Mark K Rong1, Koen van Duin1, Tom van Dijk1, Jeroen J M de Pater2, Berth-Jan Deelman2, Martin Nieger3, A W Ehlers4, J Chris Slootweg1, Koop Lammertsma4. 1. Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam , De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands. 2. Arkema B.V. , location Vlissingen, P.O. Box 70, 4380 AB Vlissingen, The Netherlands. 3. Laboratory of Inorganic Chemistry, Department of Chemistry, University of Helsinki , A. I. Virtasen aukio 1, P. O. Box 55, Helsinki, Finland. 4. Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands; Department of Chemistry, University of Johannesburg, Oakland Park 2006, South Africa.
Abstract
Highly stable iminophosphanes, obtained from alkylating nitriles and reaction of the resulting nitrilium ions with secondary phosphanes, were explored as tunable P-monodentate and 1,3-P,N bidentate ligands in rhodium complexes. X-ray crystal structures are reported for both κ1 and κ2 complexes with the counterion in one of them being an unusual anionic coordination polymer of silver triflate. The iminophosphane-based ruthenium(II)-catalyzed hydration of benzonitrile in 1,2-dimethoxyethane (180 °C, 3 h) and water (100 °C, 24 h) and under solvent free conditions (180 °C, 3 h) results in all cases in the selective formation of benzamide with yields of up to 96%, thereby outperforming by far the reactions in which the common 2-pyridyldiphenylphosphane is used as the 1,3-P,N ligand.
Highly stable iminophosphanes, obtained from alkylating nitriles and reaction of the resulting nitrilium ions with secondary phosphanes, were explored as tunable P-monodentate and 1,3-P,N bidentate ligands in rhodiumcomplexes. X-ray crystal structures are reported for both κ1 and κ2 complexes with the counterion in one of them being an unusual anioniccoordination polymer of silver triflate. The iminophosphane-based ruthenium(II)-catalyzed hydration of benzonitrile in 1,2-dimethoxyethane (180 °C, 3 h) and water (100 °C, 24 h) and under solvent free conditions (180 °C, 3 h) results in all cases in the selective formation of benzamide with yields of up to 96%, thereby outperforming by far the reactions in which the common 2-pyridyldiphenylphosphane is used as the 1,3-P,N ligand.
Hybrid ligands are
important in catalysis, as they enable ligand-
and metal-based reactivity in bifunctional systems.[1,2] In
particular, P,N ligands are of interest because of the presence of
both soft phosphorus and hard nitrogendonor sites.[3] Of the various topologies,[4−7] 1,3-P,N ligands have shown ample applicability
in transition-metal-based catalysis (Figure ). The C1 spacer provides optimal
separation[8] for metalcoordination at nitrogen,[9] phosphorus,[10,11] or both atoms,[11] as well as enabling the formation of homo-[12] and heterobimetallic[13] complexes for metal activation, photoluminescence, and cooperative
metal–metal systems (Figure ).[13a,14]
Figure 1
1,3-P,N ligand systems.
Figure 2
1,3-P,N coordination modes.
pan class="Chemical">1,3-P,N ligand systems.
pan class="Chemical">1,3-P,N pan class="Chemical">coordination modes.
Exemplary reactions that make use of 1,3-P,N ligands are
the Ru-catalyzed
hydration of nitriles,[10a,11a,11h] the Pd-catalyzed carbonylation of alkynes,[10c,11d,11e] Pd-catalyzed Buchwald–Hartwig
and Suzuki–Miyaura cross-coupling reactions,[15] dinuclear Pd- and [Fe-Rh]-catalyzed carbonylation of alcohols,[12d,12e,13c] the Ru- and Au-catalyzed hydration
of alkynes,[10d,10e] the Ru-catalyzed transfer hydrogenation
of ketones and aldehydes,[11b,11g] the Ru-catalyzed isomerization
of alkenes,[10b] the Ru-catalyzed and dinuclear
Rh-catalyzed hydrogenation of alkenes,[11f,12b] the Rh-,
Ir-, and [Ru–Rh]-catalyzed hydroformylation of alkenes,[11c,13a,13d] and the Cr-catalyzed tri- and
tetramerization of ethylene,[11i] whereas
CuI2 and [AuI3-AgI] complexes have been used in photoluminescence.[12a,12c,13b] A popular ligand in the complexes
used for these reactions is 2-pyridyldiphenylphosphane (PyPPh2),[3,16] even though it has limited flexibility for
steric and electronic tuning of particularly the N-donor site. Replacement
of the pyridine group by an imine or amine functionality provides
1,3-P,N ligands with more opportunities to tune the properties of
their derived catalysts.Repan class="Chemical">cently, we repn>orted a simple methodology,
based on pan class="Chemical">nitrilium
ions, to synthesize iminophosphanes (IUPAC: c-phosphanylimines)
and anionicphosphaamidinates that carry different substituents at
all three C, N, and P sites (Scheme ).[17] Shortly thereafter
Hanton, Dyer, and co-workers reported neutral ligands by condensing
silylated phosphanes and imidoyl chlorides, but this method is intrinsically
more limited.[11i] These 1,3-P,N ligands
coordinate to Au,[17] Rh,[17] Ir,[17b] and Cr[11i] complexes. The iminophosphanes display dynamic κ1 and κ2 coordination behavior in the case
of Rh(III), while 2-pyridylphosphanes generally give exclusively κ1 Rhcomplexes.[18a−18c]
Scheme 1
Synthesis of Iminophosphanes
and Phosphaamidinates from Nitrilium
Ions
In the present study we expand
on the synthesis of iminophosphanes
to display the diversity of the nitrilium ion methodology, report
on the chelation to Rh(III) to further exploit the intricacies of
κ1 versus κ2 coordination, and examine
the performance of the iminophosphane versus the PyPPh2 ligand for the Ru(II)-catalyzed hydration of nitriles. The last
process, first reported in 1986[19] and used
for the production of polyacrylates and pharmaceuticals,[19−21] was chosen because of the role of the N-donor site in activating
water in the catalyticcycle.
Results and Discussion
The synthesis
and stability of new pan class="Chemical">iminophosphanes with relatively
small substituents is presented first. Next, the coordination to Rh(III)
is examined. This is followed by an assessment of the performance
of the ligands in the Ru(II)-catalyzed hydration of nitriles.
