Rhodium Catalysed Asymmetric Hydrogenation indolinephos Bachelor thesis

Download 87.32 Kb.
Date conversion17.07.2018
Size87.32 Kb.

Rhodium Catalysed Asymmetric Hydrogenation


Bachelor thesis

Guido Guijt


Daily Supervisor; Jeroen Wassenaar

Supervision; Joost Reek

Second Reviewer; Jarl Ivar van der Vlugt


The goal of this project is to develop an asymmetric synthesis of the INDOLINEPhos hybrid bidentate phosphine-phosphoramidite ligand and to investigate what effect the introduction of a chiral backbone to INDOLPhos has on its scope in enantioselective homogeneous hydrogenation with a rhodium complex of this ligand. The hybrid bidentate ligand was synthesised from 2-diphenylphosphoindoline and (S)-(-)-2,2’-bisnaphtol phosphorochloridite. However, due to the unstability if the phosphinoindoline, the ligand was not obtained. It was the first time that the sparteine mediated asymmetric lithiation of indoline was used with a phosphine based electrophile, which might prove a successful way of introducing chirality in phosphine ligands in the future. When the crude ligand mixture was reacted to a rhodium precursor, it formed a successful catalyst for the asymmetric hydrogenation of several prochiral olefins with up to 93 % ee. The active catalyst was most probably a norbornadiene rhodium complex with two (S)-(-)-2,2’-bisnaphtol indole ligands. So the scope of INDOLPhos has been broadened, but not by introduction of a chiral backbone to INDOLPhos, but by making it into a monodentate phosphoramidite ligand.

  1. Populair-wetenschappelijke Samenvatting

Tijdens dit bachelorproject is de asymmetrische hydrogenering van alkenen onderzocht. Deze chemische reactie maakt van een dubbele binding een enkele binding, door er waterstof aan toe te voegen en doet dat op een zodanige manier dat er veel meer van het ene spiegelbeeld wordt gemaakt dan van het andere, wat erg belangrijk is in bijvoorbeeld medicijnen, omdat de 2 spiegelbeelden een heel andere werking kunnen hebben. Tijdens dit project is hydrogenering gedaan met het metaal rhodium, dat een katalysator is voor de hydrogenering. Het rhodium is verbonden met een type chemische verbinding dat ligand heet en die de werking van de katalysator heel erg beïnvloed. Tijdens het project is geprobeerd om een katalysator die al succesvol was in de asymmetrische hydrogenering van een aantal stoffen te verbeteren. Dit ligand greep met twee plaatsen vast op het metaal en die twee plaatsen waren niet identiek, om het ligand-metaalcomplex extra asymmetrisch te maken. Het beoogde complex is niet gemaakt, maar wel een ander complex dat maar op één plaats vastgrijpt op rhodium. Toen dit complex werd gebruikt, bleek het wel succesvol, want van een dubbele binding in een bepaalde stof werd er maar 3 % van de ongewenste spiegelbeeld gevormd en 97 % van het gewenste spiegelbeeld.

