Indomuscone-Based Sterically Encumbered Phosphines as Ligands for Palladium-Catalyzed Reactions

Abstract

The fragrance compound indomuscone is used here as a scaffold to prepare two different sterically hindered phosphines, one aromatic and another alkylic, in good yields, after four synthetic steps. The new phosphines show enhanced electronic and steric properties when compared to benchmark commercial phosphine ligands, which is reflected in the catalytic results obtained for representative palladium-catalyzed reactions such as the telomerization reaction, the Buchwald–Hartwig and Suzuki cross-coupling reactions of chloroaromatic rings, and the semi-hydrogenation reaction of an alkyne. In particular, the indomuscone-based aromatic phosphine ligand leads to the highest selectivity for the tail-to-head telomerization product between isoprene and methanol, while the indomuscone-based alkylic phosphine ligand shows extraordinary similarities with the Buchwald-type SPhos phosphine ligand.

Introduction

Phosphines are very active ligands for palladium-catalyzed organic reactions, which include cross-coupling reactions,¹ hydrogenations,² and hydro(alkoxy)formylations,^2,3 to name a few.⁴ In particular, sterically encumbered phosphines are considered privileged ligands in palladium-catalyzed reactions such as the telomerization reaction,⁵ the Buchwald–Hartwig⁶ and Suzuki⁷ cross-coupling reactions, the semi-hydrogenation reaction of alkynes,⁸ and others.⁹ The rationale behind the high catalytic activity of these phosphines with palladium lies, on one hand, on the protection of the metal site during the catalytic cycle, which avoids typical palladium deactivation processes such as reduction or aggregation, and, on the other hand, on the formation of ipso carbon-palladium bonds that pre-activates the catalytic site for reagent coordination.¹⁰ In addition, the constrained local environment provided by the phosphine around the metal site directs the selectivity of the reaction toward the desired product in many cases, and the steric shield around the P atom increases the stability toward air, which are both additional advantages in catalysis.¹

Figure 1 shows typical phosphine ligands (1a–1d) and representative examples of sterically hindered phosphines (1e–1i), most of them currently under industrial use as ligands for palladium-catalyzed reactions. It can be seen that a common, prominent feature of these ligands is the combined use of cyclohexyl and aromatic rings with methyl multisubstituted tertiary and quaternary carbon atoms (i.e., isopropyl groups), which, together, generate the desired sterically crowded structure around the palladium site. It is worthy commenting here that small structural changes in the phosphine produce dramatic changes in the final catalytic activity of the corresponding palladium complex.^1,6 For all these reasons, the search for new sterically hindered phosphines is of much interest, and even machine learning is being used for that.¹¹ However, as new phosphines are discovered, more sophisticated starting materials and synthetic routes are required, which translates into long and expensive synthetic protocols. In this way, the final price of the ligand can be even superior to the palladium atom itself. Thus, the search for precursors of phosphine ligands that contain new sterically encumbered structural features and, at the same time, are cheap and widely available is of interest from both scientific and economic point of views. Here, we report two new phosphine ligands based on a widely available and cheap fragrance compound, as shown in Figure 1 (2a and 2b).

Figure 1.

Commercially available phosphines used in metal catalysis as ligands (1a–1i), sterically encumbered or not, and the phosphine ligands reported in this work (2a and 2b).

Results and Discussion

Synthesis of Phosphines 2a and 2b

The synthesis of the sterically encumbered phosphines 2a and 2b is shown in Figure 2. The precursor is the fragrance compound indomuscone 3 (commercial name Cashmeran), which is produced in multiton amounts per year with a price <0.05 $/g. This widely available compound is easily transformed in two simple steps to the polycyclic compound 5 in good yields, either through the methallyl 4a or the hydroxyl 4b derivative. Both intermediates are obtained after deprotonation in the α-position of 3 with solid sodium amide (NaNH₂) and alkylation of the in situ-generated enolate with the corresponding electrophiles. Any other deprotonation agent tested, including sodium methoxide (NaOMe, solid or in THF solution), sodium or potassium tert-butoxide (Na or KO^tBu, solids or in solution), NaH, or lithium reagents were completely ineffective for the reaction. We tentatively ascribe the uniqueness of NaNH₂ as a base for this reaction to the release of NH₃ as a gas, which avoids the reversal of the equilibrium. Despite 4b being produced with complete selectivity and 29% of starting material could be recovered after reaction by column chromatography separation, the lower price of the methallyl electrophile made us to choose this reaction for scaling up.

Figure 2.

Synthesis of phosphines 2a and 2b. Yields refer to isolated products.

