The Influence of Substituents in Diphosphine Ligands on the Hydrogenation Activity and Selectivity of the Corresponding Rhodium Complexes as Exemplified by ButiPhane

Abstract

The influence of substituents in ButiPhane ligands (with R=Me, Et, iPr) on the hydrogenation activity and selectivity of the corresponding rhodium complexes has been quantitatively assessed. Increasingly demanding substituents negatively affect the reduction rate of 1,5-cyclooctadiene in catalyst precursors en route to catalytically active MeOH–solvent complexes. The same trend as to activity is observed in the hydrogenation of prochiral olefins [methyl-(Z)-α-acetamidocinnamate, methyl (Z)-3-N-acetylamino-3-methylacrylate, and methyl (Z)-3-N-acetylamino-3-phenylacrylate], which however is counterbalanced by an increase in the stereorecognition ability of the bulkier catalysts.

Introduction

Among the plethora of ligands designed for asymmetric homogeneous catalysis, some, such as the DuPhos introduced by Burk (Scheme 1),1 stand out as especially successful and are therefore recognized as “privileged ligands.”2

Scheme 1. DuPhos and analogue ligands.

One main feature of the DuPhos ligands is the very high selectivity they provide in asymmetric hydrogenation.

Furthermore, in contrast to what is observed with other catalysts that bear five-membered ring chelates, DuPhos complexes retain their high activity even when the hydrogenation is performed under normal pressure. They possess a modular structure that can be easily modified by varying both the phospholane ring and the PP bridge through the introduction of diverse substituents. Consequently, it is not surprising that the successful basic structure of the DuPhos family is also present in other ligand classes such as RoPhos,3 BasPhos,4 catASiumM,5 UlluPhos,6 and ButiPhane7 (Scheme 1).

For each class of ligands, a higher degree of substitution seems to be beneficial for selectivity. In the hydrogenation of furane derivatives using ButiPhane ligands also, reactivity improves when the steric bulk of substituents on the phospholane ring increases.8 However, this trend of activity is not general: For example, in the hydrogenation of COD (COD=1,5-cyclooctadiene) in methanol under normal pressure, the pseudo rate constant9 measured for ligand Et-DuPhos is 0.028 min⁻¹, lower than that measured for ligand Me-DuPhos, which is 0.115 min⁻¹.10

A similar behavior as to activity can be observed in the hydrogenation of methyl-(Z)-α-acetamidocinnamate (mac) and benzoyl-(Z)-α-acetamidocinnamate (bac): Although the hydrogenation of mac with [Rh(Et-DuPhos)(MeOH)₂]BF₄ under standard conditions11 proceeds under kinetic control, with [Rh(Me-DuPhos)(MeOH)₂]BF₄ the hydrogenation is strongly affected by diffusion.12 This is not the case with bac, the hydrogenation of which is slower with both ligands although Me-DuPhos again provides the most active of the two catalysts (see Figures S3 and S4 in the Supporting Information). On the basis of these observations, we set out to investigate the influence of the steric bulk of substituents on the phospholane ring of ButiPhane ligands on the activity and selectivity of the corresponding hydrogenation catalysts. Three ligands were considered in the following study: methyl-, ethyl-, and isopropyl-substituted ButiPhanes.

Results and Discussion

X-ray structure and NMR data of diolefin precatalysts

Complexes [Rh((R,R)-Me-ButiPhane)(COD)]BF₄ (1), [Rh((R,R)-Et-ButiPhane)(COD)]BF₄ (2), and [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ (3)13 are commercially available and were supplied by Solvias AG. The ³¹P NMR data of the complexes are summarized in Table 1.

Table 1. ³¹P NMR data of COD complexes 1–3^[a] and selected literature data.^[b]

Complex	^31P NMR δ [ppm]		^2J(P,P) [Hz]	^1J(P,Rh) [Hz]
[a] ^31P NMR (CD3OD, 122 MHz); [b] ^31P NMR (CDCl3, 122 MHz). See Ref. 7.
[Rh((R,R)-Me-ButiPhane) (COD)]BF4 (1)	63.05	64.28	25.4	148.5	149.6
[Rh((R,R)-Et-ButiPhane) (COD)]BF4 (2)	55.45 (56.16)^[b]	56.67 (56.28)^[b]	23.5 (23)^[b]	147.5 (145.7)^[b]	149.1 (147.7)^[b]
[Rh((R,R)-iPr-ButiPhane) (COD)]BF4 (3)	50.19 (50)^[b]	52.07 (51)^[b]	21.0	228.1	229.7

Crystals of [Rh((S,S)-Me-ButiPhane)(COD)]BF₄ and of the BH₃ adduct of ligand (R,R)-Me-ButiPhane suitable for X-ray analysis were successfully grown (see Figures 1 and 2 as well as the Supporting Information).