  1. Table of contents

Abstract 4

1.Populair-wetenschappelijke Samenvatting 4

4.Results and Discussion 14

Catalyst Synthesis 14

Asymmetric Hydrogenation 17

Characterisation 18

5.Conclusions and Outlook 20

6.Experimental Section 22

Synthesis of catalyst 22

Appendix 28

Ligand Synthesis 30

Acknowledgements 32

7.References 32

  1. Introduction

Homogeneous catalysis is a very important tool in chemistry, for example in the synthesis of medicines and in fine chemistry. It can be applied to a broad range of reactions, of which hydrogenation of alkenes is one of the most extensively studied. In essence it is a very simple reaction in which hydrogen from a hydrogen source, like dihydrogen, is added to an alkene. (Figure 1). Because dihydrogen has σ-symmetry and an alkene-bond has π-symmetry, the direct addition of dihydrogen to an alkene is forbidden and will not take place, unless the reaction mixture is heated considerably. But dihydrogen can form a metal hydride by oxidative addition of the dihydrogen to the metal. If to this metal hydride an alkene is coordinated, or coordinates after the metal hydride is formed, the double carbon carbon bond can be reduced to a single carbon carbon bond.1 If the R-groups of one or two of the carbons of the alkene are different, chirality is introduced into the resulting alkane. If the catalyst is not chiral, then a racemic mixture of the two possible enantiomers is formed. With a chiral catalyst on the other hand, one of the enantiomers can be favoured. For some processes the chirality of the alkane is not important, but when the chiral alkane is reacted with another chiral moiety, the reactivity differs. This can be important in chemistry with natural products, for example in medicinal chemistry, because 19 of the 20 amino acids in human proteins are chiral. Sometimes one enantiomer of a particular pharmaceutical targets the desired active site, so only half of the pharmaceutical ingredient is active, which is just an economical problem. But sometimes one enantiomer is targeting the desired protein and the other enantiomer is targeting another protein, which could cause undesired effects. The best known example is thalidomide, the active component of softenon, a racemic medicine proscribed to pregnant women from 1957 to 1961. (R)-thalidomide has the desired effect against morning sickness, but (S)-thalidomide causes birth-defects.2

Figure 1 hydrogenation of an alkene

Homogeneous hydrogenation is mostly performed with late transition metal complexes, (rhodium, iridium, ruthenium, osmium, palladium, platinum and nickel). These complexes can be stabilized by many types of ligands, with several different oxidation states and coordination numbers. Rhodium is the most used and studied late transition metal for asymmetric hydrogenation because in most cases it has outstanding coordinative properties and good selectivity and activity compared to the other late transition metals. It is also conveniently studied by NMR techniques, because 103Rh, which has spin ½, is 100% abundant in nature.Error: Reference source not found

The ligands coordinated to rhodium play a vital role in the activity and the selectivity of the complex in homogeneous hydrogenation. Already in 1939 Iguchi et al reported several rhodium complexes that were active in hydrogenation.3 In 1965 Wilkinson et al reported the first use of a phosphine rhodium complex to do homogeneous hydrogenation.4 Wilkinson’s catalyst, RhCl(PPh3)3, is still the most popular catalyst in homogeneous hydrogenation.

The mechanism of the rhodium catalyzed asymmetric hydrogenation of alkenes has been extensively studied with both computational chemistry and spectroscopy, to gain insight in this reaction. Two different mechanisms were found to be active (Figure 2), dependent on the ligand and substrate properties. These mechanisms differ in the order of dihydrogen and alkene coordination to the rhodium complex. First the chiral catalyst precursor is hydrogenated and solvated, after which the two mechanisms take different pathways.

The first pathway is called the unsaturated pathway and was proposed by Halpern et al.5 The solvent is displaced by an alkene substrate. Because the complex is chiral and the substrate is prochiral two diastereoisomers are formed. Then dihydrogen adds to one of these diastereoisomers by oxidative addition, followed by the transfer of one of the hydrides to the substrate. Lastly, the product is released by reductive elimination and the catalyst is recovered for the next cycle. It was shown that the relative reactivity of the diastereoisomers towards dihydrogen decides which of the two possible isomers of the product, rather than which one of the diastereoisomeric catalyst substrate complexes is more abundant. When dihydrogen pressure was increased for these reactions, hydrogenation selectivity was lower, because the difference in relative reactivity of the two diastereoisomers became smaller.Error: Reference source not found

The second pathway is called the dihydride pathway. Dihydrogen adds to the complex through oxidative addition. Then the solvent molecules are displaced by an alkene substrate. After that, the same steps as in the unsaturated pathway are followed; a transfer of a hydride to the alkene and reductive elimination of the product. In the dihydride pathway it is suggested that the transfer of the hydride is the enantio-determining step in this pathway, because the oxidative addition of dihydrogen and the coordination of the alkene are both reversible steps. In reality however, the boundaries of these two pathways are blurred and in some cases more than one mechanism is operating.

Figure 2 catalytic cycle via unsaturated pathway and via dihydride pathway

In order to hydrogenate new substrates in high selectivity, many ways of designing chiral ligands have been explored in the past. In 1968 Knowles and Horner found chiral phosphine ligands to induce enantioselectivity in hydrogenation. They hydrogenated some substrates with an ee up to 28 %. In 1971 Morrison found phosphine ligands with a chiral carbon attached to it that were even more selective, with ee’s up to 52 %. In that same year Kagan and Dang found that a chiral bidentate ligand was even more active, with an ee’s of 72 %. Since then, many bidentate chiral diphosphines were produced, which exhibited very high selectivity, reaching up to 99 % ee. More recently, new monodentate phosphorus ligands have been discovered and used in asymmetric hydrogenation, and in some cases have higher activity and selectivity then bidentate ligands.Error: Reference source not found In general, it differs for every substrate which ligand is best used in asymmetric hydrogenation.

