L.; Adams, H.; Sunley, G. J.; Ditzel, E.; Haynes, A. J.
Am. Chem. Soc. 2002, 124, 13597–13612.
(55) Lamb, G.; Clarke, M.; Slawin, A. M. Z.; Williams, B.; Key, L.
Dalton Trans. 2007, 5582–5589.

1726 Organometallics, Vol. 29, No. 7, 2010
Pawley et al.
Scheme 7. Suggested Reaction Pathways for Aldehyde Hydroacylation and Alkene Hydroacylation
with the POP0 and Xantphos ligands is interesting, but the
low conversions and long reaction times make us reluctant to
comment on the implications this has on mechanism apart
from the lack of available vacant site clearly slows the rate
considerably.
Conclusions
These, and previous,13,21 results show that the overall
conversion in the hydroacylation reaction can be controlled
by the hemilabile nature of the chelating phosphine in the
catalyst (e.g., DPEphos versus Xantphos), while the course
of the reaction can also be tuned by changing the bite angle of
the chelating, but not hemilabile, phosphine. These results
encourage further catalyst development into controlling this
atom-efficient and useful disconnection in synthesis.
Figure 6. Space-filling diagram (van der Waals radii) of the
complexes RhI2(L4)(COMe)54 and RhI2(L6)(COMe)55 show-
ing the effect that changing the ligand backbone has on the
accessible surface of the metal center (colored gray). Data were
generated using the coordinates from the CCDC database
(IBOZOP and HISZOZ, respectively).
Experimental Section
Chart 3
All manipulations, unless otherwise stated, were performed
under an atmosphere of argon, using standard Schlenk-line and
glovebox techniques. Glassware was oven-dried at 130 °C over-
night and flamed under vacuum prior to use. CH2Cl2, MeCN,
and pentane were dried using a Grubbs-type solvent purification
system (MBraun SPS-800) and degassed by successive free-
ze-pump-thaw cycles.56 1,2-C6H4F2 and C6H5F were distilled
˚
under vacuum from CaH2 and stored over 3 A molecular sieves.
Acetone and d6-acetone were dried over B2O3 and vacuum
distilled twice.57 NMR spectra were recorded on Varian Unityþ
500 MHz, Varian Mercury 300 MHz, or Bruker AVC 500 MHz
spectrometers at room temperature, unless otherwise stated.
Chemical shifts are quoted in ppm and coupling constants
in Hz. ESI-MS were recorded on Bruker MicrOTOF and
Bruker MicroOTOF-Q instruments.58 The starting mate-
rials {Ph2P(CH2)2}2O (POP0),59 Na[BArF4],60 [Rh(nbd)Cl]2,61
ligands L4-L6 would not be expected to undergo CH
activation as easily, and we see no evidence for such species
by NMR spectroscopy.
Following these arguments, the DPEphos ligand, with its
large bite angle of 103°,48 might be expected to give the
product of aldehyde hydroacylation, IV, which is not the
case (entry 1). However, as previously demonstrated,21 the
DPEphos ligand can bind to the metal center in a tridentate
manner (i.e., 1a), and this extra coordination thus needs to be
displaced by an incoming ligand before hydride transfer can
occur (i.e., stabilizing intermediate A, Scheme 7). Acetone
does not do this (to give an observed product), but MeCN
and alkene II do (B, Scheme 7).21 Thus we suggest that the
hemilabile ligand DPEphos, L1, not only protects the vacant
site toward decarbonylation but also appears to enforce
selectivity in the reaction toward alkene hydroacylation.
The large natural bite angle of DPEphos will also contribute
to the rate of reductive elimination of the final product. The
observation of aldehyde and alkene hydroacylation products
(56) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518–1520.
(57) Burfield, D. R.; Smithers, R. H. J. Org. Chem. 2002, 43, 3966–
3968.
(58) Lubben, A. T.; McIndoe, J. S.; Weller, A. S. Organometallics
2008, 27, 3303–3306.