Figure 1. X-ray structure of complex [Rh((S,S)-Me-ButiPhane)COD]BF₄. The anion BF₄⁻ and hydrogen atoms of the ligands are omitted for clarity (ORTEP, 30 % probability ellipsoids). Selected bond lengths (Å) and angles (°): Rh1P1=2.256(1); Rh1P2=2.302(1); Rh1C_M=2.134(1), 2.146(3); P1Rh1P2=85.13(4); C_MRh1C_M=84.77(1), where C_M is the centroid of the double bond of the COD.

Figure 2. X-ray structure of the ligand-BH₃ adduct (R,R)-Me-ButiPhane-BH₃. The hydrogen atoms are omitted for clarity (ORTEP, 30 % probability ellipsoids). Selected bond lengths (Å): P1C1=1.831(2); P2C2=1.817(2); P2B1=1.924(2), 2.146(3); C2P2B1=109.58(6).

The X-ray structures of [Rh((S,S)-Me-ButiPhane)(NBD)]BF₄,14 [Rh((S,S)-Et-ButiPhane)(COD)]BF₄, (S,S)-Et-ButiPhane,7 and mono-oxidized Me-ButiPhane15 have already been described in the literature.

Precatalyst hydrogenation: Generation of active catalytic species

In the catalytic hydrogenation of mac with complex 2, the rate initially increases as more of the active species becomes available in solution due to the progressive reduction of the diolefin in the precatalyst (Figure 3, very first part of the red curve).10, 16

Figure 3. Hydrogenation of mac with [Rh((R,R)-Et-ButiPhane)(COD)]BF₄ (2) (99.0 % ee, red) and with [Rh((R,R)-Et-ButiPhane)(MeOH)₂]BF₄ (99.0 % ee, blue) under standard conditions.11

These induction periods, which have been also described qualitatively by others,17 considerably complicate a comparison of the activity of various catalysts and the kinetic analysis of the hydrogen consumption curve.

To overcome this problem, the corresponding solvent complexes are more conveniently used: in so doing, the maximum intrinsic activity of the catalyst becomes accessible, an approach already implemented by Halpern et al.18

The practical problem then arises of determining the prehydrogenation time necessary to remove all diene complexes from the precatalyst.

As a result of the generally high stability constants of the diene complexes, the stoichiometric hydrogenation under isobaric conditions can be described as a pseudo-first-order reaction. In the presence of excess diene—Michaelis–Menten kinetics in the saturation range—it is possible to determine the rate constant of this reaction from the slope of the linear part of the hydrogen uptake curve.10, 16a, e

However, with reaction times greater than 10 h, this method does not provide reliable rate constants, as already noted in Ref. 16e: Reproducibility might be affected by the difficulty of securing rigorous anaerobic conditions for such a long reaction time.

Nevertheless, preliminary results (see Figure S1 in the Supporting Information) show that the hydrogenation rate of COD is negatively affected by the steric bulk of the substituents on the phospholane ring.

The stoichiometric hydrogenation of diolefin complexes can be more effectively followed by UV/Vis spectroscopy: This method, which is described in detail in Ref. 10, is more reliable when very low pseudo rate constants have to be measured.

Figure 4 shows the UV/Vis reaction spectrum of the stoichiometric hydrogenation of [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ (3). Several isosbestic points are visible that correspond to a kinetically uniform reaction. Accordingly, extinction diagrams show straight lines (see Figure S2 in the Supporting Information).19 Measured experimental extinction versus time data and calculated ones for a pseudo-first-order reaction at selected wavelengths are displayed in Figure 5.

Figure 4. UV/Vis reaction spectra for the stoichiometric hydrogenation of 3.236×10⁻³ mmol of [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ (3) under otherwise standard conditions11 with a cycle time of 220 s.

