Cyclometalation reactions on rhodium(I). Evidence for a chelate effect and competing C-H and C-O oxidative additions

Article abstract of DOI:10.1021/om010543x

The ligand Ph₂PC₆H₄(CH₂N(Me)CO(Et)) reacts with (bis(diphenylphosphino)butane) (2,5-norbornadiene)rhodium tetrafluoroborate (2a) and (bis(diphenylphosphino)ethane)(2,5-norbornadiene) rhodium tetrafluoroborate (2b) at room temperature in the presence of hydrogen to displace the norbornadiene and give the chelate complexes 3a,b, in which the phosphorus and the oxygen atoms are coordinated to give an eight-membered ring. At elevated temperatures these complexes are converted into cyclometalated Rh(III) benzyl hydride complexes. Rate law, activation parameters, and reactivity trends of this latter transformation indicate that displacement of one of the phosphine functionalities of the bis-chelating phosphine takes place before the C-H activation. Complex 3b was characterized by X-ray crystallography. On the other hand, under ambient conditions the ligand (±)-Ph₂PC₆H₄(CH(Me)O(CO)Et) undergoes activation of either the benzylic C-O or C-H bond, depending on the nature of the Rh precursor used. Thus, 2b gives overall elimination of propionic acid to result in a styrene complex, which was characterized by X-ray crystallography, whereas bis(2,5-norbornadiene)rhodium tetrafluoroborate gives a mixture of Rh(III) benzyl hydride stereoisomers. The difference in reactivity is discussed in terms of different mechanisms for the two processes.

Full text of DOI:10.1021/om010543x

Organometallics 2001, 20, 4919-4926
4919
Cyclom eta la tion Rea ction s on Rh od iu m (I). Evid en ce for
a Ch ela te Effect a n d Com p etin g C-H a n d C-O Oxid a tive
Ad d ition s
Sven Sjo¨vall, Carlaxel Andersson, and Ola F. Wendt*
Inorganic Chemistry, Department of Chemistry, Lund University, P.O. Box 124,
S-221 00 Lund, Sweden
Received J une 21, 2001
The ligand Ph2PC6H4(CH2N(Me)CO(Et)) reacts with (bis(diphenylphosphino)butane)(2,5-
norbornadiene)rhodium tetrafluoroborate (2a ) and (bis(diphenylphosphino)ethane)(2,5-
norbornadiene) rhodium tetrafluoroborate (2b) at room temperature in the presence of
hydrogen to displace the norbornadiene and give the chelate complexes 3a ,b, in which the
phosphorus and the oxygen atoms are coordinated to give an eight-membered ring. At
elevated temperatures these complexes are converted into cyclometalated Rh(III) benzyl
hydride complexes. Rate law, activation parameters, and reactivity trends of this latter
transformation indicate that displacement of one of the phosphine functionalities of the bis-
chelating phosphine takes place before the C-H activation. Complex 3b was characterized
by X-ray crystallography. On the other hand, under ambient conditions the ligand (()-Ph2-
PC6H4(CH(Me)O(CO)Et) undergoes activation of either the benzylic C-O or C-H bond,
depending on the nature of the Rh precursor used. Thus, 2b gives overall elimination of
propionic acid to result in a styrene complex, which was characterized by X-ray crystal-
lography, whereas bis(2,5-norbornadiene)rhodium tetrafluoroborate gives a mixture of Rh-
(III) benzyl hydride stereoisomers. The difference in reactivity is discussed in terms of
different mechanisms for the two processes.
In tr od u ction
or C-C bond activation can be controlled by a careful
choice of either the substrate or the metal complex, e.g.
by the use of strained alkanes,4a-c prearomatic sys-
tems,4d,e agostic interactions,4f,g clusters,4h,i and steri-
cally demanding systems.4j,k Due to their relatively
convenient preparations, the majority of the complexes
studied are of the cyclometalated type, especially those
that are ortho-metalated. Such complexes are charac-
terized by their high thermal stability due to the
energetically strong M-X bond and the chelate ring
formed upon the intramolecular chelate-assisted C-X
bond activation. Because of these beneficial features, as
a “side effect”, the use of metallacyclic complexes in
catalysis has grown into a major field of its own.5
Two of us have recently demonstrated that the
triphenylphosphine derivatives N-methyl-N-propionyl-
2-(diphenylphosphino)benzylamine (DPPAM; 1a ) and
1-(2-diphenylphosphino)benzyl propionate (1b) (Chart
1) undergo facile insertion of a Rh(I) or Ir(I) metal center
into one of its benzylic C-H bonds.6 These tridentately
PCO-coordinated phosphines are an important class of
ligands, since they produce a rather unusual complex
with a hydride and a σ-alkyl group in its coordination
Homogeneous C-X (X ) H, C) bond activation by
transition-metal complexes is a field of much current
interest, with the overall objective of designing new
selective and efficient processes for the utilization of
hydrocarbons.1 Not only has the normally more favor-
able C-H bond activation process been extensively
studied2 but also important synthetic and mechanistic
insights into the C-C bond activation process have been
realized recently.3,4 It has been shown by numerous
examples that the competing reactions of either C-H
* To whom correspondence should be addressed. E-mail: ola.wendt
@inorg.lu.se.
(1) (a) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.
(b) Crabtree, R. H. Chem. Rev. 1985, 85, 245.
(2) For excellent reviews see: (a) Arndtsen, B. A.; Bergman, R. G.;
Mobley, T. A.; Peterson, T. H. Acc. Chem. Res. 1995, 28, 154. (b) Canty,
A. J .; van Koten, G. Acc. Chem. Res. 1995, 28, 406. (c) Schneider, J . J .
Angew. Chem., Int. Ed. Engl. 1996, 35, 1068. (d) Ryabov, A. D. Chem.
Rev. 1990, 90, 403. (e) Stahl, S.; Labinger, J . A.; Bercaw, J . E. Angew.
Chem., Int. Ed. Engl. 1998, 37, 2181.
(3) Rybtchinski, B.; Milstein, D. Angew. Chem., Int. Ed. 1999, 38,
870.
(4) (a) Cassar, L.; Halpern, J . J . Chem. Soc. D 1970, 1082. (b) Cassar,
L.; Eaton, P. E.; Halpern, J . J . Am. Chem. Soc. 1970, 92, 3515. (c)
Periana, R. A.; Bergman, R. G. J . Am. Chem. Soc. 1986, 108, 7346. (d)
Eilbracht, P. Chem. Ber. 1980, 113, 542. (e) Urbanos, F.; Halcrow, M.
A.; Fernandez-Baeza, J .; Dahan, F.; Labroue, D.; Chaudret, B. J . Am.
Chem. Soc. 1993, 115, 3484. (f) Bennett, M. A.; Nicholls, J . C.; Rahman,
A. K. F.; Redhouse, A. D.; Spencer, J . L.; Willis, A. C. J . Chem. Soc.,
Chem. Commun. 1989, 1328. (g) Nicholls, J . C.; Spencer, J . L.