The aromatization step to generate 5 can be carried with different acid catalysts, including typical soluble organic acids such as p-toluenesulfonic acid (pTsOH) and methylsulfonic acid (MSA), and also solid acids such as zeolites and sulfonic resins (see Tables S1 and S2 in the Supporting Information), achieving a similar yield of 70% for pTsOH and H-Beta zeolite under the optimized conditions (Tables S2–S6 and Figure S1). Toluene is the best solvent for the transformation (Tables S5 and S6). The reaction is run under “open flask” conditions to use air as an oxidant for the dehydrogenation step, which gives access to the desired aromatic ring. Otherwise, the reaction stops in the corresponding dienes, detected by GC–MS, which suffer from isomerizations and other undesired transformations.

Compound 5 features a sterically shielded indane ring that can be easily brominated in the ortho position to the penta-substituted methyl cyclopentane cycle, by the inductive action of the ether substituent in the para position, to yield 6 in high yield in just a 15 min reaction time at room temperature. Multigram amounts of compound 6 were easily obtained as yellowish crystals after work-up and removal of volatiles under vacuum, without requiring any crystallization technique, and Figure 3 shows the resolved crystalline structure by monocrystal X-ray diffraction (XRD, see the Supporting Information for a summary of crystallographic data). The sterically crowded environment imparted by the aromatic indomuscone core can be seen together with the Br atom substitution, which, in principle, leaves room for the introduction of a phosphine moiety.

Figure 3.

Crystal structure of the intermediate bromoderivative 6 obtained by single-crystal XRD. Ellipsoids represent a 30% probability. Color code: Br in brown, O in red, C in blue, and H in white.

Thus, with multigram quantities of 6 in hand, we proceeded to couple 6 with two selected chlorophosphines of general structure ClPR₂, in this case, the representative phenyl (R = Ph) and cyclohexyl (R = c-hex) substituent groups, under typical reaction conditions.¹² For the former, the final targeted phosphine 2a was isolated by column chromatography in moderate yields and, as precursor 6, spontaneously crystallized at room temperature, to give white crystals. The overall yield after four linear steps is 16%. Figure 4 shows the crystalline structure of phosphine 2a, where the sterically encumbered coordination shell around the P atom can be observed.

Figure 4.

Crystal structure of phosphine 2a obtained by single-crystal XRD. Ellipsoids represent a 30% probability. Color code: P in orange, O in red, C in blue, and H in white.

In contrast to 2a, it was not possible to obtain crystals of 2b after isolation by column chromatography and recrystallization tests; however, the crystal structure of the oxidized form (2b-oxide) could be obtained. Figure 5 shows the crystal structure of 2b-oxide, which is similar to 2a, with the P atom surrounded by the aromatic indomuscone and the two cyclohexyl substituents groups.

Figure 5.

Crystal structure of phosphine oxide 2b-oxide obtained by single-crystal XRD. Ellipsoids represent a 30% probability. Color code: P in orange, O in red, C in blue, and H in white.

With the crystal structure in hand, the electronic and steric properties of the new phosphines were calculated and compared with the commercial phosphines 1a–e (Table S7). The electronic and steric properties were assessed by the vibrational frequency of the carbonyl stretch of the corresponding Ni(CO)₃L complex (2056.1–2073.0 cm^–1),¹³ which directly correlates to the phosphine lone pair charge density¹⁴ and the Tolman cone angle, respectively.¹³ It is worth noting to comment here that Tolman cone angle values often overestimate steric hindrance in elaborated asymmetric phosphines, bidentate ligands, and N-heterocyclic carbenes, and that other values such as percent buried volume (V_bur)^15a and ligand repulsive energy parameter (ER)^15b can be more suitable here. However, for the sake of simplification, we will use the well-established Tolman parameter in a first approximation. It can be seen that phosphine 2a shows an electronic value similar to triphenylphosphine 1a (2067 vs 2069 cm^–1) but a completely opposed steric value (195 vs 145°), 40° higher and in the range of very hindered phosphines. The phosphine 2b values are 2058 cm^–1 and 205° (inferred from 2b-oxide), which denotes a much higher electron donating capability and even higher steric hindrance than 2a.

Notice that the values for 2b are extraordinarily similar to SPhos 1e. The combined electronic and steric properties of phosphines 2a and 2b make them attractive for use as ligands in catalysis. In addition, the unique polysubstituted tricyclic structure of aromatic indomuscone may confer to the new phosphine ligands an enhanced catalytic action. Notice here that the final R groups in 2 can be chosen for a plethora of other available chlorophosphines (R₂PCl); thus, this family of phosphines could be easily widened if desired.

Synthesis of the Palladium–Phosphine 2a Complex

The titration of palladium acetate with phosphine 2a was accomplished in the methanol solvent by UV–vis and ³¹P-NMR. Figure 6 (top) shows the UV–vis results after adding from 0.5 to 4 equiv of 2a, and it can be seen that the typical broad absorption band of palladium acetate at ∼380 nm (black curve) disappears after adding two or more equivalents of 2a to give a new single absorption band at ∼340 nm (i.e. pink curve). This band is more pronounced for 2 equiv of 2a, although it is not discarded compared to 3 equiv of 2a that are filling the coordination sphere of the palladium cation, according to these spectra.

Figure 6.

