4,6-Bis(supermesitylphosphanylidenemethyl)dibenzofuran. Synthesis, X-ray structure and reactivity towards group 11 metals

Abstract

New kinetically stabilized mono- and bis-phosphaalkene ligands (2 and 3, respectively) were synthesized via the phospha-Wittig approach. Ligand 3 was characterized by X-ray diffraction. The coordinating behaviour of the bidentate ligand was investigated towards group 11 metal centers in order to test its capacity to bind two coordination sites located in a trans-fashion. The [Au(3)][BF₄], [Ag(3)(H₂O)][BF₄] and [Cu(3)(CH₃CN)][BF₄] complexes 4, 5 and 6, respectively, were characterized by X-ray diffraction. The peculiar geometries of 4 and 6 were rationalized by means of DFT calculations.

---

Introduction

Low-coordinate phosphorus molecules incorporating P=C double bonds are slowly emerging as a very important class of ligands with unique electronic properties that markedly differ from that of classical tertiary phosphines and sp²-hybridized nitrogen ligands.¹ Over the last decade, a great deal of effort has been devoted to the understanding both of their electronic properties and their use in coordination chemistry of low valent and reduced transition metals. Thus, it was shown that ligands such as phosphinines and their functional derivatives can act as efficient ligands for the stabilization of various electron rich or electron excessive metal fragments.² Much more recently, due to their strong π-accepting capacity, these ligands were also employed in the stabilization of gold nanoparticles presenting unique optical properties.³ On the other hand, some low-coordinate phosphorus ligands have found promising applications in homogeneous catalysis. Thus, phosphaferrocenes⁴ and kinetically stabilized phosphaalkenes⁵ were successfully employed as ligands in some catalytic processes of synthetic relevance such as C–X bond forming reactions, ethylene polymerization, nucleophilic allylic substitutions and hydrogenations. Concerning phosphalkenes, much effort have focused on the use of phosphorus equivalents of 1,4-diazabutadienes, namely 1,4-diphosphabutadienes such as 1,2-diaryl-3,4-diphosphinidene cyclobutene (DCPB) developed by the group of Yoshifuji.⁵ Though some studies were devoted to the synthesis of tridentate pincer ligands featuring two phosphaalkene moieties as pendant ligands (Mes* = 2,4,6-tris(tert-butyl)phenyl) and a carbon⁶ or nitrogen atom as central binding site (Scheme 1),⁷ little information is available on the coordination chemistry of these ligands and the catalytic activity of their corresponding complexes.

Scheme 1

As part of a program aimed at developing the use of such polydentate ligands featuring sp²-hybridized phosphorus atoms in coordination chemistry and catalysis,⁸ we recently investigated the synthesis of bidentate structures featuring the dibenzofuran skeleton. Indeed, 4,6-bis(carbaldehyde)dibenzofuran proved to be a convenient precursor for the elaboration of bidentate ligands featuring imines⁹ or oxazolines (DBFOX ligands) as pendant ligands which can bind a metal centre in a trans-fashion. Some DBFOX-based complexes proved to be particularly efficient in some catalytic transformations.¹⁰ Herein, we wish to report on these syntheses as well as on the coordinating behaviour of this new type of ligands towards group 11 metals.

Results and discussion

Syntheses of ligands 2 and 3

Various methods are available for the synthesis of P=C double bonded systems. One of the most attractive approaches, which was initially developed in our laboratories for the synthesis of phosphaalkene complexes of metal carbonyls (M = Cr, Mo, W, Fe), is the well known phospha-Wittig reaction, a transformation that mimics the ubiquitous Wittig process.¹¹ This reaction which relies on the reactivity of phosphoranylidenephosphine complexes towards aldehydes was further extended to the synthesis of uncomplexed phosphaalkenes by Protasiewicz et al.¹² In practice, free phosphoranylidenephosphine reagents (phospha-Wittig) reagents are available through the reaction of PMe₃ with dichlorophosphines in the presence of zinc as reducing agent.¹² Among all methods available this approach offers several decisive advantages: its simplicity (a one-step process), the use of mild conditions and cheap reagents (Scheme 2).

Scheme 2 The phospha-Wittig approach.

As a prerequisite to these syntheses, 4,6-bis(carbaldehyde)dibenzofuran was synthesized according to a well-known reported procedure that involves the dimetallation of dibenzofuran with n-butyllithium followed by a trapping reaction with dimethylformamide.¹³ The phospha-Wittig reagent 1 was conventionally prepared by reacting Mes*PCl₂ with PMe₃ (2.6 equiv.) in the presence of Zn as reductant in THF at 0 °C. Reaction of 1.1 equivalent of 1 with 4,6-bis(carbaldehyde)dibenzofuran readily occurred at room temperature to cleanly yield phosphaalkene 2. After 3 h of reaction, analysis of the crude mixture by ³¹P NMR spectroscopy revealed the presence of a very characteristic downfield shifted signal at 273.9 ppm. After extraction with dichloromethane, washings with water and crystallization from MeOH, compound 2 was isolated in a 60% yield as a very air stable pale yellow powder. The E-stereochemistry of 2 was confirmed by the analysis of the ¹H NMR spectrum which exhibits a doublet at 8.58 ppm (²J(H–P) = 25.3 Hz). No trace of the Z-isomer was observed in the crude mixture (Scheme 3). The formulation proposed for 2 was confirmed by NMR and mass spectrometry as well as by elemental analyses.

Scheme 3 Reagents and conditions: (i) THF, RT, 3 h.

The bis-phosphaalkene derivative 3 was prepared following a similar strategy by using the phospha-Wittig reagent 1 in excess (2.2 equivalents). A prolonged reaction time (15 h) was needed to obtain a complete conversion. Importantly, ³¹P NMR spectroscopy revealed that only one stereoisomer is formed. Compound 3 was isolated following a similar workup in a 50% yield as a very stable pale yellow solid. The structure of 3 was easily established on the basis of ¹H, ¹³C NMR data, mass spectrometry and elemental analyses. As in 2, the presence of a P=C double bond was evidenced by the presence of a downfield signal in ³¹P NMR (δ(THF) = 273 ppm) (Scheme 4). The presence of two phosphaalkene moieties is proven first by ¹H NMR. The signal of the phosphaalkene appears as a deceptively simple doublet of doublet (AA′XX′ spin system) at 8.4 ppm (∑J(H–P) = 42 Hz, ∑ means sum). In the ¹³C NMR spectrum, the corresponding signal appears as the expected second-order signal at 167.1 ppm (AXX′ spin system, ∑J(C–P) = 51.7 Hz).

