η 1 : η 2P-Pyrazolylphosphaalkene Complexes of Ruthenium ( 0 )

An extended range of novel ruthenium phosphaalkene complexes of the type [Ru{η-N:η-P,C-P(pz)=CH(SiMe2R)}(CO)(PPh3)2] (R = Tol, C6H4CF3-p; pz′ = pzMe2, pzCF3, pzMe,CF3; R = Me, C6H4CF3-p; pz′ = pzPh) have been prepared from the respective ruthenaphosphaalkenyls [Ru{P=CH(SiMe2R)}Cl(CO)(PPh3)2] upon treatment with Lipz′. Where R = C6H4CF3-p and pz′ = pzMe2 the complex is characterized by single crystal X-ray diffraction, only the second example of such species being structurally characterized. This indicates enhanced pyramidalisation of the alkenic carbon center when compared with precedent data (R = Me, pz′ = pz) implying an enhanced Ru→π*PC contribution, which can be correlated with the greater donor power of pzMe2. This is similarly reflected in spectroscopic data that reveal significant influence of the pyrazolyl substituents upon the phosphaalkene, stronger donors imparting significantly enhanced shielding to phosphorus; in contrast, a much lesser influence if noted for the silyl substituents.


Introduction
After almost half a century of study, the chemistry of low-coordinate phosphorus continues to fascinate both organic and inorganic chemists alike [1][2][3][4][5][6][7][8].Dominated by the isolobal and isoelectronic relationship between phosphorus and the "CH" fragment, the chemistry of phosphacarbons is often familiar from their carbo-centric and nitrogenous counterparts, yet they simultaneously embody appreciable dichotomy in terms of their underlying electronic and chemical nature.Nowhere is this more apparent than in the chemistry of phosphaalkenes (RP=CR R") and phosphaalkynes (RC≡P), which rank among the most heavily studied classes of phosphacarbon.Reactivity is in each case dominated by the high-energy π-systems, though in the case of phosphaalkenes this is often competitive with the lone-pair, which lies close-albeit marginally lower-in energy.In contrast, the phosphaalkyne lone-pair is appreciably stabilized, but can be engaged chemically under appropriately directing conditions.
The study of these compounds is, however, often complicated by an intrinsic lack of stability, which restricts the range of available substrates and necessitates some synthetic ingenuity.Such difficulties are often addressed by the imposition of steric bulk, as a means of imparting kinetic stability, which has proven particularly effective in precluding homo-oligomerisation of phosphaalkenes.The same approach has typically been cited in the development of kinetically stabilized phosphaalkynes (e.g., t BuC≡P, AdC≡P), however, in these cases the bulk is often sufficiently remote from the reactive π-system as to preclude it being the sole stabilizing influence.Moreover, even bulky phosphaalkynes, e.g., R 3 SiC≡P (R = Ph [9], Me [10,11]) often exhibit only limited stability, in some instances comparable to that of unencumbered systems (e.g., MeC≡P [12]).Significantly, the formally related phosphaethynyloate ion "O-C≡P − " is isolable as a sodium salt [13,14], which exhibits appreciable stability despite the lack of any steric "protection"; this fact is attributed to electronic influences, with the preferential adoption of an O=C≡P − type structure.Taken alongside computational studies of the isolated cyaphide ion ("C≡P − "), which indicate an intrinsically unstable hypovalent "[C=P] − " structure [15], this would imply that electronics serve an equally important role in imparting stability to low-coordinate phosphacarbons.Indeed, in the context of phosphaalkenes-which remain the longest and most heavily studied of the phosphacarbons-the incorporation of π-donor substituents (e.g., NR 2 , OR) and/or conjugate π-systems has long been employed as an alternative means of stabilization [16].