Titration of palladium acetate with phosphine 2a by UV–vis (top) and ³¹P-NMR (bottom), in methanol solution.

The corresponding ³¹P-NMR measurements in Figure 6 (bottom) show that the original peak of phosphine 2a at ∼20 ppm (red line) transforms into a new single peak at ∼40 ppm when 2 equiv of 2a is added to palladium acetate (light blue line), while lower amounts of 2a give a mixture of signals. These results indicate that a 1:2 Pd:phosphine complex is being formed. Indeed, the signal at∼40 ppm remains and coexists with the original peak of phosphine 2a at ∼20 ppm when >2 equiv of 2a is added, indicating that [Pd(2a)₂]²⁺ is a stable species that does not admit a third phosphine ligand in the palladium coordination shell, in accordance with the sterically encumbered structure of 2a. Thus, we must conclude here that Pd(2a)₂(OAc)₂ is the more plausible complex formed after mixing both species.

With these results in hand, we stirred 1 equiv of palladium acetate and 2 equiv of phosphine 2a in MeOH at 50 °C for 5 min, and then, we concentrated and obtained an orange solid, in which analytical results perfectly matched those expected for Pd(2a)₂(OAc)₂, including the observed UV–vis and ³¹P NMR signals, and also ICP-OES and elemental analysis values. Particularly informative are the UPLC-HRMS results (Figure S2), where the main peak of the spectrum corresponds to the [Pd(2a)₂(OAc)]⁺ cation, after losing one of the acetates, with the associated isotopic distribution for one Pd atom and the exact mass of this cationic complex. Minor signals corresponding to free phosphine 2a, oxidized palladium complex, and complex dimers can also be observed. We could obtain some small crystals after extensive crystallization tests; however, these crystals proved unsuitable for XRD.

The corresponding study with phosphine 2b was not carried out by the tendency to oxidize of the alkyl phosphine; however, in a first approximation, we can expect a stable 1:2 Pd:phosphine complex.

Catalytic Results

Telomerization Reaction

The telomerization reaction is an industrial reaction catalyzed by palladium phosphine complexes in solution, typically not only with triphenylphosphine ligands 1a but also with other phosphines.⁵ A 1:2 Pd: phosphine catalytic ratio is very common in this reaction, and sterically encumbered phosphines have been studied.¹⁶ Thus, we tested the telomerization reaction with the new phosphine ligands 2a and 2b, and the results are shown in Table 1.

Table 1. Catalytic Results for the Telomerization Reaction of Isoprene 7 and MeOH Catalyzed by 1:2 Pd(OAc)₂-Phosphine Complex Catalysts.

entry	phosphine	MeOH (equiv)	conv. (%)a	yield to 8 (%)	ratio (8a–d)
1	1a	12	100	95	10/4/67/19
2	2a	12	100	98	3/-/95/2
3	1a	5	95	91	18/3/60/19
4	2a	5	98	92	5/-/93/2
5	1a	2	95	68	25/5/41/29
6	2a	2	96	72	7/-/88/5
7	1c	12	76	2	25/-/60/15
8	1e	12	85	5	30/-/40/30
9	2b	12	100	70	10/-/70/20

^aGC results. Mass balance completed with byproducts 9.

Isoprene 7 and MeOH were used as substrates to assess not only the catalytic activity of the different phosphine ligands but also the selectivity toward the different products, which in this case arise from the head/tail potential couplings (Figure S3, top).¹⁷ The catalytic results show that phosphine 2a outperforms triphenylphosphine 1a not only in catalytic activity but also, particularly, in selectivity, under the indicated reaction conditions (compare entries 1 and 2). The yield to the tail-to-head telomerization product 8c with ligand 2a is 94%, while 1a gives 65%. The selectivity of the reaction can be clearly seen by ¹H-NMR spectroscopy (Figure S3, bottom). Remarkably, an 85% yield of 8c is obtained when decreasing the excess of MeOH from 12 to 5 equiv with ligand 2a (entries 3 and 4) and a good yield of 8c is still obtained with just 2 equiv of MeOH, in contrast to ligand 1a (entries 5 and 6). Alkyl phosphine ligands such as 1c and 1e are barely active to produce telomerization products 8a–d, giving only byproducts 9 (entries 7 and 8). However, the alkyl phosphine ligand 2b gives a 70% yield of products 8a–d with a 70% selectivity to product 8c (entry 9), clearly improving the results with other alkyl phosphines.

Figure 7 shows the kinetic results for the telomerization reaction under optimized conditions, either adding Pd(OAc)₂ and ligand 2a by separating from the beginning in a 1:2 molar ratio or adding the isolated complex Pd(2a)₂(OAc)₂. The results show an induction time when the complex is prepared in situ and the disappearance of this induction time when the isolated complex is used, supporting that Pd(2a)₂(OAc)₂ is the active pre-catalyst of the reaction.¹⁸ It has been proposed that the Pd species in the catalytic cycle are in reduced form, and that they just have one or none phosphine ligands;^5a thus, the observed acceleration by preforming Pd(2a)₂(OAc)₂ could be due to a facile reductive elimination by the phosphine and OAc ligands.