Organometallics 1994, 13, 1781. (h) Ingman, S. L.; J ohnson, B. F. G.;
Martin, C. M.; Parker, D. G. J . Chem. Soc., Chem. Commun. 1995,
159. (i) Brown, D. B.; Dyson, P. J .; J ohnson, B. F. G.; Martin, C. M.;
Parker, D. G.; Parsons, S. J . Chem. Soc., Dalton Trans. 1997, 1909. (j)
Rybtchinski, B.; Vigalok, A.; Yehoshoa, B.-D.; Milstein, D. J . Am. Chem.
Soc. 1996, 118, 12406. (k) Gandelman, M.; Vigalok, A.; Shimon, L. J .
W.; Milstein, D. Organometallics 1997, 16, 3981.
(5) For important examples see: (a) Lewis, L. J . Am. Chem. Soc.
1986, 108, 743. (b) Bergbreiter, D. E.; Osburn, P. L.; Liu, Y.-S. J . Am.
Chem. Soc. 1999, 121, 9531. (c) Donkervoort, J . G.; Vicario, G. L.;
J astrzebski, J . T. B. H.; Gossage, R. A.; Cahiez, G.; van Koten, G. J .
Organomet. Chem. 1998, 558, 61. (d) Beletskaya, I. P.; Cheprakov, A.
V. Chem. Rev. 2000, 100, 3009. (e) Steenwinkel, P.; Gossage, R. A.;
van Koten, G. Chem. Eur. J . 1998, 4, 759.
(6) (a) Sjo¨vall, S.; Kloo, L.; Nikitidis, A.; Andersson, C. Organome-
tallics 1998, 17, 579. (b) Sjo¨vall, S.; J ohansson, M.; Andersson, C.
Organometallics 1999, 18, 2198. (c) Sjo¨vall, S.; Svensson, P. H.;
Andersson, C. Organometallics 1999, 18, 8, 5412.
10.1021/om010543x CCC: $20.00 © 2001 American Chemical Society
Publication on Web 10/10/2001

4920 Organometallics, Vol. 20, No. 23, 2001
Sjo¨vall et al.
Ch a r t 1
Sch em e 2
sphere.7 Here we report a mechanistic study regarding
the C-H activation process of our PCO system based
on kinetics and structure-reactivity arguments. More-
over, an unusual example of competition between C-H
and C-O activation on a tertiary carbon center is
presented.
Resu lts a n d Discu ssion
Liga n d s. Hemilabile phosphines are advantageous
ligands for the study of oxidative processes, especially
cyclometalation, since upon coordination to a transition-
metal center the pendant side chain promotes chelate-
assisted intramolecular oxidative insertion of the metal
center. The PCO ligands previously reported by us
(Chart 1) are good examples, and with these species
cyclometalation is exceptionally facile, due to the result-
ing formation of a bis-chelating five-membered-ring
structure.6 To further investigate the C-H bond activa-
tion process with the PCO system, including a possible
competition between oxidative addition of C-H and
C-C bonds, we have synthesized the functionalized
phosphine (()-DPPES ((()-1-(2-(diphenylphosphino)-
phenyl)ethyl propionate; 1c) (Chart 1). The new phos-
phine 1c was obtained by reducing 1-(2-(diphenylphos-
phino)phenyl)ethanone to give (()-1-(2-(diphenylphos-
phino)phenyl)ethanol, which was then reacted with
propionyl chloride in THF (Scheme 1). Characteristic
2
tioned at 48.9 (1J RhP ) 188.5 Hz, J PP ) 51.1, 34.5 Hz,
2
P
P
DPPB), 39.0 (1J RhP ) 145.1 Hz, J PP ) 302.7, 51.1 Hz,
DPPB) and 30.3 ppm (1J RhP ) 132.3 Hz, J PP ) 302.7,
2
34.5 Hz, PDPPAM). The lack of a symmetry plane in
complex 3a results in diastereotopic protons in the
1
benzylic group, and these are displayed in the H NMR
spectrum as broad doublets at 8.98 (2J HH ) 14.4 Hz)
and 3.81 ppm (2J HH ) 14.4 Hz), respectively. The far
downfield shift for one of them indicates a significant
through-space interaction with the metal center. In
addition, the 13C{1H} NMR spectrum of 3a displays a
singlet for the benzylic carbon in the normal aliphatic
region (29.5 ppm). The carbonyl stretching frequency,
positioned at 1586 cm-1, is shifted to a lower wavenum-
ber compared to the free ligand (1655 cm-1),8 verifying
the eight-membered-ring structure in the complex re-
sulting from coordination of the carbonyl group to the
rhodium atom.
The inability of 3a to cyclometalate at room temper-
ature was unexpected, since the analogous mono-phos-
phine precursor complex [Rh(PPh3)2(NBD)][PF6] readily
yields a bis-chelating PCO complex with DPPAM when
using the same reaction conditions as for the synthesis
of 3a .6c Clearly, the DPPB ligand in 3a is not innocent
but influences the course of the reaction. To further
investigate the possible steric effects, the reaction was
repeated, but now with [Rh(DPPE)[NBD)][BF4] (2b)
(DPPE ) bis(1,2-diphenylphosphino)ethane) as the
diphosphine precursor (Scheme 2), bearing in mind the
differences in bite angles for DPPE (78°) and DPPB
(98°).9 The structure of 3b is identical with that for 3a ,
as determined by multinuclear NMR spectroscopy. The
31P{1H} NMR spectrum of 3b exhibits three doublets
Sch em e 1
spectroscopic features for the racemic mixture of 1c are
the singlet exhibited at -17.3 ppm (31P{1H} NMR), the
strong vibrational absorption for the carbonyl group at
1729 cm-1 (IR), and the doublet displayed at 70.0 ppm
(3J PC ) 28.1 Hz) for the tertiary benzylic carbon (13C-
{1H} NMR).
Com p lexa tion of DP P AM (1a ) w ith Bis-P h os-
p h in e Rh od iu m P r ecu r sor s. A dichloromethane solu-
tion of the rhodium bis-phosphine complex [Rh(DPPB)-
(NBD)][BF4] (2a ; DPPB ) bis(1,2-diphenylphosphino)-
butane, NBD ) 2,5-norbornadiene) and 1 equiv of ligand
1a , stirred for 40 min under an atmosphere of molecular
hydrogen at room temperature, yields after workup the
four-coordinate Rh(I) complex 3a (Scheme 2). This
complex was unambiguously characterized by NMR and
IR spectroscopy. The 31P{1H} NMR spectrum of 3a
displays three doublets of doublets of doublets posi-
2
of doublets of doublets at 79.4 (1J RhP ) 195.9 Hz, J PP
2
) 37.3, 30.9 Hz, PDPPE), 66.3 (1J RhP ) 144.7 Hz, J PP
)
308.9, 37.3 Hz, PDPPE), and 29.4 ppm (1J RhP ) 127.3 Hz,
2J PP ) 308.9, 30.9 Hz, PDPPAM). The signals for the
diastereotopic benzylic protons appear in the H NMR
1
spectrum at 8.48 (2J HH ) 13.6 Hz) and 4.07 ppm (2J HH
) 13.6 Hz) while the benzylic carbon appears at 35.7
ppm in the 13C{1H} NMR spectrum; the assignments
were verified by an HMQC NMR experiment. The
molecular structure of 3b was confirmed by X-ray
crystallography, which also revealed an agostic interac-
(8) Nikitidis, A.; Andersson, C. Phosphorus, Sulfur Silicon Relat.