Figure 5. Comparison of spectroscopic values (points) and values fitted as pseudo-first order (solid line) for several wavelengths (cf. Figure 4<xfigr4

Table 2 reports the pseudo rate constants for COD hydrogenation with complexes 1–3, as determined from the analysis of UV/Vis reaction spectra with the Specfit^®[20] software package (stoichiometric) and from the analysis of hydrogenation consumption curves (catalytic). Despite the worse reproducibility of the pseudo rate constants obtained from the catalytic hydrogenation runs, there is a good agreement between the two set of values.

Table 2. Rate constants of the pseudo-first-order diolefin hydrogenation with complexes 1–3 as determined from stoichiometric (UV/Vis) and catalytic hydrogenations and half-life time values.

Complex	k2diolefin [min^−1] (stoichiometric; UV/Vis)^[a]	k2diolefin [min^−1] (catalytic; hydrogenation)	t1/2 [min]	Diolefin hydrogenation [min] (99.2 % conversion)^[b]
[a] Average of at least three measurements; [b] calculated from rate constants measured from UV/Vis reaction spectra.
1	0.119	0.100	5.8	40.8
2	0.031	0.029	22.3	156.5
3	0.009	0.010	77.0	539.1

The differences in diolefin hydrogenation rates are quite high within the ButiPhane family and show a similar trend as observed within the DuPhos ligand class. The half-life time for the conversion of the precatalyst into the active species is about 13 times greater for the sterically demanding isopropyl-substituted complex (3) than for the methyl-substituted one (1).

A very interesting possibility to process the UV/Vis reaction spectra is offered by the so-called factor analysis. In this analysis, the spectra of each reaction partner and thus their concentration also as a function of time are accessible merely by mathematical elaboration of experimental reaction spectra—without the need for any a priori information.

If the Bouguer–Lambert–Beer law is applied, reaction spectra can be described as a linear superposition of the absorbances of the participating species in the form of a matrix. A low rank factorization of this matrix followed by a proper transformation of the matrix factors finally provides the desired information.21

Concrete applications with several mathematical algorithms and programs have been described. A representative application is the band target entropy minimization (BTEM) of Garland et al., especially used for IR spectroscopy.22

In the course of our investigations, we have tested the recently described pure component decomposition (PCD) method23 with the following results.

Figure 6 illustrates the spectra of each reaction component (diolefin and solvent complex with the iPr-ButiPhane ligand) obtained by decomposition of the reaction spectrum from Figure 4. They are superimposable on the spectra obtained from separate solutions of [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ and [Rh((R,R)-iPr-ButiPhane)(MeOH)₂]BF₄. The pseudo rate constant for COD hydrogenation with complex 3 calculated from the time-dependent pure component concentration curve obtained with PCD (Figure 7) perfectly reproduces the corresponding one in Table 2. This result nicely illustrates the potential of quantitative spectra decomposition.

Figure 6. Pure component spectra obtained by PCD decomposition of the reaction spectrum shown in Figure 4.

Figure 7. Rate constant of COD hydrogenation with precatalyst (3) by PCD decomposition of time-dependent pure component spectra.

As a control experiment, complex 1 was prehydrogenated for 14 min. A pseudo rate constant of 0.119 L min⁻¹ (as reported in Table 2) implies that, after this time, besides the solvent complex, 19 % of the COD precursor should still be present in solution. This was nicely confirmed by a ³¹P NMR spectrum in Figure 8. These results show that comparing the activities of different catalysts generated in situ might be misleading even when similar ligands are employed. Generation of the active catalytic species through hydrogenation of the COD diolefin can have a very different impact on the activity of each catalyst despite the similarity of the ligands.

Figure 8. ³¹P NMR spectrum of a solution of 0.01 mmol (1) in 1 mL CD₃OD, prehydrogenated under 1 bar total pressure for 14 min at 25 °C.