In seeking to further stabilize, and chemically exploit phosphaalkenes, the incorporation of transition metal fragments has proven to be a particularly valuable tool.Prominently, the coordination of phosphaalkenes [17] in an η 1 -fashion allows for selective reactivity of the π-system to be developed, by sequestering the lone-pair and precluding its competitive reaction [18].Somewhat less extensively studied, η 2 -coordination-common among carbo-centric analogues-has also been employed with phosphaalkenes [19,20], discrimination between η 1 and η 2 being achieved through judicious selection of the metal fragment, though bridging-η 1 :η 2 complexes have also been described [21].Moreover, both transition metal and, to a lesser extent, main-group fragments can be incorporated as discrete substituents on the phosphaalkenic core to afford a range of metallaphosphaalkenyl complexes (Figure 1) [22,23].First described in 1985 [24], such systems remain relatively rare, though-with the exception of type E-all possible motifs have been realized, with P-metalla-(type A) and C-metalla-(type B) systems the most heavily studied.
Inorganics 2016, 4, 30 2 of 11 alongside computational studies of the isolated cyaphide ion ("C≡P − "), which indicate an intrinsically unstable hypovalent "[C=P] − " structure [15], this would imply that electronics serve an equally important role in imparting stability to low-coordinate phosphacarbons.Indeed, in the context of phosphaalkenes-which remain the longest and most heavily studied of the phosphacarbons-the incorporation of π-donor substituents (e.g., NR2, OR) and/or conjugate π-systems has long been employed as an alternative means of stabilization [16].
In seeking to further stabilize, and chemically exploit phosphaalkenes, the incorporation of transition metal fragments has proven to be a particularly valuable tool.Prominently, the coordination of phosphaalkenes [17] in an η 1 -fashion allows for selective reactivity of the π-system to be developed, by sequestering the lone-pair and precluding its competitive reaction [18].Somewhat less extensively studied, η 2 -coordination-common among carbo-centric analogues-has also been employed with phosphaalkenes [19,20], discrimination between η 1 and η 2 being achieved through judicious selection of the metal fragment, though bridging-η 1 :η 2 complexes have also been described [21].Moreover, both transition metal and, to a lesser extent, main-group fragments can be incorporated as discrete substituents on the phosphaalkenic core to afford a range of metallaphosphaalkenyl complexes (Figure 1) [22,23].First described in 1985 [24], such systems remain relatively rare, though-with the exception of type E-all possible motifs have been realized, with P-metalla-(type A) and C-metalla-(type B) systems the most heavily studied.Recently, as part of an extended program investigating transition metal compounds featuring low-coordinate phosphacarbons with potential for conjugation [25][26][27][28], we have prepared and studied a range of ruthenaphosphaalkenyl complexes of the type [Ru{P=CH(SiMe2R)}Cl(CO)(PPh3)2] (R = Me 1a, Ph 1b, Tol 1c) [29,30].These are prepared by hydroruthenation of phosphaalkynes R3SiC≡P, following from methodology developed initially by Hill and Jones for t BuC≡P and related systems [31][32][33], and superficially related to Nixon's independent reduction of [(Ph3P)2Pt(η 2 -P≡C t Bu)] with Schwartz's reagent [34].Notably, we have described the first structural data for ruthenaphosphaalkenyls, which demonstrate the phosphaalkenyl moieties to behave as classical 1-electron donors, within a square-based pyramidal metal coordination sphere.Significantly, this precludes, in the ground state at least, any augmenting P→Ru donation from the lone-pair (i.e., phosphavinylidene character), which is consistent with precedent reactivity studies that demonstrate a nucleophilic phosphorus center [35][36][37][38][39]; indeed, in common with these precedent reports, we have found 1a-b to engage in reaction with electrophilic fragments at phosphorus [30,40].