Ligands
A set of iminophosphanes (3a–f)
with different electronic properties but similar steric
features was prepared from nitrilium triflates 2 (Scheme ). The procedure
is analogous to the protocol we reported recently, the difference
being that here nitriles 1 are directly alkylated to
provide the ion, whereas previously we obtained the ions from imidoyl
chlorides.[17] The choice of synthetic approach
depends on the N substituent: with i-Pr or larger
alkyl groups, imidoyl chlorides are the starting point, sometimes
necessitating stabilization of the resulting nitrilium ions with pyridine
bases,[17c] but smaller groups are only accessible
by nitrile alkylation.[22] Thus, treating 1 with MeOTf gave the corresponding N-methyl
nitrilium triflates 2 (R = Ph (a), 4-CH3-C6H4 (b), 4-CF3-C6H4 (c)). The reaction with 1b was carried out in toluene at ambient temperature to provide
after 5 days white crystalline 2b (62%). Methylations
of the less nucleophilic 1a,c were performed
without solvent over an 18 h period; in the case of 1c a slightly elevated reaction temperature of 45 °C was used
to melt the nitrile. Crystallization of the crude products provided
nitrilium salts 2a (DCM/Et2O) and 2c (DCM/pentane) as white solids in 86% and 69% yields, respectively.
Scheme 2
Synthesis of Iminophosphanes 3
Dropwise addition of diphenylphosphane and its 3-methyl,
3-CF3, and 4-methyl derivatives to a DCM solution of nitriliumtriflates 2a–c at −78 °C,
followed by deprotonation with triethylamine at room temperature,
resulted in the quantitative formation of iminophosphanes 3a–f (Scheme ). Et2O extraction, filtration over alumina,
and crystallization provided the products in 46–91% as E/Z mixtures, as determined by their 31P NMR chemical shifts (i.e., 3a 7.8 (E), −7.6 (Z); 3b 7.6
(E), −8.0 (Z); 3c 8.2 (E), −7.7 (Z); 3d 5.9 (E), −8.5 (Z); 3e 8.3 (E), −7.5 (Z); 3f 6.3 (E), −8.8
(Z)) with assignments based on related ones reported
earlier.[17b] The E/Z ratios were found to be sensitive to the experimental
setup (e.g., concentration), which is in agreement with the calculated
very small energy differences (ωB97X-D/6-31+G(d,p); |ΔE|= 0.19–0.78 kcal mol–1)[23] as well as similar behavior reported for related systems;[17]3g was included for its different
N substituent (i-Pr versus Me).[17b]
Ligand Stability
Exploring the applicability
of iminophosphanes
in transition-metal-catalyzed reactions, such as the hydration of
nitriles (vide infra), requires these P,N ligands to be stable toward
water, whereas it is quite conceivable that they are prone to hydrolysis
to the corresponding diarylphosphanes and amides. It is then comforting
to see, using 31P NMR monitoring, that the stability of E-/Z-3a in acetone toward
hydrolysis by degassed water over a period of 211 h showed only slight
degradation, which was attributed to oxidation to [3a(O)] by air rather than hydrolysis (Figure ). The oxidation sensitivity of 3a was confirmed in a separate assessment (see Supporting Information). Similar results were obtained for 3b,c (see Supporting Information). The presence of 0.10 equiv of triflic acid did not catalyze the
hydrolysis of 3a. Exposure of 3a to triflic
acid at room temperature gave only N-protonated iminophosphane without
any degradation over a 46 h period. This indicates that iminophosphanes
are reasonably rugged ligands.
Figure 3
Sensitivity of 3a (%) toward
hydrolysis (reagent H2O (degassed)) and oxidation (reagent
O2 (in air)).
Sensitivity of 3a (%) toward
hydrolysis (reagent pan class="Chemical">H2O (degassed)) and oxidation (reagent
pan class="Chemical">O2 (in air)).
Rhodium Coordination
The chelation of the pan class="Chemical">iminophosphanes
toward rhodium was explored next. Recently, we found that ligands
bulkier than 3a–f give P-monodentate
complexes with Au(I), bidentate complexes with Ir(III), and both κ1 and κ2 complexes with Rh(III), of which
only bidentate complexes were isolable as the major products.[17b] We wondered whether we could make monodentate
Rhcomplexes the prevalent product, which is of interest for cooperative
Rh(III) catalysis,[10,11,13d,24] without relying on steric effects.[25] We also wondered how silver triflate, commonly
added in excess in Rh(III) catalysis,[24b,24c] plays a role
beyond the formation of bidentate complexes by single Cl abstraction.
To address these issues, we focus on the chelation of 3a,c,e.
Treating E-/Z-3a in DCM with 1/2 equiv of [RhCp*Cl2]2 at room temperature yielded two products in
a 96:4 ratio (Scheme ). The major product (δ(31P) 34.5 ppm; 1JP,Rh = 145.7 Hz), isolated as an orange
solid (89%), is assigned to the E isomer of monodentate
Rhcomplex 4a, in analogy to reported in situ Rhcomplexes.[17c] The minor product (δ(31P)
−5.1 ppm; 1JP,Rh = 113.4
Hz) is likely bidentate complex 5a because of its similarity
in 31P NMR data to related, bulkier Rhcomplexes.[17c] Crystallization of 4a by slow
diffusion of pentane into a DCM/Et2O solution provided
crystals suitable for an X-ray structure determination. The molecular
structure confirms P-monodentate coordination and the E conformation for the imine group (Figure ), thereby illustrating the ease of Z to E isomerization of the ligand.[17c] The N1–C1 bond length of 1.2738(17)
Å is typical for a C=N bond and the 1.8659(13) Å
long P1–C1 bond is normal for a P–alkyl single bond;
the Rh1–P1–C1–N1 torsion angle amounts to −56.15(11)°,
with H-bonding interactions between Cp*–CH3 groups and the N lone pair. The 2.3222(4) Å Rh1–P1
bond length of 4a is similar to that of the 2-pyridylphosphanyl
Rh(III) κ1 complexes.[18a−18c]
Scheme 3
Synthesis of Mono-
and Bidentate Rh(III) Complexes 4 and 5
Figure 4
Displacement ellipsoid plots of rhodium
complexes 4a (left) and 4e (right) at the
50% probability level.
Hydrogen atoms and DCM are omitted for clarity. Selected bond lengths
(Å) and angles (deg): 4a, Rh1–P1 = 2.3222(4),
Rh1–Cl1 = 2.4056(3), Rh1–Cl2 = 2.4008(4), P1–C1
= 1.8659(13), N1–C1= 1.2738(17), N1–C2 = 1.4634(17),
C1–C3 = 1.4946(17), N1–C1–P1 = 112.72(9), Rh1–P1–C1–N1
= −56.15(11); 4e, Rh1–P1 = 2.3196(6), Rh1–Cl1
= 2.4013(6), Rh1–Cl2 = 2.4011(6), P1–C1 = 1.870(2),
N1–C1= 1.269(3), N1–C2 = 1.469(3), C1–C3 = 1.493(3),
N1–C1–P1 = 113.62(18), Rh1–P1–C1–N1
= −53.92(19).
Displapan class="Chemical">cement ellipsoid plots of rhodiumcomplexes 4a (left) and 4e (right) at the
50% probability level.