Asymmetric hydrogenation of prochiral olefins

Several examples of hydrogenations that proceed according to Michaelis–Menten kinetics [Eqs. (1) and (2)] have been described in the literature. Depending on the stability of the catalyst substrate adducts (catS), first-order reactions in substrate concentration (low stability) or zero-order reactions (high stability) may result as limiting cases.18, 24–26

• 

Hydrogenation of mac, methyl (Z)-3-N-acetylamino-3-methylacrylate (Z-4),27 and methyl (Z)-3-N-acetylamino-3-phenylacrylate (Z-5)28 (see Scheme 2) was performed in methanol under standard conditions11 by using the solvate complexes derived from complexes 1–3. The results are summarized in Table 3.

Scheme 2. Prochiral substrates.

Table 3. Enantioselectivity, pseudo rate constants, and Michaelis–Menten constants for the hydrogenation of mac, Z-4, and Z-5 with ButiPhane complexes 1–3 under standard conditions.11

	1			2			3
	% ee	k2 [min^−1]	KM [mol l]	% ee	k2 [min^−1]	KM [mol l]	% ee	k2 [min^−1]	KM [mol l]
[a] The change of the absolute configuration of the iPr–ButiPhane complex 3 compared to the Me- or Et-ButiPhane complexes 1 and 2 is only apparent and stems from a change in priority of the substituents in the ligand according to the Cahn–Ingold–Prelog (CIP) system rules; [b] because the initial rate is about 12 mL min^−1, the standard equipment used does not allow to exclude the influence of diffusion on the recorded data.25
mac	96.0	48.0^[b]	1.33×10^−3	99.0	14.4	1.33×10^−3	98.5	1.3	6.66×10^−3
Z-5	80.3	13.2	1.33×10^−3	85.2	1.9	1.33×10^−3	86.6	0.25	3.33×10^−3
Z-4	92.2	0.59		91.5	0.13		94.0	0.005
		(first order)			(first order)			(first order)

The enantioselectivities observed in the hydrogenation of mac and Z-4 with the Me-ButiPhane catalyst (1) are in good agreement with those reported in the literature under similar experimental conditions.7 As a general trend, bulkier substituents on the phospholane ring of the ligand improve the stereorecognition ability of the catalyst. The activity however drops, in line with the behavior shown in the hydrogenation of COD.

The hydrogenation of Z-4 with all three ButiPhane complexes is first order in substrate concentration. Thus, the equilibria that lead to the formation of the catalyst substrate adducts are shifted toward the solvent complexes. In contrast, with mac and Z-5, reactions are zero order under the described experimental conditions (compare Figure 3, blue curve).

A zero-order reaction as a limiting case of Michaelis–Menten kinetics implies the existence of stable catalyst–substrate complexes under stationary conditions. This is reflected in the relatively low values of K_M (Table 3). Consistently, all four stereoisomeric catalyst–substrate complexes expected for the system Z-5/3 were detected by means of ³¹P NMR spectroscopy at room temperature (Figure 9) while the corresponding solvent complex (Table 4) was not observed.

Figure 9. ³¹P NMR spectrum of a solution of [Rh((R,R)-iPr-ButiPhane) (MeOH)₂]BF₄ and Z-5 at room temperature.

Table 4. ³¹P NMR data of solvent complexes^[a] and substrate complexes.^[b]

Complex	^31P NMR δ [ppm]		^2J(P,P) [Hz]	^1J(P,Rh) [Hz]
[a] ^31P NMR (CD3OD, 122 MHz). [b] ^31P NMR (CD3OD, 162 MHz).
[Rh((R,R)-Me-ButiPhane)(MeOH)2]BF4	88.27	89.96	47.7	202.1	209.6
[Rh((R,R)-Et-ButiPhane)(MeOH)2]BF4	81.61	83.31	46.4	203.7	209.5
[Rh((R,R)-iPr-ButiPhane)(MeOH)2]BF4	77.02	80.77	44.1	450.5	461.5
[Rh((R,R)-iPr-ButiPhane)(MeOH)2]BF4 +Z-5 (A)	55.17	64.14	29.8	168.7	165.6
[Rh((R,R)-iPr-ButiPhane)(MeOH)2]BF4 +Z-5 (B)	49.89	60.94	29.4	162.9	161.7
[Rh((R,R)-iPr-ButiPhane)(MeOH)2]BF4 +Z-5 (C)	57.96	65.74	33.3	170.9	159.9
[Rh((R,R)-iPr-ButiPhane)(MeOH)2]BF4 +Z-5 (D)	63.63	66.10	35.1	165.4	160.1

The NMR data of solvent complexes and substrate complexes are summarized in Table 4.