Elem. 1993, 78, 141.
(9) del Rio, I.; Ruiz, N.; Claver, C.; van der Veen, L. A.; van Leuween,
P. W. N. M. J . Mol. Catal. A 2000, 161, 39.
(7) (a) Zhang, L.; Zetterberg, K. Organometallics 1991, 10, 3806. (b)
Rybtchinski, B.; Konstantinovsky, L.; Shimon, L. J . W.; Vigalok, A.;
Milstein, D. Chem. Eur. J . 2000, 6, 3287.

Cyclometalation Reactions on Rhodium(I)
Organometallics, Vol. 20, No. 23, 2001 4921
F igu r e 2. Eyring plot for the conversion of 3b to 4b. First-
order rate constants are 0.75 × 10-4, 1.42 × 10-4, 3.80 ×
10-4, and 8.14 × 10-4 s-1 at 332.7, 342.7, 352.7, and 362.7
K, respectively.
F igu r e 1. 1H NMR peak intensities of the N-CH3 protons
for complex 3b (O) and 4b (0) in 1,1,2,2-tetrachloroethane-
d2 as a function of time. The solid lines denote the best fit
to single exponentials. T ) 352.7 K.
Mech a n ism of C-H Activa tion . The reactivity
patterns, where the complexes with chelating phos-
phines display a much lower reactivity as compared to
the triphenylphosphine complex, indicates that the
phosphines are not only innocent spectator ligands.
Secondary effects, e.g. steric hindrance, can probably
also be ruled out, since DPPE and DPPB would be
expected to behave differently in this respect as opposed
to what is observed. Chelating phosphines are generally
more difficult to displace/ring-open, and this is the most
likely explanation for the lower reactivity of the chelate
complexes. Thus, phosphine displacement/opening of the
chelate ring most probably takes place prior to or during
the rate-determining step of the C-H activation. This
is in agreement with the rate law, but its simple form
of course makes it compatible with a large number of
mechanisms. Unfortunately, it was not possible to check
the effect of adding free phosphine, since this causes
extrusion of the P,O chelate. The activation parameters
clearly speak against simple dissociation as the rate-
determining step; the entropy is negative and the
enthalpy is too low to correspond to a Rh-P bond
dissociation energy. We therefore favor an explanation
in which reaction takes place via an intermediate from
which then the actual oxidative addition occurs ir-
reversibly (Scheme 3). The observed activation param-
tion for the benzylic proton appearing in the downfield
region in the H NMR spectrum (vide infra).
1
The difficulty for DPPAM to cyclometalate at room
temperature, no matter the steric property of the
rhodium bis-phosphine precursor used, could imply that
dissociation of one phosphine functionality to form an
unsaturated three-coordinate Rh(I) species takes place
before or during the rate-determining step in the
cyclometalation. To induce C-H activation, complexes
3a ,b, separately dissolved in 1,1,2,2-tetrachloroethane-
d2, were heated to 70 °C in closed NMR tubes. Both
samples were monitored by 1H NMR spectroscopy,
which revealed formation of C-H activation products
within minutes, e.g. by the appearance of hydride
signals at -17.2 (4a ) and -17.4 ppm (4b), and after 3
h both conversions were complete. Complexes 4a ,b were
synthesized on a preparatory scale by heating THF
solutions of 3a ,b at 100 °C in closed vessels for 90 min
(Scheme 2). The complexes were characterized by NMR
and IR spectroscopy. The tridentate PCO coordination
of the DPPAM ligand is manifested by the strong
downfield shifts in their 31P{1H} NMR spectra (56.1 (4a )
and 61.9 ppm (4b)), the carbonyl stretching frequencies
in the IR spectra (1571 cm-1 (4a ) and 1572 cm-1 (4b)),
and the benzylic carbon signals in their 13C{1H} NMR
spectra (multiplets at 72.6 (4a ) and 70.9 ppm (4b)).
Kin etics of F or m a tion of 4b. Qualitative experi-
ments showed that 3a ,b undergo cyclometalation at
approximately the same rate. The kinetics of the
conversion of 3b was more well-behaved, and this
complex was chosen for a more careful analysis. The
reaction was monitored at temperatures ranging from
332.7 to 362.7 K in 1,1,2,2-tetrachloroethane-d2 using
1H NMR spectroscopy. The intensities of the NCH3
singlets for both 3b and 4b were followed as a function
of time, and the data so obtained was fitted to single
exponentials. A typical fit (at 352.7 K) is shown in
Figure 1, giving rate constants of (4.0 ( 0.3) × 10-4 and
(3.6 ( 0.2) × 10-4 s-1 for the consumption of reactant
and formation of product, respectively. Only 3b and 4b
were observed during the reaction, and the conversion
is thus a single process which is first order in rhodium
complex. Average rate constants at each temperature
were computed, and the data were fitted to the Eyring
equation as shown in Figure 2. Activation parameters
are ∆Hq ) 78 ( 6 kJ mol-1 and ∆Sq ) -90 ( 16 J K-1
Sch em e 3
eters are then a sum of those relating to the individual
rate constants in the reaction scheme. The structure of
the intermediate is to some extent speculative. One
possibility is shown in Scheme 3, in which the P,O
ligand is coordinated in a trans fashion with one of the
arms of the DPPE being pendant. This opens up a
possibility for the C-H bond to coordinate in a η2
fashion, thus producing two five-membered rings. Such
a well-ordered intermediate where the orbitals are
arranged to facilitate the electron transfer would give
a negative contribution to the observed entropy of
activation, explaining its negative value. The formation
mol-1
.

4922 Organometallics, Vol. 20, No. 23, 2001
Sjo¨vall et al.
Sch em e 4
2
of this intermediate can take place either in an associa-
tive or in a dissociative fashion. Another alternative is
that the agostic interaction seen in the crystal structure
of 3b is on the reaction path and that the activation
takes place in the axial position. In this case it is less
clear why the phosphine has to dissociate to promote
reaction (except for giving a more electrophilic metal
center, thus promoting the η2-coordination), and we
favor the first alternative.
Com p lexa tion of (()-DP P ES (1c) w ith Differ en t
Rh od iu m P r ecu r sor s. With the possible objective of
gaining insight into the important C-C bond activation
process of an unstrained saturated carbon-carbon bond,
the (()-DPPES ligand (1c) (Scheme 1) and its complex-
ation behavior were also investigated. Pioneering work
by Milstein and co-workers have shown that C-C
activation can be kinetically and thermodynamically
competitive with C-H activation under appropriate
conditions.4j,k Ligand 1c, having only one benzylic C-H
bond and being bulkier than 1a due to steric hindrance
imposed by the methyl group, may allow for a competi-
tion between C-H and C-C bond activation processes.
We considered three different reaction pathways, namely
(i) C-H bond activation of the benzylic carbon-
hydrogen bond, forming a five-membered chelate, (ii)
C-H bond activation of the methyl carbon-hydrogen
bond, forming a six-membered chelate, and (iii) C-C
bond activation of the benzylic carbon-methyl bond,
forming a five-membered chelate.