Hydrogen atoms and DCM are omitted for clarity. Selected bond lengths
(Å) and angles (deg): 4a, Rh1–P1 = 2.3222(4),
Rh1–Cl1 = 2.4056(3), Rh1–Cl2 = 2.4008(4), P1–C1
= 1.8659(13), N1–C1= 1.2738(17), N1–C2 = 1.4634(17),
C1–C3 = 1.4946(17), N1–C1–P1 = 112.72(9), Rh1–P1–C1–N1
= −56.15(11); 4e, Rh1–P1 = 2.3196(6), Rh1–Cl1
= 2.4013(6), Rh1–Cl2 = 2.4011(6), P1–C1 = 1.870(2),
N1–C1= 1.269(3), N1–C2 = 1.469(3), C1–C3 = 1.493(3),
N1–C1–P1 = 113.62(18), Rh1–P1–C1–N1
= −53.92(19).
P-monodentate complexes 4c (δ(31P)
35.4 ppm, 1JP,Rh = 147.4 Hz)
and 4e (δ(31P) 34.7 ppm, 1JP,Rh = 145.8 Hz) were prepared likewise
from 3c,e and isolated in 69% and 92% yields,
respectively (Scheme ). The molecular structure of 4e, obtained by a single-crystal
X-ray structure determination, reveals structural parameters with
bond lengths Rh1–P1 = 2.3196(6) Å, N1–C1 = 1.269(3)
Å, and P1–C1 = 1.870(2) Å and a −53.92(19)°
Rh1–P1–C1–N1 torsion angle that are comparable
to those of 4a (Figure ). 31P NMR analysis of the two reactions
showed only in the case of 3e formation of a byproduct
(6%) that is assigned to P,N bidentatecomplex 5e (δ(31P) −12.1 ppm, 1JP,Rh = 114.7 Hz).The reproducible 4:5 ratios indicate
that the small extent or absence of observed bidentate complexes depends
to a degree on the aromatic substituent. In comparison to the C- and P-phenyl groups, the m-tolyl substituent at carbon (e) enhances bidentate
formation slightly (from 4 to 6%), while the p-CF3-phenyl group at phosphorus (c) blocks it (Scheme ). This trend concurs
with the N-donor strength calculated at ωB97X-D/6-31+G(d,p)
(Def2-TZVP for Rh) from the energy required to transform cationiccomplex 5 from a κ1 to a κ2 arrangement (Scheme );[17b,23] a comparison between neutral 4 and ionic 5 is not
feasible. The Rh–N bond strengths of 5a,c,e amount to −14.7, −13.4, and
−15.2 kcal mol–1, respectively, with corresponding
N lone pair orbital energies of −11.2, −11.4, and −11.0
eV. These data support the notion that the selective formation of
only P-chelated complexes of 1,3-P,N ligands can be controlled by
changing the N-donor strength.
Scheme 4
DFT Analysis of the N-Donor Capacity
of 3a,c,e
Of pan class="Chemical">course, the pan class="Chemical">1,3-P,N ligands can also undergo
clean chelation
with rhodium as shown before, but under more forcing conditions. For
example, treating 4 with 1 equiv of silver triflate yielded
bidentate complex 6 quantitatively (Scheme ).
Scheme 5
Bidentate Coordination
of 3 to Rh(III)
These ioniccomplexes, containing a triflatecounteranion
instead
of a chloride anion as in 5, were isolated as bright
orange solids and identified by their 31P NMR characteristics
(6a, −9.1 ppm, 1JP,Rh = 113.4 Hz; 6c, −5.0 ppm, 1JP,Rh = 115.0 Hz; 6e, −8.6
ppm, 1JP,Rh = 113.4 Hz). When 4a was treated with an excess of 4 equiv of AgOTf,[24b,24c] a yellow solid resulted (7a, 80%) having a 31P NMR chemical shift (−9.9 ppm; 1JP,Rh = 111.8 Hz) which differs from that of 6a (Scheme ). Single-crystal
X-ray structure determinations confirm the bidentate nature of both
complexes (Figure ) and reveal that the excess of AgOTfcaused the exchange of rhodium’s
coordinated chloride for an OTf group, without affecting the P,N ligand.
The molecular structure of 6a shows Rh1–P1, Rh1–N1,
and Rh1–Cl1 bond lengths of respectively 2.3123(14), 2.106(4),
and 2.3728(13) Å and an acute N1–C1–P1 bond angle
of 100.4(4)°. The structural features of 7a resemble
those of 6a and other bidentate 1,3-P,N Rh(III) complexes,[17,18a,18b,18d] except for its triflate group with an Rh1–O11 bond length
of 2.1957(18) Å that is similar to those reported for monodentate
Rh(III)–triflate bonds.[26] Expectedly,
the S1–O11 bond is elongated with respect to the S1–O12
and S1–O13 bonds: i.e., 1.4711(19), 1.433(2), and 1.429(2)
Å, respectively.
Figure 5
Displacement ellipsoid plots of cationic rhodium complexes 6a (left) at the 30% probability level and 7a (right) at the 50% probability level. Hydrogen atoms and counterions
(triflate and [Ag2OTf3(DCM)]–, respectively) are omitted for clarity.
Selected bond lengths (Å) and angles (deg): for 6a, Rh1–P1 = 2.3123(14), Rh1–N1 = 2.106(4), Rh1–Cl1
= 2.3728(13), P1–C1 = 1.842(5), N1–C1 = 1.274(7), N1–C2
= 1.466(7), C1–C3 = 1.461(7), N1–Rh1–P1 = 66.27(12),
N1–C1–P1 = 100.4(4); 7a, Rh1–P1
= 2.3328(7), Rh1–N1 = 2.126(2), Rh1–O11 = 2.1957(18),
P1–C1 = 1.844(3), N1–C1 = 1.294(3), N1–C2 = 1.466(3),
C1–C3 = 1.471(4), S1–O11 = 1.4711(19), S1–O12
= 1.433(2), S1–O13 = 1.429(2), N1–Rh1–P1 = 67.02(6),
N1–C1–P1 = 102.29(18).
Displacement ellipsoid plots of cationicrhodiumcomplexes 6a (left) at the 30% probability level and 7a (right) at the 50% probability level. Hydrogen atoms and counterions
(triflate and [Ag2OTf3(DCM)]–, respectively) are omitted for clarity.