In principle, the high stability of the catalyst–substrate complex should facilitate its crystallization. Crystals suitable for X-ray diffraction analysis were obtained for one of the four possible complexes of [Rh((R,R)-iPr-ButiPhane)(methyl(Z)-3-N-acetylamino-3-phenylacrylate)]BF₄; its molecular structure is shown in Figure 10. The substrate coordinates to rhodium through the CC double bond and the amide oxygen, a common fashion for different α-dehydroamino acid29 and β-dehydroamino acid.30

Figure 10. X-ray structure of substrate complex Rh((R,R)-iPr-ButiPhane)(Z-5)]BF₄. The anion BF₄⁻ and hydrogen atoms of the ligands are omitted for clarity (ORTEP, 30 % probability ellipsoids). Selected bond lengths (Å) and angles (°): Rh1P1=2.279(2); Rh1P2=2.257(1); Rh1O1=2.122(3); Rh1C_M=2.072(5); P1Rh1P2=86.16(5); O1Rh1C_M=84.73(9), where C_M is the centroid of the double bond of the substrate (Z-5).

A reliable correlation between the X-ray structure (Figure 10) and one of the four sets of NMRs (Figure 9) is not easy to provide. This would be possible only if the equilibration between the two diastereomeric species could be frozen out at very low temperature. However, because of the C1 symmetry of the ligand, any mechanistic interpretation is speculative and therefore no further investigations were performed.

Concerning the stability, a similar phenomenon as that of the well-known inhibiting properties of aromatic solvents (e.g., benzene and toluene) for the asymmetric hydrogenation could be observed.31 The investigations of stability constants for ButiPhane–arene complexes were performed by using UV/Vis spectroscopic titration, whereas the constants have been calculated through linearization (see Figures S5–S7 in the Supporting Information).32 A trend with a decreasing stability of the corresponding benzene complexes in a row from Me-ButiPhane (51 L mol⁻¹l)–Et-ButiPhane (39 L mol⁻¹)–iPr-ButiPhane (5 L mol⁻¹) could be determined.

Conclusions

In summary, it could be shown that with ButiPhane rhodium complexes 1–3, the hydrogenation rate of the diolefin to generate the active catalytic species and the stability of benzene complexes are negatively affected by the steric bulk of the chiral ligand. Monitoring of the stoichiometric diolefin hydrogenation by means of UV/Vis spectroscopy is a suitable method to measure the corresponding pseudo rate constants. A presuppositionless analysis of reaction spectra—even as a function of time—with the PCD program allows us to easily extract the spectra of each reaction component and to obtain the desired pseudo rate constants.

In the hydrogenation of prochiral olefins (mac, Z-4, and Z-5), improved selectivities were obtained by increasing the steric bulk of the substituents on the phospholane ring of the chiral ligand albeit at the expense of a major drop in activity.

Experimental Section

General

All manipulations were performed by using standard Schlenk techniques under argon.33 MeOH was distilled from Mg under argon prior to use.

NMR spectra were recorded by using a Bruker ARX-300 spectrometer or ARX-400 spectrometer (300 or 400 MHz for ¹H) at 297–298 K. Chemical shifts were reported in parts per million (ppm) referenced to the deuterated solvent for ¹H NMR and to 85 % H₃PO₄ as an external standard for ³¹P NMR.

The enantiomeric excess (ee) of hydrogenated methyl-(Z)-α-acetamidocinnamate (mac) and methyl (Z)-3-N-acetylamino-3-methylacrylate (Z-4) was analyzed on a GC System 6890 (Agilent) with a 25 m Lipodex E column (mac) and a 50 m Chiraldex β-PM column (Z-4) at 110 °C. The enantiomeric resolution of the obtained hydrogenated substrate Z-5 on an HPLC-System 1100 (Agilent) was successful on the chiral stationary phase Knauer Eurocell01 with an eluent hexane/ethanol 98:2.

Crystal data and details of the structure solution are summarized in Table S1 in the Supporting Information. Diffraction data were collected by using a STOE-IPDS II diffractometer using graphite monochromated MoKα radiation. The structure was solved by using direct methods (SHELXS-97)34 and refined by using full matrix least-squares techniques against F² (SHELXL-97).34 XP (Siemens Analytical X-ray Instruments, Inc.) was used for structure representations.

