Chemistry of transition-metal complexes containing functionalized phosphines. Part 2. Cationic rhodium(I) complexes stabilized by hemilabile diallylphosphines

Article abstract of DOI:10.1016/j.ica.2014.02.011

Cationic rhodium(I) complexes of the type [Rh(COD){κ³(P,C, C)-RP(CH₂CHCH₂)₂}][BF₄] (R = ^tBu 3a, Ph 3b, ⁱPr₂N 3c) have been synthesized starting from the chloro-bridged dimer [Rh₂(μ-Cl) ₂(COD)₂] and fully characterized by ³¹P{ ¹H}, ¹H and ¹³C{¹H} NMR. X-ray diffraction analysis of 3a and 3c showed a square planar geometry around of the rhodium atom, with allyl-phosphine ligands in a chelating fashion due to coordination of only one of the allylic double bonds while the other allyl moiety is located away from the metal center. Crystals of 3c showed the presence of the two enantiomers in the asymmetric unit. In solution, a dynamic equilibrium of exchange between the two allylic double bonds was detected by low temperature NMR analysis. Hemilabile properties of diallylphosphine ligands have been demonstrated by ligand exchange reactions and the reversible displacement of coordinated allylic double bond by acetonitrile have been observed. DFT calculations have been performed to explain the order of reactivity on removing the acetonitrile ligand under high vacuum.

Full text of DOI:10.1016/j.ica.2014.02.011

Inorganica Chimica Acta 414 (2014) 250–256
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Chemistry of transition-metal complexes containing functionalized
phosphines. Part 2. Cationic rhodium(I) complexes stabilized by
hemilabile diallylphosphines q
a
a
Juan Manuel García a, David Santiago Coll b, Edgar Ocando-Mavárez a, , Julian Ascanio , Sarah Pekerar ,
⇑
Reinaldo Atencio c, Teresa Gonzalez c, Alexander Briceño c, Edward Avila c, Merlin Rosales d
a Laboratorio de Fisico-Química Orgánica, Centro de Química, Instituto Venezolano de Investigaciones Científicas (IVIC), Caracas, Venezuela
b Laboratorio de Fisico-Química Teórica de Materiales, Centro de Química, Instituto Venezolano de Investigaciones Científicas (IVIC), Caracas, Venezuela
c Laboratorio de Síntesis y Caracterización de Nuevos Materiales, Centro de Química, Instituto Venezolano de Investigaciones Científicas (IVIC), Caracas, Venezuela
d Laboratorio de Química Inorgánica, Departamento de Química, Facultad Experimental de Ciencias, La Universidad del Zulia (LUZ), Maracaibo, Venezuela
a r t i c l e i n f o
a b s t r a c t
3(P,C,C)-RP(CH2CH@CH2)2}][BF4] (R = tBu 3a, Ph 3b,
-Cl)2(COD)2] and fully
Article history:
Cationic rhodium(I) complexes of the type [Rh(COD){
j
iPr2N 3c) have been synthesized starting from the chloro-bridged dimer [Rh2(
l
Received 10 September 2013
Received in revised form 25 January 2014
Accepted 4 February 2014
characterized by 31P{1H}, 1H and 13C{1H} NMR. X-ray diffraction analysis of 3a and 3c showed a square
planar geometry around of the rhodium atom, with allyl-phosphine ligands in a chelating fashion due
to coordination of only one of the allylic double bonds while the other allyl moiety is located away from
the metal center. Crystals of 3c showed the presence of the two enantiomers in the asymmetric unit. In
solution, a dynamic equilibrium of exchange between the two allylic double bonds was detected by low
temperature NMR analysis. Hemilabile properties of diallylphosphine ligands have been demonstrated by
ligand exchange reactions and the reversible displacement of coordinated allylic double bond by aceto-
nitrile have been observed. DFT calculations have been performed to explain the order of reactivity on
removing the acetonitrile ligand under high vacuum.
Available online 18 February 2014
Keywords:
Hybrid ligands
Hemilabile properties
Diallylphosphines
Ligand exchange reaction
Ó 2014 Published by Elsevier B.V.
1. Introduction
Functionalized phosphines containing weakly coordinating car-
bon moieties and, in particular, the olefins are of special interest
The use of hybrid ligands has attracted the attention of
organometallic chemists in recent years [1]. This growing interest
is essentially due to the fact that the different chemical functional-
ities can exert a certain influence on the metal center to which they
are coordinated, which induces novel chemical and physical
properties to the resulting metal complexes [1a–c]. Hybrid ligands
often contain functionalities that exhibit hemilabile properties.
Among these ligands, functionalized phosphines have received
considerable attention due to their ability to provide vacant coor-
dination sites on the metal center and to stabilize reactive interme-
diates; these features have been shown to enhance both, activity
and selectivity, in catalytic reactions [2]. Recently complexes
containing hemilabile phosphine ligands showing good activities
in different catalytic reaction have been reported [3].
[1b,4], due to the well-known lability of the metal-olefin bond,
which enables the easy generation of a free coordination site at
the metal center.
Nevertheless, the use of allylphosphine ligands as potential
hemilabile ligands or to stabilize coordinatively unsaturated sys-
tems has been scarcely reported In previous work we showed that
diallylphosphines were able to stabilize unsaturated Ir(III) deriva-
tives by interaction of the double bonds with the metal center
and the lability of these interactions were demonstrated by reac-
tion with CO [4d].
Allylphosphines have also been used in catalytic processes [5],
as in the case of oligomerization and hydrosilylation of styrene
an 4-methylstyrene catalyzed by the cationic nickel complex
[Ni-(
of the cationic ruthenium complex [Ru-(
Ph2PCH2CH@CH2}{
3(P,C,C)Ph2PCH2CH@CH2}][PF6] showed that
g j
3-CH2-C(R1)CH2)( 1-R22PCH2CH@CH2)][BAr4] [5]. The study
g
5-C5Me5){ 1(P)
j
j
q
Part 1 corresponding to reference [3d].
one of the allyl groups is coordinated to the metal center and that
the coordination can be dissociated reversibly by reaction with
CH3CN and irreversibly when reacted with CO [3c]. In the same
⇑
Corresponding author. Address: Laboratorio de Fisico-Química Orgánica, Centro
de Química, Instituto Venezolano de Investigaciones Científicas (IVIC), Caracas
1020-A, Venezuela. Tel.: +58 212 5041641; fax: +58 212 5041350.
E-mail address: eocando@ivic.gob.ve (E. Ocando-Mavárez).
http://dx.doi.org/10.1016/j.ica.2014.02.011
0020-1693/Ó 2014 Published by Elsevier B.V.

J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
251
way, it has been reported the study of the indenyl ruthenium
derivative
CDCl3, 25 °C): d 66.15 (d, JPRh = 158.3 Hz). 1H NMR (300.13 MHz,
3
CDCl3, 25 °C): d 1.31 (d, JHH = 6.9 Hz, 12H, CH3NiPr), 1.95 (d,
[Ru-(g
5-C9H7){
j
3(P,C,C)-Ph2PCH2CH@CH2}(PPh3)][PF6] [4f].