Figure 7.

Kinetic plot for the telomerization reaction of isoprene 7 and MeOH catalyzed by 1:2 Pd(OAc)₂-phosphine complex catalysts, either adding by separating the Pd source and the ligand (gray squares) or pre-forming the complex Pd(2a)₂(OAc)₂ (red circles). GC results. Error bars account for a 5% uncertainty.

Different alcohols, except those sterically impeded such as isopropanol, engage well in the reaction, and the tail-to-head isomer is systematically the major isomer (Figure S4). These results showcase the superiority of the fragrance-based phosphines 2a and 2b for the telomerization reaction with respect to state-of-the-art phosphine ligands, at least in selectivity to the tail-to-head isomer under the reaction conditions tested.

Buchwald–Hartwig Cross-Coupling Reaction

The Buchwald–Hartwig coupling reaction is also a well-implemented reaction procedure in fine chemical synthesis to generate new carbon–nitrogen bonds, and Buchwald-type phosphines such as 1e and 1g–i are privileged ligands for this palladium-catalyzed reaction.^1,6,19Table 2 shows the catalytic results for the coupling between p-chloroanisole 10 and morpholine 11 under typical reaction conditions.^6a Chloroaromatics are among the most difficult substrates to activate in this reaction; thus, chloroderivative 10 was used as the starting material. Bis-palladium tris-dibenzylacetonate [Pd₂(dba)₃] was employed here as a palladium source since palladium acetate proved inactive (see below).

Table 2. Catalytic Results for the Coupling between p-Chloroanisole 10 and Morpholine 11 under Typical Reaction Conditions, Catalyzed by Different Pd-Phosphine Complex Catalysts.

entry	Pd (mol %)	phosphine (mol %)	conv. (%)a	select. to 12 (%)
1	2.5	1e (5)	100	98
2		1a (5)	2	100
3		1c (5)	60	90
4		2a (5)	2	50
5		2b (5)	40	99
6	5	2b (10)	80	92
7		2b (5)	38	82
8	2.5	2b (7.5)	85	98
9		2b (10)	86	97
10b		2b (7.5)	8	97
11c		2b (7.5)	38	90

^aGC results. Mass balance completed with anisole 13.

^bPd(OAc)₂ instead of Pd₂(Dba)₃.

^cPd(Dba)₂ instead of Pd₂(Dba)₃.

The catalytic results show that, as expected, SPhos ligand 1e gives a 98% yield for the coupling product 12 (entry 1), while the pure aromatic triphenylphosphine ligand 1a is merely inactive for the reaction (entry 2), and alkyl phosphine 1c gives moderate yields of 12 (54%) with good selectivity (90%, entry 3). In accordance, we found that the aromatic phosphine 2a is completely inactive for the reaction (entry 4) while the alkyl phosphine 2b is moderately active (40%) and completely selective toward 12 (99%, entry 5). Further optimization for ligand 2b shows that, by increasing the amount of Pd₂(dba)₃ and ligand 2b, high yields of 12 can be obtained (>80%, entries 6–9). Palladium acetate is barely active for the reaction (entry 10), while palladium dibenzylacetonate gives lower yields (35%, entry 11).

The results above indicate that phosphine 2b can be used as a ligand for the Buchwald–Hartwig coupling of chloroaromatics and morpholine 11, under optimized reaction conditions. Indeed, other chloroderivatives engage well in the reaction (Figure S5). These results, together, confirm the potential use of phosphine 2b as a ligand in Buchwald–Hartwig cross-coupling reactions.

Suzuki Cross-Coupling Reaction

The Suzuki coupling of chloroaromatic derivatives with arylboronic acids is also a challenging reaction catalyzed with high efficiency for palladium complexes of sterically encumbered phosphines.^7,20Table 3 shows the catalytic results for the Suzuki coupling between p-chlorotoluene 14 and phenylboronic acid 15, under typical reaction conditions. As expected, SPhos 1e gives the best results among the commercial phosphine ligands tested when using 2.5 mol % of the palladium catalyst (entries 1–3), with a 61% yield of product 16 in our hands. However, at these catalyst loadings, phosphines 2a and 2b proved to be low efficient (<20% yield, entries 4 and 5). Toluene 17 was found as the main byproduct of the reaction, and the homocoupling of phenylboronic acid 15 was found to be marginal by an independent reaction test with phosphine ligand 2b. With these data in hand, the selectivity for cross-coupling product 16 with the phosphine ligand 2b (80%) was the highest among all phosphines tested; thus, we doubled the amount of catalyst to improve the yield. With these new conditions, a 32% yield of 16 could be obtained, which is anyway still far from the results with SPhos 1e. The results could be expanded to other chloroaromatic derivatives and arylboronic acids (Figure S6).