) 316.8, 26.8 Hz), 53.9 (1J RhP ) 153.2 Hz, J PP ) 31.1,
26.8 Hz), and 38.9 ppm (1J RhP ) 123.3 Hz, 2J PP ) 316.8,
31.1 Hz). At this stage all peaks in the 1H NMR
spectrum related to the chelating sidearm in (()-DPPES
had been replaced by three broad signals of equal
intensity located at 5.25, 4.18, and 3.65 ppm along with
additional peaks at 2.35 and 1.18 ppm. This new
compound was crystallized from a CH2Cl2/Et2O solution
of 5 (Scheme 4), and X-ray crystallography (vide infra)
revealed the formation of [Rh(DPPSTY)(DPPE)][BF4] (6;
DPPSTY ) 2-(diphenylphosphino)styrene). In an at-
tempt to detect any kind of intermediates along the
decomposition pathway by 1H NMR, only the signals
assigned to 5, 6, and propionic acid were observed.
Thus, the metal center mediates facile elimination of
carboxylic acid from an ester, a process that normally
requires very high temperatures (300-500 °C).10 The
mechanism for this unexpected transformation is not
clear, but obviously there is yet another alternative in
addition to the three we had anticipated: namely, C-O
bond activation. A possible sequence to achieve this is
oxidative addition of the C-O bond to give a Rh(III)
benzyl species, which then can undergo â-hydride
elimination followed by deprotonation to give the sty-
rene complex and propionic acid. Given that this process
is facile at room temperature, it probably does not
involve opening of the strong chelate in DPPE. This can
be understood in terms of the usual type mechanisms
for oxidative addition. In the three-center addition
displayed by nonpolar species, the coordination (in an
η2-fashion) of the C-H bond, which often dictates the
reactivity, is promoted by coordinative unsaturation.2,11
In the C-O addition, on the other hand, we have a fairly
good leaving group and the SN2 reaction usually dis-
played by polar species will be faster with a more
electron-rich metal center.
Upon reaction of 1c with [Rh(DPPE)(NBD)][BF4] (2b)
(Scheme 4) in the presence of molecular hydrogen,
complex 5 was formed and spectroscopically character-
ized. The NMR spectra of 5 were very similar to those
of 3b, except for the signals arising from the tertiary
benzylic group. The existence of the eight-membered
chelate was further verified by an IR spectrum, display-
ing a stretching frequency for the carbonyl group at
1661 cm-1 (1729 cm-1 for free (()-DPPES). However,
the stability of complex 5 deviates substantially from
that observed for 3b; an NMR sample kept at room
temperature for some hours showed that complex 5
underwent slow, clean thermolysis. After more than 48
h at room temperature a 31P{1H} NMR spectrum of the
sample taken at 263 K still displayed three doublets of
doublets of doublets, but now at completely different
Examples of transition-metal-mediated cleavage of
carboxylic C-O ester bonds are fairly abundant, e.g.,
in the reaction of palladium(0) complexes with allyl and
benzyl carboxylates,12 but we have found no examples
in the literature of room-temperature eliminations of
(10) Depuy, C. H.; King, R. W. Chem. Rev. 1960, 60, 431.
(11) McNamara, B. K.; Yeston, J . S.; Bergman, R. G.; Moore, C. B.
J . Am. Chem. Soc. 1999, 121, 6437.
(12) (a) Yamamoto, A. Adv. Organomet. Chem. 1992, 34, 111. (b)
Nagayama, K.; Shimizu, I.; Yamamoto, A. Bull. Chem. Soc. J pn. 1999,
72, 799.
2
chemical shifts located at 63.9 (1J RhP ) 120.9 Hz, J PP

Cyclometalation Reactions on Rhodium(I)
Organometallics, Vol. 20, No. 23, 2001 4923
carboxylic acids to give olefins. Further studies are
underway to explore the mechanism behind this un-
usual transformation.
The unexpected chemical behavior of the (()-DPPES
ligand calls for a different rhodium precursor than has
been used up to now.13 To favor concerted C-X (X ) C,
H) bond addition, we need a more labile rhodium
complex, with or without ancillary ligands. Accordingly,
a dichloromethane solution of [Rh(NBD)2][BF4] was
treated with 2 equiv of (()-DPPES under an atmosphere
of hydrogen at room temperature for 5 min to form
complex 7 (Scheme 4). The ratio of cis/trans isomers
varies slightly from batch to batch but is always
approximately 3:2 in favor of the cis isomer. The 31P-
{1H} NMR spectrum of 7 at 253 K displays a total of
eight doublets of doublets: at 58.5 and 19.7 ppm (2J PP
) 36.3 Hz) for 7a , 58.2 and 15.6 ppm (2J PP ) 36.0 Hz)
for 7b, 43.4 and 27.4 ppm (2J PP ) 358.3 Hz) for 7c, and
1
41.5 and 31.3 ppm (2J PP ) 371.1 Hz) for 7d . In the H
F igu r e 3. DIAMOND17 plot of cation 3b. For clarity,
atomic numbering in the phenyl rings is only shown for
the ipso carbons. Selected bond distances (Å) and angles
(deg) with estimated standard deviations: Rh1-P1 )
2.180(1); Rh1-P2 ) 2.254(1); Rh1-P3 ) 2.310(1); Rh1-
O1 ) 2.120(3); P1-Rh1-P2 ) 85.17(4); P1-Rh1-P3 )
101.68(4); P2-Rh1-P3 ) 173.11(4); P1-Rh1-O1 ) 170.74-
(8); P2-Rh1-O1 ) 87.97(7); P3-Rh1-O1 ) 85.29(7).
NMR spectrum at 253 K two multiple hydride signals
are displayed at -21.9 (7c,d ) and -23.3 ppm (7a ,b).
Evidently, in this case the C-H activation process is
the kinetically preferred one. It is undoubtedly the
benzylic carbon that becomes metalated, as evidenced
by the 31P{1H} NMR shifts for the cyclometalated (()-
DPPES ligands. Only five-membered chelate formation
results in such a strong downfield shift, compared to
the shift for the eight-membered P,O-chelating (()-
DPPES ligand.14 Each set of cis-trans isomers consists
of a 1:1 mixture of diastereomers. This is what one
would expect with one stereogenic center on each of the
ligands. In addition there is a possibility for two
configurations along the Rh-C bond, where the vicinal
hydride and methyl group can be either anti or gauche.
This could potentially give even more stereoisomers, and
obviously only one configuration is formed. In analogy
with what was earlier exclusively observed in the crystal
structures of PCO chelates of 1a ,b with Rh and Ir,
respectively, the conformation is probably gauche.6b,c
The observed reaction with the (()-DPPES ligand is,
to the best of our knowledge, the first example of a
phosphine derivative capable of cyclometalation via a
tertiary carbon at room temperature. Previously, Shaw
and co-workers have described a number of cyclometa-
lated platinum group metal complexes derived from the
3
reasonable to believe that the M-Csp bond is too weak
to be formed at the expense of an M-H bond.