Selected bond lengths (Å) and angles (deg): for 6a, Rh1–P1 = 2.3123(14), Rh1–N1 = 2.106(4), Rh1–Cl1
= 2.3728(13), P1–C1 = 1.842(5), N1–C1 = 1.274(7), N1–C2
= 1.466(7), C1–C3 = 1.461(7), N1–Rh1–P1 = 66.27(12),
N1–C1–P1 = 100.4(4); 7a, Rh1–P1
= 2.3328(7), Rh1–N1 = 2.126(2), Rh1–O11 = 2.1957(18),
P1–C1 = 1.844(3), N1–C1 = 1.294(3), N1–C2 = 1.466(3),
C1–C3 = 1.471(4), S1–O11 = 1.4711(19), S1–O12
= 1.433(2), S1–O13 = 1.429(2), N1–Rh1–P1 = 67.02(6),
N1–C1–P1 = 102.29(18).The counterion of 7a is a coordination polymer
of
silver triflate (Figure ). Whereas aggregates of Ag(I) sulfonates are known,[27] the present structure is, to the best of our knowledge,
the first of its kind in an organometalliccomplex. The molecular
structure shows a linear chain with alternating C-symmetric dinuclear Ag(I) units (Ag1–Ag2
= 3.0410(3) Å) that are bridged by triflate anions. Each Ag atom
has a distorted-octahedral geometry with bi-, tri-, and tetradentate
coordinating triflate groups labeled as 2, 3, and 4 in Figure (right); the Ag–O bond
distances range from 2.321(2) to 2.509(2) Å.[27] The weak coordination of DCM (Ag1–Cl1 = 2.9042(9)
Å) completes the octahedral coordination sphere of Ag1.[28]
Figure 6
Displacement ellipsoid plot of a dimeric unit of the [Ag2OTf3(DCM)] counterion
of 7a at the 20% probability level (left) and its simplified
numbering scheme (right). Hydrogen atoms and minor disordered part
of the solvent DCM are omitted for clarity. Selected bond lengths
(Å) and angles (deg): Ag1–Ag2 = 3.0410(3), Ag1- - -Ag2b
= 4.0607(3), Ag1- - -Ag2a = 5.9371(3), Ag1–O
= 2.345(2)-2.394(2); Ag2–O = 2.321(2)-2.509(2), Ag1- - -O33
= 2.621(2), Ag1–Cl1 = 2.9042(9), Ag1–Ag2–Ag1a
= 113.71(1), Ag2–Ag1–Ag2b = 102.17(1), Ag2–Ag1–Ag2b–Ag1b
= 6.86(2).
Displacement ellipsoid plot of a dimeric unit of the [Ag2OTf3(DCM)] counterion
of 7a at the 20% probability level (left) and its simplified
numbering scheme (right). Hydrogen atoms and minor disordered part
of the solvent DCM are omitted for clarity. Selected bond lengths
(Å) and angles (deg): Ag1–Ag2 = 3.0410(3), Ag1- - -Ag2b
= 4.0607(3), Ag1- - -Ag2a = 5.9371(3), Ag1–O
= 2.345(2)-2.394(2); Ag2–O = 2.321(2)-2.509(2), Ag1- - -O33
= 2.621(2), Ag1–Cl1 = 2.9042(9), Ag1–Ag2–Ag1a
= 113.71(1), Ag2–Ag1–Ag2b = 102.17(1), Ag2–Ag1–Ag2b–Ag1b
= 6.86(2).
Application in Catalytic
Nitrile Hydration
Having established
simple syntheticprotocols for highly stable iminophosphanes and having
demonstrated their ability to function as P monodentate and P,N bidentate
ligands, we set out to explore their potential as hydration catalysts
for nitriles. This reaction was chosen to compare iminophosphanes
to the PyPPh2 ligand for the role of the hard nitrogendonor site to activate water, as is illustrated in Figure .[10a,11a,11h] We opted for rutheniumcatalysts,[29] which are extensively used in homogeneous catalysis[30] and are more robust than those based on rhodium
(for comparison and completeness also included in our tests), while
their chemistry is comparable. This choice is corroborated by the
elevated temperatures needed.
Figure 7
Ru-catalyzed, P,N-assisted hydration of benzonitrile.
pan class="Chemical">Ru-pan class="Chemical">catalyzed, P,N-assisted hydration of benzonitrile.
To identify a suitable Ru source
for benchmarking iminophosphane 3a against PyPPh2 in the hydration of benzonitrile,
we conducted the catalytic reactions in 1,2-dimethoxyethane, in water,
and in the absence of a solvent. Table gives the precatalyst, ligand, solvent, and the yield
of benzamide. For the reactions in DME, the procedure of Takai, Oshiki,
and co-workers[11h] was followed by premixing
5 mol % of the Ru precursor complex and 3a for 30 min,
upon which the catalyzed hydration of benzonitrile was executed in
a closed vessel at 180 °C for 3 h; 2 equiv of water was used
to remain consistent with the reported protocol. The results show
that precatalyst [Ru(p-cymene)Cl2]2 (entries 1 and 2) gives better results than [Ru(C6Me6)Cl2]2 (entry 5) and [Ru(C6H5Me)Cl2]2 (entry 6), but
even these give reasonable conversions.
Determined by Gpan class="Chemical">C.
pan class="Chemical">Conditions: Ph–pan class="Chemical">C≡N
(1 mmol), H2O (2 mmol), [M]L (5 mol %); DME (0.5 mL).
2 equiv.Prepared in situ from pan class="Chemical">[Ru(p-cym)Cl2]2 and pan class="Chemical">AgOTf.
pan class="Chemical">Conditions: Ph–pan class="Chemical">C≡N
(1 mmol), [M]L (5 mol %); H2O (3.0 mL).
pan class="Chemical">Conditions: Ph–pan class="Chemical">C≡N
(3.6 mmol), H2O (7.2 mmol), [M]L (1.4 mol %).
The presenpan class="Chemical">ce of 1 equiv of pan class="Chemical">AgOTf
hindered the reaction (Table , entry 3), presumably
through the formation of inactive bidentate species. Remarkable is
the observation that the catalytic hydration using 3a gave an excellent yield of 87% (entry 1), while the same process
with the PyPPh2 ligand performed very poorly with a product
yield of a mere 3%[11h] (entry 4). Using
2 equiv of 3a further enhanced the yield to 96% (entry
2). In addition, rhodium precatalyst [RhCp*Cl2]2 was examined under the same conditions. Even this in situ generated
catalyst, presumably 4a, yields benzamide, albeit in
only in 22% yield (entry 7), which is a remarkable observation, as
no Rh(III) catalyst has so far been reported to hydrate nitriles.