The nonhydrogen atoms, except the not partially occupied atoms of the solvents, were refined anisotropically. The hydrogen atoms were placed into theoretical positions and were refined by using the riding model. The weighting schemes used in the last cycles of refinement are ω=1/[σ²(Fo²)+(0.0363P)²+0.0000P] for [Rh((R,R)-iPr-ButiPhane)(Z-5)]BF₄, ω=1/[σ²(Fo²)+(0.0388P)²+0.0000P] for [Rh((S,S)-Me-ButiPhane)(COD)]BF₄, and ω=1/[σ²(Fo²)+(0.0402P)²+0.0532P] for (R,R)-Me-ButiPhane-BH₃.

CCDC 801864http://www.ccdc.cam.ac.uk/cgi-bin/catreq.cgi for [Rh((R,R)-iPr-ButiPhane)(Z-5)]BF₄, CCDC 801862http://www.ccdc.cam.ac.uk/cgi-bin/catreq.cgi for [Rh((S,S)-Me-ButiPhane)(COD)]BF₄, and CCDC 801863http://www.ccdc.cam.ac.uk/cgi-bin/catreq.cgi for (R,R)-Me-ButiPhane-BH₃ contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Hydrogen consumptions were monitored by using the device described in the literature.25

Complexes and characterization

The diolefin precatalysts [Rh((R,R)-Me-ButiPhane)(COD)]BF₄ (1), [Rh((R,R)-Et-ButiPhane)(COD)]BF₄ (2), and [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ (3) are commercially available and were provided by Solvias AG. They were further purified by recrystallization from methanol or a dichloromethane/diethyl ether mixture, respectively.

Single crystals suitable for X-ray diffraction analysis were obtained by recrystallization from ethylacetate/n-hexane.

2,3-Bis[(2 R,5 R)-2,5-(dimethylphospholane-1-yl)]benzo[b]thiophen-BH₃, (R,R)-Me-ButiPhane-BH₃

¹H NMR (300 MHz, CDCl₃): δ=1.01–0.87 (m, 6 H; 2 CH₃), 1.35–1.27 (m, 3 H; CH₃), 1.56–1.48 (m, 3 H; CH₃), 1.72–1.57 (m, 1 H), 1.96–1.74 (m, 2 H), 2.22–2.06 (m, 2 H), 2.48–2.22 (m, 3 H), 2.94–2.73 (m, 2 H), 3.82–3.60 (m, 1 H), 7.40–7.31 (m, 2 H; 2 H_Ar), 7.81–7.77 (m, 1 H; H_Ar), 8.35–8.32 ppm (m, 1 H; H_Ar); ¹³C NMR (76 MHz, CDCl₃): δ=14.9, 15.6 (d, J(P,C)=2.0 Hz), 17.2 (d, J(P,C)=4.4 Hz), 20.3 (d, J(P,C)=35.3 Hz), 31.8 (d, J(P,C)=23.3 Hz), 32.2 (d, J(P,C)=23.2 Hz), 32.8 (d, J(P,C)=9.5 Hz), 33.5 (d, J(P,C)=8.2 Hz), 34.8, 35.3, 36.5 (d, J(P,C)=12.3 Hz), 37.0 (d, J(P,C)=1.8 Hz), 37.4 (d, J(P,C)=3.7 Hz), 41.6 (d, J(P,C)=8.8 Hz), 121.6, 124.2, 125.1, 125.4, 141.7, 142.6 ppm; ³¹P NMR (122 MHz, CDCl₃): δ=6.32 (s; P), 43.84 ppm (d, ¹J(P,B)=64.0 Hz; P-BH₃).