Notwithstanding, we have additionally observed an unusual reaction to proceed when 1a or 1b is exposed the lithium pyrazolates Li[pz′] (pz′ = pz, pz Me2 , pz CF3 , pz Me,CF3 ), giving rise to the unprecedented phosphaalkene complexes [Ru{η 1 -N:η 2 -P,C-P(pz′)=CH(SiMe2R)}(CO)(PPh3)2] (Scheme 1) [29,30], which feature an apparent 3-member (Ru-P-C) metallacyclic core, bridged by a pyrazolyl moiety.Also replicated using Hill's [Ru{P=CH( t Bu)}Cl(CO)(PPh3)2], this chemistry would perhaps imply ambiphilicty for the ruthenaphosphaalkenyl motif, if not directly of the alkenyl phosphorus center.On the basis of available data, we reasoned the resulting complexes to be best formulated as involving an η 2 -phosphaalkene moiety, in line with the Dewar-Chatt-Duncanson model, with a dominant contribution from dπ→π*(PC) retro-donation to metal-ligand binding.The precise extent of the latter might reasonably be influenced by the nature of the phosphaalkene Recently, as part of an extended program investigating transition metal compounds featuring low-coordinate phosphacarbons with potential for conjugation [25][26][27][28], we have prepared and studied a range of ruthenaphosphaalkenyl complexes of the type [Ru{P=CH(SiMe 2 R)}Cl(CO)(PPh 3 ) 2 ] (R = Me 1a, Ph 1b, Tol 1c) [29,30].These are prepared by hydroruthenation of phosphaalkynes R 3 SiC≡P, following from methodology developed initially by Hill and Jones for t BuC≡P and related systems [31][32][33], and superficially related to Nixon's independent reduction of [(Ph 3 P) 2 Pt(η 2 -P≡C t Bu)] with Schwartz's reagent [34].Notably, we have described the first structural data for ruthenaphosphaalkenyls, which demonstrate the phosphaalkenyl moieties to behave as classical 1-electron donors, within a square-based pyramidal metal coordination sphere.Significantly, this precludes, in the ground state at least, any augmenting P→Ru donation from the lone-pair (i.e., phosphavinylidene character), which is consistent with precedent reactivity studies that demonstrate a nucleophilic phosphorus center [35][36][37][38][39]; indeed, in common with these precedent reports, we have found 1a-b to engage in reaction with electrophilic fragments at phosphorus [30,40].
Notwithstanding, we have additionally observed an unusual reaction to proceed when 1a or 1b is exposed the lithium pyrazolates Li[pz ] (pz = pz, pz Me2 , pz CF3 , pz Me,CF3 ), giving rise to the unprecedented phosphaalkene complexes [Ru{η 1 -N:η 2 -P,C-P(pz )=CH(SiMe 2 R)}(CO)(PPh 3 ) 2 ] (Scheme 1) [29,30], which feature an apparent 3-member (Ru-P-C) metallacyclic core, bridged by a pyrazolyl moiety.Also replicated using Hill's [Ru{P=CH( t Bu)}Cl(CO)(PPh 3 ) 2 ], this chemistry would perhaps imply ambiphilicty for the ruthenaphosphaalkenyl motif, if not directly of the alkenyl phosphorus center.On the basis of available data, we reasoned the resulting complexes to be best formulated as involving an η 2 -phosphaalkene moiety, in line with the Dewar-Chatt-Duncanson model, with a dominant contribution from d π →π* (PC) retro-donation to metal-ligand binding.The precise extent of the latter might reasonably be influenced by the nature of the phosphaalkene substituents, or indeed those of the pyrazolyl moiety, resulting variations in the atom-specific parameters (e.g., NMR shielding effects) offering a potential means of quantification.In seeking to assess the influence of such factors, and thus potentially to elaborate means of control over reactivity, we report herein the synthesis and characterization of an extended range of these novel pyrazolylphosphaalkene complex.
substituents, or indeed those of the pyrazolyl moiety, resulting variations in the atom-specific parameters (e.g., NMR shielding effects) offering a potential means of quantification.In seeking to assess the influence of such factors, and thus potentially to elaborate means of control over reactivity, we report herein the synthesis and characterization of an extended range of these novel pyrazolylphosphaalkene complex.Reagents and conditions: (i) Lipz′ (pz′ = pz, pz Me2 , pz CF3 , pz Me,CF3 ), THF.

The reactions of 1c-d with Li[pz Me2
] each afford a single complex (3c-d respectively), which is spectroscopically comparable to those obtained similarly from 1a and 1b [29,30].Thus, the characteristic resonances for the phosphaalkenyl and PPh3 ligands are lost, being replaced by three new, mutually coupling, resonances (1:1:1 ratio) in the region 50-30 ppm, which is commonly associated with saturated (λ 3 σ 3 ) phosphorus centers.In each case, the lower frequency resonance is identified as that associated with the phosphacarbon, on the basis of a 1 H-31 P-HMBC correlation to the, now appreciably shielded, "CH(SiMe2R)" moiety (identified from 1 H- 29 Si-HMBC spectra and a significant 1 JPC coupling (~70 Hz)).The latter is observed at δH ~1.7 and δC ~42, which, though appreciably lower frequency than for a classical alkenyl "CH", remains somewhat deshielded for a fully saturated "alkyl" moiety.Indeed, as was previously noted for 2a-b and 3a-b, the magnitude of the 1 JCH coupling (~130-140 Hz) is intermediate between those of C2H4 (156 Hz) and CH4 (125 Hz) [41], while 1 JPC are essentially unperturbed from those of the parent phosphaalkenyls, consistent with the case of a π-bound phosphaalkene.This scenario is further supported by a significant Reagents and conditions: (i) Lipz (pz = pz, pz Me2 , pz CF3 , pz Me,CF3 ), THF.