Scheme 4 Reagents and conditions: (i) THF, RT, 15 h.

Single crystals of compound 3 were obtained by crystallization in MeOH at room temperature. An X-ray crystal structure analysis was carried out and a view of one molecule of 3 is presented in Fig. 1 as well as the most relevant bond distances and bond angles. Crystal data and structural refinement details are presented in Table 1. The compound has crystallographically imposed twofold symmetry. As can be seen the E-stereochemistry of both phosphaalkene moieties is confirmed. Interestingly, one can note that two P=C bonds are coplanar with the dibenzofuran skeleton probably indicating that a weak delocalization occurs between the aromatic part of the molecules and the two P=C units. Besides, this is confirmed by the short C–C connection between the C9 carbon atoms and the C13 carbon atoms of the phosphaalkenes (d(C(9)–C(13)) = 1.454(2) Å). Apart from this, the structure of 3 deserves no special comments.

Table 1 Crystal and refinement parameters for 3 and 4

	3	4
Formula	C50H66OP2	C50H66AuBF4OP2·C4H10O
M r	744.97	1102.86
Crystal system	Monoclinic	Orthorhombic
Space group	C2	P212121
a/Å	15.966(1)	11.647(1)
b/Å	10.407(1)	16.895(1)
c/Å	14.099(1)	27.329(1)
β/°	105.260(1)	90.00
Z	2	4
µ/cm^−1	0.130	2.846
Reflections measured	7551	27925
Reflections used	5376	13250
wR 2	0.0924	0.0701
R 1	0.0349	0.0340

---

Fig. 1 ORTEP view of one molecule of 3 (50% ellipsoids). The numbering is arbitrary and different from that used in NMR data. Atoms marked with a prime (′) are at equivalent position (−1 − x, y, −1 − z). Selected bond lengths (Å) and bond angles (°): C(13)–P(1) 1.674(3), C(3)–P(1) 1.852(1), C(13)–C(9) 1.454(2), C(9)–C(5) 1.405(2), C(5)–O(2) 1.379(1); C(3)–P(1)–C(13) 98.72(1), P(1)–C(13)–C(9) 128.5(1).

Reactivity of 3 towards group 11 metals

In order to test the capacity of 3 to bind metals through a trans fashion, we focused our study on the coordination of group 11 cationic metal fragments which are known to form linear 14 VE ML₂ complexes. The gold complex 4 was readily prepared in one pot via a two-step sequence (Scheme 5).

Scheme 5 Reagents and conditions: THF, RT, 3 h.

The stoichiometric reaction of [AuCl·SMe₂] and ligand 3 in THF led to a pale yellow solution. A ³¹P NMR of this mixture showed a very broad peak which revealed a weak interaction. Addition of solid AgBF₄ resulted in the very fast formation of characteristic silver salt indicating the end of chloride abstraction. ³¹P NMR spectroscopy showed the formation of a single complex characterized by a singlet at 201 ppm (δ(3) = 273 ppm, Δδ = −72 ppm). The mixture was filtered through a pad of Celite and concentrated to give the desired complex 4 in near quantitative yield. The complex was fully characterized by usual NMR techniques as well as elemental analyses. Compared to the spectra of the starting ligand 3, complex 4 varies only marginally. In particular, the phosphaalkene proton is upfield shifted to 8.2 ppm and appears as a virtual triplet (simplified AA′XX′ spin system), because the magnitude of the J(P–P) increases upon coordination. In the ¹³C spectrum the two carbon atoms which are bound to the phosphorus atom are upfield shifted by ca. 10 ppm upon coordination. Single crystals of 4 were grown by a slow diffusion of diethyl ether in a THF solution of the complex. A view of one molecule of 4 is presented in Fig. 2 as well as the most relevant bond distances and bond angles. Crystal data and structural refinement details are presented in Table 1.

Fig. 2 ORTEP view of one molecule of 4 (50% ellipsoids). The numbering is arbitrary and different from that used in NMR data. BF₄ counter ion and methyl groups of the tert-butyl substituents have been omitted for clarity. Selected bond lengths (Å) and bond angles (°): Au(1)–P(1) 2.267(1), Au(1)–P(2) 2.267(1), Au(1)⋯O(1) 2.699, C(1)–P(1) 1.669(3), C(14)–P(2) 1.663(3), C(15)–P(1) 1.804(3), C(33)–P(2) 1.807(3), C(1)–C(2) 1.449(4), C(13)–C(14) 1.443(4), C(8)–O(1) 1.396(4), C(7)–O(1) 1.383(4); P(1)–Au(1)–P(2) 167.56(3), C(1)–P(1)–Au(1) 121.8(1), C(14)–P(2)–Au(1) 120.3(1), C(1)–P(1)–C(15) 112.3(2), C(15)–P(1)–Au(1) 125.9(1), C(13)–C(14)–P(2) 134.3(3), C(2)–C(1)–P(1) 133.0(3).

The only apparent feature of this complex is the significant deviation from linearity observed for the P–Au–P angle (167.56°). Moreover, this deviation brings the gold center away from the oxygen atom of the furan moiety. As will be seen further, this can be easily rationalized by considering a simple molecular orbital diagram.