Table 3. Catalytic Results for the Cross-Coupling Reaction between p-Chlorotoluene 14 and Phenylboronic Acid 15 under Typical Reaction Conditions, Catalyzed by Different Pd-Phosphine Complex Catalysts.

entry	Pd (mol%)	phosphine (mol%)	conv. (%)a	select. to 16 (%)
1	Pd(OAc)2 (2.5)	1e (5)	83	74
2		1a (5)	27	5
3		1c (5)	60	70
4		2a (5)	20	10
5		2b (5)	25	80
6b		2b (10)	16	49
7	Pd(OAc)2 (5)	2a (10)	40	12
8		2b (10)	50	65
9c		2b (10)	48	64
10		2b (15)	55	63

^aGC results. Mass balance completed with toluene 17.

^bPd₂(dba)₃ instead of Pd(OAc)₂.

^cNaO^tBu instead of Cs₂CO₃.

In any case, these results confirm the better performance of the alkyl phosphine 2b with respect to the aromatic phosphine 2a in cross-coupling reactions, in accordance with the results observed with commercial phosphines. It is also worthy to comment here that the concordance in catalytic activity between commercial phosphines and the new phosphines 2a and 2b is also observed in the telomerization reaction, in this case with the aromatic phosphines as the more active ligands. We have to add that we tested the above commented couplings with Ni(OAc)₂ instead of palladium; however, the desired products were not found.

Semi-Hydrogenation Reaction of 3-Methyl-1-pentyn-3-ol 18

The selective semi-hydrogenation of alkynes to cis alkenes catalyzed by palladium nanoparticles is an important reaction in the industrial synthesis of nutraceuticals, pheromones, and vitamins, among others.²¹ Supported bare palladium nanoparticles are not selective, and thus, they are treated with modifiers to achieve the desired selectivity. The catalyst of choice here is the classical Lindlar catalyst, composed of Pd nanoparticles supported on CaCO₃ and poisoned with lead and quinoline.²² Alternatively, different supported palladium nanoparticles can be poisoned with other agents such as sulfides,²³ amines,²⁴ and phosphines.^8,25 Thus, we tested the action of new phosphines 2a and 2b in the commercial solid Pd/C, an unselective catalyst for the semi-hydrogenation reaction of alkynes. This solid is composed by palladium nanoparticles of ∼10 nm average size, as assessed by dark field scanning transmission electron microscopy (DF-STEM; Figure S7). Figure 8 shows the catalytic results for the semi-hydrogenation reaction of 3-methyl-1-pentyn-3-ol 18, a representative industrial reaction.

Figure 8.

Top: correlation between turnover frequencies in the alkyne to alkene hydrogenation reactions and the electronic properties of the phosphines, expressed in terms of ν(CO) (cm^–1) of the corresponding Ni(CO)₃(PR₃) complex. Bottom: same values for the alkene to alkane transformation. The reactions were performed with Pd/C (1 wt %, 0.01 mol %) at 3 bar and 1:20 molar ratio Pd:phosphine. Vertical error bars represent ±10% uncertainty, and horizontal error bars represent the ±0.3 cm^–1 uncertainty reported.^13a,14 The phosphines synthesized in this work (2a and 2b) are encircled. GC results.

The semi-hydrogenation reaction of 18 proceeds with the best selectivity if the Pd/C catalyst (0.01 mol %) is treated with JohnPhos 1d,²⁶ SPhos 1e,^10a or phosphine 2b (0.02 mol %) due to their ability to prevent the alkene to alkane reaction while enabling the desired semi-hydrogenation of 18. Phosphines 2a and 1a are selective as well but hinder the semi-hydrogenation rates. In contrast, phosphine ligands 1b–c are much less selective for the alkene product 19, giving high amounts of alkane 20 (Figure 8, bottom). This effect is observed at different H₂ pressures (Figure S8).²⁷ These results illustrate again the similarity of phosphine 2b with Buchwald-type phosphines. In addition, it must be noticed here that the aromatic phosphine 2a is, in any case, more active than triphenylphosphine 1a, thus further confirming the apparent superiority of 2a to 1a due to the larger volume and higher inductive effect of phosphine 2a’s Buchwald-type ligand.^27c

Conclusions

The synthesis of the indomuscone-based, sterically crowded phosphines 2a and 2b has been accomplished in good yields after four synthetic steps. The new phosphine ligands have been characterized by different techniques including X-ray diffraction and show combined electronic and Tolman cone angle values, which differ significantly from most of the commercial phosphines regularly used as ligands in metal catalysis. The Pd(2a)₂(OAc)₂ complex has also been prepared and characterized and shows enhanced catalytic activity and selectivity to the tail-to-head product 8c during the telomerization of isoprene 7 in MeOH. Phosphine 2b shows catalytic similarities to the Buchwald-type SPhos phosphine ligand 1e during the Buchwald–Hartwig and Suzuki cross-coupling reactions and the semi-hydrogenation reaction of alkyne 18. These results provide new phosphine ligands for metal catalysis based on a sterically hindered but widely available fragrance compound and the possibility of generating a whole phosphine family since only the phenyl and cyclohexyl derivatives have been employed here.