Because of the often specific catalyst substrate match, many different rhodium complexes are needed for the asymmetric hydrogenation of new substrates; thus it is vital to prepare a library of different ligands, coordinate them to rhodium and screen them for activity and stereo selectivity.

To create novel ligands for rhodium catalyzed hydrogenation, two different strategies were followed in this bachelor project; a supramolecular approach and a hybrid donor atom approach. The supramolecular approach will be discussed in the appendix and we will now focus on the hybrid donor atom approach.

A hybrid donor atom ligand is synthesised with a relatively easy condensation between two molecules. As these two molecules can be differentiated, a modular synthesis is achieved, resulting in a library of ligands with a wide variety of structural and functional diversity. Together with an automated parallel screening methodology on the asymmetric hydrogenation of several substrates, successful catalysts for specific substrates can be identified in a high-throughput fashion, using a designed set of experiments.

The hybrid donor atom approach that will be discussed here will be a bidentate phosphine-phosphoramidite ligand. Previous work in this group already resulted in the development of a hybrid bidentate phosphine-phosphoramidite ligand. INDOLPhos ligands 1 (Figure 3) were found to lead to a successful enantioselective hydrogenation catalyst for several substrates, especially the important synthesis of the Roche ester with 98 % ee.6,7 The enantioselectivity was induced by a chiral phosphoramidite. The two starting compunds that build up the phosphine-phosphoramidite ligand could both be varied, facilitated by the modular synthesis of INDOLPhos. The R-groups on the phosphine could be varied by reacting differently substituted chlorophosphines to 3-methylindole (Figure 4). The other donor fragment, (s)-BINOL, is commercially available with several different R’-groups. These variations of substituents of INDOLPhos already resulted in a wide substrate scope for INDOLPhos.Error: Reference source not found

Figure 3 INDOLPhos

Another approach to differentiate INDOLPhos is to replace the P-N-C-P’ backbone. In the past it was shown that by varying the backbone of a ligand, the scope of a catalyst was often broadened. A chiral backbone is especially interesting to induce enantioselectivity in hydrogenation.8 The carbon atom in the P-N-C-P’ backbone is best suited for inducing chirality in the backbone. To induce asymmetry in this carbon atom it has to be an sp3 hybrid, so the double bond between the 2- and the 3-position in indole has to be a single bond, making the double ring structure an indoline, which after reaction to a chlorophosphine and reaction to (s)-BINOL results in INDOLINEPhos 2 (Figure 5). In this way chirality is introduced in a ligand in both the backbone and the phosphoramidite, providing opportunities to investigate different diastereoisomers.

Figure 4 reaction of a chlorophosphine and 3-methylindole to an indolylphosphine

Figure 5 hybrid bidentate phosphine-phosphoramidite ligand; INDOLINEPhos

The goal of this project is to develop an asymmetric synthesis of the INDOLINEPhos hybrid bidentate phosphine-phosphoramidite ligand and to investigate what effect the introduction of a chiral backbone to INDOLPhos has on its scope in enantioselective homogeneous hydrogenation with a rhodium complex of this ligand.

  1. Results and Discussion

Catalyst Synthesis

The phosphine-phosphoramidite-ligand 8 was synthesized in a five-step synthesis (Figure 6), starting from indoline 3. First the nitrogen of indoline was Boc-protected using di-tert-butyldicarbonate to result in 4. Then the 1-Boc-indoline was asymmetrically lithiated on the 2-position using (s-BuLi)/(-)-sparteine and subsequently substituted with chlorodiphenylphosphine resulting in 5. A procedure from Beak et al was used for the asymmetric lithiation of indoline.9 This procedure was, to the best of our knowledge, hitherto not used for phosphine chloride electrophiles. The reaction proceeded with 90 % ee, which was determined using a chiral HPLC technique. The phosphine group was very sensitive to oxidation, so the phosphorus was protected before work-up, using borane. The protective Boc-group was removed using trimethyl siliciumchloride and phenol, resulting in 6.