Cr ysta l Str u ctu r es. Yellow prismatic crystals of
complex 3b were obtained upon crystallization by layer-
ing Et2O on top of a CH2Cl2 solution of 3b. The structure
consists of cationic rhodium complexes with tetrafluo-
roborate anions. The molecular structure of the cation
is shown in Figure 3, including selected bond lengths
and angles. The coordination geometry around rhodium
is distorted square-planar with the three phosphorus
atoms and the carbonyl oxygen in the corners of the
square. Inter alia, the distortion results from ring strain
in both chelate rings (P1-Rh1-P2 ) 85.17(4)° and P3-
Rh1-O1 ) 85.29(7)°). The largest deviation from a
least-squares plane through Rh1-P1-P2-P3-O1 is
0.082 Å, displayed by O1. The difference in Rh-P bond
distances in the DPPE ligand manifests the different
trans influences exerted by a phosphorus and an oxygen
atom, respectively.16 Interestingly, the Rh1-P3 bond
length of 2.310(1) Å in the DPPAM ligand is almost
identical with that observed in the cyclometalated rho-
dium(III) compound [RhH(DPPAM)(PPh3)2][PF6] (Rh-
PDPPAM ) 2.317(2) Å).6c Furthermore, the crystal struc-
ture reveals a bridging benzylic group (C6 positioned
3.090(6) Å away from the metal center). The positions
of the hydrogen atoms on C6 were located in the
difference Fourier map and successfully refined. Thus,
H6B is located in an apical position 2.16(5) Å away from
Rh, indicating a relatively weak agostic interaction.18
Slow vapor diffusion of Et2O into a CH2Cl2 solution
of 5 at room temperature resulted in formation of deep
red crystals of complex 6. The asymmetric unit of the
structure consists of two independent cationic rhodium
t
t
sterically hindered Bu2P(o-iPrC6H4) and Bu2PCH2CH2-
CHMeCH2CH2PtBu2 ligands, all synthesized at elevated
temperatures.15 There can be several reasons for why
(()-DPPES follows the C-H oxidative addition pathway
exclusively in complexes 7a -d . Clearly, the P,O chelate
is very fluxional, thus enabling a similar trans arrange-
ment of the (other) P,O ligand just as proposed for 3a ,b
in the formation 4a ,b, and this promotes concerted over
SN2 addition. Even though C-C bond activation has
been shown to kinetically compete with C-H activa-
tion,4j,k all examples so far involve Csp -Csp bonds and
on the basis of the exclusive formation of 7 it seems
2
3
(13) Using [Rh(PPh3)2(NBD)][BF4] as the precursor gave the same
thermolysis reaction as the corresponding DPPE complex, although
at a higher rate.
(14) (a) Garrou, P. E. Chem. Rev. 1981, 81, 229. (b) Lindner, E.;
Fawzi, R.; Mayer, H. A.; Eichele, K.; Hiller, W. Organometallics 1992,
11, 1033.
(15) (a) Gill, D. F.; Shaw, B. L. J . Chem. Soc., Chem. Commun. 1972,
65. (b) Crocker, C.; Errington, R. J .; Markham, R.; Moulton, C. J .; Odell,
K. J .; Shaw, B. L. J . Am. Chem. Soc. 1980, 102, 4373.
(16) Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. Rev.
1973, 10, 335.
(17) Brandenburg, K., DIAMOND, Program for Molecular Graphics;
Crystal Impact, Bonn, Germany, 2000.
(18) Crabtree, R. H.; Holt, E. M.; Lavin, M.; Morehouse, S. M. Inorg.
Chem. 1985, 24, 1986.

4924 Organometallics, Vol. 20, No. 23, 2001
Sjo¨vall et al.
Exp er im en ta l Section
Gen er a l P r oced u r es. All experiments with metal com-
plexes and phosphine ligands were carried out under an
atmosphere of nitrogen in a Braun glovebox equipped with an
inert gas purifier or using standard Schlenk or high-vacuum
techniques. All nondeuterated solvents were refluxed over
sodium/benzophenone ketyl or CaH2 and freshly distilled under
a nitrogen atmosphere. Commercially available reagents were
used as received. The complexes [Rh(PP)(NBD)][BF4]20 (PP )
DPPB, DPPE), and [Rh(NBD)2][BF4]21 and the phosphines
2-(diphenylphosphino)phenyl methyl ketone22 and DPPAM8
(1a ) were prepared according to literature procedures.
NMR spectra were recorded either on a Varian Unity 300
MHz instrument (1H, 31P) or a Bruker ARX 500 MHz spec-
trometer (13C) in CDCl3 solvent unless otherwise stated. 1H
and 13C{1H} NMR chemical shifts are reported in ppm down-
field from tetramethylsilane but were measured relative to the
solvent residuals. 31P NMR chemical shifts are reported in ppm
downfield from an external 85% solution of phosphoric acid.
NMR multiplicities are abbreviated as follows: s ) singlet, d
) doublet, t ) triplet, q ) quartet, m ) multiplet, b ) broad.
Fast atom bombardment (FAB+) mass spectroscopic data were
obtained on a J EOL SX-102 spectrometer using 3-nitrobenzyl
alcohol as matrix and CsI as calibrant. Infrared spectra were
recorded on a Bio-Rad FTS 6000 FT-IR spectrometer. Mikro-
Kemi AB, Uppsala, Sweden, performed elemental analyses.
Syn th esis of (()-DP P ES (1c). A yellow suspension of
2-(diphenylphosphino)phenyl methyl ketone (3.02 g, 9.91
mmol) in 40 mL of THF/MeOH (1:1) was vigorously stirred.
The mixture was cooled with an ice bath, and a pellet of NaBH4
(1.02 g, 26.4 mmol) was added. After 1 h the ice bath was
removed and the reaction mixture was stirred overnight.
Addition of 40 mL of CH2Cl2 was followed by careful addition
of degassed water (40 mL). The colorless organic phase was
isolated, and the water phase was further extracted with CH2-
Cl2 (2 × 15 mL). The combined organic phases were dried over
silica. Decanting and removal of the solvent resulted in a
F igu r e 4. DIAMOND17 plot of cation 6. Only one of the
asymmetric units is shown. For clarity, atomic numbering
in the phenyl rings is only shown for the ipso carbons.
Ta ble 1. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) in 6
Bond Lengths
Rh1-C3
Rh1-C4
Rh1-P1
Rh1-P3
Rh1-P5
C3-C4
2.216(6)
Rh2-C7
Rh2-C8
Rh2-P6
Rh2-P4
Rh2-P2
C7-C8
2.201(6)
2.225(5)
2.221(6)
2.2904(18)
2.2831(19)
2.2958(18)
1.372(7)
2.2968(19)
2.2911(19)
2.2928(19)
1.408(7)
1
colorless, viscous oil. 31P{1H} NMR: δ -15.5 (s). H NMR: δ
Bond Angles
7.73-6.87 (m region, 14H, HAr), 5.62 (m, 1H, ArCH(CH3)), 1.33
(d, 3J HH ) 6.30 Hz, 3H, ArCH(CH3)). The crude oil and pyridine
(2.37 g, 30.0 mmol) were dissolved in 10 mL of THF. Propionyl
chloride (0.897 mL, 10.3 mmol) was added carefully, and the
suspension was stirred for 1.5 h. The white precipitate was
removed by filtration through a plug of silica on a glass frit.