Experimental Methods

Materials

Glassware was dried in an oven at 175 °C before use. Cyclization reactions were performed in vials or round bottom flasks equipped with a magnetic stirrer and open to the air. Reagents and solvents were obtained from commercial sources and were used without further purification unless otherwise indicated. All zeolites are commercially available from Zeolyst. Alkyne 18 and phosphines were purchased from Merck-Millipore Sigma (95–99% purity), except for 1e (98% purity), purchased from Abcr. Pd/C (1 wt %) was purchased from Merck-Millipore Sigma and used as received. Alkene 19 was independently synthesized by selectively hydrogenating 18 under 1 bar of H₂ with colloidal Pd nanoparticles supported on TiS (c-Pd/Tis, BASF) at room temperature. Compounds 2a, 2b-oxide, and 6 were assigned in the Cambridge Structural Database with deposition numbers CCDC 2240260, 2240261, and 2240259, respectively.

Physical Techniques

Products were characterized by GC–MS, ¹H-NMR_,³¹P-NMR, ¹³C-NMR, and DEPT (distortionless enhancement by polarization transfer). Gas chromatographic (GC) analyses were performed in an instrument equipped with a 25 m capillary column of 5% phenylmethylsilicone, and n-dodecane was used as an external standard. Gas chromatography coupled to mass spectrometry (GC–MS) analyses were performed on a spectrometer equipped with the same column as the GC and operated under the same conditions. ¹H-NMR, ³¹P-NMR, ¹³C-NMR, and DEPT measurements were recorded in a 300 MHz instrument using CDCl₃ as a solvent, containing TMS as an internal standard.

The metal content of the complexes was determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES, Thermo Scientific ICAP Pro) after disaggregation in aqua regia and later diluted in water before analysis. Attenuated total reflection infrared spectroscopy, performed in a JASCO FT/IR-4000, was employed to record the IR spectra (400–4000 cm^–1) by dropping a small sample of the compound in dihcloromethane solution on the ATR crystal. Absorption ultraviolet–visible spectrophotometry (UV–vis) measurements were recorded on an Agilent Cary 60 UV–vis spectrophotometer, in 1 cm wide cuvettes and a xenon source lamp. Ultra-pressure liquid chromatography coupled to high-resolution mass spectroscopy (UPLC-HRMS) was performed using the electrospray ionization technique without previous column separation and TOF MS ES+ as a mass analyzer. Monocrystal X-ray diffraction (XRD) measurements were carried out in a Bruker D8 VENTURE PHOTON-III instrument with Mo radiation at −173 °C. Images of the Pd/C catalyst were obtained on a JEM-F2100 operated at 200 kV in dark field scanning transmission electron microscopy (DF-STEM mode), after dispersion in ethanol, supporting in a copper grid and evaporation overnight. Compounds were purified by flash column chromatography or thin-layer chromatography (TLC) under standard conditions with the appropriate mixture of solvents (typically n-hexane:ethyl acetate).

Synthesis of 4a

A suspension of sodium amide (242 mg, 5.6 mmol) in toluene (1.8 mL, 3.1 M) was slowly added for 5–10 min to a toluene solution (1.9 mL, 3 M) of indomuscone 3 (1.15 g, 5.6 mmol) in a 50 mL round bottom flask equipped with a magnetic stir bar at room temperature under continuous stirring. After 20 min, a toluene solution (1.9 mL, 3 M) of methallyl chloride (507 mg, 5.6 mmol) was added into the flask, heated in an oil bath, and continuously stirred for 12 h while the reaction mixture was refluxed. After cooling, the mixture was neutralized with HCl aqueous solution, extracted with ether, and washed with brine. The combined organic phases were dried over MgSO₄, filtered, and concentrated under vacuum. Flash column chromatography (2% AcOEt in hexane) gave 946 mg (65% yield) of methallyl indomuscone 4a as a yellow oil.

Synthesis of 4b

Indomuscone 3 (9.3 g, 45 mmol) was slowly added during 15 min to a suspension of sodium amide (2.5 g, 66 mmol, 1.5 equiv) in toluene (45 mL, 1 M) in a 100 mL round bottom flask equipped with a magnetic stir bar at room temperature under continuous stirring (750 rpm) and under a N₂ atmosphere. Then, the mixture was heated to 45 °C in an oil bath. After 30 min at 45 °C heated with an oil bath, isobutylene oxide (4.76 g, 66 mmol, 1.5 equiv) was slowly added into the flask and the stirring continued for 6 h. After cooling, the reaction mixture was neutralized with NH₄Cl. The aqueous phase was extracted with ethyl acetate and washed with brine. The combined organic phases were dried over MgSO₄, filtered, and concentrated under vacuum. Flash chromatography (1% AcOEt in hexane) gave 8.8 g (70% yield) of hydroxyisobutyl indomuscone 4b as a yellow oil.