Figure 6 synthetic route to the bidentate phosphine-phosphoramidite ligand

Subsequently, borane was removed using 1,4-Diazabicyclo[2.2.2]octane resulting in 7. When 7 was analysed using 31P-NMR, the oxidised phosphine and the dissociated diphenylphosphine 9 (Figure 7) were observed. When diphenylphosphine dissociates a double bond is formed in indoline. This structure readily tautomerises into indole, a stable aromatic compound. The formation of this stable compound is the driving force of this dissociation reaction, because it is lower in energy than the phosphino indoline. These side products from the dissociation and the oxidation reaction were not removed when it was filtered over silica, so the phosphine-indoline was coupled to (S)-(-)-2,2’-bisnaphthol phosphorochloridite to prevent these side reactions. However, also in the hybrid phosphine-phosphoramidite ligand 8 the oxidised phosphine and the dissociated diphenylphosphine 9 were still present.

Figure 7 dissociation of diphenylphosphine from 7

Purification was unsuccessfull due to the sensitivity of the phosphine, so the crude ligand was reacted with a standard rhodium-precursor for H2 catalysis; [Rh(nbd)2]BF4 (Figure 8). When the products and side products present in the ligand mixture that was reacted with rhodium are taken into account, several complexes can be identified in the 31P-NMR spectrum of 11. The doublet signal at 133.44 ppm with a coupling constant of 264.2 Hz can be identified as a bis-phosphoramidite complex, [Rh(nbd)(12)2]BF4. 12 can be formed when the indole 10 is reacted with (S)-(-)-2,2’-bisnaphthol phosphorochloridite to form 12 (Figure 9). The doublet signal at 50.79 ppm with a coupling constant of 179.7 Hz can be identified as [Rh(9)2(nbd)]BF4. So both side products from the dissociation reaction have formed a complex with rhodium. The singlet signal at -15.70 ppm can be identified as an oxidised phosphine. The desired complex 11 was not identified in the spectrum.

Figure 8 complexation of 8 to Rh(nbd)2BF4

Figure 9 formation of 12 from diphenylphosphine and (s)-BINOL-PCl

Asymmetric Hydrogenation

Using the ligand mixture asymmetric hydrogenation of several benchmark substrates was conducted. First a low pressure was conducted on substrate A (Figure 10). The catalyst based on the crude ligand mixture, was dissolved in dichloromethane to a 1 mM concentration. Also substrate A was added to a 100 mM concentration. Dihydrogen was bubbled through for 1.5 h and the product was analysed on chiral HPLC, which showed that the hydrogenation was completed with 100% conversion and 91 % ee. Then substrates A-G (Figure 10) were subjected to high pressure hydrogenation. Again the concentration of the catalyst was 1mM and the concentration of the substrates was 100 mM in dichloromethane. The solutions were put under 10 bar of dihydrogen pressure for 16 h in an autoclave. The resulting reaction mixtures were analyzed using chiral GC (Table 1; uneven numbered entries). The hydrogenation of A now gave a lower ee. Greater enantioselectivity at lower hydrogen pressure has been reported for other homogeneously catalysed hydrogenation reactions and is typical for the unsaturated pathway.10, 11 The hydrogenation of E resulted in an excellent ee value, which indicates the potential of this catalyst. Also hydrogenation of B resulted in a good ee. The hydrogenation reaction of F went to full conversion within 16 h reaction time, but unfortunately the product was racemic. Hydrogenation of C did not go to full conversion in 16 h and the product that formed was racemic. Hydrogenation of D did not occur.