substituents, or indeed those of the pyrazolyl moiety, resulting variations in the atom-specific parameters (e.g., NMR shielding effects) offering a potential means of quantification.In seeking to assess the influence of such factors, and thus potentially to elaborate means of control over reactivity, we report herein the synthesis and characterization of an extended range of these novel pyrazolylphosphaalkene complex.

Synthesis and Characterization of η 2 -Pyrazolylphosphaalkene complexes
As we have previously described [29,40], the ruthenaphosphaalkenyls 1c-d were prepared via hydroruthenation of the respective phosphalkynes RMe2SiC≡P with [RuHCl(CO)(PPh3)3] (Scheme 2).Each system is readily identified on the basis of characteristic spectroscopic signatures, viz.(i) the heavily deshielded 31   Me2 ] each afford a single complex (3c-d respectively), which is spectroscopically comparable to those obtained similarly from 1a and 1b [29,30].Thus, the characteristic resonances for the phosphaalkenyl and PPh3 ligands are lost, being replaced by three new, mutually coupling, resonances (1:1:1 ratio) in the region 50-30 ppm, which is commonly associated with saturated (λ 3 σ 3 ) phosphorus centers.In each case, the lower frequency resonance is identified as that associated with the phosphacarbon, on the basis of a 1 H-31 P-HMBC correlation to the, now appreciably shielded, "CH(SiMe2R)" moiety (identified from 1 H- 29 Si-HMBC spectra and a significant 1 JPC coupling (~70 Hz)).The latter is observed at δH ~1.7 and δC ~42, which, though appreciably lower frequency than for a classical alkenyl "CH", remains somewhat deshielded for a fully saturated "alkyl" moiety.Indeed, as was previously noted for 2a-b and 3a-b, the magnitude of the 1 JCH coupling (~130-140 Hz) is intermediate between those of C2H4 (156 Hz) and CH4 (125 Hz) [41], while 1 JPC are essentially unperturbed from those of the parent phosphaalkenyls, consistent with the case of a π-bound phosphaalkene.This scenario is further supported by a significant The reactions of 1c-d with Li[pz Me2 ] each afford a single complex (3c-d respectively), which is spectroscopically comparable to those obtained similarly from 1a and 1b [29,30].Thus, the characteristic resonances for the phosphaalkenyl and PPh 3 ligands are lost, being replaced by three new, mutually coupling, resonances (1:1:1 ratio) in the region 50-30 ppm, which is commonly associated with saturated (λ 3 σ 3 ) phosphorus centers.In each case, the lower frequency resonance is identified as that associated with the phosphacarbon, on the basis of a 1 H-31 P-HMBC correlation to the, now appreciably shielded, "CH(SiMe 2 R)" moiety (identified from 1 H- 29 Si-HMBC spectra and a significant 1 J PC coupling (~70 Hz)).The latter is observed at δ H ~1.7 and δ C ~42, which, though appreciably lower frequency than for a classical alkenyl "CH", remains somewhat deshielded for a fully saturated "alkyl" moiety.Indeed, as was previously noted for 2a-b and 3a-b, the magnitude of the 1 J CH coupling (~130-140 Hz) is intermediate between those of C 2 H 4 (156 Hz) and CH 4 (125 Hz) [41], while 1 J PC are essentially unperturbed from those of the parent phosphaalkenyls, consistent with the case of a π-bound phosphaalkene.This scenario is further supported by a significant reduction in ν CO for the retained carbonyl (1900-1910 cm −1 ), indicative of increased density at metal and, on the basis of calculated force constants, consistent with reduction of the metal to Ru(0) [42][43][44][45][46].