(59) Thewissen, D.; Timmer, K.; Noltes, J. G.; Marsman, J. W.;
Laine, R. M. Inorg. Chim. Acta 1985, 97, 143–150.
(60) Buschmann, W. E. M., J. S.; Bowman-James, K.; Miller, C. N. In
Inorganic Syntheses; Coucouvnis, D., Ed.; Wiley-Interscience: 2002; Vol. 33.
(61) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J. Organometallics 1995,
14, 3081–3089.

Article
Organometallics, Vol. 29, No. 7, 2010 1727
[Rh(COD)Cl]2 and [Rh(dppe)Cl]2;63 the catalyst precursors
[Rh(nbd)(L5)][BArF4],4 [Rh(nbd)(L6)][BArF4],4 [Rh(nbd)(L7)]-
[BArF4],28 [Rh(nbd)(DPEphos)][BArF4],21 and [Rh(nbd)(POP0)]-
[BArF4];36 and the bis-acetone complexes [Rh(L4)(acetone)2]-
[BArF ],4 [Rh(L7)(acetone)2][BArF4],4 [Rh(L6)(acetone)2][BArF ],4
Microanalysis for C78H52BF24OP2Rh (1636.9): calcd C 57.2, H
3.20; obsd C 57.5, H 3.19.
62
[Rh(L)(H)2(acetone)][BArF4] (L = POP0, 8; Xantphos, 12).
The title compounds were prepared in situ by placing a solu-
tion of [Rh(nbd)(L)][BArF4] (6.7 ꢁ 10-6 mol) in d6-acetone
(0.3 mL) under 4 atm of hydrogen (298/77 K = 3.9). These com-
pounds where characterized in situ by NMR spectroscopy and
ESI-MS.
4
4
and [Rh(DPEphos)(acetone)2][BArF
]
4
21 were all prepared by pub-
lished literature methods or variations thereof using Na[BArF4] as
the chloride extracting agent. Aldehyde I was purchased from
Aldrich and distilled before use. All other chemicals are commercial
products and were used as received. Hydroacylation products III
and IV were identified by comparison with previously reported
data.15 Elemental analyses for the new compounds, apart from 3,
were not satisfactorily obtained due to the fact that all crystallized
with small amounts of intractable oil. NMR spectra of all the new
compounds that could be isolated as solids are provided in the
Supporting Information.
[Rh(POP0)(H)2(acetone)][BArF4], 8. 1H NMR (500 MHz, d6-
acetone): δ 8.09-6.96 [m, 32H, Ar-H þ BArF4 {7.79 (8H, BArF4),
7.67 (4H, BArF4)}], 4.27 (br, 4H, O-CH2), 3.18 (br, 4H, P-CH2),
-19.84 (br, fwhm = 106 Hz, 2H, Rh-H) (coordinated d6-acetone
is not observed). 31P{1H} NMR (202 MHz, d6-acetone): δ 48.4 [d,
J(RhP) = 121]. 1H NMR (500 MHz, d6-acetone, 225 K): δ 8.02-
7.38 [m, 32H, Ar-H þ BArF {7.82 (8H, BArF4), 7.73 (4H,
4
BArF4)}], 4.35 (m, 2H, O-CH2), 04.16 (m, 2H, O-CH2 ), 3.42 (m,
0
Crystallography. Data were acquired on a Nonius Kappa
CCD diffractometer using graphite-monochromated Mo KR
2H, P-CH2), 2.94 (m, 2H, P-CH2 ), -19.25 [dtd, J(RhH) = 24.5,
J(PH) = 13.9, J(HH) = 10.7, 1H, Rh-H], -20.27 [dtd, J(RhH) =
24.3, J(PH) = 13.4, J(HH) = 10.7, 1H, Rh-H]. 31P{1H} NMR
(202 MHz, d6-acetone, 225 K): δ 49.0 [d, J(RhP) = 121]. ESI-MS
(acetone, 60 °C, 4.5 kV) positive ion: (minor peak) [M]þ m/z =
605.1 (calcd 605.1); (major peak) [M - acetone]þ m/z = 547.1
(calcd 547.1). ESI-MS (acetone, 60 °C, 4.5 kV) positive ion: m/z,
773.2 (Mþ calcd 773.2).
