Reactivity of novel N, N′-diphosphino-silanediamine-based rhodium(I) derivatives

Article abstract of DOI:10.1021/ic2004408

The coordination abilities of the novel N,N′-diphosphino- silanediamine ligand of formula SiMe₂(NtolPPh₂) ₂ (SiNP, 1) have been investigated toward rhodium, and the derivatives [RhCl(SiNP)]₂ (2), [Rh(SiNP)(COD)]

Full text of DOI:10.1021/ic2004408

ARTICLE
pubs.acs.org/IC
Reactivity of Novel N,N0-Diphosphino-Silanediamine-Based
Rhodium(I) Derivatives
Vincenzo Passarelli*,† and Franco Benetollo‡
†Centro Universitario de la Defensa, Ctra. Huesca s/n, ES-50090 Zaragoza, Espa~na
‡Instituto di Chimica Inorganica e delle Superfici (ICIS), Consiglio Nazionale delle Ricerche, Corso Stati Uniti 4, I-35127 Padova, Italia
S Supporting Information
b
ABSTRACT: The coordination abilities of the novel N,N0-diphosphino-silanediamine
ligand of formula SiMe2(NtolPPh2)2 (SiNP, 1) have been investigated toward rhodium,
and the derivatives [RhCl(SiNP)]2 (2), [Rh(SiNP)(COD)][BF4] (3), and Rh(acac)-
(SiNP) (4) have been synthesized. The stability of the dinuclear frame of [RhCl(SiNP)]2
(2) toward incoming nucleophiles has been shown to be dependent on their π-acceptor
ability. Indeed, the mononuclear complexes RhCl(SiNP)(L) (L = CO, 5; CNtBu, 6) have
been isolated purely and quantitatively upon reaction of 2 with CO and CNtBu,
respectively. Otherwise, PPh3 and RhCl(SiNP) equilibrate with Rh(Cl)(SiNP)(PPh3)
(7). Carbon electrophiles such as MeI and 3-chloro-1-proprene afforded the oxidation of
rhodium(I) to rhodium(III) and the formation of RhCl2(η3-C3H5)(SiNP) (8) and
Rh(Me)(I)(SiNP)(acac) (10), respectively. The methyl derivative 10 is thermally stable
and does not react either with CO or with CNtBu even in excess. Otherwise, RhCl2(η3-
C3H5)(SiNP) (8) is thermally stable but reacts with CO, affording 3-chloro-1-proprene
and RhCl(SiNP)(CO) (5). Finally, upon reaction of Rh(acac)(SiNP) (4) and 3-chloro-1-proprene, RhCl(acac)(η1-C3H5)(SiNP)
(9a) and [Rh(acac)(η3-C3H5)(SiNP)]Cl (9b) could be detected at 233 K. At higher temperatures, 9a and 9b smoothly decompose,
affording the dinuclear derivative [RhCl(SiNP)]2 (2) and the CC coupling product 3-allylpentane-2,4-dione.
’ INTRODUCTION
promote their catalytic activity5,6 and (b) the reported route to
SiNpyP suggests that, in principle, the appropriate choice of the
starting materials could allow easy incorporation in the PꢀNꢀ
SiꢀNꢀP frame of a great variety of substituents (both at silicon,
nitrogen, and/or phosphorus), we decided to contribute to
the so far barely explored chemistry of the N,N0-diphosphino-
silanediamines by preparing a new ligand of this family and
investigating its interaction with rhodium as well as the reactivity
of the newly prepared complexes.
The rational design of novel ligands or the structural mod-
ification of known ones, either at the donor atoms or of their
overall structure, is a basic strategy used to modulate the
reactivity of metal complexes. Indeed, the change in the electro-
nic and/or steric properties of the donor atoms as well as of the
ligand itself can induce a tailored reactivity at the coordinated
metal center. Relevant to this paper, ligands containing PꢀN
bonds have shown great versatility in both organometallic
chemistry and catalysis,1 and a few years ago, Woolins and co-
workers reported2,3 the stepwise synthesis of SiMe2[(C5H4N-2)-
NPPh2]2 (from now onward SiNpyP, Scheme 1A) via the amino-
lysis of the Ph2PꢀCl bond by 2-aminopyridine, NH deprotonation
of the resulting phosphinoamine (C5H4N-2)NHPPh2, and final
chloride substitution in SiMe2Cl2 (Scheme 1A). Also, the palladium
and platinum complexes of formula MX2(SiNpyP) (X = Cl, Me)
were reported along with the observation3 that the reaction of SiNpyP
with [RhCl(COD)]2 or [RhCp*Cl2]2 results in the hydrolysis of the
SiꢀN bond (Scheme 1A) probably due to adventitious water.
On the other hand, a survey of the literature showed that,
besides the above-mentioned MX2(SiNpyP) (M = Pd, Pt; X = Cl,
Me), very few further examples4 (namely with Ni, Pd, Pt, Mo,
and W, Scheme 1B) are reported, in which a transition metal is
bonded to a PꢀNꢀSiꢀNꢀP scaffold.
’ RESULTS AND DISCUSSION
The ligand N,N0-bis(diphenylphosphino)-1,1-dimethyl-
N,N0-di-p-tolyl-silanediamine SiMe2(NtolPPh2)2 (SiNP, 1) is cleanly
obtained by the two-step synthesis reported in Scheme 2: 1,1-
diphenyl-N-(p-tolyl)phosphinoamine (NHtolPPh2, A) is pre-
pared purely and with high yields through the reaction of NH2tol
with PClPh2 (2:1 molar ratio); the deprotonation of A by nBuLi
and the following reaction with SiMe2Cl2 yields SiNP (1) as a
pure microcrystalline colorless solid (Scheme 2). Relevant NMR
data for 1 are reported in Table 1.
SiNP is thermally stable toward O2; indeed oxidation was not
observed even after prolonged (up to 2 days) exposure to O2
(either atmospheric or pure), both in toluene solution or in the
Thus, provided that (a) diphosphines are versatile ligands
extensively used in order to both stabilize transition metals and
Received: March 4, 2011
Published: September 22, 2011
r
2011 American Chemical Society
9958
dx.doi.org/10.1021/ic2004408 Inorg. Chem. 2011, 50, 9958–9967
|

Inorganic Chemistry
ARTICLE
Scheme 1
Scheme 2
and 13C NMR spectra are in agreement with the proposed
symmetric square planar arrangement of the donor atoms around
the metal center (i.e., equivalent methyl groups of the SiMe2
moiety, equivalent tolyl groups, equivalent PPh2 groups, each
containing equivalent phenyl rings).
When dealing with the dinuclear structure of 2, relevant data
are in order. The diffusion coefficients of 2 and 4 were measured
in C6D6 solutions (D2 = 4.42ꢁ 10ꢀ6 cm2/s; D4 = 5.88 ꢁ
10ꢀ6 cm2/s), showing that 2 diffuses slower than 4 and as a
consequence should be bigger than 4. Furthermore the ratio
D4/D2 (1.33) is very close to the calculated value (1.26) for two
diffusing species, one (2) with a 2-fold volume with respect to the
other (4).7 Also, a sharp lorenztian profile was observed for
the 31P NMR signal of the mononuclear derivative 4, whereas the
line shape of the 31P NMR signal of 2 appeared to be a bit more
complex (Figure 1). Definitely, in agreement with the proposed
dinuclear structure, the line shape analysis showed that 2 should
contain two magnetically nonequivalent phosphorus nuclei, and
there exist long-range P-to-P (2.0 Hz) and P-to-Rh couplings
(ꢀ0.4 Hz; Figure 1, cf. Supporting Information).
solid state, and eventually heating at 80 °C. On the other hand,
water may afford the hydrolisis of the SiꢀN bond, but heat and a
large excess of water are necessary (water/acetone 50%vol, 70 °C,
6 h) in order to observe approximately 40% conversion of 1 to
NHtolPPh2 and uncharacterized silane derivatives.