2JPH = 8.9 Hz, 4H, PCH2), 2.29 (m, 4H, CH2COD), 2.49 (m, 2H, CH2COD),
2.86 (m, 2H, CH2COD), 3.63 (brs, 2H, CH@CHCOD), 4.04 (m, 2H,
CHNiPr), 5.11 (brs, 2H, @CH2), 5.14 (brs, 2H, @CH2), 5.33 (brs, 2H,
CH@CHCOD), 6.08 (m, 2H, CH@).13C{1H} NMR (75.47 MHz, CDCl3,
25 °C): d 28.42 (s, 4C, CH3NiPr), 32.52 (s, 2C, CH2COD), 32.81 (s, 2C,
In the present work we report on the synthesis, characterization
and dynamic behavior of cationic rhodium complexes of the type
[Rh(COD){j
3(P,C,C)-RP(CH2CH@CH2)2}][BF4] (R = tBu 3a, Ph 3b,
iPr2N 3c). The hemilabile properties of diallylphosphine ligands
are showed by the reversible displacement of the coordinated allylic
double bond by acetonitrile addition, which can be removed under
vacuum. DFT calculations have been performed in order to better
explain the experimental observations.
2
CH2COD), 33.27 (d, JCP = 2.4 Hz, 2C, PCH2), 51.90 (d, JCP = 7.4 Hz,
2C, CHNiPr), 67.32 (d, JCRh = 14.2 Hz, 2C, CH@CHCOD), 103.46 (dd,
3
2JCP = 6.8 Hz, JCRh = 12.9 Hz, 2C, CH@CHCOD), 118.24 (d, JCP = 10.8 -
Hz, 2C, @CH2), 131.84 (s, 2C, CH@). Anal. Calc. for C20H36ClNPRh:
C, 52.25; H, 7.83; N, 3.04. Found: C, 51.93; H, 8.03; N, 3.07%.
2. Experimental
2.3. Synthesis of [Rh(COD){
3a, Ph 3b, Pr2N 3c)
j
3(P,C,C)-RP(CH2CH@CH2)2}][BF4] (R = tBu
i
2.1. General information
A solution of the complexes2a–c (0.22 mmol) in THF (20 mL) was
treated with AgBF4 (42.4 mg, 0.22 mmol), and the mixture was stir-
red for 1 h at ꢀ80 °C. The solution was evaporated to dryness. The
resulting solid was extracted with dichloromethane (2 ꢁ 20 mL)
and vacuum-dried to give the complex as orange solid which was
washed with diethyl ether (20 mL) and vacuum-dried. R = tBu 3a:
(88 mg, 86%); 31P{1H} NMR (202.41 MHz, CD2Cl2, 25 °C): d ꢀ 68.22
(d, JPRh = 110.1 Hz). 1H NMR (500.03 MHz, CD2Cl2, 25 °C): d 1.21 (d,
All manipulations were performed under an inert atmosphere
of argon by using standard Schlenk techniques. Dry, oxygen-free
solvents were employed. The elemental analysis was performed
(C, H, N) on
a model EA1108 Fisons elemental analyzer.
1H, 13C{1H}, 31P{1H} NMR spectra were recorded on Bruker, Avance
500 or Avance 300 spectrometers. 1H, 13C{1H} NMR chemical shifts
are reported in ppm relative to Me4Si as external standard. 31P{1H}
NMR chemical shifts are expressed in ppm relative to 85% H3PO4.
Diallylphosphines were prepared by published methods [6,7]. Di-
2JPH = 16.1 Hz, 9H, CH3 Bu), 2.27–2.59 (m, 8H, CH2COD), 2.96–3.07
t
(m, 4H, PCH2), 4.84 (dd, 3JHH = 9.5 Hz, 4JPH = 3.5 Hz, 2H, @CH2), 5.06
mer complex [Rh2(l-Cl)2(COD)2] was purchased from Strem
Chemicals. Other chemicals were purchased from Sigma–Aldrich
Co.
3
4
(dd, JHH
= 16.5 Hz, JPH = 2.9 Hz, 2H, @CH2), 5.25 (brs, 2H,
CH@CHCOD), 5.79–5.89 (m, 2H, CH@). 13C{1H} NMR (125.74 MHz,
CD2Cl2, 25 °C): d 25.99 (d, JCP = 20.5 Hz, 2C, PCH2), 26.91 (d,
2JCP = 3.8 Hz, 3C, CH3 Bu), 29.23 (s, 2C, CH2COD), 31.69 (d, JCP = 17.0 Hz,
t
2.2. Synthesis of complexes
PCtBu), 32.03 (s, CH2COD), 32.05 (s, CH2COD), 90.99 (d, JCRh = 10.3 Hz,
2
2C, CH@CHCOD), 105.86 (dd, JCP = 6.0 Hz, JCRh = 8.5 Hz, 2C,
CH@CHCOD), 107.23 (brs, @CH2), 107.94 (d, JCP = 7.6 Hz, @CH2),
113.07 (d, JCP = 7.4 Hz, 2C, CH@). Anal. Calc. for C18H31BF4PRh: C,
3
2.2.1. Synthesis of [RhCl(COD){
Ph 2b, Pr2N 2c)
j
1(P)-RP(CH2CH@CH2)2}] (R = tBu 2a,
i
2
To a solution of [Rh2(
l
-Cl)2(COD)2] (100 mg, 0.203 mmol) in
46.18; H, 6.62. Found: C, 45.20; H, 6.56%. R = Ph Mixture 3b and
3b-THF: The more important signals in 31P{1H} NMR (202.41 MHz,
CDCl3, 25 °C): d 2.9 (d, JPRh = 143.6 Hz for 3b) and 15.4 (d,
JPRh = 159.9 Hz for 3b-THF) and 13C{1H} NMR (125.74 MHz, CDCl3,
25 °C): d = 25.6 (CH2THFuncoordinated), 26.5 (CH2THFcoordinated), 67.9
(OACH2THFuncoordinated), 70.6 (OACH2THFcoordinated), 107.0 (d,
3JCP = 7.6 Hz, @CH2), 105.6 (d, 2JCP = 2.4 Hz, CH@), 120.7
CH2Cl2 (20 mL) was added dropwise a solution of the correspond-
ing diallylphosphine 1a–c (0.426 mmol) in 10 mL of CH2Cl2. The
solution was stirred for 1 h at ꢀ80 °C. The solution was evaporated
to dryness to give a yellow solid, which was washed with hexane
(5 mL) and vacuum-dried. R = tBu 2a: (172 mg, 97%); 31P{1H}
NMR (121.49 MHz, CDCl3, 25 °C): d 24.71 (d, JPRh = 146.8 Hz). 1H
2
2
NMR (300.13 MHz, CDCl3, 25 °C): d 1.25 (d, JPH = 13.4 Hz, 9H,