Removal of the solvent resulted in a colorless, viscous oil.
n-Pentane (100 mL) was added, and the cloudy solution was
kept at -18 °C. The resulting white needles were isolated and
dried under vacuum. Overall isolated yield: 2.73 g (76%). 31P-
{1H} NMR: δ -17.3 (s). 1H NMR: δ 7.72-6.91 (m region, 14H,
C3-Rh1-P3
C4-Rh1-P3
C3-Rh1-P1
C4-Rh1-P1
P3-Rh1-P1
C3-Rh1- P5
C4-Rh1-P5
P3-Rh1-P5
142.89(18)
173.22(18)
99.26(17)
90.46(17)
83.18(7)
82.65(17)
87.41(17)
99.16(7)
C7-Rh2-P4
C8-Rh2-P4
C7-Rh2-P6
C8-Rh2-P6
P4-Rh2-P6
C7-Rh2-P2
C8-Rh2-P2
P4-Rh2-P2
140.99(17)
172.10(17)
101.26(18)
90.01(16)
83.00(7)
82.20(18)
87.00(16)
100.52(7)
complexes, each with a tetrafluoroborate anion and one
molecule of CH2Cl2. The molecular structure of one of
the independent rhodium cations is shown in Figure 4.
Selected bond lengths and angles are given in Table 1.
The coordination geometry around the rhodium cen-
ters can be rationalized as distorted square planar,
where the three phosphorus atoms and the â-carbon in
the vinyl group occupy the corners. The mean deviation
from these planes is 0.0726 and 0.1133 Å for Rh1 and
Rh2, respectively. The R carbons are located 1.34 and
1.39 Å above these planes in the two asymmetric units,
respectively. Despite this apparent asymmetry of the
vinyl group, it is clearly an η2 coordination with M-C
distances being slightly (but not significantly in asym-
metric unit 1) longer for the â-carbons and the C-C axis
being approximately perpendicular to the coordination
plane (cf. Figure 4). As expected, the vinylic bonds are
longer than those in, for example, ethylene (1.339(5)
Å).19 The counteranions show large disorder in some of
their atoms; no attempts were made to resolve this.
H
Ar), 6.64 (m, 1H, ArCH(CH3)), 2.26-1.96 (m region, 2H, CH2-
3
3
CH3), 1.37 (d, J HH ) 6.3 Hz, 3H, ArCH(CH3)), 1.04 (t, J HH
)
7.2 Hz, 3H, CH2CH3). 13C{1H} NMR: δ 173.2 (s, CdO), 146.9-
3
125.8 (m region, CAr), 70.0 (d, J PC ) 28.1 Hz, ArCH(CH3)),
27.5 (s, CH2CH3), 21.9 (s, ArCH(CH3)), 9.0 (s, CH2CH3). IR
(KBr): 1729 cm-1 (s, CdO). Anal. Calcd for C23H23O2P: C, 76.2;
H, 6.4. Found: C, 76.0; H, 6.6.
Rea ction of [Rh (DP P B)(NBD)][BF 4] (2a ) w ith P h os-
p h in e 1a . F or m a tion of [Rh (DP P AM)(DP P B)][BF 4] (3a ).
Dichloromethane (10 mL) was added to a Schlenk tube
containing [Rh(DPPB)(NBD)][BF4] (118 mg, 0.166 mmol),
DPPAM (1a ; 60.6 mg, 0.167 mmol), and a magnetic stirring
bar, and this resulted in an orange solution. A freeze-pump-
thaw cycle replaced the argon atmosphere with molecular
hydrogen. The closed Schlenk tube was allowed to reach room
temperature and then stirred for 40 min, resulting in a yellow-
tinted solution. Once again the atmosphere was replaced, this
(20) Fairlie, D. P.; Bosnich, B. Organometallics 1988, 7, 936.
(21) Uson, R.; Oro, L. A.; Garralda, M. A.; Claver, M. C.; Lahuerta,
P. Transition Met. Chem. 1979, 4, 55.
(19) Costain, C. C.; Stoicheff, B. P. J . Chem. Phys. 1959, 30, 777.
(22) Schiemenz, G.; Kaack, H. Liebigs Ann. Chem. 1973, 1494.

Cyclometalation Reactions on Rhodium(I)
Organometallics, Vol. 20, No. 23, 2001 4925
time from molecular hydrogen to argon, and the solution was
allowed to stand for 5 min. Removal of the solvent under
vacuum gave a yellow powder, which was washed with 1 ×
10 mL of Et2O and 2 × 10 mL of n-hexane. Finally, the air-
stable compound 3a was dried under vacuum. Yield: 148 mg
(KBr): 1572 cm-1 (s, CdO). MS (FAB+): m/z 862 [C49H48NOP3-
Rh+]. Anal. Calcd for C49H48BF4NOP3Rh: C, 62.0; H, 5.1.
Found: C, 61.7; H, 5.1.
Rea ction of [Rh (DP P E)(NBD)][BF 4] (2b) w ith P h os-
p h in e 1c. F or m a t ion of [R h ((()-DP P E S)(DP P E )][BF 4]
(5). Dichloromethane (10 mL) was added to a Schlenk tube
containing [Rh(DPPE)(NBD)][BF4] (2b; 63.8 mg, 93.8 µmol),
(()-DPPES (1c; 34.1 mg, 94.1 µmol), and a magnetic stirring
bar, and this resulted in an orange solution. A freeze-pump-
thaw cycle replaced the argon atmosphere with molecular
hydrogen. The closed Schlenk tube was allowed to reach room
temperature, and its contents were then stirred for 5 min,
resulting in a light yellow solution. Once again the atmosphere
was replaced, this time from molecular hydrogen to argon, and
the solution was allowed to stand for 2 min. Concentration of
the solution to ca. 1 mL, followed by addition of 10 mL of Et2O/
n-hexane (1:1), gave complex 5 as a light yellow precipitate.
The powder was isolated by filtration and dried under vacuum.
The complex was too unstable for microanalysis. Yield: 74.0
1
(91%). 31P{1H} NMR (CD2Cl2): δ 48.9 (ddd, J RhP ) 188.5 Hz,
1
2
2J PP ) 51.1, 34.5 Hz, PDPPB), 39.0 (ddd, J RhP ) 145.1 Hz, J PP
1
2
) 302.7, 51.1 Hz, PDPPB), 30.3 (ddd, J RhP ) 132.3 Hz, J PP
)
)
302.7, 34.5 Hz, PDPPAM). 1H NMR (CD2Cl2): δ 8.98 (bd, 2J HH
14.4 Hz, 1H, ArCH2), 8.35-6.03 (m region, 34H, HAr), 3.81 (bd,
2J HH ) 14.4 Hz, 1H, ArCH2), 2.53 (s, 3H, NCH3), 2.50-0.90
3
(m region, 10H, CH2CH3 and CH2), -0.13 (t, J HH ) 7.0 Hz,
3H, CH2CH3). 13C{1H} NMR (CD2Cl2): δ 175.6 (s, CdO),
139.6-129.3 (m region, CAr), 37.3 (s, NCH3), 36.2 (m, CH2),
30.2 (bs, CH2), 29.9 (bs, CH2), 29.5 (s, ArCH2), 26.0 (s, CH2-
CH3), 25.1 (bs, CH2), 9.4 (s, CH2CH3). IR (KBr): 1586 cm-1 (s,
CdO). Anal. Calcd for C51H52BF4NOP3Rh: C, 62.6; H, 5.4.