Finally, it is noteworthy that the structural motif (P2)Rh(μ-Cl)2-
Rh(P2) has already been observed in rhodium chemistry (P2 =
two monodentate phosphines or a bidentate one).8
The reactions of 2 with CO, CNtBu, and PPh3 afford the
fragmentation of the dinuclear compounds yielding the corre-
sponding square planar mononuclear rhodium(I) derivative 5ꢀ7
(Scheme 4).
It is worth mentioning that the carbonyl (5) and isonitrile (6)
derivatives are thermally stable and are formed quantitatively
upon the reaction of 2 with CO and CNtBu, respectively. Relevant
NMR data are given in Table 1. Similar to 2ꢀ4, the 31P NMR
signals are observed to be downfield shifted with respect to the
uncomplexed ligand, and an AX system and a strong dependence
of both the chemical shifts and the 1JPRh coupling constant on the
trans ligand are observed. Interestingly, the carbonyl stretching
IR band of 5 is observed at 2023 cmꢀ1, just between the values
observed for PPh2-based diphosphino9 or diphosphinito10 deri-
vatives of formula RhCl(CO)(P2), thus indicating an intermediate
A family of square planar rhodium(I) complexes is obtained by
the reaction of SiNP with classical rhodium(I) precursors
(Scheme 3). Indeed, acetonitrile in [Rh(COD)(MeCN)2]+ is
readily displaced by SiNP, affording the cationic mononuclear
complex [Rh(SiNP)(COD)]+ (3). In a similar way, CO and
COD are substituted in Rh(acac)(CO)2 and [RhCl(COD)]2,
respectively, affording the bis-chelate complex Rh(acac)(SiNP) (4)
and the chloro-bridged derivative [RhCl(SiNP)]2 (2) (Scheme 3).
Relevant NMR data for the compounds 2ꢀ4 are given in
Table 1. In general, the coordination of SiNP to rhodium(I)
causes the downfield shift of its 31P signal, and the observed 1JPRh
coupling constants are in the range typical of rhodium(I) com-
plexes (Table 1). Furthermore, the observed patterns in the 1H
9959
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Table 1. Selected 31P NMR Data for Compounds 1ꢀ10
= 2.0 Hz, 3JPRh = ꢀ0.4 Hz (cf. Figure 1 and Supporting Information). b 2JPPtrans = 365.0 Hz. c 2JPPcis = 33.5 Hz.
a 4
J
PP
The mononuclear RhCl(PPh3)(SiNP) (7) derivative was
observed in equilibrium with PPh3 and [RhCl(SiNP)]2
(2; Scheme 4). The thermodynamic parameters of the reac-
tion “2 + 2 PPh3 h 2 7” were obtained through variable
temperature 31P NMR measurements in the range 303ꢀ333 K
Scheme 3
(cf. Supporting Information). In agreement with the lower
11
BDERhꢀCl(bridging) vs BDERhꢀP
,
the enthalpic variation is
negative (ΔHr = ꢀ81.5 ( 0.3 kJ/mol), and ΔSr is ꢀ283 ( 112
J/molK, reasonably as a consequence of the variation of the
overall number of moles (Δn = ꢀ1).
Similar to 5 and 6, the 31P NMR spectrum of 7 (Table 1)
shows downfield shifted signals of the two nonequivalent SiNP
phosphorus nuclei which afford an AX system. Furthermore,
as expected, the 31P NMR resonance for coordinated PPh3 is
observed as a doublet of doublets of doublets at 32.7 ppm
(2JPPcis = 33.5, 2JPPtrans = 365.0, 1JPRh = 132.6 Hz).
Finally, no reaction was observed between 2 and weaker π-
acceptor ligands such as 1-hexene, styrene, and acrylonitrile, even
after prolonged reaction times (up to 72 h) and heating (60 °C),
thus reasonably indicating that the stability of the square planar
rhodium(I) derivatives of formula RhCl(L)(SiNP) raises on
increasing the π-acceptor ability of L.
electron density at the metal center of 5 compared with the
mentioned complexes.
9960
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Figure 1. (left) Experimental (A) and simulated (B, δP = 86.0 ppm; 1JPRh = 201.4; 3JPRh = ꢀ0.4, 4JPP = 2.0 Hz) 31P NMR spectrum of 2 at 298 K (121.45
MHz). (right) 31P NMR (202.47 MHz) spectrum of 4.
enantiomers, namely Δ and Λ (Figure 2), are present in the
crystal (1:1 ratio; cf. Supporting Information).
Scheme 4
As far as the PNSiNP scaffold is concerned, only the platinum
complex PtCl2(SiNpyP)3 and the azasilatrane derivative PtCl2-
(L), L = EtOSi(Ph2PNCH2CH2)2(HNCH2CH2)N,4 have been
structurally characterized so far (Scheme 1A, B). When
comparing14 PtCl2(SiNpyP) and 8, similar intrannular distances
are observed (Table 2); furthermore, the boat conformations of
the six-membered [PNSiNPM] ring of the two complexes are
quite similar.
Finally, it is worth mentioning that the observed bite angle
(91.35°) and distances (2.266, 2.306 Å) for the coordinated
SiNP ligand in 8 are very close to the average values reported
by Dierkes and van Leeuwen15 for the related 1,3-bis-
(diphenylphosphino)propane (average bite angle, 91.56°;
average MꢀP distances, 2.29 Å).
The oxidative addition of carbon electrophiles such as organic
halides is one of the most general routes in order to obtain
metalꢀcarbon bonds, which are the key functional groups on the
way to the functionalization of organic substrates. Since π-allylic
derivatives are well established intermediates in CC bond for-
mation via allylation (vide infra), the interaction of 3-chloro-
1-propene with selected rhodium(I) derivatives among those
described above was investigated. The reaction of 2 with
3-chloro-1-propene readily and cleanly affords the oxidative
addition product RhCl2(η3-C3H5)(SiNP) (8; eq 1). Its solid
state structure is shown in Figure 2, along with the coordination
polyhedron and a view of the six-membered [PNSiNPRh] ring;
selected bond distances and angles are given in Table 2.
The 31P NMR spectrum of 8 shows two nonequivalent
phosphorus nuclei as an AX system, indicating that the non-
symmetric solid state structure should be preserved in solution
(vide infra). As far as the allyl moiety is concerned, the expected
1H and 13C patterns were observed with two nonequivalent
methylene and one methyne moiety whose assignment is based
1
13
on the 1Hꢀ H NOESY, 1Hꢀ C HSQC, and 13C NMR spectra,
cf. Experimental Section. In this regard, the most relevant
observations are the following: the CaH2 methylene protons
(δH = 2.57, 1.64) show a NOE with the o-PPh2 protons (δH =
7.99), while Cc is observed as a doublet (δC = 86.6, 2JPC = 29 Hz).
Thus, as far as the enantiomer Δ is concerned, P1 should be
pseudo-cis to Ca and pseudo-trans to Cc (Scheme 5B).
When dealing with the SiNP ligand, similar to the solid state
structure, both the SiMe2 methyls and the two SiNP “arms” are
nonequivalent (1H, 13C, cf. Experimental Section), in agreement
with the existence of two pairs of nonequivalent semispaces
(Scheme 5A) separated by the σv (left/right) and σh (up/down)
planes, respectively.