Table 1 Screening of rhodium catalyzed hydrogenation of various substrates using INDOLPhos ligands 1a, 1b and 1c and INDOLINEPhos ligand 8




% conv

% ee (configuration)





75 (nd)





73 (S)





60 (nd)





13 (S)





3 (R)





68 (nd)










45 (nd)





93 (R)





99 (nd)





3 (nd)

  • Reaction time of 62 h

Results for the asymmetric hydrogenation of some of these substrates, using the analogous INDOLPhos ligands 1a, 1b and 1c (Figure 11), can be found in (Table 1; even numbered entries)Error: Reference source not found When the results of the INDOLPhos ligands 1a, 1b and 1c are compared to the results using the ligand 8, big differences are observed. The enantioselectivity for substrate B is much greater for INDOLINEPhos, than INDOLPhos but they perform similar for A. For D there is no activity using 8, but considerable activity when using 1a. The activity is greatly improved for asymmetric hydrogenation of C, using ligand 8, but unfortunately no selectivity was observed. Activity is much greater for hydrogenation of E, when the long reaction time of 1b is considered.

Figure 10 Substrates

Figure 11 INDOLPhos ligands


As mentioned in the catalyst synthesis section the 31P-NMR spectrum of the synthesised catalyst shows 3 complexes, two based on side products 9 and 10 and the oxidised phosphine, but only in a small amount. When determining on which ligand the active catalyst is based, it is clear that [Rh(nbd)(9)2]BF4 is not chiral, and cannot induce ee in asymmetric hydrogenation. The oxidised ligand was only present in a small quantity.

[Rh(nbd)(12)2]BF4 could induce ee in asymmetric hydrogenation, because 12 is chiral. Also, previously chiral monodentate phosphoramidite ligands based on BINOL have been proven to be very effective in the asymmetric rhodium catalyzed homogeneous hydrogenation of alkenes.12

  1. Conclusions and Outlook

The goal of this project is to attempt the synthesis of the INDOLINEPhos hybrid bidentate phosphine-phosphoramidite ligand and to investigate what effect the introduction of a chiral backbone to INDOLPhos has on its scope in enantioselective homogeneous hydrogenation with a rhodium complex of this ligand. The question whether this goal was reached has a twofold answer.

Firstly, it is unclear whether the synthesis of the bidentate ligand containing both a chiral phosphine and a chiral phosphoramidite was successful. In the characterisation by 31P-NMR spectroscopy of the rhodium complex with the ligand the desired complex 11 was not shown. [Rh(nbd)(9)2]BF4 and [Rh(nbd)(12)2]BF4 were shown in the complex as well as the oxidation product. The possibility exists that the desired complex 11 was present in a quantity under the detection limit. However, it was the first time that the asymmetric lithiation of indoline was used with a phosphine based electrophile, which might prove a successful way of introducing chirality in phosphine ligands in the future.

Secondly, on the substrates A and B the selectivity of INDOLPhos in hydrogenation has been surpassed with the novel ligand and for substrate E the selectivity of INDOLPhos was approached. [Rh(nbd)(12)2]BF4 is most likely to induce the enantioselectivity, because previous chiral monodentate phosphoramidite ligands based on BINOL used in asymmetric rhodium have had good results in homogeneous hydrogenation of alkenes before.

So the scope of INDOLPhos has been broadened, but not by introduction of a chiral backbone in INDOLPhos, but by making it into a monodentate phosphoramidite ligand. To reach the research goal and introduce a chiral backbone in INDOLPhos, there are several things that can be done.

The first is suppressing the decomposition reaction of both the rhodium complex and the free ligand. This might be done by using a different phosphine, for instance diisopropylphosphine chloride instead of diphenylphosphine chloride. Also the oxidation reaction can be suppressed by improving the methods to exclude air and water during synthesis.

Secondly, when the desired INDOLINEPhos complex is present in the ligand mixture, but not in a high concentration, a successful way to purify the complex or ligand has to be found.

When the catalyst is obtained and identified in a pure form, further research can be done on this catalyst. Substrates can be screened with the catalyst, to see how active it is in hydrogenation reactions. Also reaction conditions can be optimized by differentiating the dihydrogen pressure and the temperature. In addition, the turnover number and turnover frequency can be determined for these substrates to investigate the activity. Furthermore, research can proceed into the advanced development of this hybrid ligand into a large library. For example, different BINOLs can be reacted to the phosphinoindoline and also the phosphines can be differentiated, e.g. by diisopropylphines. Differentiating the phosphines might lead to a more stable phosphine, but also reactivity and selectivity for some substrates might be improved, like it does for INDOLPhos.Error: Reference source not found

  1. Experimental Section

Synthesis of catalyst

tert-butyl indoline-1-carboxylate (4)