(brs, @CH2uncoordinated), 129.4 (d, JCP = 10.1 Hz, CH@). R = iPr2N 3c:
CH3 Bu), 1.88 (brs, 4H, CH2COD), 2.15–2.36 (m, 6H, CH2COD, PCH2),
(90 mg, 80%). 31P{1H} NMR (202.41 MHz, CD2Cl2, 25 °C): d – 23.79
(d, JPRh = 104.4 Hz). 1H NMR (500.03 MHz, CD2Cl2, 25 °C): d 1.17 (d,
t
2.64–2.73 (m, 2H, PCH2), 3.66 (brs, 2H, CH@CHCOD), 5.03–5.10
(m, 4H, @CH2), 5.22 (brs, 2H, CH@CHCOD), 5.87–6.02 (m, 2H,
3JHH = 6.6 Hz, 12H, CH3N Pr), 2.24–2.44 (m, 8H, CH2COD), 2.93 (t, 4H,
i
CH@). 13C{1H} NMR (75.47 MHz, CDCl3, 25 °C):
d
25.42 (d,
JPH = 7.1 Hz, PCH2), 3.34 (d, JHH = 5.9 Hz, 2H,CHN Pr), 4.53 (dd,
3
i
JCP = 18.2 Hz, 2C, PCH2), 28.33 (s, 2C, CH2COD), 29.27 (d, 2JCP = 3.9 Hz,
3JHH = 16.3 Hz, JHH = 9.0 Hz, 4H, @CH2), 5.00 (brs, 1H, CH@COD),
3
3C, CH3 Bu), 33.24 (s, 2C, CH2COD), 34.02 (d, JCP = 17.9 Hz, PCtBu),
5.28 (brs, 1H, CH@COD), 5.87 (brs, 2H, CH@). 13C{1H} (125.74 MHz,
CD2Cl2, 25 °C): d 24.93 (s, CH3NiPr), 29.05 (s, 2C, CH2COD), 31.97 (s,
CH2COD), 32.00 (s, CH2COD), 34.51 (d, JCP = 27.5 Hz, PCH2), 47.30 (s,
CHNiPr), 89.15 (d, JCRh = 7.6 Hz, CH@CHCOD), 100.18 (brs, @CH2),
107.56 (d, 2JCP = 7.6 Hz, CH@), 107.88 (dd, 2JCP = 5.0 Hz, JCRh = 9.9 Hz,
2C, CH@CHCOD). Anal. Calc. for C20H36NBF4PRh: C, 46.99; H, 7.04; N,
2.74. Found: C, 44.91; H, 7.06; N, 2.86%. 31P{1H} NMR
(202.41 MHz, CD2Cl2, ꢀ90 °C): d – 22.79 (d, JPRh = 100.7 Hz). 1H
NMR (500.03 MHz, CD2Cl2, ꢀ90 °C): d 0.99 (brs, 9H, CH3NiPr), 1.41
(brs, 3H, CH3NiPr), 2.07–2.63 (m, 10H, PCH2uncoordinated, CH2COD),
t
2
68.00 (d, JCRh = 13.8 Hz, 2C, CH@CHCOD), 103.36 (dd, JCP = 7.2 Hz,
3
JCRh = 11.9 Hz, 2C, CH@CHCOD), 118.41 (d, JCP = 9.6 Hz, 2C, @CH2),
2
132.16 (d, JCP = 4.7 Hz, 2C, CH@). Anal. Calc. for C18H31ClPRh: C,
51.88; H, 7.44. Found: C, 51.72; H, 7.17%. R = Ph 2b: (182 mg,
98%); 31P{1H} NMR (121.49 MHz, CDCl3, 25 °C): d = 14.15
(d, JPRh = 148.4 Hz). 1H NMR (300.13 MHz, CDCl3, 25 °C): d = 1.85–
1.99 (m, 4H, CH2COD), 2.29 (brs, 4H, PCH2), 2.78–3.01 (m, 4H,
CH2COD), 3.23 (brs, 2H, CH@CHCOD), 5.09–5.16 (m, 4H, @CH2),
5.36 (brs, 2H, CH@CHCOD), 5.79–5.95 (m, 2H, CH@), 7.33–7.35 (m,
3H, CHAr), 7.50–7.57 (m, 2H, CHAr). 13C{1H} NMR (75.47 MHz,
CDCl3, 25 °C): d = 28.46 (s, 2C, CH2COD), 29.98 (d, JCP = 23.7 Hz, 2C,
PCH2), 32.89 (s, 2C, CH2COD), 69.42 (d, JCRh = 13.8 Hz, 2C,
CH@CHCOD), 104.54 (dd, 2JCP = 6.8 Hz, JCRh = 11.9 Hz, 2C,
3
3.20 (brs, 4H, PCH2coordinated, CHNiPr), 3.67 (d, JHH = 13.9 Hz, 1H,
@CH2coordinated), 3.77 (brs, 1H, @CH2coordinated), 4.48 (brs, 1H,
CH@COD), 4.69 (brs, 1H, CH@COD), 5.20 (brs, 2H, @CH2uncoordinated),
5.52 (brs, 1H, CH@COD), 5.64 (brs, 2H, CH@COD, CH@uncoordinated),
5.98 (brs, 1H, CH@Coordinated). 13C{1H} NMR (125.74 MHz, CD2Cl2,
ꢀ90 °C): d 20.80 (brs, CH3NiPr), 21.49 (brs, CH3NiPr), 25.84 (brs,
CH3NiPr), 27.11 (s, CH2COD), 28.73 (s, CH2COD), 30.05 (s, CH2COD),
31.07 (brs, CH3NiPr), 31.97 (s, CH2COD), 32.35 (d, JCP = 36.3 Hz,
PCH2uncoordinated), 34.63 (s, CH2COD), 35.50 (d, JCP = 20.3 Hz,
3
CH@CHCOD), 118.78 (d, JCP = 10.3 Hz, 2C, @CH2), 128.17 (d,
3JCP = 8.8 Hz, 2C, CHAr), 129.70 (s, CHAr), 130.71 (brs, Cipso), 131.09
2
2
(d, JCP = 4.7 Hz, 2C, CH@), 131.36 (d, JCP= 8.7 Hz, 2C, CHAr). Anal.
Calc. for C20H27ClPRh: C, 54.99; H, 6.18. Found: C, 52.76; H,
6.17%. R = iPr2N 2c: (192 mg, 98%); 31P{1H} NMR (121.49 MHz,

252
J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
PCH2coordinated), 41.74 (brs, CHNiPr), 50.23 (brs, CHNiPr), 76.99
(s, @CH2coordinated), 79.59 (s, CH@ COD), 83.18 (s, CH@Coordinated),
93.75 (s, CH@ COD), 103.56 (s, CH@ COD), 111.10 (s, CH@ COD),
120.66 (s, @CH2uncoordinated), 127.73 (s, CH@uncoordinated).
56:44, respectively. In structure 3c, the systematic absences were
compatible with space groups Pca21 (29) and Pbcm (57), attempts
to solve and refinement this structure for centrosymmetrical group
were unsuccessfully, which was discarded. The final refinement
was carried-out using space group Pca21, which two independent
enantiomers metal moieties were found in contrast to that ob-
served in 3a. Such complexes are slightly different, in this structure
one of the non-coordinated allyl moiety also was found disordered
over to orientations with complementary occupancies of 59:41,
respectively. No large correlations between pseudo symmetry-re-
lated atoms were observed for this space group. All the refinement
calculations were made using SHELXTL-NT [9].