{2,3-Bis[(2S,5 R)-2,5-dimethylphospholane-1-yl]benzo[b]thiophene}(1,5-cyclo-octadiene)rhodium(I) tetrafluoridoborate, [Rh((R,R)-Me-ButiPhane)(COD)]BF₄ (1)

¹H NMR (300 MHz, CD₃OD): δ=1.08–0.94 (dd, ³J(H,H)=7.2 Hz, ³J(P,H)=15.6 Hz, 3 H; CH₃), 1.17–1.10 (dd, ³J(H,H)=7.2 Hz, ³J(P,H)=15.6 Hz, 3 H; CH₃), 1.53–1.36 (2 dd, ³J(H,H)=7.2 Hz, ³J(P,H)=18.9 Hz, ³J(P,H)=18.5 Hz, 6 H; 2 CH₃), 1.70–1.57 (m, 1 H), 1.98–1.81 (m, 1 H), 2.21–2.08 (m, 1 H), 2.58–2.26 (m, 14 H), 2.81–2.60 (m, 2 H), 3.19–3.11 (m, 1 H; CHP), 4.92–4.81, 5.20–5.08, 5.59–5.49, 5.81–5.72 (4 m, 1 H each; COD-CH), 7.54–7.44 (m, 2 H; 2 H_Ar), 7.97–7.94 (m, 1 H; H_Ar), 8.05–8.02 ppm (m, 1 H; H_Ar); ³¹P NMR (122 MHz, CD₃OD): δ=64.28 (ddd, ²J(P,P)=25.4 Hz, ¹J(P,Rh)=148.5 Hz; P), 63.05 ppm (ddd, ¹J(P,Rh)=149.6 Hz; P).

{2,3-Bis[(2S,5 R)-2,5-diethylphospholane-1-yl]benzo[b]thiophene}(1,5-cyclo-octadiene)rhodium(I) tetrafluoridoborate, [Rh((R,R)-Et-ButiPhane)(COD)]BF₄ (2)

¹H NMR (300 MHz, CD₃OD): δ=0.77 (t, ³J(H,H)=7.4 Hz, 3 H; CH₃), 0.94 (t, ³J(H,H)=7.4 Hz, 6 H; 2 CH₃), 1.10 (t, ³J(H,H)=7.4 Hz, 3 H; CH₃), 1.34–1.21 (m, 2 H), 1.59–1.42 (m, 4 H), 2.22–1.77 (m, 7 H), 2.65–2.25 (m, 14 H), 3.19–3.03 (m, 1 H; CHP), 4.81–4.69, 5.12–5.00, 5.57–5.47, 5.81–5.72 (4 m, 1 H each; COD-CH), 7.54–7.44 (m, 2 H; 2 H_Ar), 7.97–7.94 (m, 1 H; H_Ar), 8.05–8.02 ppm (m, 1 H; H_Ar); ³¹P NMR (122 MHz, CD₃OD): δ=55.45 (ddd, ²J(P,P)=23.5 Hz, ¹J(P,Rh)=147.5 Hz; P), 56.67 ppm (ddd, ¹J(P,Rh)=149.1 Hz; P).

{2,3-Bis[(2S,5 R)-2,5-diisopropylphospholane-1-yl]benzo[b]thiophene}(1,5-cyclo-octadiene)rhodium(I) tetrafluoridoborate, [Rh((R,R)-iPr-ButiPhane)(COD)]BF₄ (3)

¹H NMR (300 MHz, CD₃OD): δ=0.57 (d, ³J(H,H)=6.6 Hz, 3 H; CH₃), 0.72 (d, ³J(H,H)=6.6 Hz, 3 H; CH₃), 0.75 (d, ³J(H,H)=6.6 Hz, 3 H; CH₃), 0.82 (d, ³J(H,H)=6.6 Hz, 3 H; CH₃), 1.02–0.99 (m, 3 H; CH₃), 1.11–1.06 (m, 6 H; 2 CH₃), 1.43 (d, ³J(H,H)=6.6 Hz, 3 H; CH₃), 1.87–1.48 (m, 4 H), 2.15–1.96 (m, 5 H), 2.60–2.18 (m, 14 H), 3.17–3.01 (m, 1 H; CHP), 5.00–4.89, 5.20–5.09, 5.57–5.48, 5.76–5.67 (4 m, 1 H each; COD-CH), 7.54–7.44 (m, 2 H; 2 H_Ar), 7.96–7.93 (m, 1 H; H_Ar), 8.11–8.08 ppm (m, 1 H; H_Ar); ³¹P NMR (122 MHz, CD₃OD): δ=50.19 (ddd, ²J(P,P)=21.0 Hz, ¹J(P,Rh)=304.1 Hz; P), 52.07 ppm (ddd, ¹J(P,Rh)=306.3 Hz; P).