Next, we explored the protocol of Cadierno et al. by heating benzonitrile
and 5 mol % of precatalyst [Ru(p-cym)Cl2]2 and the 1,3-P,N ligand in boiling water for 24 h using
a closed vessel.[11a,31] Also in this case an outstanding
yield of 94% was obtained for the catalytic reaction with 3a (Table , entry 8),
which is far better than the 57% that resulted with the PyPPh2 ligand (entry 9).[11a] This performance
is in line with that with DME as solvent (entries 1 and 4) by also
showing that the catalytic reaction with ligand 3a outperforms
that with Ph2PPy. Finally, we compared the influence of
these ligands in the solvent-free hydration of benzonitrile by heating
1.4 mol % of the catalyst in a 1:2 molar mixture of benzonitrile and
water at 180 °C for 3 h in a closed vessel. Again, the reaction
of [Ru(p-cym)Cl2]2 and 3a (78%, entry 10) gave a far better performance than that
with the Ph2PPy ligand (6%, entry 11), which, in fact,
seems to have no effect, as 6% was also obtained when no P,N ligand
was present (entry 12). It appears that all three Ru-catalyzed approaches
for the hydration of benzonitrile underscore the potential of iminophosphanes
as ligands. The observed differences in the activity of 3a and PyPPh2 were attributed to their difference in N-donor
strength.[10a]Finally, we address
briefly the influenpan class="Chemical">ce of the P,pan class="Chemical">C,N substituents
of 3 on the yield of the Ru(II)-catalyzed hydration using
solvent-free conditions. The results are given in Table .
Table 2
Solvent-Free
[(3)Ru(p-cym)Cl2]-Catalyzed
Hydration of Benzonitrilea
pan class="Chemical">Conditions: Ph–pan class="Chemical">C≡N
(3.6 mmol), H2O (7.2 mmol), [Ru(p-cym)Cl2]L (1.4 mol %).
Determined by Gpan class="Chemical">C.
Whereas
the iminophosphanes are only modestly different from each
other, some distinctly affected the benzamide yield. As a reference
we used 3a, which has Ph substituents on the P and Ccenters and a Me group on the N atom. The catalyzed hydration yield
of benzonitrile amounted to 78% (Table , entry 1). Changing the N substituent from Me to the
bulkier i-Pr group (3g) decreased the
yield by only 3% (entry 5). A similar marginal effect (−2%)
was observed on introducing a 4-CF3 substituent on the C-phenyl group (3c, entry 2). Slightly larger
effects were found on introducing 3-CH3 and 3-CF3 substituents on the P-phenyl groups, causing modest
changes in yield of +4% (entry 3) and −8% (entry 4), respectively.
These may possibly be attributed to a difference in electron-donating
properties, which affect ruthenium’s Lewis acidity via the
Ru–P bond.[14d] All of this leads
to the observation that the iminophosphanes 3 are all
effective ligands enabling the solvent-free Ru(II)-catalyzed hydration
of benzonitrile and that these 1,3-P,N ligands can be modified at
the P, C, and N centers to effect the hydration to different degrees.
Conclusion
This study shows iminophosphanes to be readily
synthesized by simple
alkylation of nitriles and reaction of the resulting nitrilium ions
with a secondary phosphane. They are very stable in water but do show
some sensitivity to oxidation in air. The 1,3-P,N ligands are tunable
by substitution at their P, C, and N centers. Their coordination behavior
was explored for rhodium(III) complexes, using [RhCp*Cl2]2 as precursor. Both P monodentate and 1,3-P,N bidentate
ligands can be obtained in high yield. Silver triflate is needed to
enforce the formation of the κ2 chelates, and when
an excess is used, also the second chloride is exchanged. Several
X-ray crystal structures of rhodiumcomplexes are reported, including
one with a polymeric aggregate of Ag(I) sulfonate with embedded DCM
solvent molecules. The ligands were found to be effective in the Ru(II)-catalyzed
hydration of benzonitrile, be it in an organic solvent, in water,
or under solvent-free conditions. Its performance appears to be better
than that of 2-pyridyldiphenylphosphane, with yields of up to 96%
on using the commercial [Ru(p-cym)Cl2]2 as precatalyst. Modifying the substitution pattern of the
iminophosphanes affects the yield of the ruthenium-catalyzed hydration
of benzonitrile as well as the coordination to the rhodiumcomplex.
Clearly, more extensive studies have to be performed for more catalytic
processes, but the easily synthesized iminophosphanes seem to have
great prospects as new 1,3-P,N ligands.
Experimental
Section
Preparation of Compounds
All experiments were performed
under an atmosphere of dry nitrogen using standard Schlenk-line and
glovebox techniques, unless stated otherwise. Solvents were distilled
under nitrogen over the appropriate drying agent; CaCl2 (DCM), benzophenone/NaK (Et2O, THF, triethylamine), Na
(toluene), LiAlH4 (pentane), P2O5 (CD2Cl2, CDCl3). C6D6 was dried over Na at room temperature. H2O was
degassed ultrasonically in vacuo. Silver salts were handled with minimum
light exposure. Diphenylphosphane was purchased from Sigma-Aldrich,
and tris(3-methylphenyl)phosphane and bis(4-methylphenyl)chlorophosphane
were obtained from STREM Chemicals Inc. Bis(3-trifluoromethylphenyl)chlorophosphane
was provided by Arkema B.V. The known compounds 2a,[22b] bis(3-trifluoromethylphenyl)phosphane, bis(3-methylphenyl)phosphane,
and bis(4-methylphenyl)phosphane are reported here, since their syntheses
were revised and/or new analytical data were obtained. All phosphanes
and chlorophosphanes were distilled under reduced pressure before
use. Solids were predried in vacuo for at least 30 min. All other
reagents were used as received. NMR spectra were recorded on a Bruker
Avance 250 (1H, 250.13 MHz; 19F, 235.36 MHz; 31P, 101.25 MHz; room temperature), a Bruker Avance 400 (1H, 400.13 MHz; 13C{1H}, 100.61 MHz; 31P, 161.98 MHz; room temperature) ,or a Bruker Avance 500
instrument (1H, 500.23 MHz; 13C{1H}, 125.78 MHz; room temperature). 1H spectra and 13C{1H} spectra were internally referenced to residual
solvent resonances (CDCl3, δ(1H) 7.26,
δ(13C{1H}) 77.16; CD2Cl2, δ(1H) 5.32, δ(13C{1H}) 53.84; C6D6, δ(1H) 7.16, δ(13C{1H}) 128.06), and 31P spectra were referenced externally to H3PO4. Melting points were measured using a Büchi M-565
melting point apparatus (sealed capillaries) and are uncorrected.
High-resolution electrospray ionization (ESI) mass spectrometry was
carried out with a Bruker micrOTOF-Q instrument in positive ion mode
(capillary potential of 4500 V). Infrared spectra were recorded on
a Shimadzu FT-IR 8400S spectrophotometer.
Computational Procedure
Density functional calculations
were performed at the ωB97X-D[32] level
of theory using Gaussian09, revision A.02.[23] Geometry optimizations were performed using the 6-31+G(d,p)[33] basis set (Def2-TZVP for Rh),[34] and the nature of each stationary point was confirmed by
frequency calculations.