We next investigated the coordination to silver, which was readily achieved by the stoichiometric reaction of ligand 3 with AgBF₄ (Scheme 6). After a few minutes, the ³¹P NMR of the crude mixture proves the formation of a single complex at 204 ppm. Unlike many Ag⁺ complexes, coupling of the two phosphorus atoms with both isotopes of Ag, namely Ag¹⁰⁷ and Ag¹⁰⁹ which both possess a spin of 1/2, was observed resulting in two sets of doublets (¹J(Ag¹⁰⁹–P) = 686 Hz and ¹J(Ag¹⁰⁷–P) = 596 Hz). After isolation, complex 5 was obtained as an air-stable solid. It was fully characterized by usual NMR techniques. In fact, a first surprising piece of information was given by a broad peak at 2.5 ppm in the ¹H NMR spectrum. This peak integrated for two protons which suggested the coordination of a water molecule. We verified that the same complex is formed when the reaction is carried out in air. Therefore, it seems likely that the water-free complex adopts a distorted linear geometry like the gold analogue described above. The silver centre then picks up water from the solvent to complete the coordination sphere yielding the observed complex which possesses a trigonal planar geometry. The presence of a water molecule was confirmed by elemental analyses. In the ¹H NMR spectrum also, the phosphaalkene proton appears as the expected signal for an AA′XX′ spin system (δ = 8.23 ppm, ∑J = 40 Hz). In fact, the silver analogue gives very similar spectra than the gold complex.

Scheme 6 Reagents and conditions: CH₂Cl₂, air (traces of water), RT, 5 min.

Single crystals were obtained by a slow diffusion of hexanes into a CH₂Cl₂ solution of the complex, in air. We were both quite surprised and pleased to see in the structure (Fig. 3) that the coordination sphere of the metal had been completed by water during crystallisation. Indeed, this definitely proves that, unlike classical phosphaalkenes, the bulky derivatives are very robust toward hydrolysis, even once coordinated to a metal centre.

Fig. 3 ORTEP view of one molecule of 5 (50% ellipsoids). The numbering is arbitrary and different from that used in NMR data. BF₄ counter ion and methyl groups of the tert-butyl substituents have been omitted for clarity. Selected bond lengths (Å) and bond angles (°): Ag(1)–P(1) 2.388(1), Ag(1)–P(2) 2.403(1), Ag(1)–O(2) 2.316(3), Ag(1)⋯O(1) 3.095, C(14)–P(2) 1.681(4), C(13)–P(1) 1.672(5), C(2)–C(13) 1.456(6), C(14)–C(11) 1.441(7), C(1)–O(1) 1.383(6), C(12)–O(1) 1.387(6), C(33)–P(2) 1.840(5), C(15)–P(1) 1.825(5); P(1)–Ag(1)–P(2) 145.14(4), P(1)–Ag(1)–O(2) 109.8(1), P(2)–Ag(1)–O(2) 103.3(1), C(14)–P(2)–Ag(1) 131.1(2), C(13)–P(1)–Ag(1) 130.6(2), C(11)–C(14)–P(2) 132.6(4), C(2)–C(13)–P(1) 132.1(4), C(13)–P(1)–C(15) 102.5(2).

The P–Ag bond lengths Ag(1)–P(1) 2.388(1) Å, Ag(1)–P(2) 2.403(1) Å) are significantly longer than the P–Au bond lengths (Au(1)–P(1) 2.267(1) Å, Au(1)–P(2) 2.267(1) Å) as expected. The geometry at silver is trigonal planar (∑angles = 358.24°) with a wide P(1)–Ag(1)–P(2) angle at 145.14(4)°. This increase in the angle (compared to the gold complex) is accompanied by an increase in the C=P–M angle from ca. 120 to ca. 130° (C(13)–P(1)–Ag(1) 131.1(2)). The fact that phosphaalkene ligands can accommodate readily such changes in coordination angles reflects the diffuse nature of the lone pair of sp² hybridized phosphorus.

To complete this investigation we examined the coordinating behaviour of 3 toward the catalytically relevant Cu(I) centre. Acetonitrile ligand was readily displaced from the [Cu(CH₃CN)₄][BF₄] precursor to form a single complex 6 (Scheme 7). It appears as a singlet at 207 ppm in the ³¹P NMR spectrum. This complex was fully characterized by usual NMR techniques as well as elemental analysis.

Scheme 7 Reagents and conditions: CH₂Cl₂, RT, 5 min.

Single crystals of complex 6 were obtained by slow diffusion of hexanes in a CDCl₃ solution of complex 6. A view of one molecule of 6 is presented in Fig. 4 that also lists the most significant bond lengths and angles. Crystal data and structural refinement details are presented in Table 2.

Table 2 Crystal and refinement parameters for 5 and 6

	5	6
Formula	C50H66AgBF4O2P2·1/2CH2Cl2	C52H69BCuF4NOP2·2CHCl3
M r	1040.57	1175.11
Crystal system	Monoclinic	Triclinic
Space group	C2/c	P1̄
a/Å	35.058(3)	11.931(1)
b/Å	11.084(1)	14.527(1)
c/Å	30.013(3)	19.169(1)
α/°	90.00	107.920(1)
β/°	116.700(2)	92.030(1)
γ/°	90.00	109.710(1)
Z	8	2
µ/cm^−1	0.602	0.748
Reflections measured	17591	16842
Reflections used	7576	13870
wR 2	0.2282	0.1806
R 1	0.0724	0.0628

---

Fig. 4 ORTEP view of one molecule of 6 (50% ellipsoids). The numbering is arbitrary and different from that used in NMR data. BF₄ counter ion and methyl groups of the tert-butyl substituents have been omitted for clarity. Selected bond lengths (Å) and bond angles (°): Cu(1)–P(1) 2.221(1), Cu(1)–P(2) 2.230(1), Cu(1)–N(1) 1.958(4), Cu(1)⋯O(1) 2.456, C(1)–P(1) 1.654(5), C(14)–P(2) 1.671(4), C(1)–C(2) 1.456(6), C(13)–C(14) 1.453(6), C(1)–O(1) 1.393(4), C(8)–O(1) 1.389(4), C(17)–P(1) 1.834(4), C(35)–P(2) 1.827(4); P(1)–Cu(1)–P(2) 132.60(4), P(1)–Cu(1)–N(1) 113.5(1), P(2)–Cu(1)–N(1) 113.7(1), C(1)–P(1)–Cu(1) 116.5(2), C(14)–P(2)–Cu(1) 120.6(2), C(2)–C(1)–P(1) 130.7(3), C(13)–C(14)–P(2) 127.3(3), Cu(1)–P(1)–C(17) 101.6(2).