2.3.1. Reaction of [Rh(COD){
j
3(P,C,C)-RP(CH2CH@CH2)2}] [BF4]
(R = tBu 3a, Ph 3b, Pr2N 3c) with acetonitrile
To a solution of [Rh(COD){
3(P,C,C)-RP(CH2CH@CH2)2}][BF4]
(R = tBu 3a, Ph 3b, Pr2N 3c, 0.117 mmol) in CD2Cl2 (50
L) was
added a solution of CH3CN (12.5 L, 0.234 mmol). The solution
i
j
i
l
l
was stirred for 15 min at RT, and then was evaporated to dryness
to give a pasty solid. R = tBu 4a: RMN 31P{1H} (121.49 MHz, CD2Cl2,
25 °C): d 27.26 (d, JPRh = 142.4 Hz). RMN 1H (300.13MHz, CD2Cl2,
25 °C): d 1.35 (d, 2JPH = 14.7 Hz, 9H, CH3 Bu), 2.17 (brs, 4H, CH2COD),
t
2.5. Theoretical details
2.23 (s, 5H, CH3CN), 2.43 (brs, 4H, CH2COD), 2.53 (m, 4H, PCH2), 4.27
(brs, 2H, CH@CHCOD), 5.28–5.32 (m, 6H, CH@CHCOD, @CH2), 5.95
(m, 2H, CH@). RMN 13C{1H} (75.47 MHz, CDCl3, 25 °C): d 3.42
3
All structures were optimized using DMOL [10–12]. This DFT
based program allows us to determine the relative stability of all
studied species based on their electronic structure. The calcula-
tions were performed using Kohn–Sham Hamiltonian with the Per-
dew–Wang 1991 gradient correction [13] and the double-zeta plus
(DNP) numerical basic set [10–12]. The utilization of the numerical
basics sets combined with DFT allows the program to obtain a high
accuracy by keeping a relative low computational cost, compared
with ab-initio methods. DMOL3 calculates variational self-consistent
solutions to the density functional theory (DFT) equations. The
solutions to these equations provide the molecular electron densi-
ties, which can be integrated among the atomic volume (defined by
the interatomic surfaces and/or the Van der Walls envelope) in or-
der to obtain the Bader charge on each atom of the system [14,15].
2
(s, CH3CN), 25.38 (d, JCP = 19.6 Hz, 2C, PCH2), 28.17 (d, JCP = 2.9 Hz,
t
3C, CH3 Bu), 28.64 (s, 2C, CH2COD), 32.37 (s, 2C, CH2COD), 33.92
(d, JCP = 19.01 Hz, PCtBu), 80.64 (d, JCRh = 10.3 Hz, 2C, CH@CHCOD),
2
104.19 (t, JCP = 8.0 Hz, 2C, CH@CHCOD), 116.93 (brs, 2C, @CH2),
125.73 (brs, 2C, CH@), 127.97 (brs, NC). R = Ph 3b: RMN 31P{1H}
(121.49 MHz, CD2Cl2, 25 °C): d 13.25 (d, JPRh = 144.9 Hz). RMN 1H
(300.13 MHz, CDCl3, 25 °C): d 2.05 (m, 4H, CH2COD), 2.27 (s, 8H,
CH3CN, PCH2), 2.78 (m, 4H, CH2COD), 3.76 (m, 2H, CHNiPr), 3.84
(brs, 2H, CH@CHCOD), 5.19–5.24 (m, 6H, @CH2, CH@CHCOD), 5.72
(m, 2H, CH@), 7.43–7.47 (m, 3H, CHAr), 7.51–7.56 (m, 2H, CHAr).
RMN 13C{1H} (75.47 MHz, CDCl3, 25 °C): d 3.09 (s, CH3CN), 27.81,
28.07 (2xs, 2C, PCH2), 28.64 (s, CH2COD), 32.34 (s, CH2COD), 78.22
(brs, 2C, CH@CHCOD), 104.77 (brs, 2C, CH@CHCOD), 120.60
3
TS were determined with the same code. DMOL uses the nudged
3
elastic band (NEB) [16,17] method for minimum energy path cal-
culations. The NEB method introduces a fictitious spring force that
connects neighbouring points on the path to ensure continuity of
the path and projection of the force so that the system converges
to the minimum energy path (MEP). The NEB method has been
used widely in solid-state physics and has recently been applied
to molecules as well. The advantage of the NEB algorithm is that
it provides a fast qualitative examination of the MEP [10].
(d, JCP = 9.97 Hz, 2C, @CH2), 126.78 (brs, NC), 128.82 (brs, CHAr),
128.91 (s, CHAr), 129.03 (s, CHAr), 129.33 (brs, Cipso), 130.87 (brs,
2
2C, CH@), 131.54 (d, JCP = 9.6 Hz, 2C, CHAr). R = iPr2N 3c: RMN
31P{1H} (121.49 MHz, CD2Cl2, 25 °C): d 56.40 (d, JPRh = 150.4 Hz).
3
RMN 1H (300.13 MHz, CD2Cl2, 25 °C): d 1.33 (d, JHH = 6.8 Hz,
12H, CH3NiPr), 2.08 (m, 4H, CH2COD), 2.20 (s, 5H, CH3CN), 2.38–2.75
(m, 8H, CH2COD, PCH2), 3.76 (m, 2H, CHNiPr), 4.10 (brs, 2H,
CH@CHCOD), 5.20–5.32 (m, 6H, @CH2, CH@CHCOD), 6.05 (m, 2H,
CH@). RMN 13C{1H} (75.47 MHz, CDCl3, 25 °C): d 3.26 (s, CH3CN),
25.37 (s, 2C, CH3NiPr), 25.40 (s, 2C, CH3NiPr), 28.92 (brs, 2C, PCH2),
32.86 (s, CH2COD), 32.89 (s, CH2COD), 33.22 (s, CH2COD), 33.54 (s,
3. Result and discussions
2
CH2COD), 50.61 (d, JCP = 4.4 Hz, 2C, CHNiPr), 78.55 (d, JCRh = 13.6 Hz,
Two equivalents of diallylphosphines 1a–c reacted in dichloro-
2
2C, CH@CHCOD)d, 104.53 (dd, JCP = 6.4 Hz, JCRh = 10.6 Hz, 2C,
methane with one equivalent of the dimer complex [Rh2(
(COD)2] leading to the quantitative formation of the neutral
complexes [RhCl(COD){
1(P)-RP(CH2CH@CH2)2}] 2a–c (Scheme 1),
l-Cl)2
CH@CHCOD), 118.20 (brs, 2C, @CH2), 123.26 (brs, NC), 127.56 (s,
2C, CH@).
j
isolated as air sensitive yellow solids in 98% yield, fully character-
ized by 1H, 13C{1H}, and 31P{1H} NMR methods and elemental anal-
ysis. 31P{1H} NMR spectra at RT of complexes 2a–c (given in Table 1)
showed doublet signals shifted to downfield with respect to those of
the corresponding free ligands [6,7,4d], and coupling constants in
the range of those observed in analogous rhodium(I) complexes
[18,19]. Meanwhile, 13C{1H} NMR spectra of complexes (2a–c) are
2.4. X-ray structures determinations
Intensity data were recorded at room temperature on a Rigaku
AFC-7S diffractometer using monochromated Mo K
a radiation
(k = 0.71073 Å). Experimental details on unit cell and intensity
measurements can be found in the CIF files deposited with the
Cambridge Crystallographic Data center [CCDC-843258 for 3a
and CCDC-843259 for 3c]. Crystal data, intensity data collection
parameters and final refinement results are summarized in Table 5.