radiation (λ = 0.71073 A) and a low-temperature device;64 data
˚
were collected using COLLECT; reduction and cell refinement
were performed using DENZO/SCALEPACK.65 Structures
were solved by direct methods using DIRDIF99 (11)66 or
SIR2004 (9, 13, 15)67 and refined with full-matrix least-squares
on F2 using SHELXL-97.68 All non-hydrogen atoms were
refined anisotropically. The hydride ligands (H0a and H0b in
9 and 13, H10 in 11, H0 in 15) were located on the difference map
and freely refined; all other hydrogen atoms were placed in
calculated positions using the riding model. Disorder of the
phosphine ligand in 9 was treated by modeling one of the
substituents over two sites and restraining its geometry; this
included using rigid body restraints. Disorder of the thioether-
acyl ligand in 15 was treated by modeling part of it over two sites
and restraining its geometry. Disorder of the phosphine back-
bone in 15 was treated by modeling the Xantphos methyl groups
over two sites and restraining their geometry. Rotational dis-
order of the anion CF3 groups was treated by modeling the
fluorine atoms or the entire CF3 group over two or three sites
and restraining their geometry. Problematic solvent disorder in
9 and 15 was treated using the SQUEEZE algorithm.68 Re-
straints to thermal parameters were applied where necessary in
order to maintain sensible values. See Table S1 in the Supporting
Information for further crystallographic data.
[Rh(Xantphos)(H)2(acetone)][BArF4], 12. 1H NMR (500 MHz,
d6-acetone): δ 8.34-6.85 [m, 38H, Ar-H þ BArF {7.79 (8H,
4
BArF4), 7.68 (4H, BArF4)}], 2.25-1.54 (br m, 6H, CH3), -18.81 (1
H, br, Rh-H), -19.00 (br, 1H, Rh-H). 31P{1H} NMR (202 MHz,
d6-acetone): δ 45.4 [d, J(RhP) = 121]. 1H NMR (500 MHz, d6-
acetone, 225 K): δ 8.26-7.20 [m, 38H, Ar-H þ BArF4 {7.83 (8H,
BArF4), 7.73 (4H, BArF4)}], 1.94 (s, 3H, CH3), 1.70 (s, 3H, CH3 ),
0
-18.44 [dtd, J(RhH)=22.1, J(PH)=12.8, J(HH)=9.5, 1H, Rh-
H], -19.17 [dtd, J(RhH) = 31.4, J(PH) = 14.5, J(HH) = 9.2, 1H,
Rh-H]. 31P{1H} NMR (202 MHz, d6-acetone, 225 K): δ 46.0 [d,
J(RhP) = 121]. ESI-MS (acetone, 60 °C, 4.5 kV) positive ion:
(minor peak) [M]þ m/z = 741.15 (calcd 741.15); (major peak)
[M - acetone]þ m/z = 683.11 (calcd 683.10).
[Rh(L)(H)2(NCMe)][BArF4] (L = POP0, 9, Xantphos, 13).
[Rh(L)(nbd)][BArF4] (1.5 ꢁ 10-2 mmol) in d6-acetone (0.3 mL)
was freeze-thaw-degassed and backfilled with hydrogen at
77 K. The solution was shaken for 30 min at room temperature.
The solution was then freeze-thaw-degassed again and back-
filled with argon at room temperature. Acetonitrile (7.8 μL,
1.5 ꢁ 10-1 mmol, 10 equiv) was added and the solution shaken
for 30 min. The solution was transferred into a Young’s crystal-
lization tube. Solvent was removed in vacuo and the resulting
residue washed with pentane (3 ꢁ 2 mL), then recrystallized by
diffusion of hexane into a fluorobenzene solution of the crude
product to yield colorless crystals. Yield: 43%, 9.3 mg and 73%,
17.4 mg, respectively.