As can be seen, the geometry at the copper center is trigonal planar (∑angles = 359.8°) with a wide P(1)–Cu(1)–P(2) angle at 132.60(4)°. The P–Cu bond distances are normal at 2.221(1) and 2.230(1) Å. The other bond distances and angles are quite similar than the above mentioned ones, and therefore will not be further commented on. This complex possesses however a very significant feature, namely the metal centre is located well above the plane of the ligand, as clearly shown on the ORTEP plot (dihedral angle C2–C1–P1–Cu1 = 9°). This geometry results in a bond distance of 2.456 Å, more consistent with a weak interaction than a true bond. Indeed, typical Cu–O bond distances of 2.208 Å are found with neutral ligands such as carbaldehyde–pyridine coordination to the [Cu(PPh₃)₂]⁺ fragment.¹⁴ For several carboxylate complexes of the type [Cu(PPh₃)₂(O₂CR)], Hart et al. have shown that the Cu–O=C linkage can vary from a true bond for R = CH₃ (2.257(7) Å) to a weak interaction for R = CHF₂ (2.465(6) Å) or R = CF₃ (2.545(5) Å).¹⁵ In fact, for these two last cases, addition of ethanol resulted in the breaking of this carboxylate interaction and ethanol coordination. How can the geometry at the copper centre of our complex be rationalized? Being in sharp contrast to the other geometries of the Au and Ag complexes, this fact can not be fortuitous. Like in the case of 4 and 5, this particular feature can be easily rationalized first through a qualitative molecular orbital diagram, and more quantitative DFT calculations.

Theoretical calculations

As previously noted, both Au and Ag complexes adopt a non linear coordination mode. Looking into the simple molecular orbital diagram of linear ML₂ complexes provided a lead toward the understanding of this peculiar geometry. Indeed, as in d¹⁰ ML₂ complexes all five d orbitals are filled, four-electron destabilizing interactions can occur between the d_x²−y² and d_xy orbitals at the metal and the 2p_x (lone pair) and 2p_y (π-system) orbitals at the oxygen atom respectively for obvious symmetry reasons (Scheme 8). Note that the d_z² also possesses the appropriate symmetry to interact with the 2p_y AO at oxygen. Thus in order to diminish these repulsive interactions, a bending of the P–Au–P is observed.

Scheme 8 Qualitative interaction diagram showing the four-electron destabilizing interactions occurring between the filled d-orbitals at the metal and 2p_x and 2p_y orbitals at the oxygen atom.

Theoretical calculations were carried out on a model complex I in which the Mes* groups were replaced by xylyl groups. A view of the optimized structure is presented in Fig. 5. Though the substitution scheme of the phosphorus atom is different, a reasonable agreement was found between theoretical and experimental data (Fig. 5). Thus, the P–Au bonds falls at 2.311 Å for 2.267(1) Å in 4 and the P–Au–P bending is relatively well reproduced (165.83° for 167.56(3)° in 4). Examination of MOs allowed to check our hypothesis and as can be seen in the following figure several MOs account for this four-electron destabilizing interaction (see Fig. 6). The H-11 MO and H-13 correspond to the antibonding interaction between the d_x²−y² and the d_z² with the 2p_y AO at oxygen, respectively, whereas the H-12 is antibonding between the d_xy at Au and the 2p_x at O.

Fig. 5 Optimized geometry of the model complex I. Mes* groups at phosphorus have been replaced by xylyl groups. Selected bond lengths (Å) and bond angles (°): Au(44)–P(25) 2.311, P(25)–C(22) 1.682, C(22)–C(3) 1.445, C(3)–C(4) 1.409,C(4)–O(19) 1.371; P(24)–Au(44)–P(25) 165.826, Au(44)–P(25)–C(22) 121.050, P(25)–C(22)–C(3) 134.597, C(22)–C(3)–C(4) 128.947, C(3)–C(4)–O(19) 126.254.

Fig. 6 H-11, H-12 and H-13 MO of complex I.

Having rationalized the geometry of the gold and silver complexes, we then turned our attention to the copper complex. The particular structural arrangement in complex 6 can be explained as follows. In d¹⁰ ML₃ 16 VE complexes, the five d orbitals are also filled and but a vacant 3p_y orbital which is perpendicular to the plane defined by the three ligands is found as the LUMO. In complex 6, this orbital points towards the lone pair at the oxygen atom. Note that the d_y²−z² orbital which is involved in the bonding with the two phosphorus atoms has the right symmetry to interact with the oxygen lone pair (Scheme 9). Therefore the bonding between Cu and O can be described as a classical four-electron interaction involving three orbitals.

Scheme 9 Interaction diagram showing the two-electron stabilizing interaction occurring between the vacant 3p_y-orbital at the metal and 2p_y orbital at the oxygen atom.

The theoretical structure of complex 6 was calculated. A first calculation carried out on the model complex featuring xylyl groups at phosphorus yielded a structure in which the O–Cu bond distance was found to be by far too long (2.638 Å in IIvs. 2.456(3) Å in 6). Therefore, supposing that in this case, the presence of the two very bulky tert-butyl groups at the ortho positions would probably tend to pucker the copper metal towards the oxygen atom of the furan ligand, calculations were carried out on a complex featuring these two t-Bu groups. The optimization was carried out using the ONIOM method (B3PW91/UFF). A view of the optimized structure is presented in Fig. 7. As can be seen upon examining the theoretical data, there is a very good agreement with experimental metric parameters. Importantly, the Cu–O bond distance at 2.486 Å is rather well reproduced as well as the P–Cu–P angle (132.00 in IIvs. 132.60° in 6).

Fig. 7 Optimized geometry of the model complex II. Selected bond lengths (Å) and bond angles (°): Cu(26)–P(25) 2.288, P(25)–C(22) 1.683, C(22)–C(3) 1.448, C(3)–C(4) 1.406, C(4)–O(19) 1.380, Cu(26)–N(27) 2.054; P(24)–Cu(26)–P(25) 132.004, Cu(26)–P(25)–C(22) 118.546, P(25)–C(22)–C(3) 129.602, C(22)–C(3)–C(4) 127.453, C(3)–C(4)–O(19) 125.805, P(24)–Cu(26)–N(27) 115.341, P(25)–Cu(26)–N(27) 112.340, P(24)–Cu(26)–P(25) 132.003.