1
A distorted η3-coordination [RhꢀC(1), 2.138(6) Å; RhꢀC(2),
2.204(6) Å; RhꢀC(3), 2.331(6) Å; C(1)ꢀC(2), 1.428(9) Å;
C(2)ꢀC(3), 1.368(9) Å] of the allyl moiety is observed with
rhodium-to-carbon distances (Table 2) similar to those observed
in known rhodium(III) allyl complexes.13
The coordination sphere of rhodium is completed by two
mutually pseudo-cis chloride ligands [RhꢀCl(1), 2.466(2) Å;
RhꢀCl(2) 2.441(2) Å; Cl(1)ꢀRh-Cl(2), 90.03(6)°] and by the
phosphorus atoms from the chelating SiNP ligand [RhꢀP(1),
2.266(2) Å; RhꢀP(2), 2.306(2) Å; P(1)ꢀRh-P(2), 91.35(5)°].
As a consequence, the complex is asymmetric, and indeed two
Finally, the room temperature H NMR spectrum contains
broad aromatic resonances (cf. Experimental Section), so a
fluxional process should be going on in solution. Indeed, the
1
1Hꢀ H EXSY spectrum contains exchange crosspeaks between
the methyl signals of the tolyl groups, pointing at the fact that the
left and right semispaces are exchanging. On the other hand,
crosspeaks have not been detected between the SiMe2 methyls,
so the up and down semispaces do not exchange (Scheme 5A). On
these bases, the fluxionality should rely on the interconversion
between the two enantiomers Δ and Λ with the concomitant
exchange between P1 and P2 (Scheme 5B).
9961
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Figure 2. (A) View of the molecular structure of 8 with the numbering scheme adopted (only ipso carbons are shown for clarity) with ellipsoids at 50%
probability. (B) Coordination polyhedron of 8. (C) View of the [PNSiNPRh] ring of 8. (D) Representation of Δ and Λ configurations of 8 as seen in the
solid state (cf. Supporting Information).
Table 2. Selected Angles and Bond Distances of 8 and
PtCl2(SiNpyP)3 with Its Numbering Scheme
(cf. Experimental Section and Supporting Information), and the
activation parameters were obtained by the Eyring plot,
indicating an intramolecular nondissociative mechanism
(ΔHact = 77 ( 3 kJ/mol, ΔSact = 2 ( 916 J/molK). Thus,
the ΔꢀΛ interconversion could occur via a turn-style mech-
anism through a trigonal prism intermediate (Scheme 5C).
Finally, as far as 8 is concerned, it should be mentioned that
it is indefinitely thermally stable (when stored under an argon
atmosphere) but readily reacts with CO (C6D6 solution,
1 atm, room temperature), affording 3-chloro-1-propene (1H,
GC-MS) and 5 (1H, 31P; eq 1), thus indicating the ability of
the [Rh(SiNP)] moiety to promote coupling reactions with
the coordinated allyl group.
The organic chloride 3-chloro-1-propene also reacts with 4 at
room temperature, and the clean formation (3 h) of 2 is observed
along with stoichiometric amounts (1:1) of 3-allylpentan-2,4-
dione (Scheme 6). In order to clarify the outcome of this
reaction, it was carried out at 223 K (acetone-d6) and two labile
intermediates were observed, namely, Rh(η1-C3H5)(Cl)(acac)-
(SiNP) (9a) and [Rh(η3-C3H5)(acac)(SiNP)]+ (9b). Selected
NMR data are given in Figure 3.
It is noteworthy that 9b shows a 31P NMR pattern virtually
1
2
identical to that of 8 (δP, JPRh and JPP), thus indicating the
presence of two nonequivalent phosphorus atoms and suggesting
a similar arrangement around the metal center (Figure 3).
Furthermore, the 1H signals of the allyl moiety are indicative of
an η3-allyl moiety (Figure 3; cf. data for 8).
At 228 K (in CDCl3), the exchange slows down, and two
nonequivalent PPh2 moieties are observed also in the H
1
On the other hand, 9a should contain an η1-allyl group and a
symmetrically coordinated bidentate SiNP ligand (Figure 3). In
this regard, 10 was prepared as a model compound for 9a by
reacting MeI with 4.17
NMR spectrum, each featuring two nonequivalent Ph rings.
Also, at 228 K, the rotation of the tolyl rings around the
NꢀCipso bond is inhibited. The complete assignment of the
1H, 13C, and 31P NMR signals was carried out using standard
1D and 2D NMR techniques, and reported in the Supporting
Information.
Additionally, as far as the equilibrium between 9a and 9b is
concerned, it is worth mentioning that the η3ꢀη1 rearrangement
of metallic allylic derivatives has been proposed as the key step
responsible for the loss of regioselectivity in allylic alkylation
reactions.18
The kinetic constants for the ΔꢀΛ exchange were calcu-
1
1
lated from H H EXSY spectra in the range 283ꢀ303 K
9962
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Scheme 5
Scheme 6
’ CONCLUSIONS
As far as the reactivity of the above-mentioned rhodium(I)
complexes is concerned, this study indicates that selected rhodium-
(I) species can be cleanly oxidized by organic halides, namely,
methyl iodide and allyl chloride. In the resulting compounds 8ꢀ10,
the coordination of the SiNP ligand is retained, and no side-
reactions have been detected (e.g., electrophilic attack to the
uncoordinated nitrogen atoms). Furthermore, the stability/behav-
ior of the oxidized organometallic species is observed to depend on
the ancillary ligands. The chloride allyl derivative 8 is chiral (Δ, Λ),
and the enantiomers interconvert by an intramolecular pathway
through of a prism-like activation state (Scheme 5). On the other
hand, the η3-allyl cation 9b, isoelectronic with 8, could be observed
only at low temperatures in equilibrium with the related η1-allyl
complex 9b. To the best of our knowledge, this η1ꢀη3 equilibrium
has been claimed in order to justify the loss of regioselectivity in
rhodium-catalyzed allylation reactions but never directly observed.18
The novel ligand SiMe2(NtolPPh2)2 (SiNP) featuring a PNSiNP
scaffold could be prepared by a step-by-step synthesis based on
the aminolysis of the PꢀCl bond of PClPh2 and the following
Cl/amido metathesis in SiMe2Cl2. The presence of nitrogen and
phosphorus atoms suggests that the SiNP molecule could act as
either a k2-N,N0, k2-P,P0, or k-N/k-P ligand. Nevertheless, the
reactivity of SiNP toward rhodium(I) indicates that diphosphine-
like coordination is preferred; indeed, complexes 2ꢀ7 exhibit
a k2-P,P0 coordinated SiNP ligand. When compared with
the structurally related 1,3-bis(diphenylphosphino)propane,
SiNP features similar bite angles and distances. Nevertheless,
the overall structure appears to be more hindered, probably due
to the congestion within methyl, N-tolyl, and P-phenyl substit-
uents. Definitely, low temperature NMR measurements of
selected compounds showed the hindered rotation of these rings.
On the other hand, the presence of the nitrogen atom attached to
the phosphorus atom reduces the basicity of this atom, as
confirmed by the frequency of the stretching band of the CO
ligand of 5.