Synthesis of Complexes. [Rh(Xantphos)(nbd)][BArF4], 3.
Xantphos (125.5 mg, 2.17 ꢁ 10-4 mol, 2 equiv) in CH2Cl2 (13
mL) was added dropwise over 30 min to [Rh(nbd)Cl]2 (50 mg,
1.08 ꢁ 10-4 mol, 1 equiv) in CH2Cl2 (1 mL). The resulting
reaction mixture was then added dropwise to Na[BArF4] (192.2
mg, 2.17 ꢁ 10-4 mol, 2 equiv) in CH2Cl2 (1 mL), and the mixture
was stirred for 2 h. The solution was filtered, and the filtrate was
reduced to ∼5 mL. Pentane was added to the filtrate until a
precipitate was observed, then layered with pentane (15 mL).
The solution was stored overnight at -18 °C to give orange
crystals, yield = 53%, 189 mg. 1H NMR (300 MHz d6-acetone):
δ 7.79 (m, 8H, BArF4), 7.68 (s, 4H, BArF4), 7.46-6.82 (m, 26H,
Ar), 3.78 (s, 4H, nbd CHdCH), 3.52 (s, 2H, nbd CH), 1.70
(s, 6H,CH3), 1.16 (s, 2H, nbd CH2). 31P NMR (122 MHz
d6-acetone): δ 9.74 [d, J(RhP) = 142]. ESI-MS (acetone,
60 °C, 4.5 kV) positive ion: [M]þ m/z, 773.2 (Mþ calcd 773.2).
[Rh(POP0)(H)2(NCMe)][BArF4], 9. 1H NMR (500 MHz,
CD2Cl2): δ 7.85-7.30 [m, 32 H, Ar-H þ BArF {7.72 (8H,
4
BArF4)0, 7.56 (4H, BArF4)}], 4.10 (m, 2H, O-CH2), 3.53 (m, 2H,
O-CH2 ), 2.92 (m, 4H, P-CH2), -16.69 [dtd, J(RhH) = 21.3,
J(PH)=12.3, J(HH)=8.9, 1H, Rh-H], -19.39 [dtd, J(RhH) =
21.9, J(PH) = 13.1, J(HH) = 8.8, 1H, Rh-H]. 31P{1H} NMR
(202 MHz, CD2Cl2): δ 47.1 [d, J(RhP) = 120 Hz]. ESI-MS
(fluorobenzene, 60 °C, 4.5 kV) positive ion: (minor peak) [M]þ
m/z = 588.0922 (calcd 588.1087); (major peak) [M - MeCN]þ
m/z = 547.0705 (calcd 547.0821).
(62) Giordano, G.; Crabtree, R. H. Inorg. Synth. 1990, 28, 88.
(63) Uehara, A.; Bailar, J. C. J. Organomet. Chem. 1982, 239, 11–15.
(64) Cosier, J.; Glazer, A. M. J. Appl. Crystallogr. 1986, 19, 105–107.
(65) Otwinowski, Z.; Minor, W. In Macromolecular Crystallography,
Part A; Carter, C. W., Sweat, R. M., Eds.; Academic Press: New York, 1997;
Vol. 276.
(66) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garcia-Granda,
S.; Gould, R. O.; Israel, R.; Smits, J. M. M. The DIRDIF-99 program
system; University of Nijmegen: Netherlands, 1999.
[Rh(Xantphos)(H)2(NCMe)][BArF4], 13. 1H NMR (500 MHz,
CD2Cl2): δ 7.92-7.29 [m, 38H, Ar-H þ BArF4 {7.72 (8H, BArF4),
7.55 (4H, BArF4)}], 1.91 (s, 3H, CH3), 1.51 (s, 3H, CH3 ), 1.16 (s,
0
3H, coordinated MeCN CH3), -15.36 [dtd, J(RhH) =20.3,
J(PH)=12.3, J(HH)=7.8, 1H, Rh-H], -19.36 [dtd, J(RhH) =
30.8, J(PH) = 13.9, J(HH) = 7.8, 1H, Rh-H]. 31P{1H} NMR (202
MHz, CD2Cl2): δ 44.8 [d, J(RhP) = 117 Hz]. ESI-MS (fluoro-
benzene, 60 °C, 4.5 kV) positive ion: (minor peak) [M]þ m/z =
724.0896 (calcd 724.1400); (major peak) [M - MeCN]þ m/z =
683.0745 (calcd 683.1134).