The reactions of 1c-d with Li[pz
Though augmented by computational data, the direct structural characterization of η 2 -pyrazolylphosphaalkene complexes has been limited previously to the single crystal X-ray structure of pyrazole derivative 2a and a low-quality structure of 3a that served to confirm connectivity [30].However upon prolonged standing at ambient temperature, a concentrated CDCl 3 solution of 3d yielded X-ray quality single crystals of the chloroform solvate (Figure 2).reduction in νCO for the retained carbonyl (1900-1910 cm −1 ), indicative of increased density at metal and, on the basis of calculated force constants, consistent with reduction of the metal to Ru(0) [42][43][44][45][46].
Though augmented by computational data, the direct structural characterization of η 2 -pyrazolylphosphaalkene complexes has been limited previously to the single crystal X-ray structure of pyrazole derivative 2a and a low-quality structure of 3a that served to confirm connectivity [30].However upon prolonged standing at ambient temperature, a concentrated CDCl3 solution of 3d yielded X-ray quality single crystals of the chloroform solvate (Figure 2).In common with the precedent structure of 2a, compound 3d exhibits distorted trigonal-bipyramidal geometry about ruthenium, with PPh3 ligands and the η 2 -phosphaalkene lying in the equatorial plane.The carbonyl adopts an axial position, along with the metal-pyrazolyl bond, the latter being marginally distorted from linearity (∠CCO-Ru-N 170.2(2)°) due to the strain of bridging the Ru-phosphacarbon linkage.Internal angles for the phosphacarbon moiety indicate partial pyramidalization of the carbon center, while the unusually tight angles about phosphorus can be attributed to the constraint of the bridging pyrazolyl group.Taken together with the P-C linkage 1.782(5) Å, which is intermediate between a single and double bond [47,48], these data are wholly consistent with our previous conclusion of η 2 -binding, rather than a discrete metallacyclic species.The C≡O linkage is marginally elongated relative to the structurally characterized ruthenium(II) phosphaalkenyls [30], consistent with a more electron rich metal center, and is comparable to those of other established Ru(0) carbonyls, including that previously reported for 2a.Indeed, within the bounds of uncertainty, the molecular geometry of 3d is fully comparable to that of 2a, with the exception of a marginally contracted P-N linkage (1.788(4) Å, cf.1.809(5) Å 2a), which might reasonably be attributed to the increased donor strength of pz Me2 relative to pz, as is reflected in the spectroscopic features (vide infra).Moreover, marginally greater pyramidalization about carbon is apparent in 3d (∠Si-C-P 113.4(2)° cf.116.7(3)° 2a), which might similarly reflect the enhanced donor strength of pz Me2 leading to an increase in retro-donation from the metal.
In analogous fashion to the synthesis of 3, treatment of phosphaalkenyls with Li[pz′] (pz′ = pz CF3 , pz Me,CF3 , pz Ph ) led to formation of the respective η 2 -phosphaalkene complexes 4c-d, 5c-d In common with the precedent structure of 2a, compound 3d exhibits distorted trigonal-bipyramidal geometry about ruthenium, with PPh 3 ligands and the η 2 -phosphaalkene lying in the equatorial plane.The carbonyl adopts an axial position, along with the metal-pyrazolyl bond, the latter being marginally distorted from linearity (∠C CO -Ru-N 170.2(2) • ) due to the strain of bridging the Ru-phosphacarbon linkage.Internal angles for the phosphacarbon moiety indicate partial pyramidalization of the carbon center, while the unusually tight angles about phosphorus can be attributed to the constraint of the bridging pyrazolyl group.Taken together with the P-C linkage 1.782(5) Å, which is intermediate between a single and double bond [47,48], these data are wholly consistent with our previous conclusion of η 2 -binding, rather than a discrete metallacyclic species.The C≡O linkage is marginally elongated relative to the structurally characterized ruthenium(II) phosphaalkenyls [30], consistent with a more electron rich metal center, and is comparable to those of other established Ru(0) carbonyls, including that previously reported for 2a.Indeed, within the bounds of uncertainty, the molecular geometry of 3d is fully comparable to that of 2a, with the exception of a marginally contracted P-N linkage (1.788(4) Å, cf.1.809(5) Å 2a), which might reasonably be attributed to the increased donor strength of pz Me2 relative to pz, as is reflected in the spectroscopic features (vide infra).Moreover, marginally greater pyramidalization about carbon is apparent in 3d (∠Si-C-P 113.4(2) • cf.116.7(3) • 2a), which might similarly reflect the enhanced donor strength of pz Me2 leading to an increase in retro-donation from the metal.