’ EXPERIMENTAL SECTION
All of the operations were carried out using standard Schlenk-tube
techniques under an atmosphere of prepurified argon or in a Braun
9963
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
1
6.23; N, 4.81. H NMR (CDCl3, 298 K): δ 7.52 (m, 4H, o-PPh),
7.45ꢀ7.40 (6H, m-PPh + p-PPh), 7.04 (d, 2H, 3JHH = 8.0 Hz, C3Htol),
3
6.83 (d, 2H, JHH = 8.0 Hz, C2Htol), 4.37 (br, 1H, NH), 2.29 (s, 3H,
CH3tol). 13C NMR (CDCl3, 298 K): δ 144.1 (d, 1JCP = 17 Hz, ipso-PPh),
140.4 (d, 2JCP = 12 Hz, C1-tol), 131.2 (d, 2JCP = 20 Hz, o-PPh), 129.82 (s,
C
3-tol), 129.03 (s, p-PPh), 128.79 (s, CCH3tol), 128.59 (d, 3JCP = 5 Hz,
3
m-PPh), 116.1 (d, JCP = 12 Hz, C2-tol), 20.5 (s, CH3tol). 31P NMR
(CDCl3, 298 K): δ, 28.7 (s).
b. Reaction of NH(tol)PPh2 with BuLi and SiMe2Cl2. A solution of
NH(tol)PPh2 (1.74 g, 5.97 mmol) in THF (8 mL) + toluene (20 mL) at
198 K was treated wih nBuLi (1.6 M in hexane, 3.70 mL, 5.92 mmol).
The mixture was allowed to warm up to room temperature (2 h), and
SiMe2Cl2 (0.360 mL, 2.98 mmol, d=1.07 g/mL) was added. After
stirring overnight at room temperature, the suspension was heated at
333 K for 3 h. All of the volatiles were removed in vacuo, and the residue
was extracted with CH2Cl2. The liquid phase was concentrated and
hexane added, thus affording the precipitation of a colorless solid, which
was filtered off, washed with Et2O, dried in vacuo, and finally identified as
SiMe2[N(tol)PPh2]2 (1, SiNP, 1.30 g, 68% yield). Found: C, 75.25; H,
6.18; N, 4.42. Calcd for C40H40N2P2Si (638.79): C, 75.20; H, 6.31; N,
4.39. 1H NMR (CDCl3, 298 K): δ 7.35ꢀ7.23 (10H, PPh), 6.79 (d, 2H,
3JHH = 8.2 Hz, C3Htol), 6.54 (d, 2H, 3JHH = 8.2 Hz, C2Htol), 2.26 (s, 3H,
CH3tol), 0.54 (s, 3H, SiCH3). 13C NMR (CDCl3, 298 K): δ 141.6
(m, ipso-PPh), 139.4 (m, C1-tol), 134.1 (s, C4-tol), 133.6 (m, o-PPh),
130.4 (s, C2-tol), 128.6 (s, C3-tol), 128.4 (s, p-PPh), 127.6 (m, m-PPh),
20.9 (s, CH3tol), 2.3 (t, 3JCP = 9.6 Hz, SiCH3). 31P NMR (CDCl3, 298
K): δ, 49.6 (s).
Synthesis of [RhCl(SiNP)]2 (2). A yellow solution of [RhCl(COD)]2
(80.8 mg, 0.164 mmol) in CH2Cl2 (5 mL) was treated with SiNP
(209 mg, 0.327 mmol). The solution readily changed to red-orange. After
30 min of stirring, the solution was concentrated and Et2O (5 mL) added,
thus yielding the precipitation of a deep orange solid, which was filtered
off, washed with Et2O (2 mL), dried in vacuo, and finally identified as
[RhCl(SiNP)]2 (2, 194 mg, 76% yield). Found: C, 62.00; H, 5.21; N,
3.45. Calcd for C40H40ClN2P2RhSi (777.15): C, 61.82; H, 5.19; N, 3.60.
1H NMR (C6D6, 298 K): δ 7.69 (m, 4H, o-PPh), 7.04 (t, 3JHH = 7.2 Hz,
2H, p-PPh), 6.92 (t, 3JHH = 7.3 Hz, 4H, m-PPh), 6.72 (d, 3JHH = 8.2 Hz,
2H, C2Htol), 6.56 (d, 3JHH = 8.2 Hz, 2H, C3Htol), 1.88 (s, 3H, CH3tol),
0.73 (s, 3H, SiCH3). 13C NMR (C6D6, 298 K): δ 141.8 (m, C1-tol), 137.5
(m, ipso-PPh), 134.9 (m, o-PPh), 134.5 (s, C4-tol), 131.1 (s, C2-tol), 129.0
(s, C3-tol), 128.5 (s, p-PPh), 126.7 (m, m-PPh), 20.7 (s, CH3tol), 4.1 (t,
3JCP = 2.95, SiCH3). 31P NMR: 86.0 (m, 1JPRh = 201.4 Hz, 4JPP = 2.0 Hz,
3JPRh = ꢀ0.4 Hz), cf. Results and Discussion, and Supporting Informa-
tion for further details. MS (MALDI+, DCTB): m/z 1552.2 (M+, 2%),
791.1 ([M/2+CH3]+, 15%), 775.1 ([M/2 ꢀ H]+, 50%), 761.2 ([M/2 ꢀ
CH3]+, 55%), 757.2 ([M/2 ꢀ CH3 ꢀ 4H]+, 100%).
Figure 3. 31P NMR spectrum (acetone-d6, 233 K) of the reaction
mixture obtained after adding 3-chloro-1-propene to Rh(acac)(SiNP)
(4), with the proposed assignment. Selected 1H NMR data are in order:
9a, δH = 1.13 (CaH2), 5.24 (CbH), 4.21 and 4.30 (CcH2), 5.23 (CHacac).
9b, δH = 1.49 and 2.40 (CaH2), 5.10 (CbH), 4.35 and 4.82 (CcH2), 5.41
(CHacac).
glovebox under dinitrogen. The solvents were dried and purified accord-
ing to standard procedures. NH2tol (Aldrich), PClPh2 (Aldrich),
SiMe2Cl2 (Aldrich), CH2CHCH2Cl (Aldrich), PPh3 (Aldrich), CNtBu
(Aldrich), and CO (Praxair) were commercially available and were used
as received if not otherwise stated.
NMR spectra were measured with Bruker spectrometers (AV300,
AV400, AV500) and are referred to SiMe4 (1H, 13C) and H3PO4 (31P).
Infrared spectra were recorded on a ThermoNicolet Avatar 360 FT-IR
spectrometer. Elemental analyses were performed by using a Perkinꢀ
Elmer 2400 microanalyzer. Mass spectra (MALDI) were recorded with a
Bruker MicroFlex spectrometer using DCTB (1,1-dicyano-4-t-butyl-
phenyl-3-methylbutadiene) as a matrix.
The diffusion experiments were performed using the stimulated echo
pulse sequence19 without spinning. The shape of the gradient pulse was
rectangular, and its strength varied automatically in the course of the
experiments. The diffusion coefficients (D) were determined from the
slope of the regression line ln(I/I0) versus G2 according to ln(I/I0) =
ꢀ(γδ)2(Δꢀδ/3)DG2, in which I and I0 are the observed spin echo
intensity with and without gradients, respectively, G is the gradient
strength, Δ is the delay between the midpoints of the gradients, D is the
diffusion coefficient, and δ is the gradient length. The calibration of the
gradients was carried out by a diffusion measurement of HDO in D2O.
The experimental error in the D values was estimated to be 2%. All of the
data leading to the reported D values afforded lines with correlation
coefficients of >0.999, and 10ꢀ16 points were used for regression
analysis. The value for the δ parameter was set to 1.75ꢀ2 ms.
Additionally, three different measurements with different diffusion
parameters (Δ=25ꢀ100 ms) were always carried out to check the
reproducibility of the data obtained. The gradient strength was increased
in 5ꢀ8% steps. A measurement of the 1H NMR spectrum and T1 was
carried out before each diffusion experiment and the recovery delay
(aprox. 5ꢀ8 s) set to 5 times T1. The number of scans varied between 32
and 64 per increment. The typical experiment duration was 1.5ꢀ3 h.
Synthesis of SiMe2[N(tol)PPh2]2 (1, SiNP). a. Synthesis of
NH(tol)PPh2. Freshly sublimed NH2tol (3.38 g, 31.5 mmol) was treated
with PClPh2 (2.80 mL, d = 1.194 g/mL, 15.2 mmol) in THF (100 mL).