(67) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.;
Cascarano, G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 2005, 38, 381–388.
(68) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.

1728 Organometallics, Vol. 29, No. 7, 2010
Pawley et al.
[Rh(L)(η2-H2CdCHCOOMe))][BArF4] (L=POP0, 10; Xant-
phos, 14). The title compounds were prepared in situ by placing a
solution of [Rh(L)nbd][BArF4] (6.7 ꢁ 10-3 mmol) in d6-acetone
(0.3 mL) under 4 atm of hydrogen, followed by addition of
methyl acrylate (1.2 mL, 1.3 ꢁ 10-2 mmol, 2 equiv) under an
atmosphere of argon. These compounds where characterized
in situ by NMR spectroscopy and ESI-MS.
BArF4), 7.72 (4H, BArF4)}], 2.80 (m, 1H, CH2), 2.37 (m, 1H,
CH2), 2.10 (m, 10H, CH2), 1.94 (s, 3H, CH3), 1.82 (m, 1H, CH2)
1.47 (s, 3H, CH3 ), 1.38 (s, 3H, SCH3), -8.65 [br dd, J(RhH) =
23 Hz, J(PH) = 7 Hz, 1H, Rh-H]. 31P{1H} NMR (202 MHz, d6-
acetone, 257 K): ABX coupling pattern observed: δ 35.22 (dd,
J(PP) = 300, J(RhP) = 124), 32.04 [dd, J(PP) = 300, J(RhP) =
127]. ESI-MS (acetone, 60 °C, 4.5 kV) positive ion: [M]þ m/z =
785.1221 (calcd 785.1274).
[Rh(POP0)(η2-H2CdCHCOOMe)][BArF4], 10. 1H NMR
(500 MHz, d6-acetone): δ 8.11-7.34 (m, 32H, Ar-H þ BArF
[Rh(nbd)(L4)][BArF4], 16 (improved synthetic route). [Rh-
(dppe)Cl]2 (50 mg, 4.9 ꢁ 10-2 mmol) and Na[BArF4] (86.8 mg,
9.8 ꢁ 10-2 mmol, 2 equiv) were stirred together in C6H4F2
(10 mL) for 20 min. C7H8 (norbornadiene) (24.9 μ, 2.5 ꢁ 10-1
mmol, 5 equiv) was added to the mixture and the mixture stirred
for 2 h. The mixture was filtered and pentane (25 mL) added
to the filtrate. The solution was stored overnight at -18 °C. The
solution was reduced to 10 mL and filtered, and hexane (25 mL)
was added to the filtrate. The solvent was removed in vacuo,
4
{7.79 (8H, BArF4), 7.68 (4H, BArF4)}), 4.41 (m, 2H, O-CH2),
0
4.31 (m, 2H, O-CH2 ), 3.60 (br m, 1H H2CdCH), 3.38 (s, 3H,
0
OMe), 3.05 (m, 2H, P-CH2), 2.90 (br m, 2 þ 1 H, P-CH2
þ
H2CdCH), 2.64 (m, 1H, H2CdCH). Also observed: 6.41-5.82
(br m, free acrylate, H2CdCH), 3.70 (br s, free acrylate OMe).
31P{1H} NMR (202 MHz, d6-acetone): δ 39.1 [d, J(RhP) = 129
Hz]. ESI-MS (acetone, 60 °C, 4.5 kV) positive ion: [M]þ m/z,
631.1 (calcd 631.1). ESI-MSMS (631.1): m/z = 545.1 (Mþ
-
1
leaving an orange powder. Yield: 45%, 65 mg. H NMR (300
CH2CHCO2CH3 calcd 545.1).