An empirical absorption correction (multi-scan) was applied to all
the data using the CrystalClear crystallographic software package
[8]. The structures were solved by Direct Methods and refined by
full-matrix least-squares on F2. The H-atoms on C were placed in
calculated positions using a riding atom model with fixed C–H dis-
tances [0.93 Å for C(sp2), 0.96 Å for C(sp3, CH3) and 0.97 Å for C(sp3,
CH2)]. All the H atoms were refined with isotropic displacement
parameters set to 1.2 ꢁ Ueq for C(sp2) and 1.5 for C(sp3) of the at-
in agreement with the j
1(P) coordination mode of the diallylphos-
phine ligands showing a very poor or no interaction of the double
bond with the metallic center, as indicated by the presence of
resonances of the carbon-carbon double bonds of allyl groups at
values similar to those observed for free ligands, which means
around 118 ppm for CH2 and 128–130 for CH carbon atoms (Table 2).
Treatment of a solution of 2a–c complexes in THF with one equiv of
AgBF4 at ꢀ80 °C led to the formation of cationic complexes
[Rh(COD){j
3(P,C,C)-RP(CH2CH@CH2)2}][BF4] 3a–c (Scheme 1),
isolated as air-sensitive orange solids in 80–85% yield and fully
characterized by spectroscopic methods.
ꢀ
tached atom. In structure 3a, BF4 anion was found disordered,
which was modeled over two orientations with restraints in the
B–F and Fꢂ ꢂ ꢂF distances and with complementary occupancies of
NMR analysis of complexes 3a and 3c showed in the 31P{1H}
spectra doublet resonances shifted toward higher fields with
respect to observed for respective complexes 2a–c (Table 2),

J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
253
R
P
In contrast, reaction of 2b complex in THF with one equiv of
AgBF4 leads to reversible coordination of THF solvent and allylic
double bond, as indicated the presence of two doublet signal
[d 2.9 (JPRh = 143.6 Hz) for 3b and 15.4 (JPRh = 159.9 Hz) for
3b-THF] in NMR 31P{1H} at 25 °C. While, NMR 13C{1H} at 25 °C
showed the resonances corresponding for both the coordinated [d
107.0 (@CH2), 105.6 (CH@)] and uncoordinated [d 120.7 (@CH2),
129.4 (CH@)] alkene carbons, as well as, the resonances corre-
sponding for both the coordinated [d 70.6 (OACH2), 26.5 (CH2)]
and uncoordinated [d 67.9 (OACH2), 25.6 (CH2)] tetrahydrofuran
carbons. It is interesting to note that the 31P{1H} NMR spectra in
CH2Cl2 remain unchanged within a wide range of temperature
(ꢀ90 to +50 °C). Additionally, all attempts to isolate 3b were
unsuccessful, even when trying to place under high vacuum to
eliminate of THF solvent. The results indicate that in 3b the
rhodium center is better stabilized by THF solvent than by the allyl
double bonds, which might be due to poorer donor potential of the
P-Ph moiety (compared to P-Bu and P-NiPr2), as well as, by lower
coordinating power of olefins in comparison to THF ligand.
In addition, the existence of a dynamic behavior of ‘‘coordina-
tion-decoordination’’ of allyl fragments was observed by NMR
measurements at low temperature. A rapid exchange between
coordinated allyl moieties, which was found to be fast on the
NMR time scale in CD2Cl2 at room temperature but enough slow
at ꢀ90 °C, was observed for 3c (Scheme 2). 1H NMR spectrum in
CD2Cl2 at ꢀ90 °C of 3c clearly showed two sets of resonances
[d 3.67 and 3.77 (@CH2), 5.98 (CH@)] and [d 5.20 (@CH2), 5.64
(CH@)] corresponding to protons of coordinated and uncoordi-
nated alkenes moieties (see Section 2). Moreover, 13C{1H} NMR
spectrum in CD2Cl2 at ꢀ90 °C of 3c unambiguously showed the res-
onances corresponding for both the coordinated [d 77.02 (@CH2),
83.19 (CH@)] and uncoordinated [d 120.73 (@CH2), 127.77
(CH@)] alkene carbons (see Section 2). While for 3a, the exchange
rate between both allyl moieties continues to be too fast at ꢀ90 °C
and could not be observed.
Orange crystals of 3a and 3c suitable for X-ray analysis were ob-
tained by slow diffusion of diethyl ether into a concentrated solu-
tion of the respective complexes in CH2Cl2. A selected set of
relevant bond lengths and bond angles is presented in Table 3,
molecular structures of 3a and 3c are shown in Figs. 1 and 2
respectively and crystallographic data presented in Table 1.
It can be observed that, in the solid state, diallylphosphines
showed a chelate behavior by coordination of only one of the
allylic double bonds while the other allyl moiety is located away
from the metal center. Both crystal structures contain an enantio-
meric mixture. Since 3a showed only one of the isomers in the unit
cell, in contrast, the structure of 3c displays two independent crys-
tallographic enantiomers into the asymmetric unit. The presence of
enantiomeric mixture in the asymmetric cell is reasonable since
the phosphorus atom must be considered chiral when an allyl moi-
ety is coordinated while the other remains free, furthermore, this
Cl
Cl
CH2Cl2
-80°C
Rh
+ 2
R P
Rh
Rh
Cl
1a R= tBu
1b R= Ph
1c R= iPr2N
2a-c
AgBF4, THF
-80°C
-AgCl
R
P
BF4
Rh
3a-c
Scheme 1.
indicating a chelating coordination mode of the diallylphosphine
1a and 1c [20]. Similar chemical shift toward high field have been
observed for ruthenium cationic complexes bearing allyldiphenyl-
phosphine as ligand [4c,f]. 13C{1H} NMR spectra also showed the
olefin coordination, displaying the expected resonances for
CH@ and CH2@ fragments which are shifted to higher field than
those observed for complexes 2a–c (Table 2).
Table 1
Crystallographic data for 3a and 3c.