The ready precipitation of a colorless solid was observed. After 30 min of
stirring, the solid was filtered off and the filtrate evaporated, thus yielding
a colorless solid identified as NH(tol)PPh2 (3.40 g, 77% yield). Found:
C, 78.51; H, 6.45; N, 4.75. Calcd for C19H18NP (291.33): C, 78.33; H,
Synthesis of [Rh(SiNP)(COD)][BF4] (3). A solution of [Rh(COD)-
(CH3CN)2][BF4] (129 mg, 0.339 mmol) in CH2Cl2 (10 mL) was
treated with SiNP (220 mg, 0.344 mmol). After 30 min of stirring, the
solution was concentrated and hexane added, thus yielding the
precipitation of a yellow solid, which was filtered off, dried in vacuo,
and finally identified as [Rh(COD)(SiNP)][BF4] (3, 251 mg, 79%
yield). Found: C, 61.89; H, 5.48; N, 2.85. Calcd for C48H52BF4-
N2P2RhSi (936.68): C, 61.55; H, 5.60; N, 2.99. 1H NMR (CD2Cl2,
298 K): δ 7.50ꢀ7.40 (12H, o-PPh + p-PPh), 7.33 (t, 3JHH = 7.3 Hz,
3
3
8H, m-PPh), 6.84 (d, JHH = 8.1 Hz, 4H, C3Htol), 6.66 (d, JHH
=
8.1 Hz, 4H, C2Htol), 4.66 (br, 4H, CHCOD), 2.40ꢀ2.24 (8H,
CH2COD), 2.19 (s, 6H, CH3tol), 0.74 (s, 6H, SiCH3). 13C NMR
(CD2Cl2, 298 K): δ 139.3 (m, C1-tol), 136.4 (s, C4-tol), 133.5 (m, o-
PPh), 131.3 (s, p-PPh), 130.8 (m, ipso-PPh), 129.4 (C2-tol + C3-tol),
128.8 (m, m-PPh), 100.4 (dt, J = 7.3, 4.9; CHCOD), 30.5 (s, CH2COD),
20.46 (s, CH3tol), 3.29 (t, 3JCP = 3.8, SiCH3). 31P NMR (CD2Cl2, 298
K): δ 66.1 (d, 1JPRh = 153.5 Hz). MS (MALDI+, DCTB): m/z 849.2
(M+, 17%), 741.1 ([M ꢀ COD]+, 100%).
9964
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Scheme 7
Scheme 8
6.67 (d, 3JHH = 8.20 Hz, 2H, C3Htol1), 2.04 (s, 3H, CH3tol2), 1.99 (s, 3H,
CH3tol1), 0.76 (s, 9H, CH3tBu), 0.69 (s, 6H, SiCH3). 13C NMR (C6D6,
298 K): δ 141.9 (dd, 2JCP = 11.0, 4JCP = 2.1, C1-tol2), 141.4 (dd, 2JPP
=
4
1
Synthesis of Rh(acac)(SiNP) (4). A toluene (10 mL) solution of
Rh(acac)(CO)2 (145 mg, 0.562 mmol) was treated with SiNP (360 mg,
0.564 mmol). The prompt evolution of gas was observed along with a
color change from yellow-green to orange. After 30 min of stirring, the
solution was concentrated and hexane added (20 mL), thus yielding the
precipitation of an orange solid, which was filtered off, dried in vacuo, and
identified as Rh(acac)(SiNP) (4, 308 mg, 65% yield). Found: C, 64.01;
H, 5.55; N, 3.48. Calcd for C45H47N2O2P2RhSi (840.81): C, 64.28; H,
5.63; N, 3.33. 1H NMR (C6D6, 298 K): δ 8.03 (m, 8H, o-PPh), 7.23 (t,
3JHH = 7.2 Hz, 4H, p-PPh), 7.18 (t, 3JHH = 7.3 Hz, 8H, m-PPh), 6.93 (d,
3JHH = 8.2 Hz, 4H, C2Htol), 6.74 (d, 3JHH = 8.2 Hz, 4H, C3Htol), 5.30 (s,
1H, CHacac), 2.03 (s, 6H, CH3tol), 1.63 (s, 6H, CH3acac), 0.98 (s, 6H,
SiCH3). 13C NMR (C6D6, 298 K): δ 184.2 (s, CO), 141.5 (m, C1-tol),
137.2 (m, ipso-PPh), 134.7 (m, o-PPh), 134.6 (s, C4-tol), 131.1 (s, C2-tol),
129.0 (s, C3-tol), 128.6 (s, p-PPh), 126.7 (m, m-PPh), 99.7 (d, J = 1.9 Hz,
10.1, JPP = 1.7, C1-tol1), 139.6 (d, JCP = 51.4, ipso-P1Ph), 135.33 (d,
1JCP = 46.7, ipso-P2Ph), 135.30 (d, 2JCP = 11.9, o-P2Ph), 134.95 (d, 5JCP
=
5
2
1.5, C4-tol2), 134.67 (d, JCP = 1.4, C4-tol1), 134.45 (d, JCP = 12.5, o-
P1Ph), 131.1 (d, 4JCP = 1.9, C3-tol2), 130.9 (d, 4JCP = 1.8, C3-tol1), 129.1
(s, C2-tol2), 129.0 (d, 3JCP = 1.7, p-P2Ph), 128.9 (C2-tol1 + p-P1Ph), 127.0
(d, JCP = 10.0, m-PPh), 55.5 (br, CtBu), 29.4 (s, CH3tBu), 20.58
3
(s, CH3tol2), 20.50 (s, CH3tol1), 3.8 (s, SiCH3). 31P NMR (C6D6, 298 K):
δ 80.8 (dd, 1JPRh = 184.7, 2JPPcis = 43.3, P1), 76.7 (dd, 1JPRh = 139.4,
2JPPcis = 43.3, P2).
c. PPh3. In a NMR tube, [RhCl(SiNP)]2 (2, 15.3 mg, 9.84 μmol) was
dissolved in C6D6 (0.379 g, d = 0.950 g/mL, 0.399 mL). Sequentially,
PPh3 (5.8 mg, 22 μmol) was added, thus affording an orange solution.
The mixture was analyzed by 1H and 31P NMR spectroscopy, proving
to be an equilibrium mixture of [RhCl(SiNP)]2 (δP = 86.1, d, 1JPRh
=
201.8 Hz), PPh3 (δP = ꢀ5.1, s) and RhCl(SiNP)(PPh3) (7; vide infra).
The determination of the thermodynamic parameters of the equilibrium
(ΔHr = ꢀ81.5 ( 0.3 kJ/mol; ΔSr = ꢀ282 ( 1 J/molK) was carried out
through variable temperature 31P NMR measurements in the range
303ꢀ333 K (cf. Supporting Information). Selected NMR data of
RhCl(SiNP)(PPh3) (7) are in order. 31P NMR (C6D6, 293 K): δ
32.7 (ddd, 2JPPtrans = 365.0, 1JPRh = 132.6, 2JPPcis = 33.5, 1P, PPh3), 78.6
(ddd, 2JPPtrans = 365.0, 1JPRh = 152.3, 2JPPcis = 41.7, 1P, P trans to PPh3),
87.6 (ddd, 1JPRh = 195.7, 2JPPcis = 41.7, 33.5, 1P, P cis to PPh3). Scheme 7
CHacac), 26.9 (td, J = 3.5, 0.8, CH3acac), 20.6 (s, CH3tol), 3.86 (t, 3JCP
3.1, SiCH3). 31P NMR (C6D6, 298 K): δ 93.0 (d, 1JPRh = 198.7 Hz).
=
Reaction of [RhCl(SiNP)]2 with PPh3, CO, CNtBu. a. CO. A
toluene solution (5 mL) of [RhCl(SiNP)]2 (2, 150 mg, 96.5 μmol) was
bubbled with CO over 10 min, thus turning deep yellow. On standing
(approximately 1 h), the precipitation of a yellow solid was observed; it
was filtered off, dried in vacuo, and identified as RhCl(SiNP)(CO) (5,
116 mg, 75% yield). Found: C, 61.25; H, 4.95; N, 3.47. Calcd for
1
1
shows the assigned δH of 7 based on 1Hꢀ H COSY, 1Hꢀ H NOESY,
1
C41H40ClN2OP2RhSi (805.16): C, 61.16; H, 5.01; N, 3.48. H NMR
31
and 1Hꢀ P HMBC spectra.