To a solution of indoline (3) (5.0 g; 41.9 mmol) in THF (40 mL) was added di-tert-butyldicarbonate (10.6 g; 46.09 mmol). Effervescence was immediate. The solution was stirred for 22 h at room temperature. The solvent was removed by rotary evaporation, whereafter the product was purified over a flash SiO2-column, using a 20 % mixture of ethyl acetate in hexane as eluens. When the purified product was kept overnight under reduced atmosphere, the product crystallized into a light pink solid. Yield = 11.438g (99+ %). 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 7.25 – 6.89 (m, 4 H), 3.96 (t, J = 8.2 Hz, 2 H), 3.07 (t, J = 8.7 Hz, 2 H), 1.57 (s, 9 H).

tert-butyl 2-(boranediphenyl)phosphinoindoline-1-carboxylate (5)

tert-butyl indoline-1-carboxylate (4) (1.036 g; 4.713 mmol) and (-)sparteine are dissolved in cumene (30 mL) and cooled to -780C. S-butyllithium (4.713 mL of a 1.3 M solution in cyclohexane/hexane, 6.127 mmol) was added dropwise to the cumene solution and the solution was stirred for 7 h at -780C. Diphenylphosphinechloride (897 μL, 4.713 mmol) was added dropwise. The solution was stirred overnight during which the solution was allowed to slowly warm to RT. The solution was then cooled to 00C and the BH3.THF (6.127 mL of a 1 M solution in THF, 6.127 mmol) was added drop-wise and it was stirred for 2 h at 00C. 25 mL of H2O was added to the solution and the aqueous layer and the cumene layer were separated. The aqueous layer was extracted with diethyl ether (3 . 25 mL). The cumene layer and the diethyl layers were combined and washed with an aqueous 5% H3PO4 solution (25 mL). The organic extracts were dried over MgSO4, filtrated and the solvent was removed under vacuum. The product was purified over a flash SiO2-column, using a 10 % mixture of ethyl acetate in hexane as eluens, resulting in a white solid. Yield = 0.92 g (47 %). 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 8.05 – 6.92 (m, 17 H) 5.54 (br s, 1 H), 3.54 (m, 1 H) 3.19 (t, J = 0.22 Hz, 1 H) 1.56 (s, 9 H). 31P{1H}-NMR (CDCl3, 121.5 MHz, 298 K) δ (ppm) 28.65 (br s). The ee = 90 % determined by chiral HPLC analysis (Chiralcel AD-H, flow-rate: 0.7 mL/min, eluent: hexane/isopropanol (99.5/0.5), detection at 254 nm, tR (S) = 16.4 min., tR (R) = 18.8 min.).

2-(boranediphenyl)phosphinoindoline (6)

Phenol (1.90 g, 20.1 mmol) was dissolved in dichloromethane (10 mL). Subsequently trimethylsilane chloride (726 mg, 6.69 mmol) and tert-butyl 2-(boranediphenyl)phosphinoindoline-1-carboxylate (5) (279 mg, 0.669 mmol) were added. Immediately gas escaped the solution. The solution was stirred for 40 h at RT, after which the solvent was removed under vacuum. 25 mL of an aqueous solution of 10 % sodium hydroxide was added to the product. The product was extracted with diethyl ether (3 . 25 mL). The combined organic layers were washed with H2O (25 mL) and then dried over MgSO4, after which the solvent was removed under vacuum. The product was then purified over a flash SiO2-column, using a 10 % mixture of ethyl acetate in hexane as eluens. The product was washed with hexane and dried under vacuum. Yield = 130 mg (61 %). 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 8.00-6.55 (m, 17 H), 4.73 (dt, J1 = 7.2 Hz, J2 = 9.8 Hz, 1 H), 3.42 (m, 1 H), 3.19 (m, 1 H). 31P{1H}-NMR (CDCl3, 121.5 MHz, 298 K) δ (ppm) 21.74 (br s).