Complexes
3a
3c
Formula
C
18H31BF4PRh
C20H36BF4NPRh
511.19
FW (gmolꢀ1
)
468.12
Crystal size (mm)
Crystal system
Space group
a (Å)
0.46 ꢁ 0.23 ꢁ 0.22 0.49 ꢁ 0.32 ꢁ 0.25
triclinic
orthorhombic
Pca21
ꢀ
P1
7.823(3)
9.890(5)
14.034(6)
105.467(9)
91.059(13)
94.402(15)
298
0.71070
1042.5(8)
2
16.644(6)
11.351(5)
23.436(7)
b (Å)
c (Å)
a
(°)
b (°)
(°)
Temperature (K)
c
150
Mo K
a
(Å)
071073
4428(3)
8
1.534
2112.0
0.883
27780
7962
Volume (Å3)
Z
qcalc (g cmꢀ3
F(000)
)
1.491
480.0
0.928
11518
3899
0.033
1.095
l
(mmꢀ1
)
Reflections collected
Unique reflections
Rint
0.045
1.06
Goodness of fit (GOF) on F2
R1 [I > 2
wR2
r
(I)]
0.0703
0.2002
0.66/ꢀ0.79
0.045
0.121
1.94/ꢀ0.59
Residual electron density peaks
(e Åꢀ3
)
Table 2
31P{1H}, 1H and 13C{1H} NMR data neutral, 2a–c, and cationic, 3a–c, rhodium derivatives.
Complexes
d
31P{1H} NMR
JPRh (Hz)
d
13C{1H} NMR (@CH)
d
13C{1H} NMR (@CH2)
2
3
2aa
3ab
2ba
3ba,c
2ca
24.71
ꢀ68.22
14.15
2.9
66.15
ꢀ23.79
146.8
110.1
148.4
143.6
158.3
104.4
132.16 (d, JCP = 4.7Hz)
118.41 (d, JCP = 9.6 Hz)
2
3
113.07 (d, JCP = 7.4 Hz)
107.23 (brs), 107.94 (d, JCP = 7.6 Hz)
2
3
131.09 (d, JCP = 4.7 Hz)
118.78 (d, JCP = 10.3 Hz)
2
3
105.60 (d, JCP = 2.4 Hz)
107.49 (d, JCP = 7.6 Hz)
3
131.84 (s)
118.24 (d, JCP = 10.8 Hz)
2
3cb
107.56 (d, JCP = 7.6 Hz)
100.18 brs
a
b
c
Spectra recorded in CDCl3 at 25 °C.
Spectra recorded in CD2Cl2 at 25 °C.
Spectra data corresponding to rhodium(I) complex with THF uncoordinated.

254
J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
R
P
R
P
BF4
BF4
Rh
Rh
3
3'
Scheme 2.
Table 3
Selected bond distances (Å) and angles (°) 3a and 3c.
3a
3c
3c0
Rh1ꢀP1
2.287(2)
2.339(8)
2.287(8)
1.34(1)
1.31(1)
1.48(1)
1.49(1)
–
1.842(7)
1.825(7)
109.3(4)
107.4(3)
–
2.269(2)
2.258(8)
2.237(6)
1.38(1)
1.19(1)
1.52(1)
1.52(1)
1.656(5)
1.838(7)
1.847(7)
–
2.266(2)
2.264(8)
2.259(7)
1.39(1)
1.37(2)
1.51(1)
1.527(2)
1.656(5)
1.819(8)
1.833(9)
–
Rh1ꢀC10
Fig. 2. Molecular structures of 3c (left) and 3c0 (right). Thermal ellipsoids represent
30% probability. H atoms and counter-ions BF4 have been omitted for clarity.
Rh1ꢀC11
ꢀ
C11ꢀC10
C13ꢀC14
C9ꢀC10
C12ꢀC13
N1ꢀP1
P1ꢀC9
P1ꢀC12
C15ꢀP1ꢀC9
C15ꢀP1ꢀC12
N1ꢀP1ꢀC9
N1ꢀP1ꢀC12
C12ꢀP1ꢀC9
Rh1ꢀP1ꢀC9
P1ꢀC9ꢀC10
C11ꢀC10ꢀC9
C10ꢀRh1ꢀP1
C11ꢀRh1ꢀP1
–
–
108.7(3)
112.4(3)
107.9(4)
91.4(2)
97.8(5)
121.0(7)
68.2(2)
83.4(2)
107.9(3)
111.7(3)
107.8(4)
90.7(2)
99.5(5)
122.7(7)
68.4(2)
83.6(2)
–
105.5(4)
92.2(3)
99.3(5)
123.0(9)
66.7(2)
89.8(3)
observation agrees with equilibrium observed in solution for com-
plexes 3c.
Fig. 3. Molecular structure of 3a showing the ‘‘envelope-like’’ five membered
metallacycle Rh1P1C9C10C11.
In both cases, the five-membered ring formed by the coordina-
tion of the allyl moiety is presented as a ‘‘distorted envelope’’
(Fig. 3) where the two planes of the ‘‘bent square’’ (Rh1P1C10 and
Rh1C9C10) form an angle of 12.07°, 13,68° and 15,99° for 3a, 3c
and 3c0 respectively (3c and 3c0 corresponding to both the mole-
cules presents in the unit cell for 3c), while ‘‘the envelope folding
edge’’ (C11 atom) is located at a distance to the average plane of
the ‘‘square’’ of 0.94Å, 1.26Å and 1.28 Å for 3a, 3c and 3c0
respectively.
complexes (Table 3) [4k,21]. The Rh1ꢀC10 and Rh1ꢀC11 bond dis-
tances as well as the elongation of the C10ꢀC11 double bond, are
agree with the coordination of the alkene moiety, in the same
range than observed for coordinated allyl phosphines in cationic
ruthenium derivatives [4f].
Addition of a slight excess of acetonitrile to complexes 3a–c
produced the displacement of the coordinated allylic double bond
from the metal center, leading to the formation of the acetonitrile
substituted cationic rhodium complexes 4a–c, as pasty solids, fully
characterized by spectroscopic methods (see Section 2). Complexes
4a–c showed in the 31P{1H} spectra doublet resonances shifted to-
ward low fields with respect to observed for respective complexes
3a–c, d = 27.3 [JPRh = 142.4 Hz], 13.3 [JPRh = 144.9 Hz] and d = 54.4
[JPRh = 150.4 Hz] for 4a, 4b and 4c respectively, in accord with a
non metallacycle compound [19]. Moreover, in the 13C{1H} NMR
spectra the reappearance of the signals corresponding to the unco-
ordinated double bonds, can also be observed (see Section 2). An
attempt for isolating the complexes 4a–c, as solids, revealed the
existence of an equilibrium between the cationic rhodium species
3 and 4 (Scheme 3).
Moreover, the Rh1ꢀP1 bond distances are agree with the RhꢀP
bond distances observed in similar rhodium metallacycle
Interestingly, the 31P{1H} NMR spectra at ꢀ80 °C showed that
after 4 days under high vacuum (10ꢀ3 mmHg) acetonitrile ligand
R
P
R
P
CH3CN
BF4
BF4
Rh
Rh
- CH3CN
NCCH3
Vacuum (10-3 mmHg)
3a-c
4a-c
R= tBu (a), Ph (b), iPr2N (c)
Fig. 1. Molecular structure of 3a. Thermal ellipsoids represent 30% probability. H
ꢀ
atoms and counter-ions BF4 have been omitted for clarity.
Scheme 3.

J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
255
Table 4
Calculated energies (in kcal/mol) for complexes 3a–c, 4a–c and (TS) and selected distance bonds (Å).