(acetone-d6, 298 K): δ 7.65 (m, 4H, o-P1Ph), 7.55 (m, 4H, o-P2Ph), 7.44
(t, 2JHH = 7.4, 4H, p-P1Ph + p-P2Ph), 7.30 (t, 3JHH = 7.6, 8H, m-P1Ph +
Synthesis of RhCl2(η3-C3H5)(SiNP) (8). A toluene solution
(15 mL) of [RhCl(SiNP)]2 (2, 242 mg, 0.156 mmol) was treated with
3-chloro-1-propene (25.5 μL, 0.939 g/mL, 0.313 mmol). The solution
was allowed to stand overnight, yielding the abundant precipitation of a
pale yellow microcrystalline solid which was filtered off, washed with
Et2O (2 ꢁ 3 mL), dried in vacuo, and identified as RhCl2(η3-C3H5)-
(SiNP) (8, 189 mg, 71% yield). Found: C, 61.01; H, 5.42; N, 3.15. Calcd
for C43H45Cl2N2P2RhSi (853.67): C, 60.50; H, 5.31; N, 3.28. 1H NMR
3
3
m-P2Ph), 6.87 (d, JHH = 8.2, 2H, C3Htol2), 6.80 (d, JHH = 8.2, 2H,
3
3
C3Htol1), 6.73 (d, JHH = 8.2, 2H, C2Htol2), 6.57 (d, JHH = 8.2, 2H,
C2Htol1), 2.18 (s, 3H, CH3tol2), 2.14 (s, 3H, CH3tol1), 0.48 (s, 6H,
SiCH3). 13C NMR (acetone-d6, 298 K): δ 187.7 (ddd, J = 14.6, 65.6,
122.2 Hz, CO), 140.1 (d, JCP = 9.8, C1-tol2), 139.8 (d, JCP = 9.8,
2
2
C
1-tol1), 137.1 (d, 1JCP = 56.9, ipso-PPh), 135.8 (s, C4-tol2), 135.5 (s, C4-tol1),
134.8 (d, JCP = 11.9, o-P1Ph), 133.5 (d, JCP = 12.4, o-P2Ph), 132.6
(d, 1JCP = 48.7, ipso-PPh), 130.8 (s, C2-tol2), 130.6 (s, C2-tol1), 130.2 (d,
4JCP = 2.0, p-P2Ph), 130.0 (d, 4JCP = 2.0, p-P1Ph), 129.1 (s, C3-tol2), 128.9
2
2
(CD2Cl2, 298 K, δC as obtained from H 13C HSQC spectrum, cf.
1
Scheme 8 for labeling): δ 9.07 (very br, no integrable), 7.99 (m, 2H,
PPh, δC = 132.8), 7.94 (t, 2JHH = 7.5, 1H, PPh, δC = 131.5), 7.86 (m, 2H,
PPh, δC = 128.9), 7.53 (t, 2JHH = 7.6, 1H, PPh, δC = 130.5), 7.43 (d,
2JHH = 7.9, 1H, C2Hr, 1H, δC = 133.8), 7.31 (b, 2H, PPh, δC = 126.3),
7.22 (t, 2JHH = 6.7, 1H, PPh, δC = 131.1), 7.18 (d, 2JHH = 7.9, 2H, C2Hr +
(s, C3-tol1), 127.5 (d, JCP = 10.6, m-P1Ph), 127.2 (d, JCP = 10.6,
3
3
m-P2Ph), 19.94 (s, CH3tol2), 19.87 (s, CH3tol1), 2.63 (t, JCP = 3.3,
3
SiCH3). 31P NMR (acetone-d6, 298 K): δ 75.8 (dd, 2JPP = 42.1, 1JPRh
=
168.8, P2), 67.6 (dd, JPP = 42.1; JPRh = 132.3, P1). IR (CH2Cl2):
2
1
2023 cmꢀ1 (νCO).
C2Hl, δC = 129.1, 131.8), 7.12 (br, 2H, PPh, δC = 131.1), 7.04 (d, 2JHH
8.5, 1H, C3Hl, δC = 128.9), 7.00 (m, 2H, PPh, δC = 126.8), 6.80 (d, 2JHH
=
=
b. CNtBu. A toluene solution (10 mL) of [RhCl(SiNP)]2 (2, 160 mg,
0.103 mmol) was treated with CNtBu (23.3 μL, d = 0.735 g/mL,
0.206 mmol). The solution readily turned pale yellow. After 30 min of
stirring, the volatiles were removed in vacuo and the residue recrystalized
from toluene/hexane and identified as RhCl(SiNP)(CNtBu) (6, 152 mg,
86% yield). Found: C, 63.10; H, 5.70; N, 4.91. Calcd for C45H49ClN3-
P2RhSi (860.28): C, 62.83; H, 5.74; N, 4.88. 1H NMR (C6D6, 298 K): δ
8.20 (m, 4H, o-P2Ph), 7.96 (m, 4H, o-P1Ph), 7.21ꢀ7.11 (12H, m-P1Ph +
m-P2Ph + p-P1Ph + p-P2Ph), 6.86 (d, 3JHH = 7.7 Hz, 2H, C2Htol2), 6.76
(d, 3JHH = 8.2 Hz, 2H, C2Htol1), 6.74 (d, 3JHH = 7.7 Hz, 2H, C3Htol2),
8.50, 1H, C3Hr, δC = 129.1), 6.67ꢀ6.61 (3H, PPh + C3Hl + C3Hr, δC =
130.4, 128.4), 5.47 (d, 2JHH = 8.2, 1H, C2Hr/l, δC = 131.8), 5.09 (m, 1H,
CbH, δC = 109.5), 5.02 (dd, J = 8.3, 13.7, 1H, CcHtrans, δC = 86.6, 2JCP
=
2
29 Hz), 4.63 (t, J = 7.1, CcHcis, δC = 86.6, JCP = 29 Hz), 2.57 (d,
J = 11.3, 1H, CaHtrans, δC = 55.7), 2.22 (s, 3H, CH3tol,l, δC = 20.4), 2.08 (d,
J = 2.08, 3H, CH3tol,r, δC = 20.3), 1.64 (dq, J = 6.6, 2.0, CaHcis, δC = 55.7),
d
u
0.89 (s, 3H, Si CH3 , δC = 2.19), ꢀ0.27 (s, 3H, SiCH3 , δC = 2.29). 31
P
=
1
NMR (CD2Cl2, 298 K, cf. Scheme 8 for labeling): δ 79.3 (dd, JPRh
117.8, 2JPPcis = 21.6, P2), 65.7 (dd, 1JPRh = 151.1, 2JPPcis = 21.6, P1).