MeOTf (1.0 equiv) was added dropwise to
a solution of arylnitrile (1.0 equiv) in pan class="Chemical">toluene (0.27 mL/mmol of
arylnitrile). After the mixture was stirred for 18 h at room temperature,
a microcrystalline solid precipitated. Volatiles were removed in vacuo,
and after washing with pentane (2 × 0.4 mL/mmol), 2 was obtained as an off-white solid (2a, 62%; 2b, 62%; 2c: 14%).
A solution of freshly distilled bis(3-trifluoromethylphenyl)chlorophosphane
(5.88 g, 16.50 mmol, 1.00 equiv) in 30 mL of Et2O was added
dropwise at −78 °C to a suspension of LiAlH4 (0.25 g, 6.59 mmol, 0.40 equiv) in 50 mL of Et2O, during
which gas evolved from the mixture. After addition, the gray suspension
was warmed to room temperature. After it was stirred for 1.5 h, the
mixture was cooled to 0 °C and H2O (0.95 mL, 52.63
mmol, 3.19 equiv) was added dropwise. After the mixture was stirred
for 0.5 h at room temperature, volatiles were removed in vacuo and
the resulting gray-white suspension was extracted into 3 × 10
mL of pentane. The combined extracts were evaporated to provide bis(3-trifluoromethylphenyl)phosphane
as a colorless liquid (4.94 g, 15.33 mmol, 93%). 1H NMR
(500.23 MHz, C6D6): δ 7.61 (d, 3JH,P = 6.9 Hz, 2H; PC–CH-CCF3), 7.17 (d, 3JH,H = 6.3 Hz, 2H; p-ArH), 7.05 (t, 3JH,H = 6.9 Hz,
2H; PC–CH-CH), 6.71 (t, 3JH,H = 7.7 Hz, 2H; m-ArH), 4.81 (d, 1JH,P = 219.7 Hz, 1H; PH). 13C{1H} NMR (125.78 MHz, C6D6): δ 137.2 (d, 2JC,P = 17.3 Hz; PC–CH-CH), 135.9 (d, 1JC,P = 13.6 Hz; ipso-ArC), 131.2 (qd, 2JC,F = 32.7 Hz, 3JC,P = 6.4 Hz; PC-CH-CCF3), 130.5 (dq, 2JC,P = 18.2 Hz, 3JC,F = 4.5 Hz;
PC-CH-CCF3), 129.3 (d, 3JC,P = 5.5 Hz; m-ArC), 125.7 (q, 3JC,F = 3.6 Hz; p-ArC), 124.6 (q, 1JC,F = 272.5 Hz; CF3). 19F NMR (235.36 MHz, C6D6): δ −63.3
(s; C6H4CF3). 31P NMR (161.98 MHz, C6D6): δ −41.1
(dq, 1JP,H = 218.3 Hz, 3JP,H = 6.7 Hz).
Bis(3-methylphenyl)phosphane.[36]
A colorless solution of tris(3-methylphenyl)phosphane
(4.18 g,
13.73 mmol, 1.00 equiv) in 30 mL of THF was prepared in a separate
vessel and added dropwise at 0 °C to an excess of finely cut
Li (0.21 g 30.23 mmol, 2.20 equiv) suspended in 5 mL of THF. Afterward,
the vessel was washed with 2 × 10 mL of THF, which was also added
to the reaction mixture. After addition, the mixture was warmed to
room temperature. After this mixture was stirred for 26 h at room
temperature, a deep red mixture was obtained, which was decanted to
remove excess Li. The resulting solution was cooled to 0 °C,
and H2O (1.0 mL, 55.40 mmol, 4.04 equiv) was added dropwise
to give a white-gray mixture. After 1 h of stirring at room temperature,
volatiles were removed in vacuo to provide a white oil, which was
extracted sequentially into 50 mL of Et2O for 24 h and
40 mL of Et2O for 20 h. The combined extracts were evaporated
to give a light yellow liquid, which was distilled (3.2 × 10–2 mbar, heated with Bunsen burner) to provide bis(3-methylphenyl)phosphane
(1.71 g, 7.99 mmol, 58%) as a colorless liquid. 1H NMR
(500.23 MHz, C6D6): δ 7.30 (t, 3JH,H = 8.2 Hz, 4H; o-ArH), 7.00 (t, 3JH,H = 7.4 Hz, 2H; m-ArH),
6.87 (d, 3JH,H = 6.6 Hz, 2H; p-ArH), 5.27 (d, 1JH,P = 215.0 Hz, 1H; PH), 1.99 (s, 6H;
CH3). 13C{1H} NMR
(125.78 MHz, C6D6): δ 138.3 (d, 2JC,P = 6.4 Hz; m-ArC-CH3), 135.4 (d, 1JC,P = 10.9 Hz; ipso-ArC), 135.0 (d, 2JC,P = 18.2
Hz; o-ArC), 131.5 (d, 2JC,P = 16.4 Hz; o-ArC), 129.5 (s; p-ArC),
128.8 (d, 3JC,P = 6.4 Hz; m-ArC), 21.2 (s; CH3). 31P NMR (161.98 MHz, C6D6): δ −40.3 (dq, 1JP,H = 215.4 Hz, 3JP,H = 7.9 Hz).
Bis(4-methylphenyl)phosphane.[35]
A solution of bis(4-methylphenyl)chlorophosphane (1.81 mL, 8.01
mmol, 1.00 equiv) in 15 mL of Et2O was added dropwise to
a suspension of LiAlH4 (0.094 g 2.48 mmol, 0.31 equiv)
at −78 °C. After addition, the gray suspension was warmed
to room temperature. After it was stirred for 1.5 h at room temperature,
the mixture was cooled to 0 °C and H2O (0.15 mL, 8.31
mmol, 1.04 equiv) was added dropwise. After it was stirred for 1 h
at room temperature, the gray-white suspension was filtered onto anhydrous
Na2SO4 (2.71 g), after which it was filtered
again. Volatiles were removed in vacuo to provide a white suspension,
which was extracted into 20 mL of pentane. After evaporation of the
extract, the obtained white suspension was distilled (2.3 × 10–2 mbar, 100 °C). Bis(4-methylphenyl)phosphane
(0.81 g, 3.78 mmol, 47%) was obtained as a colorless liquid. 1H NMR (500.23 MHz, C6D6): δ 7.38
(t, 3JH,H = 7.8 Hz, 4H; o-ArH), 6.89 (d, 3JH,H = 7.5 Hz, 4H; m-ArH), 5.28 (d, 1JH,P = 214.1
Hz, 1H; PH), 2.02 (s, 6H; CH3). 13C{1H} NMR (125.78 MHz, CDCl3): δ 138.4 (s; p-ArC), 134.5 (d, 2JC,P = 17.3
Hz; o-ArC), 132.2 (d, 2JC,P = 9.1 Hz; ipso-ArC), 129.7 (d, 3JC,P = 6.4 Hz; m-ArC), 21.1 (s; CH3). 31P NMR (161.98 MHz, C6D6): δ −42.6 (dq, 1JP,H = 215.4 Hz, 3JP,H = 3.2 Hz).