Found: C, 62.3; H, 5.5.
Rea ction of [Rh (DP P E)(NBD)][BF 4] (2b) w ith P h os-
p h in e 1a . F or m a tion of [Rh (DP P AM)(DP P E)][BF 4] (3b).
Using a procedure identical with that for 3a using [Rh(DPPE)-
(NBD)][BF4] (49.9 mg, 73.4 µmol) and DPPAM (1a ; 26.7 mg,
73.9 µmol) gave complex 3b as an air-stable yellow powder.
Yield: 57.9 mg (83%). 31P{1H} NMR: δ 79.4 (ddd, 1J RhP ) 195.9
1
mg (83%). 31P{1H} NMR (CD2Cl2): δ 78.2 (ddd, J RhP ) 205.3
2
1
Hz, J PP ) 35.7, 31.9 Hz, PDPPE), 65.8 (ddd, J RhP ) 142.0 Hz,
2J PP ) 298.1, 35.7 Hz, PDPPE), 25.9 (ddd, J RhP ) 127 Hz, J PP
1
2
1
) 298.1, 31.9 Hz, PDPPES). H NMR (CD2Cl2): δ 9.05 (m, 1H,
H
Ar), 8.12-6.62 (m region, 33H, HAr), 2.51-1.26 (m region, 5H,
2
1
CH2 and ArCH(CH3)), 1.70 (d, 3J HH ) 4.9 Hz, 3H, ArCH(CH3)),
1.19-0.92 (m region, 2H, CH2CH3), 0.41 (t, 3H, 3J HH ) 7.4 Hz,
CH2CH3). IR (KBr): 1661 cm-1 (s, CdO). MS (FAB+): m/z 863
[C49H47O2P3Rh+].
Hz, J PP ) 37.3, 30.9 Hz, PDPPE), 66.3 (ddd, J RhP ) 144.7 Hz,
2J PP ) 308.9, 37.3 Hz, PDPPE), 29.4 (ddd, 1J RhP ) 127.3 Hz, 2J PP
1
2
) 308.9, 30.9 Hz, PDPPAM). H NMR: δ 8.48 (bd, J HH ) 13.6
Hz, 1H, ArCH2), 8.17-6.34 (m region, 34H, HAr), 4.07 (bd, 2J HH
) 13.6 Hz, 1H, ArCH2), 2.75 (s, 3H, NCH3), 2.51-0.86 (m
Con ver sion of Com p lex 5 to [Rh (DP P STY)(DP P E)]-
[BF 4] (6). A tube containing [Rh((()-DPPES)(DPPE)][BF4] (5;
50.2 mg, 52.8 µmol), dissolved in 3 mL of CH2Cl2, was placed
inside a Schlenk vessel containing Et2O. After 7 days deep red,
air-stable crystals of 6 were harvested from the inner tube.
Yield: 33.3 mg (72%). 31P{1H} NMR (263 K): δ 63.9 (ddd, 1J RhP
3
region, 6H, CH2CH3 and CH2), 0.25 (t, J HH ) 7.3 Hz, 3H,
CH2CH3). 13C{1H} NMR: δ 175.6 (s, CdO), 137.2-128.2 (m
region, CAr), 52.8 (m, CH2), 40.8 (m, CH2), 36.6 (s, NCH3), 35.7
(s, ArCH2), 28.0 (s, CH2CH3), 8.8 (s, CH2CH3) (assignments of
1H and 13C{1H} NMR signals were confirmed by an HMQC
NMR experiment). IR (KBr): 1601 cm-1 (s, CdO). Anal. Calcd
for C49H48BF4NOP3Rh: C, 62.0; H, 5.1. Found: C, 61.8; H, 5.1.
Con ver sion of Com p lex 3a to [Rh H(DP P AM)(DP P B)]-
[BF 4] (4a ). A Schlenk tube was charged with [Rh(DPPAM)-
(DPPB)][BF4] (3a ; 72.2 mg, 73.9 µmol), a magnetic stirring bar,
and 10 mL of THF. The vessel was closed, and the stirred
solution was heated at 100 °C for 90 min, resulting in a color
change from bright to light yellow. Evaporation of the solvent
under vacuum gave 4a as an off-white powder in a quantitative
2
1
) 120.9 Hz, J PP ) 316.8, 26.8 Hz, PDPPE), 53.9 (ddd, J RhP
)
153.2 Hz, 2J PP ) 31.1, 26.8 Hz, PDPPE), 38.9 (ddd, 1J RhP ) 123.3
Hz, 2J PP ) 316.8, 31.1 Hz, PDPPSTY). 1H NMR (263 K): δ 7.72-
6.59 (m region, 34H, HAr), 5.25 (m, 1H, ArCHCH2), 4.18 (m,
1H, ArCHCH2), 3.65 (m, 1H, ArCHCH2), 2.58-2.02 (m region,
4H, CH2). Anal. Calcd for C46H41BF4P3Rh‚1/2CH2Cl2: C, 60.8;
H, 4.6. Found: C, 60.6; H, 4.8.
Rea ction of [Rh (NBD)2][BF 4] w ith P h osp h in e 1c. F or -
m a tion of [Rh H((()-DP P ES)2][BF 4] (7a -d ). Using a pro-
cedure identical with that described for the formation of 5 but
using [Rh(NBD)2][BF4] (28.3 mg, 75.6 µmol) and (()-DPPES
(1c; 55.3 mg, 0.152 mmol) gave complexes 7a -d as yellow
powders. Yield: 69.1 mg (81%). 31P{1H} NMR (253 K): δ 58.5
1
yield. 31P{1H} NMR (CD2Cl2): δ 56.1 (ddd, J RhP ) 109.5 Hz,
1
2J PP ) 372.6, 22.5 Hz, PDPPAM), 40.0 (ddd, J RhP ) 110.7 Hz,
1
2
2J PP ) 372.6, 28.8 Hz, PDPPB), 17.7 (ddd, J RhP ) 83.7 Hz, J PP
) 28.8, 22.5 Hz, PDPPB). 1H NMR (CD2Cl2): δ 7.82-6.47 (m
2
region, 34H, HAr), 3.81 (bd, J RhH ) 17.7 Hz, 1H, ArCHRh),
1
2
1
(dd, J RhP ) 169.2 Hz, J PP ) 36.3 Hz, Pcis, 7a ), 58.2 (dd, J RhP
2.52-0.70 (m region, 10H, CH2CH3 and CH2), 2.09 (s, 3H,
2
1
) 185.1 Hz, J PP ) 36.0 Hz, Pcis, 7b), 43.4 (dd, J RhP ) 126.2
3
NCH3), 0.54 (t, J HH ) 7.4 Hz, 3H, CH2CH3), -17.2 (m, 1H,
2
1
Hz, J PP ) 358.3 Hz, Ptrans, 7c), 41.5 (dd, J RhP ) 120.4 Hz,
RhH). 13C{1H} NMR (CD2Cl2): δ 179.4 (s, CdO), 135.9-128.5
1
2
2J PP ) 371.1 Hz, Ptrans, 7d ), 31.3 (dd, J RhP ) 120.2 Hz, J PP
)
3
(m region, CAr), 72.6 (m, ArCHRh), 35.1 (d, J RhC ) 4.2 Hz,
1
2
371.1 Hz, Ptrans, 7d ), 27.4 (dd, J RhP ) 115.5 Hz, J PP ) 358.3
1
NCH3), 31.3 (d, J PC ) 21.0 Hz, CH2), 28.5 (m, CH2), 26.9 (s,
1
2
Hz, Ptrans, 7c), 19.7 (dd, J RhP ) 152.0 Hz, J PP ) 36.3 Hz, Pcis
,
2
7a ), 15.6 (dd, J RhP ) 154.4 Hz, J PP ) 36.0 Hz, Pcis, 7b). 1H
NMR (253 K): δ -21.9 (m, RhH, 7c,d ), -23.3 (m, RhH, 7a ,b)
(ratio 2:3) (assignments in the normal region impossible due
to severe overlapping). IR (KBr): 1683 cm-1 (bs, CdO). Anal.