3
4
TS
a
a
E
dRh–N
dRh–C
E
dRh–N
dRh–C
ETS
D
E1
DE2
a, tBu
b, Ph
c, Pr2N
0.0
0.0
0.0
4.807
5.790
5.597
2.374
2.306
2.340
ꢀ3.22
ꢀ9.62
ꢀ5.30
2.126
2.127
2.129
4.927
4.913
4.797
15.20
13.05
9.45
15.20
13.05
9.45
18.42
22.67
14.75
i
a
Rh–CH2allyl distance bond.
coordinated to the rhodium atom. In this case, for tBu and Ph sub-
stituents, 3a and 3b respectively, the pentacoordinated complexes
are relatively stable but not for the i-Pr2N 3c, then it seems not to
be the pathway of the reaction. A second explored pathway reac-
tion was one which involves the decoordination of the allyl moiety
before the coordination of the acetonitrile, leading to the formation
of a tricoordinated rhodium intermediate, which showed to be less
stable than the corresponding tetra or pentacoordinated ones. The
last possible pathway reaction evaluated involves the concerted
rapprochement of the acetonitrile and the decoordination of the
double bond from the metal center. This single step mechanism
for the ligand exchange reaction showed to be energetically fa-
voured for the three phosphines and reproduces the experimental
observations and was then considered as the more probable reac-
tion pathway for this reaction.
Table 5
Atomic charge (q) and electron density (
q
) at the BCP (in a.u.).
a
b
c
d
d
e
e
Complexes
qRh
qN
D
qRh–N
qTP
qTCN
qRh–N
qRh–P
4a
4b
4c
0.9
ꢀ1.20
ꢀ1.23
ꢀ1.35
ꢀ0.27
+ 0.23
ꢀ0.84
0.88
0.89
0.89
0.17
0.19
0.17
0.074
0.075
0.075
0.081
0.086
0.087
1.43
0.51
a
b
c
Rh atomic charge.
N atomic charge of nitrile.
Difference between qRh and qN.
Overall charge for the phosphine and acetonitrile ligands (qTP) and (qTCN).
Electron density at the Rh–N and Rh–P bond critical points.
d
e
was almost fully displaced in the case of 4c, whereas for 4a and 4b
was observed a mixture corresponding to allyl coordinated 3 and
acetonitrile coordinated 4 in a ratio of 7.8:2.2 and 6:4 respectively.
These results indicate that allylic double bond can be reversibly
displaced from rhodium center in presence of acetonitrile, which
demonstrates the hemilabile properties of the diallylphosphine li-
gands. Moreover, the results indicate that the removing of acetoni-
trile ligand under vacuum follow the order 4c > 4a > 4b, as
expected.
In order to gain more understanding on the observed results, a
theoretical study of the reaction of complexes 3a–c with acetoni-
trile at DFT level of theory has been performed.
Three different pathways have been explored for the reaction of
acetonitrile with cationic complexes 3a–c. First, we have evaluated
the possible formation of a pentacoordinated intermediate in
which the double bond and the acetonitrile are simultaneously
In the Table 4 are shown the relative calculated energy values
for 3a–c, 4a–c and for the obtained transition state geometries,
as well as, selected bond distances corresponding to the structures
3a–c, 4a–c. Meanwhile, in Table 5 are shown the calculated Bader
charges (q) of some atoms and the value of the electronic density at
the bond critical point (BCP) for some interactions (q).
The theoretical calculations predict a single-step mechanism for
the ligand exchange reaction, which proceeds exothermically
through the formation of a five-coordinate cationic rhodium specie
in the TS, in which the allylic double bond can be considered as no
longer coordinated and have been displaced by the acetonitrile
(Fig. 4).
As a first observation, and as expected, the rhodium atom is
electron deficient in both 3a–c and 4a–c set of molecules (Table 4),
which can explain the observed interaction of the nitrile with the
TSa
TSb
TSc
ΔE1
ΔE2
3a-b
4a
4c
4b
Fig. 4. Energy diagram for reaction of complexes 3a–c with acetonitrile calculated at DFT level.

256
J.M. García et al. / Inorganica Chimica Acta 414 (2014) 250–256
metal center when the acetonitrile is introduced on the system,
even if is located relatively far away (dRhN ꢃ 5–6 Å) from the metal
center, inducing a slight elongation of bond distance between the
rhodium and the carbon atoms of coordinated allylic double bond,
if it is compared with experimental values obtained by X-ray dif-
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via http://www.ccdc.cam.ac.uk/data_request/
cif. Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.ica.2014.02.
011.
fraction (Table 4). Moreover, the energy barrier values,
DE1 and
D
E2, shown in Table 4, indicate that theoretical calculations are
References
in agreement with the experimental observations. In this sense,
complexes 4a–c are more stable than the corresponding 3a–c
[1] (a) A. Bander, E. Lindner, Coord. Chem. Rev. 108 (1991) 27;
(b) C.S. Slone, D.A. Weinberger, C.A. Mirkin, Prog. Inorg. Chem. 48 (1999) 233;
(c) P. Braunstein, F. Naud, Angew. Chem., Int. Ed. 40 (2001) 680;
(d) C.J. Elsevier, J. Reedijk, P.H. Walston, M.D. Ward, Dalton Trans. (2003) 1869;
(e) X. Morise, P. Braunstein, R. Welter, Inorg. Chem. (2003) 7752;
(f) A.L. Gravilova, B. Bosnish, Chem. Rev. 104 (2004) 349;
(g) V.V. Grushin, Chem. Rev. 104 (2004) 1629;
(h) M. Bassetti, Eur. J. Inorg. Chem. (2006) 4473.
[2] (a) S. Bischoff, A. Weigt, H. Miessner, B. Lücke, Energy Fuels 10 (1996) 520;
(b) A.D. Burrows, Sci. Prog. 85 (2002) 199;
and
D
E1 barriers are enough lower to explain the equilibrium dis-
E2 bar-
placement to 4a–c at room temperature, furthermore, the
D
riers increase in the order 4c < 4a < 4b, which agrees with the order
the reactivity on removing the acetonitrile ligand under vacuum.
The calculated values of the whole charge on both diallylphos-
phine and acetonitrile ligands in complexes 4a–c are similar
(Table 5), showing that covalent character of bonds RhꢀP and
RhꢀN do not significantly changes when we go from 4a to 4b
(c) P. Espinet, K. Soulantica, Coord. Chem. Rev. 193–195 (1999) 499;
(d) P. Braunstein, J. Organometal. Chem. 689 (2004) 3953;
(e) M. Diéguez, O. Pàmies, A. Ruiz, Y. Díaz, S. Castillón, C. Claver, Coord. Chem.
Rev. 248 (2004) 2165;
and from 4b to 4c. Nevertheless, the low values of
BCP, suggests that the Rh–N bonds are polarized, showing a
marked ionic character. Indeed, these values of
q
at the RhꢀN
(f) H.M. Lee, C.-C. Lee, P.-Y. Cheng, Curr. Org. Chem. 11 (2007) 1491;
(g) M. Sauthier, P. Zinck, A. Mortreux, C. R. Chim. 13 (2010) 304.