As expected, molecular orbital analysis revealed that a bonding interaction develops between the oxygen and the copper atom. As can bee seen in Fig. 8 the most important contribution to the bonding is provided by the interaction between the vacant 3p_y and 3d_y²−z² AOs at Cu and the 2p_y AO of the oxygen atom (HOMO-19).

Fig. 8 A view of one of the MO describing the Cu–O bond in the model complex II. The methyl groups of the t-Bu substituents at supermesityl groups have been omitted for clarity.

Conclusion

In conclusion, we have synthesized a new trans spanning bidentate bis phosphaalkene ligand 3 in a one pot procedure from the readily available 4,6-bis(carbaldehyde)dibenzofuran. Using this same starting material, the mono-phosphaalkene mono-aldehyde 2 could also be synthesized, which opens the way to mixed ligands bearing one phosphaalkene moiety. This area is being pursued in our laboratories. Using the bidentate ligand 3, the series of complexes of group 11 metals has been synthesized. In the gold and silver complexes 4 and 5, respectively, a distortion form the linearity of the initial ML₂ fragment was observed, which in the case of complex 5 resulted in the coordination of a water molecule. This complex proves the robustness of the phosphaalkene ligands towards hydrolysis. The observed distortion, which has been rationalized by DFT calculations, results from a destabilizing interaction between the appropriate filled d_x²−y² and d_z² with the 2p_y AO at oxygen, and d_xy at Au with the 2p_x at oxygen. The catalytically relevant Cu(I) complex 6 was also synthesized. The geometry given by an X-ray structure analysis was rationalized by ONIOM calculations. In this case, a bonding interaction between, on one hand, empty 3p_y-orbital and filled 3d_y²−z² at the copper and, on the other hand, filled 2p_y orbital at the oxygen atom, is responsible for the peculiar geometry. Catalytic tests with this complex are currently underway and will be reported in due course.

Experimental

General

All reactions were routinely performed under an inert atmosphere of argon or nitrogen by using Schlenk and glove-box techniques and dry deoxygenated solvents. Dry THF and hexanes were obtained by distillation from Na/benzophenone and dry ether from CaCl₂ and then NaH and dry CH₂Cl₂ from P₂O₅. CDCl₃ was dried from P₂O₅ and stored on 4 Å Linde molecular sieves. CD₂Cl₂ was used as purchased and stored in the glove-box. Nuclear magnetic resonance spectra were recorded on a Bruker 300 AVANCE spectrometer operating at 300.0 MHz for ¹H, 75.5 MHz for ¹³C and 121.5 MHz for ³¹P. Solvent peaks are used as internal reference relative to Me₄Si for ¹H and ¹³C chemical shifts (ppm); ³¹P chemical shifts are relative to a 85% H₃PO₄ external reference. Coupling constants are given in hertz. The following abbreviations are used: s, singlet; d, doublet; t, triplet; q, quadruplet; m, multiplet; v, virtual; b, broad. Mass spectra were obtained at 70 eV with a HP 5989B spectrometer couple to a HP 5980 chromatograph by the direct inlet method. [Cu(CH₃CN)₄BF₄]¹⁶ and [AuCl(SMe₂)]¹⁷ were prepared according to reported procedures.

Phospha-Wittig reagent 1

To a mixture of Mes*PCl₂ (1.0 g, 2.9 mmol), Zn (1.0 g, 15.0 mmol) in THF (10 mL) at 0 °C, was syringed PMe₃ (1 M, 7.5 mmol, 7.5 mL). The crude mixture was allowed to warm to room temperature and was stirred for 3 h. The ³¹P NMR spectrum showed the formation of the desired reagent which was used without further purification.

³¹P NMR (121.5 MHz, THF, 298 K): δ 6.6 ppm (¹J(P–P) = 577.9, Mes*P), −132.5 ppm (¹J(P–P) = 577.9, PMe₃).

Synthesis of 4′-(supermesitylphosphanylidenemethyl)-6-(carbaldehyde)dibenzofuran 2. Reagent 1 was cooled to 0 °C and cannulated on a suspension of 4,6-bis(carbaldehyde)dibenzofuran (0.75 g, 3.3 mmol, 1.1 equivalent) at 0 °C in THF (10 mL). The crude mixture was stirred during 3 h and cooled to 0 °C. CH₂Cl₂, water and ice were then added. The organic phase was separated, dried over MgSO₄ and filtered. The solvent was then evaporated, MeOH was added which resulted in the formation of a precipitate which was filtered off and dried under vacuum. The title compound was obtained as a white solid. Yield: 50%, 0.80 g

¹H NMR (300 MHz, CD₂Cl₂, 298 K): δ 1.48 (s, 18H, para C(CH₃)₃), 1.67 (s, 36H, ortho C(CH₃)₃), 7.43 (t, 1H, ³J(H–H) = 7.5, H₈), 7.51 (t, 1H, ³J(H–H) = 7.3, H₂), 7.60 (s, 2H, H₁₂), 7.85 (dd, 1H, ³J(H–H) = 7.5, ⁴J(H–H) = 3.5, H₃), 7.92 (dd, 1H, ³J(H–H) = 7.7, ⁴J(H–H) = 1.4, H₇), 7.99 (dd, 1H, ³J(H–H) = 7.7, ⁴J(H–H) = 1.2, H₉), 8.22 (dd, 1H, ³J(H–H) = 7.6, ⁴J(H–H) = 1.2, H₁), 8.58 (d, 1H, ²J(H–P) = 25.3, H₁₀) 10.75 (s, 1H, CHO). ³¹P NMR (121.5 MHz, CD₂Cl₂, 298 K): δ 273.9 ppm. ¹³C NMR (75.5 MHz, CD₂Cl₂, 298 K): δ 30.34 (s, para C(CH₃)₃), 32.96 (d, ³J(C–P) = 7.2, ortho C(CH₃)₃), 34.12 (s, paraC(CH₃)₃), 37.42 (s, orthoC(CH₃)₃), 119.01 (d, ³J(C–P) = 7.1, C₇), 120.31 (s, C₄ or C_9b), 121.20 (s, C₁₃, C₁₅), 122.23 (s, C₂), 122.38 (d, ⁴J(C–P) = 2.8, C_9a), 122.98 (d, ⁴J(C–P) = 3.9, C₈), 124.49 (d, ³J(C–P) = 23.0, C₃), 124.66 (d, ²J(C–P) = 14.5, C₆), 124.89 (s, C₉), 124.96 (s, C_9b or C₄), 125.80 (s, C₁), 138.96 (d, ¹J(C–P) = 54.4, C₁₁), 149.18 (s, C₁₂, C₁₆), 151.87 (d, ³J(C–P) = 11.3, C_5a), 153.28 (s, C₁₄), 155.67 (s, C_4a), 166.60 (d, ¹J(C–P) = 39.0, C₁₀).