MHz, d6-acetone): δ 7.50-7.88 (m, 32H, BArF and Ph), 5.51
[Rh(Xantphos)(η2-H2CdCHCOOMe)][BArF ], 14. 1H NMR
4
4
(500 MHz, d6-acetone): δ 8.20-7.12 [m, 38H, Ar-H þ BArF4 {7.79
(8H, BArF4), 7.68 (4H, BArF4)}], 3.53 (s, 3H, OMe), 3.38 (m, 1H,
H2CdCH), 2.67 (m, 1H, H2CdCH), 2.17 (m, 1H, H2CdCH,
coincident with a norbornane peak), 1.90 (s, 6H, CH3). Also
observed: 6.40-5.76 (br m, free acrylate, H2CdCH), 3.70 (br s,
free acrylate OMe). 31P{1H} NMR (202 MHz, d6-acetone): δ 34.23
[br s, fwhm = 1187 Hz]. ESI-MS (C6H5F, 60 °C, 4.5 kV) positive
ion: [M]þ m/z, 767.1 (calcd 767.1).
(m, 4H, nbd CHdCH), 4.26 (m, 2H, nbd CH), 2.58 (m, 4H,
PCH2), 1.82 (s, 2H, nbd CH2). 31P NMR (122 MHz, d6-acetone):
δ 55.9 [d, J(RhP) = 157].
[Rh(L7)(acetone)2][BArF4], 19a, [Rh(PPh2CH2CH2CHCH2-
CH2PPh2)(H)(acetone)2][BArF4], 19b*, and [Rh(L7)(H)2-
(acetone)2][BArF4], 19c$. The title compounds were prepared
in situ by placing a solution of [Rh(nbd)(L7)][BArF4] (6.7 ꢁ 10-6
mol) in d6-acetone (0.3 mL) under 4 atm of hydrogen. These
compounds where characterized in situ by NMR spectroscopy and
ESI-MS. 1H NMR (500 MHz, d6-acetone): δ 7.91-7.31 [m, 32H,
[Rh(L)(MeSCH2CH2CO)(H)][BArF4] (L = POP0, 11; Xant-
phos, 15). [Rh(L)(nbd)][BArF4] (6.1 ꢁ 10-2 mmol) in acetone
(3 mL) was freeze-thaw-degassed and backfilled with hydrogen
at 77 K. The solution was stirred for 1 h at room temperature.
The solution was then freeze-thaw-degassed again and back-
filled with argon at room temperature. Methyl acrylate (8.2 μL,
9.2 ꢁ 10-2 mmol, 1.5 equiv) was added and the solution stirred
for 4 h, after which 3-(methylthio)propionaldehyde (9.1 μL,
9.2 ꢁ 10-2 mmol, 1.5 equiv) was added and the solution stirred
for a further hour. The solvent was removed in vacuo and the
resulting residue washed with pentane (2 ꢁ 3 mL), then recrys-
tallized by diffusion of pentane into a fluorobenzene solution of
the crude product to yield colorless crystals. Yield: 63%, 69.4 mg
and 35%, 31.8 mg, respectively.
Ar-H þ BArF {7.80 (8H, BArF4), 7.68 (4H, BArF4)}], 2.92 (m,
4
2H, PCH2CH2), 2.47 (m, 4H, PCH2CH2CH2), 1.86 (m, 4H,
PCH2), -21.68 [dt, J(RhH) = 27, J(PH) = 16, Rh-H]*, -22.30
[dt, J(RhH) = 35, J(PH) = 11, Rh-H]$ (coordinated d6-acetone is
not observed). 31P{1H} NMR (202 MHz, d6-acetone): δ 55.3 [d,
relative 0.8 P, J(RhP)=133]$, 37.3 [d, relative 10P, J(RhP) = 201],
35.8 [d, relative 2P, J(RhP) = 115]*. *Denotes selected signals
$
due to 19b. Denotes selected signals due to 19c. ESI-MS (ace-
tone, 60 °C, 4.5 kV) positive ion: (minor peak) [M]þ m/z =
659.2 (calcd 659.2); (major peak) [M - acetone]þ m/z = 601.1
(calcd 601.1); (major peak) [M - (acetone)2]þ m/z = 543.1 (calcd
543.1).