9965
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
Table 3. Crystal Data and Structure Refinement Parameters
for RhCl2(η3-C3H5)(SiNP) 2CH2Cl2 (8 2CH2Cl2)
Scheme 9
3
3
formula
fw
C45H49Cl6N2P2RhSi
1023.50
triclinic
P1
cryst syst
space group
a, Å
9.778(2)
10.863(2)
22.465(3)
87.49(3)
87.66(3)
88.64(3)
2381.3(7)
2
(10, 155 mg, 63% yield). Found: C, 56.15; H, 5.22; N, 2.64. Calcd for
1
b, Å
C46H50IN2O2P2RhSi (982.74): C, 56.22; H, 5.13; N, 2.85. H NMR
(C6D6, 298 K, δC from 1H 13C HSQC NMR, cf. Scheme 9 for labeling):
δ 8.57 (br, 4H, PPh, C not observed), 7.91 (br, 4H, PPh, δC = 133.5),
7.34 (dd, J = 8.2, 1,8; 2H, C2Hd, δC = 131.7), 7.17 (dd, J = 8.2, 1.9, 2H,
c, Å
α, deg
β, deg
γ, deg
C2Hu, δC = 135.7), 7.14 (6H, PPh, δC = 129.4, 126.4), 7.10 (t, 2JHH
=
7.3, 2H, PPh, δC = 130.1), 7.02 (br, 4H, PPh, δC = 125.6), 6.72 (dd, J =
8.2, 1.9, 2H, C3Hd, δC = 128.1), 6.47 (dd, J = 8.2, 1.9, 2H, C3Hu, δC =
129.1), 5.63 (s, 1H, CHacac, δC = 99.8), 1.97 (s, 6H, Meacac, δC = 27.7),
1.94 (s, 6H, Metol, δC = 20.5), 1.17 (s, 3H, SiMeu, δC = 3.1), 0.71 (q, J =
2.2, 3H, RhMe, δC = 32.3, dt, J = 23.0, 4.8), 0.69 (s, 3H, SiMed, δC = 5.3).
δC (from 13C NMR spectrum): 184.9 (CO), 139.1 (m, C1), 135.8 (C4),
134.0 (m, ipso-PPh). 31P NMR (C6D6, 298 K): δ 90.7 (d, 1JPRh = 143.9).
1H NMR (acetone-d6, 203 K, cf. Scheme 9 for labeling): δ 9.83 (br,
1H, o0-PPhd, δC = 141.4), 8.61 (br, 1H, o0-PPhu, δC = 132.6), 7.78
(t, 3JHH = 6.8 Hz, 1H, m0-PPhu, δC = 127.7), 7.43ꢀ7.31 (2H, p-PPhu,
130.2, + C2Hu, δC = 131.9), 7.13 (t, 3JHH = 6.9 Hz, 1H, m00-PPhu, δC =
126.0), 6.95 (d, 3JHH = 7.4 Hz, 1H, C3Hu, δC = 128.4), 6.64 (br, o00-PPhu,
δC = 133.3), 6.57 (t, 3JHH = 7.4 Hz, 1H, m00-PPhd, δC = 125.3), 6.53 (d,
3JHH = 8.5 Hz, 1H, C3Hd, δC = 128.4), 6.32 (d, 3JHH = 8.2 Hz, 1H, C2Hd,
ꢀ3
V, Å
Z
Dcalcd, g cmꢀ3
1.427
T, K
293(2)
0.821
μ(Mo Kα), mmꢀ1
no. reflns measd
no. obsd reflns [I g 2σ(I)]
8843
7403
R = ∑|Fo| ꢀ |Fc|/∑|Fo|
0.064
Rw ={∑[w(Fo ꢀ Fc2)2]/∑[w(Fo2)2]}1/2
0.146
2
δC = 129.66, + m0-PPhd, δC = 125.5), 7.28ꢀ7.18 (2H, p-PPhd, δC
=
GOF
1.252
The fluxional behavior of 8 was investigated via 1Hꢀ H EXSY and the
activation parameters obtained from the Eyring plot over the temperature
range 283ꢀ303 K (cf. Supporting Information; ΔHact = 77 ( 3 kJ/mol;
ΔSact = 2 ( 9 J/mol K). The crosspeaks between signals of the two
nonequivalent tolyl C4 Me groups were used to calculate the exchange
rate constants (k) given in the Supporting Information; k = IK/[tM(IK +
ID)], where ID and IK are the diagonal and crosspeak integrals,
respectively, and tM = mixing time (1.0 s).
1
δC = 134.8), 6.04 (br, 1H, o00-PPhd, δC = 135.3), 5.41 (s, 1H, CHacac
,
δC = 99.1), 2.05 (s, 6H, Metol, δC = 20.0), 1.70 (s, 6H, Meacac, δC
=
27.1), 0.43 (s, 3H, SiMed, δC = 1.5), 0.41 (s, 3H, SiMeu, δC = 4.4), ꢀ0.08
(br, 3H, RhMe, δC = 32.5). δC (from H 13C HSQC NMR): 184.1
1
(CO), 138.2 (C1), 136.0 (C4). 31P NMR (acetone-d6, 203 K): δ 89.6
(d, 1JPRh = 143.3 Hz).
31P NMR (CDCl3, 228 K): δ 79.5 (dd, 1JPRh = 118.3, 2JPP = 21.1 Hz,
1P, P1), 64.9 (dd, 1JPRh = 149.9, 2JPP = 21.1 Hz, 1P, P2).
Reaction of Rh(acac)(SiNP) (4) with 3-Chloro-1-propene.
3-Chloro-1-propene (24.6 μL, 0.939 g/mL, 0.302 mmol) was added to a
toluene solution (15 mL) of Rh(acac)(SiNP) (4, 254 mg, 0.302 mmol).
The mixture was allowed to stir at room temperature for 3 h. A portion of
the liquid was analyzed by GC-MS, and the presence of 3-allylpentane-
2,4-dione was observed along with traces (<5%) of 3,3-diallylpentane-
2,4-dione. All of the volatiles were removed in vacuo and the sticky
residue washed with hexane (2 ꢁ 5 mL), thus yielding a deep orange
solid spectroscopically (1H, 31P) identified as [RhCl(SiNP)]2 (2, 208
mg, 89% yield). As a confirmation, when the reaction was carried out on
a NMR-tube scale (12.0 mg of 4, 1.2 μL of allylchloride), the clean
formation of 3-allylpentane-2,4-dione and 3,3-diallylpentane-2,4-dione
(traces) along with [RhCl(SiNP)]2 was observed.
X-Ray Measurements and Structure Determination of 8.
Single crystals of RhCl2(η3-C3H5)(SiNP) (8) were obtained by the slow
diffusion of hexane into a CH2Cl2 solution of the compound. Crystal
data, collected reflections, and parameters of the final refinement are
reported in Table 3. Intensity data were collected using a Philips
PW1100 diffractometer (FEBO system) with graphite-monochromated
(Mo Kα) radiation, following the standard procedures. There were no
significant fluctuations of intensities other than those expected from
Poisson statistics. All intensities were corrected for Lorentz polarization
and absorption.20 The structures were solved by standard direct
methods.21 All non-H atoms were located in the subsequent Fourier
maps. Refinement was carried out by full-matrix least-squares proce-
When the reaction was carried out at 203 K, the intermediates
RhCl(acac)(η1-C3H5)(SiNP) (9a) and [Rh(acac)(η3-C3H5)(SiNP)]Cl
(9b) were detected and spectroscopically characterized (cf. Figure 3 for
selected NMR data).
2
dures (based on Fo ) using anisotropic temperature factors for all non-
hydrogen atoms. The H atoms were placed in calculated positions with
fixed, isotropic thermal parameters (1.2Uequiv) of the parent carbon
atom. Calculations were performed with the SHELX-97 program22
implemented in the WinGX package.23
Reaction of RhCl2(η3-C3H5)(SiNP) (8) with CO. In a NMR
tube, RhCl2(η3-C3H5)(SiNP) (8, 10.7 mg, 12.5 μmol) was dissolved in
C6D6 (0.4 mL). The resulting solution was saturated with CO and
stirred under a CO atmosphere for 2 h. The mixture was analyzed via 1H
and 31P NMR, showing the clean and quantitative formation of RhCl-
(CO)(SiNP) (5) and 3-chloro-1-propene (confirmed by GC-MS).
Synthesis of Rh(Me)(I)(acac)(SiNP) (10). A solution of MeI
(15.6 μL, 2.28 g/mL, 0.251 mmol) in toluene was added to Rh(acac)-
(SiNP) (4, 211 mg, 0.251 mmol) dissolved in toluene (15 mL). The
mixture was allowed to stir at room temperature for 2 h. Then, all
volatiles were removed and the residue crystallized from toluene/hexane
’ ASSOCIATED CONTENT
S
Supporting Information. Additional NMR data, details
b
about the activation and thermodynamic parameters’ determina-
tions, and a CIF file. This material is available free of charge via
the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: passarel@unizar.es.