((N-Methyl)arylimidoyl)diarylphosphanes 3a–f
Protocol 3a–d
pan class="Chemical">Diarylphosphane
(1.0 equiv) was added dropwise to a solution of 2a (1.1
equiv) in pan class="Disease">DCM (4.0 mL/mmol) at −78 °C to give a bright
yellow solution, which was warmed to room temperature and stirred
for 15 min. Triethylamine (1.1 equiv) was added to give a yellow solution,
which was stirred for 1 h. Volatiles were removed in vacuo to give
a yellow oil, which was extracted into Et2O (3 × 4
mL/mmol). The extract was concentrated to saturation and filtered
over neutral alumina. Evaporation provided 3 as an off-white
solid (mixture of E and Z isomers; 3a 91%, 3b 75%, 3c 85%, 3d 45%*) (*for crystallization, a saturated Et2O solution
was cooled to −80 °C).
3 (0.02 g, 1.00 equiv) was dissolved in 0.6 mL of
acetone in an NMR tube. Next, a solution of H2O in acetone
(10%) was added. The mixture was vigorously shaken and subsequently
kept at room temperature, during which the reaction mixture was monitored
using 31P NMR spectroscopy. Using this methodology, systematically
H2O was added to the mixture, which was kept at room temperature
during the described time span. 3a: (1) 0.50 equiv, 21
h; (2) 0.50 equiv, 68.5 h; (3) 1.00 equiv, 29 h; (4) 2.00 equiv, 26.5
h; (5) 4.00 equiv, 18.5 h; (6) 8.03 equiv, 24.5 h. (7) 481.7 equiv,
22.5 h. Observed hydrolysis (% 3a:% PhPH, time of measurement): (1) 97:0, 21 h; (2) 95:0, 89 h; (3)
91:0, 118 h; (4) 89:0, 144 h; (5) 85:1, 163 h; (6) 85:1, 187 h; (7)
62:4, 210 h. 3b: (1) 0.50 equiv, 18 h; 92) 0.50 equiv,
49 h; (3) 3.00 equiv, 24 h; (4) 12.1 equiv, 25 h; (5) 461.67 equiv,
16 days. Observed hydrolysis (% 3b:% Ph2PH,
time of measurement): (1) 95:0, 18 h; 92) 91:0, 65 h; 93) 88:1, 91
h; (4) 84:2, 115 h; (5) 36:6, 524 h. 3c: (1) 0.46 equiv,
47.5 h; 92) 2.25 equiv, 24 h; (3) 16.9 equiv, 23.5 h; (4) 563.34 equiv,
72 h). Observed hydrolysis (% 3c:% Ph2PH,
time of measurement): (1) 95:0, 41 h; (2) 94:0, 69 h; (3) 92:1, 91
h; (4) 72:3, 165 h.
Acid Stability Study of 3a
3a (0.0203 g, 0.066 mmol, 1.00 equiv) was dissolved
in 0.4 mL of pan class="Disease">DCM
to give a yellowish solution. Addition of a solution of pan class="Chemical">TfOH in DCM
(0.23 mL of a 0.28 M solution, 0.064 mmol, 0.98 equiv) provided a
bright yellow solution, which was vigorously shaken and kept at room
temperature for 20 h, during which the reaction was monitored using 31P NMR spectroscopy. Next, again TfOH in DCM (0.23 mL of a
0.28 M solution, 0.064 mmol, 0.98 equiv) was added. The mixture was
vigorously shaken and kept at room temperature for 24 h, during which
the reaction was monitored using 31P NMR spectroscopy.
Excess triethylamine (0.20 mL, 1.43 mmol, 21.74 equiv) was added to
the mixture at 0 °C, resulting in a light yellow solution of
pure 3a.
Acid-Catalyzed Hydrolysis Study of 3a
3a (0.0267 g, 0.088 mmol, 1.00 equiv)
was dissolved
in 0.6 mL of pan class="Disease">DCM to give a yellowish solution. Addition of a solution
of pan class="Chemical">TfOH in DCM (31 μL of a 0.28 M solution, 0.009 mmol, 0.10
equiv) provided a more intensely yellow solution, which was vigorously
shaken and kept at room temperature for 1 h. Subsequently, H2O (0.8 μL, 0.044 mmol, 0.50 equiv) was added to the mixture,
which then was vigorously shaken and kept at room temperature for
2 days, during which the reaction was monitored using 31P NMR spectroscopy. Next, excess H2O (5.6 μL, 0.310
mmol, 3.52 equiv) was added to the mixture. After vigorous shaking,
the mixture was kept at room temperature for 24 days. Observed hydrolysis
(% 3a:% Ph2PH, time of measurement): (1) 86:1,
17 h; (2) 76:3, 60 h; (3) 74:3, 108 h; (4) 68:5, 248 h; (5) 58:9,
603 h; (6) 49:12, 1251 h.
Oxidation Study of 3a
Under an N2 atmosphere, 3a (0.012 g, 0.04
mmol, 1.00 equiv) was
dissolved in 0.6 mL of acetone. Stepwise, O2 (21% in air)
was bubbled manually through the mixture with a rate of approximately
6 bubbles per second using a needle with an inner diameter of 1.194
mm. During additions, the mixture was kept in a water bath at room
temperature to inhibit acetone evaporation. After each addition, the
NMR tube was sealed using a screw cap and the reaction mixture was
monitored using 31P NMR spectroscopy. Using this methodology,
systematically O2 (21% in air) was added to the mixture
at room temperature over the described intervals: (1) O2 (21% in air, 1 mL, 8.73 μmol, 0.22 equiv), 30 s; (2) O2 (21% in air, 1 mL, 8.73 μmol, 0.22 equiv), 30 s; (3)
O2 (21% in air, 2 mL, 17.47 μmol, 0.44 equiv), 60
s; (4) O2 (21% in air, 2 mL, 17.47 μmol, 0.44 equiv),
60 s; (5) O2 (21% in air, 4 mL, 34.94 μmol, 0.87
equiv), 2 min; (6) O2 (21% in air, 4 mL, 34.94 μmol,
0.87 equiv), 2 min; (7) O2 (21% in air, 8 mL, 69.88 μmol,
1.75 equiv), 4 min; (8) O2 (21% in air, 8 mL, 69.88 μmol,
1.75 equiv), 4 min; (9) O2 (21% in air, 16 mL, 0.140 mmol,
3.49 equiv), 8 min; (10) O2 (21% in air, 16 mL, 0.140 mmol,
3.49 equiv), 8 min. From addition 7 onward, a white solid started
to precipitate. After the additions, the mixture was fully oxidized
by dissolving the mixture in 10 mL of acetone (technical grade) and
stirring the resulting solution under a closed air atmosphere for
42 h at room temperature. Subsequently, the reaction vessel was opened
to the external air atmosphere and stirred for an additional 173.5
h. The resulting solution was evaporated to provide a white oil. Observed
oxidation (% 3a, time of measurement): (1) 95, 17 min;
(2) 86, 24 min; (3) 82, 30 min; (4) 74, 37 min; (5) 68, 46 min.; (6)
61, 55 min; (7) 56, 66 min; (8) 52, 77 min; (9) 49, 91 min; (10) 44,
107 min; (11) 4, 46 h; (12) 0, 220 h.