MS: 485 (M + H, 100%).

Synthesis of 4,6-bis(supermesitylphosphanylidenemethyl)dibenzofuran 3. Reagent 1 was cooled to 0 °C and cannulated on a suspension of 4,6-bis(carbaldehyde)dibenzofuran (0.27 g, 1.2 mmol, 0.4 equivalent) at 0 °C in THF (10 mL). The crude mixture was stirred during 18 h and cooled to 0 °C. CH₂Cl₂, water and ice were then added. The organic phase was separated, dried over MgSO₄ and filtered. The solvent was then evaporated, MeOH was added which resulted in the formation of a precipitate which was filtered off and dried under vacuum. The title compound was obtained as a white solid. Yield: 55%, 0.50 g.

¹NMR (300 MHz, CD₂Cl₂, 298 K): δ 1.40 (s, 18H, para C(CH₃)₃), 1.56 (s, 36H, ortho C(CH₃)₃), 7.37 (t, 2H, ³J(H–H) = 7.6, H₂, H₈), 7.50 (s, 4H, H₁₃), 7.77 (d, 2H, ³J(H–H) = 5.3, H₃, H₇), 7.90 (d, 2H, ³J(H–H) = 7.4, H₁, H₉), 8.48 (m, AA′XX′, m, ∑J = 41.9, H₁₀, H_10′). ³¹P NMR (121.5 MHz, CD₂Cl₂, 298 K): δ 271.5 ppm. ¹³C NMR (75.5 MHz, CD₂Cl₂, 298 K) δ 30.4 (s, para C(CH₃)₃), 33.0 (d, ⁴J(C–P) = 3.0, ortho C(CH₃)₃), 34.1 (s, paraC(CH₃)₃), 37.4 (s, orthoC(CH₃)₃), 118.94 (s, C₁, C₉), 121.01 (s, C₁₃), 122.36 (s, C₂, C₈), 123.70 (m, C₄, C₆, C₃, C₇), 124.57 (vt, AXX′, ∑J(C–P) = 14.4, C_9a, C_9b), 140.71 (m, AXX′, ∑J(C–P) = 62.7, C₁₁, C_11′), 148.90 (s, C₁₂), 151.55 (vt, AXX′, ∑J(C–P) = 12.0, C_4a, C_5a), 153.17 (s, C₁₄), 167.09 (m, AXX′, ∑J(C–P) = 51.7, C₁₀, C_10′). MS: 746 (M + H, 100%).

Synthesis of gold complex 4. [AuCl(SMe₂)] (30 mg, 0.10 mmol) and ligand 3 (75 mg, 0.10 mmol) were dissolved in THF (3 mL). The solution was stirred for 20 min. AgBF₄ was added (19.8 mg, 0.10 mmol) and the solution was stirred for 12 h. The ³¹P NMR spectrum of the crude mixture shows the formation of a unique product. The mixture was filtered on Celite and the solution was taken to dryness. The product was washed with hexanes and dried to yield a yellow powder. Yield: 92%, 87 mg. X-Ray quality crystals were obtained after diffusion of ether into a THF solution of the complex.

¹H NMR (300 MHz, CDCl₃, 298 K): δ 1.40 (s, 18H, para C(CH₃)₃), 1.68 (s, 36H, ortho C(CH₃)₃), 7.58 (t, 2H, ³J(H–H) = 7.6, H₂, H₈), 7.67 (s, 4H, H₁₃), 8.17 (t, ³J(H–H) = 8.8, 2H, H₃, H₇), 8.37 (m, AA′XX′, m, ∑J = 7.0, H₁₀, H_10′). ³¹P NMR (121.5 MHz, CDCl₃, 298 K): δ 200.2. ¹³C NMR (75.5 MHz, CDCl₃, 298 K): δ 32.9 (s, para C(CH₃)₃), 36.5 (s, ortho C(CH₃)₃), 37.2 (s, paraC(CH₃)₃), 40.6 (s, orthoC(CH₃)₃), 118.2 (s, C_9a,9b), 121.1 (m, C_1,2,8,9), 122.3 (s, C₁₃), 122.4 (s, C_4,6), 131.1 (m, C_3,7,11), 148.2 (AA′XX′, m, ∑J = 8.0, C_4a,5a), 151.7 (s, C₁₂), 152.9 (s, C₁₄), 157.9 (AA′XX′, m, ∑J = 69, C_10,10′).

Synthesis of silver complex 5. Ligand 3 (57 mg, 0.07 mmol) was dissolved in CH₂Cl₂ (1 mL) and in the glove-box, solid AgBF₄ (15 mg, 0.07 mmol) was added. The initially colorless solution rapidly turned bright yellow. The ³¹P NMR spectrum of the crude mixture shows the formation of a unique product. The solution was taken to dryness. The product was washed with hexanes and dried to yield a yellow powder. Yield: 95%, 57 mg. X-Ray quality crystals were obtained after diffusion of hexanes into a CH₂Cl₂ solution of the complex.