In analogous fashion to the synthesis of 3, treatment of phosphaalkenyls with Li[pz ] (pz = pz CF3 , pz Me,CF3 , pz Ph ) led to formation of the respective η 2 -phosphaalkene complexes 4c-d, 5c-d (Scheme 2), 6a and 6d (Scheme 3); the SiMe 3 (a) and SiMe 2 Ph (b) derivatives of 4 and 5 have been previously described [29].In contrast to 2 and 3, the asymmetrically substituted pyrazolates offer potential for positional isomerism; however, only a single isomer is observed in each case.Though non-trivial in lieu of crystallographic data, assignment of the specific isomer obtained can be achieved with recourse to spectroscopic data.Thus, for each of 4 and 5 a notable coupling (J FP ~20 Hz) is apparent between the CF 3 substituent and the phosphacarbon only; this would imply proximity of the two moieties, strongly suggesting that the bulkier CF 3 substituent (cf.H or Me) is more favorably positioned away from the sterically encumbered metal center.This is supported by significant deshielding of the phosphacarbon center, consistent with proximity to an electron withdrawing substituent.Indeed, we find the resonant frequency of the phosphacarbon center to be very sensitive to the pyrazolyl substituents (vide infra), most significantly so for that at the proximal site.
(Scheme 2), 6a and 6d (Scheme 3); the SiMe3 (a) and SiMe2Ph (b) derivatives of 4 and 5 have been previously described [29].In contrast to 2 and 3, the asymmetrically substituted pyrazolates offer potential for positional isomerism; however, only a single isomer is observed in each case.Though non-trivial in lieu of crystallographic data, assignment of the specific isomer obtained can be achieved with recourse to spectroscopic data.Thus, for each of 4 and 5 a notable coupling (JFP ~20 Hz) is apparent between the CF3 substituent and the phosphacarbon only; this would imply proximity of the two moieties, strongly suggesting that the bulkier CF3 substituent (cf.H or Me) is more favorably positioned away from the sterically encumbered metal center.This is supported by significant deshielding of the phosphacarbon center, consistent with proximity to an electron withdrawing substituent.Indeed, we find the resonant frequency of the phosphacarbon center to be very sensitive to the pyrazolyl substituents (vide infra), most significantly so for that at the proximal site.Assignment of the isomeric preference in 6a and 6d is a more complex undertaking, given the lack of any direct spectroscopic handle, though the same steric arguments can reasonably be applied.A more electron-withdrawing character is apparent, relative to pyrazole itself (δP 64.4 6a, cf.58.7 2a), which conforms to expectation and is again consistent with the precedent systems, though the modest level of deshielding is inadequate to unequivocally confirm substituent proximity to the phosphacarbon.Nonetheless, we believe this to be the most likely scenario.

Spectroscopic Features and Trends
The key NMR spectroscopic data for all compounds 2, 3, 4 and 5, along with 6a and 6d are summarized in Table 1.As one would anticipate, the pz Me2 derivatives impart significant shielding to the phosphacarbon "P" center, which resonates around 25 ppm to lower frequency than for the pz systems; conversely, the presence of an electron-withdrawing CF3 substituent leads to appreciable deshielding of this site (ΔδP ~+20), though this is moderated in the pz Me,CF3 derivatives by competitive induction from the methyl.All complexes featuring a methyl substituent proximal to the metal also exhibit additional shielding of one PPh3 ligand (ΔδP −3) compared to their non-alkylated analogues; significantly, the extent of this effect is quantitatively comparable whether the second substituent is methyl or trifluoromethyl, which would imply a localized effect, and also offers further support for the isomeric assignment.The presence of a methyl proximal to the metal also leads to an appreciable decrease in the magnitude of the alkenic C-H coupling constant.This implies reduced s-character in this linkage, consistent with an increased contribution from dπ→π * (CP) retrodonation (the result of increased electron density at the metal) and thus greater pyramidalization; indeed, this concurs with structural data for 3d (vide supra).