Synthesis of 5 from 4a

In a 8 mL vial equipped with a magnetic stir bar and containing the catalyst (140 mg of pTsOH, 0.81 mmol, 20 mol % or 1.0 g of zeolite Beta-H), a solution of methallyl indomuscone 4a (1.04 g, 4 mmol) in dry toluene (2 mL, 2 M) was added. The reaction mixture was open to the air, heated in an oil bath, and stirred at 70 °C for 24 h. After that time, the mixture was cooled (the zeolite filtered off if present) and neutralized with 10% aqueous sodium hydrogen carbonate solution. The aqueous phase was extracted with hexane and washed with brine. The combined organic phases were dried over MgSO₄, filtered, and concentrated under vacuum. An aliquot was dissolved in AcOEt (1 mL) and filtered through a 20 μm nylon filter, and the resulting filtrate solution was analyzed by GC and GC–MS. The product was purified by flash column chromatography (100% hexane) to give 5 as a colorless oil (703 mg, 70%).

Synthesis of 6

The aromatic compound 5 (1.95 g, 7.5 mmol) was placed in a 250 mL round bottom flask equipped with a magnetic stirrer. Dichloromethane (50 mL, 0.15 M) was added into the flask. After the solid was dissolved at room temperature under magnetic stirring, Br₂ (1.2 g, 7.5 mmol, 1 equiv) was added dropwise into the flask. After the addition of Br₂ was completed, the resulting reaction mixture was further stirred for 15 min and then treated with Na₂S₂O₃ aqueous solution (aq.), NaHCO₃ (aq.), and brine. The organic phase was dried over Na₂SO₄ and filtered. Volatiles were removed from the filtrate under vacuum to give the aromatic bromide compound 6 as a yellowish solid after cooling (2.1 g, 87% yield).

Synthesis of 2a and 2b

The aromatic bromide compound 6 (340 mg, 1.0 mmol) was placed in a dried 10 mL round bottom flask equipped with a magnetic stirrer, dissolved in anhydrous THF (2 mL, 0.5 M) under a nitrogen atmosphere, and cooled to −78 °C. Then, n-BuLi 2.5 M in hexane (0.5 mL, 1.25 mmol, 1.25 equiv) was added dropwise into the flask, and the resulting reaction mixture turned from yellow to an orange color. The reaction mixture was magnetically stirred at −78 °C for an additional 15 min before chlorodiphenylphosphine (Ph₂PCl, 180 μL, 1.0 mmol, 1 equiv) or chlorodicyclohexylphosphine (c-Hex₂PCl, 265 μL, 1.2 mmol, 1.2 equiv) was added into the flask at once. The reaction mixture was then warmed to room temperature for 90 min under stirring. The reaction mixture was then quenched with NH₄Cl (aq.) and washed with water and brine consecutively. The organic phase was dried over Na₂SO₄ and filtered. The desired phosphine derivative compound product [(2,2,6,6,7,8,8-heptamethyl-3,6,7,8-tetrahydro-2H-indeno[4,5-b]furan-5-yl)diphenyl phosphane for 2a; (2,2,6,6,7,8,8-heptamethyl-3,6,7,8-tetrahydro-2H-indeno[4,5-b]furan-5-yl) dicyclohexyl phosphane for 2b] was purified by column chromatography and then by preparative thin-layer chromatography [TLC, 3% AcOEt in n-hexane; R_f (10% AcOEt in n-hexane) = 0.65 for 2a; R_f (5% AcOEt in n-hexane) = 0.5 for 2b] to give the desired phosphine derivative compound (colorless solid, 150 mg, 34% yield for 2a; colorless solid, 182 mg, 40% yield for 2b) after removing volatiles under vacuum.

Synthesis of the Oxidized Phosphine 2b-oxide

2b (91 mg, 0.2 mmol) was dissolved in wet THF (0.1 M) in a round bottom flask equipped with a stirring bar, heated in an oil bath, and magnetically stirred at 50 °C under 3 atm of O₂. The mixture was monitored by GC–MS. After a 2 h reaction time, the reaction was stopped, O₂ was removed, and the mixture was submitted to crystallization by slow evaporation of the solvent to obtain a colorless solid, 75 mg, 80% yield.

Titration of Palladium and Phosphine 2a by UV–Vis and ³¹P-NMR

In different round bottom flasks equipped with a stirring bars, Pd(OAc)₂ (3 mg, 0.0133 mmol), 5 mL of MeOH (MeOD was used for ³¹P-NMR), and different amounts of phosphine (0.5–4 equiv) were added. The mixture was heated in an oil bath and allowed to stir for 10 min at 50 °C until a uniform color, without change, was achieved. This dissolution with the complex formed was measured directly by UV–vis or ³¹P NMR. The data were processed and plotted for analysis.