¹H NMR (300 MHz, CD₂Cl₂, 298 K): δ 1.39 (s, 18H, para C(CH₃)₃), 1.63 (s, 36H, ortho C(CH₃)₃), 2.52 (s, H₂O), 7.41 (d, 2H, ³J(H–H) = 7.7, H₁,₉), 7.53 (t, ³J(H–H) = 7.7, 2H, H_2,8), 7.63 (s, 4H, H₁₃), 8.18 (dd, 2H, ³J(H–H) = 7.7, ⁴J(H–P) = 1.3, H_3,7), 8.23 (AA′XX′, m, 2H, ∑J = 41.9, H_10,10′). ³¹P NMR (121.5 MHz, CD₂Cl₂, 298 K): δ 204.9 (d + d, ¹J(P–Ag) = 686, ¹J(P–Ag) = 596). ¹³C NMR (75.5 MHz, CD₂Cl₂, 298 K): δ 31.7 (s, para C(CH₃)₃), 35.4 (d, ⁴J(C–P) = 3.0, ortho C(CH₃)₃), 36.1 (s, paraC(CH₃)₃), 39.7 (s, orthoC(CH₃)₃), 122.2 (s, C_9a,9b), 123.8 (m, C_1,2,8,9), 125.1 (s, C₁₃), 125.7 (s, C_4,6), 129.2 (d, ¹J(P–C) = 15.5, C₁₁), 133.8 (C), 153.4 (AXX′, m, ∑J(C–P) = 7.5, C_4a,5a), 154.2 (s, C₁₂), 156.3 (s, C₁₄), 169.6 (AXX′, m, ∑J(C–P) = 40.3, C₁₀).

Synthesis of copper complex 6. [Cu(CH₃CN)₄][BF₄] (34.5 mg, 0.13 mmol) and ligand 3 (100 mg, 0.13 mmol) were dissolved in CH₂Cl₂ (7 mL). The ³¹P NMR spectrum of the crude mixture shows the formation of a unique product. The volume of the solution was reduced, hexanes were added which resulted in the precipitation of a yellow solid. It was filtered, washed with hexanes then dried. Yield: 95%, 116 mg. X-Ray quality crystals were obtained after diffusion of hexanes into a CDCl₃ solution of the complex.

¹H NMR (300 MHz, CD₂Cl₂, 298 K): δ 1.38 (s, 18H, para C(CH₃)₃), 1.61 (s, 36H, ortho C(CH₃)₃), 7.34 (m, H_2,8), 7.50 (m, 2H, H_1,9), 7.61 (s, 4H, H₁₃), 8.10 (m, 4H, H_3,7,10,10′). ³¹P NMR (121.5 MHz, CD₂Cl₂, 298 K): δ 207.6. ¹³C NMR (75.5 MHz, CD₂Cl₂, 298 K): δ 31.3 (s, para C(CH₃)₃), 34.9 (s, ortho C(CH₃)₃), 35.6 (s, paraC(CH₃)₃), 39.0 (s, orthoC(CH₃)₃), 123.1 (s, C_9a,9b), 123.3 (s, C_1,2,8,9), 123.6 (s, C₁₃), 125.4 (s, C_4,6), 129.4 (d, ¹J(P–C) = 20, C₁₁), 131.4 (m, ∑J = 26, C_3,7), 152.4 (AXX′, m, ∑J(C–P) = 14, C_4a,5a), 153.7 (s,C₁₂), 156.2 (s, C₁₄), 163.8 (AXX′, m, ∑J(C–P) = 46, C₁₀).

Computational details

All computations were performed using the Gaussian 03 suite of programs and gradient corrected density functional theory using the B3PW91 functional.^18,19 All optimizations were carried out using the 6-31 G(d) basis set for H, C, N, O, P. The basis set employed for the Cu and Au atoms incorporate the Hay and Wadt small-core relativistic effective core potential and double-ζ valence basis set.²⁰ For complex II, the ONIOM method was employed with the UFF force field (for methyl groups of the xylyl substituents).²¹ Full optimizations were followed by analytical computation of the Hessian matrix to confirm the nature of the located minima on the potential energy surface. Minima were characterized by no imaginary frequency.

Crystallography

Data were collected at 150.0(1) K on a Nonius Kappa CCD diffractometer using a Mo-Kα (λ = 0.71073 Å) X-ray source and a graphite monochromator. All data were measured using phi and omega scans. Experimental details are given in Tables 1 and 2. The crystal structure was solved using SIR 97²² and SHELXL-97.²³ Molecular drawings were made using ORTEP III for Windows then POV-Ray.²⁴ contain the supplementary crystallographic data for this paper.

CCDC reference numbers 275394–275397.

For crystallographic data in CIF or other electronic format see DOI: 10.1039/b508678g

Acknowledgements

The CNRS and the Ecole Polytechnique are thanked for supporting this work. The use of computational facilities from the IDRIS institute (University Paris XI, Orsay, France) is also gratefully acknowledged.