Under an argon atmosphere,
to a solution of 3a (92 mg 0.30 mmol, 2.0 equiv) in 15
mL of DCM was added [RhCp*Cl2]2 (94 mg 0.15
mmol, 1.0 equiv) to provide a red solution, which was stirred for
30 min at room temperature. Next, AgOTf (317 mg, 1.23 mmol, 8.2 equiv)
was added and the resulting suspension was stirred for 60 min at room
temperature in the absence of light, during which the mixture turned
yellow. Filtration provided an orange-red solution, which was evaporated
to provide an orange-red solid. After washing with 20 mL of pentane, 7a was obtained as a yellow solid (306 mg, 0.21 mmol, 71%).
For crystallization, slow diffusion of pentane (55 mL/mmol of compound)
into a DCM solution (45 mL/mmol of compound) at room temperature provided
red blocks. Mp: 178.6 °C. 1H NMR (500.23 MHz, CDCl3): δ 7.76–7.63 (m, 6H; P-o,p-ArH), 7.49–7.46 (m, 5H; P-m-ArH, C-p-ArH), 7.40 (t, 3JH,H = 7.7 Hz,
2H; C-m-ArH), 7.33 (d, 3JH,H = 7.6 Hz, 2H; C-o-ArH), 4.01 (d, 3JH,Rh = 4.4 Hz, 3H; CN-CH3), 1.72
(d, 3JH,Rh = 5.0 Hz, 15H; Cp*-CCH3). 13C{1H} NMR (125.78
MHz, CDCl3): δ 193.9 (dd, 1JC,P = 48.1 Hz, 2JC,Rh = 3.6 Hz; CN-CH3), 136.2 (dd, 3JC,P = 11.8 Hz, 4JC,Rh = 3.6 Hz; P-m-ArC), 134.4 (d, 4JC,P = 8.2 Hz;
P-p-ArC), 133.6 (d, 4JC,P = 7.3 Hz; P-p-ArC), 133.3 (d, 2JC,P = 10.9 Hz; P-o-ArC), 132.3 (s;
C-p-ArC), 132.1 (d, 2JC,P = 6.4 Hz; C-ipso-ArC), 130.8 (d, 2JC,P = 8.2 Hz; P-o-ArC), 129.7
(d, 3JC,P = 10.7 Hz; P-m-ArC), 129.3 (s; C-m-ArC), 127.7 (d, 3JC,P = 1.8 Hz; C-o-ArC), 100.8 (dd, 1JC,Rh = 8.2 Hz, 2JC,P = 2.7 Hz; Cp*-CCH3), 46.1 (d, 2JC,Rh = 17.3
Hz; CN-CH3), 9.8 (s; Cp*-CCH3), signals for P-ipso-ArC are unresolved. 19F NMR (235.36 MHz, CDCl3): δ −78.2 (d, 4JF,Rh = 12.1 Hz; Rh-OS(O2)CF3). 31P NMR (161.98 MHz, CDCl3): δ −9.9
(d, 1JP,Rh = 111.2 Hz). FT-IR
(cm–1): ν 3065 (w), 2995 (w), 1580 (w), 1483
(w), 1437 (m), 1379 (w), 1304 (s), 1261 (s), 1229 (s), 1202 (s), 1159
(s), 1144 (s), 1099 (m), 1082 (m), 1030 (s), 1005 (s), 951 (m), 926
(w), 771 (m), 754 (m), 733 (m), 689 (s), 635 (s), 573 (m), 557 (m),
536 (w), 515 (s), 476 (s), 463 (m), 442 (m). HR-MS (ESI-Q-TOF): calcd
for C31H33F3NO3PRhS+ 690.0920; found 690.0915.
[(1,3-P,N)RuII]-Catalyzed Hydrations of Benzonitrile
pan class="Chemical">[Ru(Ar)Cl2]2 (Ar = p-cym, C6Me6, C6H5Me, Cp*; 0.025 mmol, 5 mol %)
and ligand 3a or Ph2PPy (0.05 mmol, 5 mol
%) were dissolved in DME (0.5 mL) under
an argon atmosphere and stirred for 30 min. Benzonitrile (105 μL,
1.02 mmol, 1.0 equiv) and H2O (36 μL, 1.99 mmol,
2.0 equiv) were added and the sealed vessel was stirred at 180 °C
for 3 h.
pan class="Chemical">[Ru(p-cym)Cl2]2 (16 mg, 0.026 mmol, 5 mol % [pan class="Chemical">Ru]) and ligand 3a or Ph2PPy (0.052 mmol, 5 mol %) were dissolved
in benzonitrile (105 μL, 1.02 mmol, 1.0 equiv) under an argon
atmosphere and stirred for 30 min. H2O (3.0 mL, 166 mmol,
163 equiv) was added and the sealed vessel was stirred at 100 °C
for 24 h.
Protocol 3
pan class="Chemical">[Ru(p-cym)Cl2]2 (15 mg, 0.024 mmol, 1.4 mol % of pan class="Chemical">Ru) and ligand (0.050
mmol, 1.4 mol %) were dissolved in benzonitrile (370 μL, 3.58
mmol, 1.0 equiv) under an argon atmosphere and stirred for 30 min.
H2O (130 μL, 7.2 mmol, 2.0 equiv) was added and the
sealed vessel was stirred at 180 °C for 3 h.
Analysis
After they were cooled to room temperature,
the mixtures were extrapan class="Chemical">cted with i-PrOH under atmosphericconditions and analyzed by GC (internal standard naphthalene).
Authors: Adrien P Côté; Michael J Ferguson; Kashif A Khan; Gary D Enright; Angela D Kulynych; Sean A Dalrymple; George K H Shimizu Journal: Inorg Chem Date: 2002-01-28 Impact factor: 5.165
Authors: Andreas Decken; Carsten Knapp; Grigori B Nikiforov; Jack Passmore; J Mikko Rautiainen; Xinping Wang; Xiaoqing Zeng Journal: Chemistry Date: 2009-06-22 Impact factor: 5.236
Authors: Tom van Dijk; Sebastian Burck; Mark K Rong; Amos J Rosenthal; Martin Nieger; J Chris Slootweg; Koop Lammertsma Journal: Angew Chem Int Ed Engl Date: 2014-06-27 Impact factor: 15.336