[Rh(POP0)(MeSCH2CH2CO)(H)][BArF4], 11. 1H NMR (400
MHz CD2Cl2): δ 8.23-7.30 [m, 32 H, Ar-H þ BArF4 {7.72 (8H,
BArF4), 7.61 (4H, BArF4)}], 4.01 (m, 2H, CH2), 3.05-2.79 (m,
6H, CH2), 2.37 (br m, 2H, CH2), 1.79 [d, J(HH) 1.8 Hz, 3H,
SCH3], 1.67 (br m, 2H, CH2), -9.58 [br d, J(RhH) 24 Hz, 1H,
General Method for Catalysis. In a typical experiment
[Rh(L)(nbd)][BArF4] (1.5 ꢁ 10-2 mmol) was placed into a
New Era high-pressure NMR sample tube. d6-Acetone
(0.3 mL) was added by cannula and the solution placed under
an H2 atmosphere for 5 min to generate the active catalyst.
Under an argon atmosphere I (15 mL, 1.5 ꢁ 10-1 mmol,
10 equiv) and II (13.5 mL, 1.5 ꢁ 10-1 mmol, 10 equiv) were
added by syringe. The solution was kept at ambient temperature
1
RhH]. H-1H COSY NMR (400 MHz, CD2Cl2): Correlation
observed between SCH3 and hydride. 31P{1H} NMR (162 MHz,
CD2Cl2): δ 31.1 (br s, fwhm = 513 Hz). 31P{1H} NMR (202
MHz, d6-acetone, 225 K): ABX coupling pattern observed: δ
34.6 [dd, J(PP) = 309 Hz, J(RhP) = 128 Hz], 28.5 [dd, J(PP) =
309 Hz, J(RhP) = 133 Hz]. ESI-MS (acetone, 60 °C, 4.5 kV)
positive ion: [M]þ m/z = 649.1022 (calcd 649.0961).
1
(22 °C) and the reaction followed in situ by H NMR spectro-
scopy. Compounds III and IV were identified by comparison
with published data.15
[Rh(Xantphos)(MeSCH2CH2CO)(H)][BArF4], 15. 1H NMR
Acknowledgment. We thank the EPSRC, the University
of Oxford, and the John Fell Fund (University of Oxford)
for support and the referees for insightful comments.
(500 MHz, d6-acetone): δ 8.48-7.36 [m, 38H, Ar-H þ BArF
4
{7.79 (8H, BArF4), 7.68 (4H, BArF4)}], 2.62 (br s, fwhm = 77
Hz, 2H, CH2), 2.04 (br, 2H, CH2, coincident with CD2HCOCD3
0
peak), 1.97 (s, 3H, CH3), 1.51 (s, 3H, CH3 ), 1.47 [d, J(HH) = 1.8
Hz, 3H, SCH3], -8.72 [br d, J(RhH) = 23 Hz, 1H, Rh-H].
31P{1H} NMR (202 MHz, d6-acetone): δ 32.8 (br s, fwhm = 478
Hz). 1H-1H COSY NMR (500 MHz, d6-acetone): Correlation
observed between SCH3 and hydride. 1H NMR (500 MHz, d6-
acetone, 240 K): δ 8.58-7.24 [m, 38H, Ar-H þ BArF4 {7.81 (8H,
Supporting Information Available: Simulated and experimen-
tal NMR spectra for 8 and 12; experimental NMR spectra for 9,
11, 13, and 15; and crystallographic collection and refinement
parameters for 9, 11, 13, and 15. This material is available free of
charge via the Internet at http://pubs.acs.org.