Influence of the pyrazole upon the alkenic carbon center is also apparent, but exhibits a less defined trend; in contrast, the nature of the silyl substituent clearly holds significance, the SiMe3 derivatives being notably more deshielded than their SiMe2Ar analogues, distinctions between the latter being ill defined.This is consistent with 13 C-NMR studies of vinyl- [49] and aryl-silanes [50], for which enhanced deshielding of the α-carbon follows from increasing methyl substitution at silicon, presumably reflecting the relative "inductive effect", given the greater electronegativity of carbon relative to silicon.While carbocentric systems show the reverse shielding trend for the β-center, in the case of 2-6 enhanced deshielding of phosphorus is again noted for the SiMe3 Assignment of the isomeric preference in 6a and 6d is a more complex undertaking, given the lack of any direct spectroscopic handle, though the same steric arguments can reasonably be applied.A more electron-withdrawing character is apparent, relative to pyrazole itself (δ P 64.4 6a, cf.58.7 2a), which conforms to expectation and is again consistent with the precedent systems, though the modest level of deshielding is inadequate to unequivocally confirm substituent proximity to the phosphacarbon.Nonetheless, we believe this to be the most likely scenario.

Spectroscopic Features and Trends
The key NMR spectroscopic data for all compounds 2, 3, 4 and 5, along with 6a and 6d are summarized in Table 1.As one would anticipate, the pz Me2 derivatives impart significant shielding to the phosphacarbon "P" center, which resonates around 25 ppm to lower frequency than for the pz systems; conversely, the presence of an electron-withdrawing CF 3 substituent leads to appreciable deshielding of this site (∆δ P ~+20), though this is moderated in the pz Me,CF3 derivatives by competitive induction from the methyl.All complexes featuring a methyl substituent proximal to the metal also exhibit additional shielding of one PPh 3 ligand (∆δ P −3) compared to their non-alkylated analogues; significantly, the extent of this effect is quantitatively comparable whether the second substituent is methyl or trifluoromethyl, which would imply a localized effect, and also offers further support for the isomeric assignment.The presence of a methyl proximal to the metal also leads to an appreciable decrease in the magnitude of the alkenic C-H coupling constant.This implies reduced s-character in this linkage, consistent with an increased contribution from d π →π * (CP) retrodonation (the result of increased electron density at the metal) and thus greater pyramidalization; indeed, this concurs with structural data for 3d (vide supra).
Influence of the pyrazole upon the alkenic carbon center is also apparent, but exhibits a less defined trend; in contrast, the nature of the silyl substituent clearly holds significance, the SiMe 3 derivatives being notably more deshielded than their SiMe 2 Ar analogues, distinctions between the latter being ill defined.This is consistent with 13 C-NMR studies of vinyl- [49] and aryl-silanes [50], for which enhanced deshielding of the α-carbon follows from increasing methyl substitution at silicon, presumably reflecting the relative "inductive effect", given the greater electronegativity of carbon relative to silicon.While carbocentric systems show the reverse shielding trend for the β-center, in the case of 2-6 enhanced deshielding of phosphorus is again noted for the SiMe 3 derivatives.Though the relatively lower electronegativity of phosphorus (cf.carbon) might reasonably be invoked in accounting for this disparity, indirect effects via the synergic metal-ligand binding mode (such complexes have not been considered for the carbocentric silanes) cannot be discounted.Indeed, it is noted that analogues of 2 and 3 featuring a tert-butyl [29], rather than silyl, substituent exert a significant shielding effect upon the phosphorus center (δ P 38.8 (pz); 14.2 (pz Me2 )), yet deshield the α-carbon nucleus (δ C 81.6 (pz), 79.8 (pz Me2 )) even more effectively than does SiMe 3 , which would appear counter-intuitive on the basis of purely electronegativity arguments.The relative contributions of ligand→metal σ donation, and d π →π* (CP) retrodonation for each center must thus be considered contributory, though inadequate data are currently available to quantify this.