9966
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967

Inorganic Chemistry
ARTICLE
’ ACKNOWLEDGMENT
chelating acac ligand in the equatorial plane and the iodide in the
remaining axial position. As a confirmation, the two SiMe2 methyls are
not equivalent, one facing RhMe and the other facing iodide. As far as the
PPh2 moieties are concerned, the H NMR spectrum shows several
broad signals indicative of a fluxional behavior, which reasonably could
Funding was provided by MICINN (CTQ 2008-03860) and
Fondazione G. Donegani—Accademia dei Lincei (G. Donegani
Programme). V.P. thanks CSIC for the “I3P-Doctores” grant
(2006-09).
1
be the exchanging between the up and down semispace (cf. Experimental
1
Section). Indeed, the H NMR at 203 K shows well-resolved sharp
resonances which could be assigned to two equivalent PPh2 moieties,
each containing two nonequivalent phenyl rings.
’ REFERENCES
(1) Gopalakrishnan, J. Appl. Organomet. Chem. 2009, 23, 291ꢀ318
(18) Wucker, B.; Moser, M.; Schumacher, S. A.; Rominger, F.; Kunz,
D. Angew. Chem., Int. Ed. 2009, 48, 4417. Evans, P. A.; Nelson, J. D. J. Am.
Chem. Soc. 1998, 120, 5581. Muraoka, T.; Matsuda, I.; Itoh, K.; Ueno, K.
Organometallics 2007, 26, 387.Leahy, D. K.; Evans, P. A., Rhodium(I)-
catalyzed Allylic Substitution Reactions and their Apllications to Target
Directed Synthesis. In Modern Rhodium-Catalyzed Reactions; Evans,
P. A., Ed.; Wiley-VCH: Weinheim, Germany, 2005; pp 191ꢀ214.
(19) Stilbs, P. Prog. Nucl. Magn. Reson. Spectrosc. 1987, 19, 1–45.
Price, W. S. Concepts Magn. Reson. 1997, 9, 299. Price, W. S. Concepts
Magn. Reson. 1998, 10, 197. Johnson, C. S., Jr. Prog. Nucl. Magn. Reson.
Spectrosc. 1999, 34, 203. Cohen, Y.; Avram, L.; Frish, L. Angew. Chem.
2005, 117, 524. Cohen, Y.; Avram, L.; Frish, L. Angew. Chem., Int. Ed.
2005, 44, 520. Pregosin, P. S.; Kumar, P. G. A.; Fernandez, I. Chem. Rev.
2005, 105, 2977. Macchioni, A.; Ciancaleoni, G.; Zuccaccia, C.;
Zuccaccia, D. Chem. Soc. Rev. 2008, 37, 479.
(20) North, A. T. C.; Philips, D. C.; Mathews, F. S. Acta Crystallogr.
1968, A24, 351–359.
(21) Burla, A. 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.
(22) Sheldrick, G. M. SHELXL-97; University of G€ottingen:
G€ottingen, Germany, 1997.
(23) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837–838.
and references therein.
(2) Clarke, M. L.; Slawin, A. M. Z.; Woolins, J. D. Phosphorus Sulfur
Silicon Relat. Elem. 2001, 168ꢀ169, 329–332.
(3) Aucott, S. M.; Clarke, M. L.; Slawin, A. M. Z.; Woolins, J. D.
J. Chem. Soc., Dalton Trans. 2001, 972–976.
(4) Gudat, D.; Daniels, L. M.; Verkade, J. G. Organometallics 1990,
9, 1464–1470.
(5) Dowing, J. H.; Smith, M. B. In Comprehensive Coordination
Chemistry, Clever, J. A., Meyer, T. J., Eds.; Elsevier Pergamon Press:
Oxford, U.K., 2004; Vol. 1, pp 253ꢀ296.
(6) Selected references are: Balakrishna, M. S.; Chandrasekaran, P.;
George, P. P. Coord. Chem., Rev. 2003, 241, 87–117. Dahlenburg, L.
Coord. Chem. Rev. 2005, 249, 2962–2992. Birkholz, M. N.; Freixa, Z.;
Van Leeuwen, P. W. N. M. Chem. Soc. Rev. 2009, 38, 1099–1118.Zhang,
X.; Chi, Y.; Tang, W. In Comprehensive Organometallic Chemistry III;
Crabtree, R. H., Mingos, D. M. P., Eds.; Elsevier: Oxford, U.K., 2007;
Vol. 10, p 1.
(7) According to the StokesꢀEinstein equation for spherical diffus-
ing species [DX = (kBT)/(6πηrX), kB is the Boltzman constant, T is the
temperature, and η is the viscosity of the solvent]. The diffusion
coefficient (DX) is inversely proportional to the hydrodynamic radio
(rX). If we consider two species (A and B), one (A) with a 2-fold
hydrodynamic volume compared to the other (B), the ratio between the
radii is rA/rB = 21/3 = 1.26; therefore, DB/DA = rA/rB = 1.26.
(8) Jardin, F. H. In Encyclopedia of Inorganic Chemistry, 2nd ed.; King,
R. B., Ed.; John Wiley and Sons: New York, 2006; Vol. 7, pp 4693ꢀ4722.
(9) 2006 cmꢀ1, for the dppe complex RhCl(CO)(dppe). See: Lamb,
G. W.; Law, D.; Slawin, A. M. Z.; Clarke, M. L. Inorg. Chim. Acta 2009,
362, 4263.
(10) 2037 cmꢀ1, for RhCl(CO)(P2), P2 = bis(3-(diphenyl-
phosphinooxy)naphthalen-2-yl)methane. See: Punji, B.; Mague,
J. T.; Balakrishna, M. S. Dalton Trans. 2006, 1322.
(11) Serron, S.; Nolan, S. P.; Moloy, K. G. Organometallics 1996,
15, 4301–4306. Pribula, A. J.; Drago, R. S. J. Am. Chem. Soc. 1976, 98,
2784–2788.
(12) The entropy error was estimated from the linear least-squares
analysis without taking into account the error associated with each
equilibrium constant measurement.
(13) For instance, see [Rh(μ-Cl)(η3-C3H5)2]2 (CCDC refcode
ALYLRH01) and [Rh3(μ3-tz)(μ2-Cl)Cl(η3-C3H5)2(CO)4] (Htz =
thriazole) (CCDC refdcode COYYEU).
(14) The structure of the azasilatrane complex PtCl2(L), L =
EtOSi(Ph2PNCH2CH2)2(HNCH2CH2)N, is highly disordered (cf. ref
4); therefore its comparison with 8 would be meaningless and has been
omitted.
(15) Dierkes, P.; Van Leeuwen, P. W. N. M. J. Chem. Soc., Dalton
Trans. 1999, 1519–1529.
(16) The entropy error was estimated from the linear least-squares
analysis (cf. Supporting Information) without taking into account the
error associated with each rate constant measurement.
(17) The 31P NMR doublet at 90.7 ppm of 10 indicates a symmetric
bidentate coordination of the SiNP ligand to the metal center. Further-
more, the smaller 1JPRh of 10 when compared with 4 is in agreement with
the higher oxidation state of the metal center. The RhMe signals are
observed as a pseudoquartet (1H) and doublet of triplets (13C) in
agreement with the proposed cis arrangement of the CH3 ligand with
respect to both phosphorus atoms (cf. Experimental Section). The
coordination sphere of the metal is completed by a symmetrically
9967
dx.doi.org/10.1021/ic2004408 |Inorg. Chem. 2011, 50, 9958–9967