References

1. K. B. Dillon, F. Mathey and J. F. Nixon, Phosphorus: The Carbon Copy, Wiley, Chichester, 1998.
2. See, for example: (a) N. Avarvari, P. Le Floch and F. Mathey, J. Am. Chem. Soc., 1996, 118, 11978; (b) N. Avarvari, P. Le Floch, L. Ricard and F. Mathey, Organometallics, 1997, 16, 4089; (c) P. Rosa, N. Mézailles, L. Ricard, F. Mathey, P. Le Floch and Y. Jean, Angew. Chem., Int. Ed., 2001, 40, 1251; (d) N. Mézailles, P. Rosa, L. Ricard, F. Mathey and P. Le Floch, Organometallics, 2000, 19, 2942; (e) N. Mézailles, N. Avarvari, N. Maigrot, L. Ricard, F. Mathey, P. Le Floch, L. Cataldo, T. Berclaz and M. Geoffroy, Angew. Chem., Int. Ed., 1999, 38, 3194.
3. A. Moores, F. Goettmann, C. Sanchez and P. Le Floch, Chem. Commun., 2004, 2842.
4. See, for example: (a) S. Qiao and G. C. Fu, J. Org. Chem., 1998, 63, 4168; K. Tanaka and G. Fu, J. Org. Chem., 2001, 66, 8177; (b) S. Bellemin-Laponnaz, M. M.-C. Lo, T. H. Peterson, J. M. Allen and G. C. Fu, Organometallics, 2001, 20, 3453; (c) R. Shintani and G. C. Fu, Org. Lett., 2002, 4, 3699; (d) R. Shintani and G. Fu, Angew. Chem., Int. Ed., 2003, 42, 4082; (e) R. Shintani and G. Fu, J. Am. Chem. Soc., 2003, 125, 10778; (f) C. Ganter, C. Glinsböckel and B. Ganter, Eur. J. Inorg. Chem., 1998, 1163; (g) C. Kaulen, C. Pala, C. Hu and C. Ganter, Organometallics, 2001, 20, 1614; (h) C. Ganter, L. Brassat, C. Glinsböckel and B. Ganter, Organometallics, 1997, 16, 2862; (i) C. Ganter, L. Brassat and B. Ganter, Tetrahedron: Asymmetry, 1997, 8, 2607; (j) C. Ganter, C. Kaulen and U. Englert, Organometallics, 1999, 18, 5444; (k) C. Ganter, L. Brassat and B. Ganter, Chem. Ber., 1997, 130, 1771; (l) D. Carmichael, F. Mathey, L. Ricard and N. Seeboth, Chem. Commun., 2002, 2976; (m) D. Carmichael, J. Klankermayer, L. Ricard and N. Seeboth, Chem. Commun., 2004, 1144; (n) X. Sava, L. Ricard, F. Mathey and P. Le Floch, Organometallics, 2000, 19, 4899; (o) X. Sava, M. Melaimi, N. Mézailles, L. Ricard, F. Mathey and P. Le Floch, New J. Chem., 2002, 26, 1378; (p) M. Melaimi, L. Ricard, F. Mathey and P. Le Floch, J. Organomet. Chem., 2003, 684, 189.
5. See, for example: (a) S. Ikeda, F. Ohhata, M. Miyoshi, R. Tanaka, T. Minami, F. Ozawa and M. Yoshifuji, Angew. Chem., Int. Ed., 2000, 39, 4512; (b) T. Minami, H. Okamoto, S. Ikeda, R. Tanaka, F. Ozawa and M. Yoshifuji, Angew. Chem., Int. Ed., 2001, 40, 4501; (c) F. Ozawa, S. Yamamoto, S. Kawagishi, M. Hiraoka, S. Ikeda, T. Minami, S. Ito and M. Yoshifuji, Chem. Lett., 2001, 972; (d) F. Ozawa, H. Okamoto, S. Kawagishi, S. Yamamoto, T. Minami and M. Yoshifuji, J. Am. Chem. Soc., 2002, 124, 10968; (e) H. Murakami, T. Minami and F. Ozawa, J. Org. Chem., 2004, 69, 4482.
6. A. Jouaiti, M. Geoffroy, G. Terron and G. Bernardinelli, J. Am. Chem. Soc., 1995, 117, 2251.
7. A. S. Ionkin and W. J. Marshall, Heteroat. Chem., 2002, 13, 662.
8. (a) M. Doux, C. Bouet, N. Mézailles, L. Ricard and P. Le Floch, Organometallics, 2002, 21, 2785; (b) M. Doux, N. Mézailles, M. Melaimi, L. Ricard and P. Le Floch, Chem. Commun., 2002, 1566; (c) M. Dochnahl, M. Doux, E. Faillard, L. Ricard and P. Le Floch, Eur. J. Inorg. Chem., 2005, 125.
9. J. R. Hagadorn, Chem. Commun., 2001, 2144.
10. U. Iserloh, D. P. Curran and S. Kanemasa, Tetrahedron: Asymmetry, 1999, 10, 2417.
11. (a) P. Le Floch, A. Marinetti, L. Ricard; and F. Mathey, J. Am. Chem. Soc., 1990, 112, 2407; (b) P. Le Floch and F. Mathey, Synlett, 1990, 171; (c) A. Marinetti, L. Ricard and F. Mathey, Organometallics, 1990, 9, 788; (d) A. Marinetti, S. Bauer, L. Ricard and F. Mathey, Organometallics, 1990, 9, 793; (e) A. Marinetti, P. Le Floch and F. Mathey, Organometallics, 1991, 10, 1190.
12. (a) S. Shah and J. D. Prostasiewicz, Chem. Commun., 1998, 1585; (b) S. Shah, M. C. Simpson, R. C. Smith and J. D. Protasiewicz, J. Am. Chem. Soc., 2001, 123, 6925; (c) R. C. Smith, X. Chen and J. D. Protasiewicz, J. Am. Chem. Soc., 2004, 126, 2268; (d) S. Shah and J. D. Prostasiewicz, Coord. Chem. Rev., 2000, 210, 181.
13. S. Kanemasa, Y. Oderaotoshi, S.-I. Sakaguchi, H. Yamamoto, J. Tanaka, E. Wada and D. P. Curran, J. Am. Chem. Soc., 1998, 120, 3074.
14. D. Saravanabharathi, M. Nethaji and A. Samuelson, Proc. Indian Acad. Sci. (Chem. Sci.), 2002, 114, 347.
15. R. Hart, P. C. Healy, G. A. Hope, D. W. Turner and A. H. White, J. Chem. Soc., Dalton Trans., 1994, 773.
16. G. Kubas, Inorg. Synth., 1979, 19, 90.
17. R. Uson and A. Laguna, in Organometallic Syntheses, ed. R. B. King and J. Eisch, Elsevier Science, Amsterdam, 1986, p. 324.
18. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 98 (Revision A.11), Gaussian, Inc., Pittsburgh, PA, 2001.
19. A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
20. P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
21. A. K. Rappé, C. J. Casewitt, K. S. Colwell, W. A. Goddard and W. M. Skiff, J. Am. Chem. Soc., 1992, 114, 10024.
22. A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. Spagna, SIR 97, an integrated package of computer programs for the solution and refinement of crystal structures using single crystal data, J. Appl. Crystallogr., 1999, 32, 115.
23. G. M. Sheldrick, SHELXL-97, Program for refinement of crystal structures, University of Göttingen, Germany, 1997.
24. M. N. Burnett and C. K. Johnson, ORTEP-III: Oak Ridge Thermal Ellipsoid Plot Program for Crystal Structure Illustrations, Report ORNL-6895, Oak Ridge National Laboratory, Oak Ridge, TN, USA, 1996; L. J. Farrugia, ORTEP-3, Department of Chemistry, University of Glasgow, 1996.

---

Footnote

† Electronic supplementary information (ESI) available: View, cartesian coordinates and frequencies of the theoretical structures I and II. See http://dx.doi.org/10.1039/b508678g

---