[3] Same recent examples: (a) J. Andrieu, J.M. Camus, P. Richard, R. Poli, L.
Gonsalvi, F. Vizza, M. Peruzzini, Eur. J. Inorg. Chem. (2006) 51;
(b) M.V. Jiménez, J.J. Pérez-Torrente, M.I. Bartolomé, E. Vispe, F.J. Lahoz, L. Oro,
Macromolecules 42 (2009) 8146;
q
at the RhꢀN
BCP, are in the same order to calculated values of
inorganic salts [14,15].
q for some
In this way, a comparison of charge on the rhodium atom and
(c) E. Drinkel, A. Briceño, R. Dorta, R. Dorta, Organometallics 29 (2010) 2503;
(d) M.V. Jiménes, J.J. Pérez-Torrente, M.I. Bartolomé, F.J. Lahoz, L. Oro, Chem.
Commun. 46 (2010) 5322;
(e) M.V. Jiménez, J.J. Pérez-Torrente, M.I. Bartolomé, F.J. Lahoz, L. Oro, Chem.
Catal. Chem. 4 (2012) 1298.
nitrogen atom of acetonitrile ligand in complexes 4a–c shows that
D
qRh–N increases in the order 4c < 4a < 4b, this trend indicates that
the electrostatic interaction between the acetonitrile ligand and
the rhodium atom increases in the same order, which explains
why is experimentally easier to remove under vacuum the acetoni-
trile in the complex 4c that in the complexes 4a and 4b.
We can then conclude that, since the covalent component of the
interaction between the acetonitrile ligand and rhodium atom are
almost equivalents, the electrostatic component of the interaction
is the responsible of the experimentally observed behavior.
[4] (a) H.-L. Ji, J.H. Nelson, A. De Cian, J. Fischer, L. Solujic´, E.B. Milosavljevic´,
Organometallics 11 (1992) 401;
(b) L.P. Barthel-Rosa, K. Maitra, J. Fischer, J.H. Nelson, Organometallics 16
(1997) 1714;
(c) L.P. Barthel-Rosa, K. Maitra, J. Fischer, J.H. Nelson, Inorg. Chem. 37 (1998)
633;
(d) E. Ocando-Mavarez, M. Rosales, N. Silva, Heteroatom Chem. 9 (1998) 253;
(e) P. Alvarez, E. Lastra, J. Gimeno, M. Bassetti, L.R. Falvello, J. Am. Chem. Soc.
125 (2003) 2386;
(f) P. Alvarez, E. Lastra, J. Gimeno, P. Braña, J.A. Sordo, J. Gomez, L.R. Falvello, M.
Bassetti, Organometallics 23 (2004) 2956;
4. Conclusions
(g) M.A. Esteruelas, A.I. González, A.M. López, E. Oñate, Organometallics 23
(2004) 4858;
(h) M. Bassetti, P. Alvarez, J. Gimeno, E. Lastra, Organometallics 23 (2004)
5127;
(i) J. Díez, M.P. Gamasa, J. Gimeno, E. Lastra, A. Villar, Organometallics 24
(2005) 1410;
(j) J. Díez, M.P. Gamasa, E. Lastra, A. Villar, E. Pérez-Carreño, Organometallics
26 (2007) 5315;
[Rh(COD){
iPr2N 3c), synthesized starting from the chloro-bridged dimer
[Rh2( -Cl)2(COD)2], has been fully characterized by spectroscopic
j
3(P,C,C)-RP(CH2CH@CH2)2}][BF4] (R = tBu 3a, Ph 3b,
l
methods. Molecular structures of complexes 3a and 3c showed
the coordination of one of allyl groups on the rhodium demonstrat-
ing the bidentate character of the diallylphosphine ligands. The
hemilabile character of the diallylphosphine has been demon-
strated by the reversible displacement of coordinated allylic
double bond with acetonitrile, this behavior of allylphosphines
has never been reported in rhodium chemistry. Theoretical
calculations are agreed with the order of reactivity on removing
the acetonitrile under vacuum, revealing that the electrostatic
interaction between the acetonitrile ligand and the rhodium center
is responsible of the experimentally observed behavior, which
would be related to the donor character of the diallylphosphine
ligands.
(k) M. Bernechea, J.R. Berenguer, E. Lalinde, J. Torroba, Organometallics 28
(2009) 312.
[5] I. Hyder, M. Jiménez-Tenorio, M.C. Puerta, P. Valerga, Dalton Trans. (2007)
3000.
[6] G. Martín, E. Ocando-Mavarez, A. Osorio, M. Laya, M. Canestrari, Hetereoatom
Chem. 3 (1992) 395.
[7] E. Ocando-Mavarez, G. Martín, A. Andrade, Heteroatom Chem. 8 (1997) 97.
[8] CRYSTALCLEAR 1.3.6 SP3 r1, Rigaku and Rigaku/MSC, The Woodlands, TX.
[9] G.M. Sheldrick, SHELXTL NT Ver. 5.1, Bruker Analytical X-ray Systems Inc.,
Madison, WI, 2001.
[10] DMOL3 is available as part of Material Studio. Accelrys Inc., San Diego, USA,
2010.
[11] B. Delley, J. Chem. Phys. 92 (1990) 508.
[12] B. Delley, J. Chem. Phys. 113 (2000) 7756.
[13] J.P. Perdew, Y. Wang, Phys. Rev. B 45 (1992) 13244.
[14] R.F.W. Bader, Atoms in Molecules: A Quantum Theory, Oxford University Press,
New York, 1990.
Acknowledgments
[15] R.J. Gillespie, P.L.A. Popelier, Chemical Bonding and Molecular Geometry: From
Lewis to Electron Densities, Oxford University Press, 2001.
[16] H. Jonsson, G. Mills, K.W. Jacobsen, B.J. Berne, G. Ciccotti, D.F. Coker, Classical
and Quantum Dynamics in Condensed Phase Simulations, World Scientific,
Singapore, 1998. p. 385.
[17] G. Henkelman, H. Jónsson, J. Chem. Phys. 113 (2000) 9978.
[18] V.R. Landaeta, M. Peruzzini, V. Herrera, C. Bianchini, A. Sánchez-Delgado, A.E.
Goeta, F. Zanobini, J. Organometal. Chem. 691 (2006) 1039.
[19] B.R. James, R.H. Morris, K.J. Reimer, Can. J. Chem. 55 (1977) 2353.
[20] P.E. Garrou, Chem. Rev. 81 (1981) 229.
We thanks the Venezuelan Minister for Sciences, Technology
and Intermediate Industries for financial support (FONACIT
G-2005000433, G-2005000447, grant LAB-97000821 and
PCP-2007001996) Drs. Antoine Baceiredo from LHFA-Toulouse
and Yosslen Aray from Centro de Química-IVIC for helpful
discussions.
Appendix A. Supplementary material
[21] L.P. Barther-Rosa, V.J. Catalano, K. Maitra, J.H. Nelson, Organometallics 15
(1996) 3924.
CCDC 843258 and 843259 contains the supplementary crystal-
lographic data for compounds 3a and 3c, respectively. These data