Synthesis of Palladium Complexes with 2a and 2b

In a 100 mL round bottom flask, Pd(OAc)₂ (29.8 mg, 0.133 mmol), phosphine 2a (0.266 mmol, 2 equiv, 118 mg) or 2b (121 mg), and 50 mL of the solvent (MeOH for 2a and THF for 2b) were added, and the mixture was heated in an aluminum plate and allowed to stir at 50 °C for 10 min until complete formation of the complex (stable orange color). Then, the mixture was concentrated in vacuo to obtain the orange solid complex (147 mg for 2a-complex and 150 mg for 2b-complex) to be used as a catalyst in the subsequent reactions.

Typical Procedure for the Telomerization Reaction

Isoprene 7 (68.1 mg, 1 mmol), 0.05 mmol of NaOMe, palladium complex (11.0 mg, 0.01 mmol) with 2a (11.4 mg, 1 mol %), and MeOH (250 μL, 4 M) were introduced into a 2 mL glass vial, which was then sealed. The mixture was stirred for 1 h at 60 °C heated in an aluminum plate, taking 5 μL aliquots for analysis by GC (the aliquot was diluted in DCM with n-dodecane as an external standard to monitor the reaction process). Once finished, 1 mL of water was added, and the mixture was extracted with 2 mL of ethyl acetate (three times), washed with brine, dried over MgSO₄, and filtered into a 50 mL flask and volatiles were removed in vacuo. To purify the complex, flash chromatography on silica (with a mixture 1:100, EtOAc:hexane) was performed, obtaining 165 mg of the corresponding ethers 8 with a yield of 98% as a yellow oil.

Typical Procedure for the Buchwald–Hartwig Reaction

Aryl chloride (0.25 mmol), amine (0.5 mmol), NaO^tBu (36 mg, 0.374 mmol), Pd(OAc)₂ (3.6 mg, 2.5 mol %), and phosphine 2b (7.7 mg, 7.5 mol %) were mixed in dioxane (2.0 mL, 0.13 M). The mixture was stirred at 80 °C heated in an aluminum plate in a sealed vial and monitored by GC/GC–MS. When the reaction ends, water was added and the mixture was extracted with Et₂O (three times). The combined organic layers were washed with H₂O and brine, dried over anhydrous MgSO₄, filtered, and concentrated in vacuo. The residue was purified by column chromatography with a gradient eluent of 5–30% ethyl acetate in hexanes to give the desired product.

Typical Procedure for the Suzuki Cross-Coupling Reaction

Aryl chloride (0.12 mmol), boronic acid (0.19 mmol), Cs₂CO₃ (81.45 mg, 0.25 mmol), Pd(OAc)₂ (1.7 mg, 3 mol %), and phosphine 2b (3.4 mg, 5 mol %) were mixed in dioxane (0.4 mL, 0.3 M). The mixture was stirred at 80 °C heated in an aluminum plate in an ambient atmosphere and monitored by GC/GC–MS. When the reaction ends, water was added and the mixture was extracted with Et₂O (three times). The combined organic layers were washed with H₂O and brine, dried over anhydrous MgSO₄, filtered, and concentrated in vacuo. The residue was purified by column chromatography with a gradient eluent of 5% ethyl acetate in hexanes to give the desired product.

Typical Procedure for the Semi-Hydrogenation Reaction of 18

All the semi-hydrogenation reactions were performed in batch in a 6 mL round bottom vial, with 0.5 mL of reaction volume and a stirring magnet. The solvent was ethanol, at a 0.3 M concentration when the pressure was set at 1 bar of H₂, and at a 0.5 M concentration under higher pressures of H₂. The reactions were conducted at 25 °C heated in an aluminum plate and stirred at 450 rpm. The H₂ pressure was kept constant by refilling the reactor periodically, so enough H₂ was always present to fully hydrogenate the alkyne to the corresponding alkane. Yields were obtained by GC, and GC-coupled mass spectrometry and NMR were used to identify the products, besides comparison with pure product samples.

Data Availability Statement

The data underlying this study are available in the published article and its Supporting Information.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.joc.3c00314.

• Additional experimental data, Figures S1–S8, Tables S1–S7, compound characterization, NMR and FT-IR copies, and summary of crystallographic data (PDF)

Author Contributions

F.G.-P. carried out the catalytic experiments, except the hydrogenation reactions. S.S.-N. synthesized phosphines 2a and 2b. J.B.-S. performed the semi-hydrogenation reactions. A.C.-P., J.S.-Q., and E.E.-F. brought the industrial analysis and supervised the project. A.L.-P. designed the synthetic route and supervised the whole work. All authors have participated in the writing of the manuscript.

The authors declare the following competing financial interest(s): The following text has been added to the MS as a note: "Patent number EP22383018 has been presented to protect the synthesis of phosphines 2a and 2b. S.S.-N. and A.L.-P. appear in the patent and declare no other competing interests. The rest of authors declares no competing interests.".