Calcd for C46H46BF4O4P2Rh: C, 60.4; H, 5.1. Found: C, 60.0;
H, 5.2.
CH2CH3), 25.3 (d, J PC ) 4.8 Hz, CH2), 21.7 (bs, CH2), 8.4 (s,
1
2
CH2CH3). IR (KBr): 1571 cm-1 (s, CdO). MS (FAB+): m/z 890
[C51H52NOP3Rh+]. Anal. Calcd for C51H52BF4NOP3Rh: C, 62.6;
H, 5.4. Found: C, 62.3; H, 5.5.
Con ver sion of Com p lex 3b to [Rh H(DP P AM)(DP P E)]-
[BF 4] (4b). Using a procedure identical with that for 4a using
[Rh(DPPAM)(DPPE)][BF4] (3b; 50.2 mg, 52.9 µmol) gave
complex 4b as an off-white powder in quantitative yield. 31P-
Str u ctu r e Deter m in a tion s. Crystal data and details
about data collection are given in Table 2. The intensity data
sets for both 3b and 6 were collected at 293 K with a Bruker
SMART CCD system using ω-scans and a rotating anode with
Mo KR radiation (λ ) 0.710 73 Å).23 The intensity was
corrected for Lorentz, polarization, and absorption effects using
SADABS.24 In both structures the first 50 frames were
1
2
{1H} NMR (CD2Cl2): δ 69.4 (ddd, J RhP ) 108.4 Hz, J PP
)
)
1
2
370.8, 15.6 Hz, PDPPE), 61.9 (ddd, J RhP ) 108.7 Hz, J PP
370.8, 20.3 Hz, PDPPAM), 51.7 (ddd, 1J RhP ) 83.2 Hz, 2J PP ) 20.3,
15.6 Hz, PDPPE). 1H NMR (CD2Cl2): δ 7.97-6.78 (m region,
34H, HAr), 4.94 (m, 1H, ArCHRh), 2.92-2.12 (m region, 4H,
CH2), 2.25 (s, 3H, NCH3), 1.10-0.96 (m region, 2H, CH2CH3),
0.36 (t, 3J HH ) 7.2 Hz, 3H, CH2CH3), -17.4 (m, 1H, RhH). 13C-
{1H} NMR (CD2Cl2): δ 179.3 (s, CdO), 135.6-126.1 (m region,
(23) BrukerAXS, SMART, Area Detector Control Software; Bruker
Analytical X-ray Systems, Madison, WI, 1995.
(24) Sheldrick, G. M. SADABS, Program for Absorption Correction;
University of Go¨ttingen, Go¨ttingen, Germany, 1996.
C
Ar), 70.9 (m, ArCHRh), 46.4 (m, CH2), 36.2 (m, CH2), 34.8 (d,
3J RhC ) 7.6 Hz, NCH3), 28.7 (s, CH2CH3), 7.6 (s, CH2CH3). IR

4926 Organometallics, Vol. 20, No. 23, 2001
Sjo¨vall et al.
Ta ble 2. Cr ysta l Da ta for 3b a n d 6
3b
Non-H atoms were refined with anisotropic displacement
parameters. Hydrogen atoms (except on methyl groups and
on C6 in 3b) were constrained to parent sites, using a riding
model.
6
formula
C
49H48BF4NOP3Rh
C46H41BF4P3Rh‚
1/2CH2Cl2
918.88
Kin etics Mea su r em en ts. A J . Young NMR tube was
charged with ca. 10 mg of complex 3b, and 0.75 mL of 1,1,2,2-
tetrachloroethane-d2 was added. The conversion of 3b was
fw
949.51
space group
a/Å
P1h
P1h
11.4318(6)
13.0682(6)
15.9971(8)
85.4390(10)
84.3880(10)
71.0460(10)
2246.55(19)
2
4.804(3)
14.977(3)
20.943(4)
75.61(3)
87.31(3)
71.07(3)
4251.4(15)
4
1
monitored by H NMR spectroscopy at four different temper-
b/Å
c/Å
atures ranging from 332.7 to 362.7 K until at least 90%
conversion. Temperatures were measured using ethylene
glycol. No deuterium incorporation was observed during the
reaction or in the final product upon standing.
R/deg
â/deg
γ/deg
V/Å3
Z
D
calcd/g cm-3
1.404
1.436
Ack n ow led gm en t. We thank Prof. Åke Oskarsson
for valuable discussions of the X-ray diffraction work.
We are also grateful to Dr. Torsten Malmstro¨m, Astra-
Zeneca, for running an HMQC experiment. We thank
one reviewer for pointing out the possibility of different
configurations with respect to the Rh-C bond in 7.
Financial support from the Swedish Natural Science
Research Council and the Swedish Foundation for
Strategic Research is gratefully acknowledged.
µ/mm-1
0.541
0.628
θ range/deg
no. of rflns collected
no. of unique rflns
Ra
1.28-31.82
1.00-32.11
23 520
44 146
13 576
0.0550
26 150
0.0622
b
Rw
0.1070
0.1213
Sc
0.892
0.763
Rint
0.0613
0.1161
a
b
2
R ) ∑(|Fo| - |Fc|)/∑|Fo|, Rw ) [∑w(|Fo| - |Fc|)2/∑|Fo| ]1/2
,
c S
) [∑w(|Fo| - |Fc|)2/(m - n)]1/2
Su p p or tin g In for m a tion Ava ila ble: Complete crystal-
lographic data in the CIF format. This material is available
free of charge via the Internet at http://pubs.acs.org. Data for
complexes 3b (CCDC 169397) and 6 (CCDC 169398) have also
been deposited with the Cambridge Crystallographic Data
Centre. Tables of structure factors are available from the
authors upon request.
collected again at the end to check for decay and no decay was
observed. All reflections were integrated using SAINT.25 Both
structures were solved by direct methods and refined by full-
matrix least-squares calculations on F2 using SHELXTL 5.1.26
(25) BrukerAXS, SAINT, Integration Software; Bruker Analytical
X-ray Systems, Madison, WI, 1995.
(26) Sheldrick, G. M. SHELXTL5.1, Program for Structure Solution
and Least Squares Refinement; University of Go¨ttingen, Go¨ttingen,
Germany, 1998.
OM010543X