2-(diphenyl)phosphinoindoline (7)

2-(boranediphenyl)phosphinoindoline (6) (86 mg, 0.27 mmol) and

1,4-Diazabicyclo[2.2.2]octane (33.6 mg, 0.30 mmol) were dissolved in toluene (5 mL). The mixture was stirred overnight at RT and subsequently the solvent was removed under reduced vacuum. The product was then dissolved in diethyl ether and filtered over SiO2. Then the solvent was removed under reduced pressure, resulting in a white solid. 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 7.76 – 6.83 (m, 14 H), 5.30 (br s, 1 H) 3.50 (m, 3 H). 31P{1H}-NMR (CDCl3, 121.5 MHz, 298 K) δ (ppm) -4.58 ( s).

1-(s)-Binol-P -2-(diphenyl)phosphinoindoline (8)

To a solution of 2-diphenylphosphinoindoline (7) (81.9 mg, 0.27 mmol) in toluene (5 mL) was added triethylamine (41.8 μL, 0.30 mmol) at 00C, followed by (S)-(-)-2,2’-bisnaphthol phosphorochloridite (94.7 mg, 0.27 mmol) and it was stirred overnight while allowed to warm to RT. Then hexane (5 mL) was added and it was filtered over celite. Subsequently the solvent was removed under reduced pressure. 1H-NMR shows the product to be unpure, but the diphenylphosphine is very sensitive to oxidation, which makes it unadvisable to purify the product. Due to the impurity 1H-NMR-signals are hard to assign. 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 8.1 – 6.6 (m, 26 H), 2.8-3.7 (m, 3 H). 31P{1H}-NMR (CDCl3, 121.5 MHz, 298 K) δ (ppm) 137 (s), 15 (s), -5 (br s)

Rh(1-(s)-Binol-P -2-(diphenyl)phosphinoindoline)(nbd)BF4 (11)

1-(s)-Binol-P -2-(diphenyl)phosphinoindoline (8) (135.9 mg, 0.22 mmol (80 % purity is assumed)) and bis(norbornadiene)rhodium(I) tetrafluoroborate (82.3 mg, 0.22 mmol) are stirred in dichloromethane (5 mL). The mixture was stirred for 10 minutes at RT. Then the solvent is removed under reduced pressure, resulting in a red solid. 1H-NMR (CDCl3, 300 MHz, 298K): δ (ppm) 8.2 – 6.6 (m, 26 H), 4.2 – 3 (m, 3 H). 31P{1H}-NMR (CDCl3, 121.5 MHz, 298 K) δ (ppm) 133 (d), 52 (d)

Hydrogenation Experiments
Procedure for atmospheric hydrogenation of dimethyl itaconate.

Rh(1-(s)-Binol-P -2-(diphenyl)phosphinoindoline)(nbd) (11) (2.79 mg, 3 μmol) and dimethyl itaconate (A) (47.4 mg, 0.30 mmol) were dissolved in Dichloromethane (3 mL). H2-gas was bubbled through for 1.5 h and subsequently analyzed on a chiral GC-column.

Procedure for pressurized hydrogenation screening experiments.

The hydrogenation experiments were carried out in a stainless steel autoclave (150 mL) charged with an insert suitable for 14 reaction vessels (including Teflon mini stirring bars) for conducting parallel reactions. 7 reaction vessels were charged with 1.0 μmol of Rh(1-(s)-Binol-P -2-(diphenyl)phosphinoindoline)(nbd) (11) and 0.10 mmol of alkene substrate (AF and N-(1-phenylvinyl)acetamide) in 1.0 mL of dichloromethane. 7 other reaction vessels were charged with 1.0 μmol of Rh(nbd)2BF4, 2.2 μmol of triphenylphosphine and 0.10 mmol of alkene substrate (AF and N-(1-phenylvinyl)acetamide)in 1.0 mL of dichloromethane. Before starting the catalytic reactions, the charged autoclave was purged three times with 15 bar of dihydrogen and then pressurized at 10 bar of dihydrogen. The reaction mixtures were stirred at RT for 16 hours. After catalysis the pressure was reduced to 1.0 bar and the conversion and enatiomeric purity were determined by chiral GC.


For this bachelor thesis the asymmetric lithiation of indoline, to attempt the synthesis of INDOLINEPhos, was investigated, but also the asymmetric synthesis of monodentate ligands. The synthesis of the ligands is based on the same technique used for INDOLINEPhos, which is based on the work of Beak et alError: Reference source not found. This research is not related to the main goal of the project, regarding the scope of INDOLPhos, therefore it is presented as the appendix.

In this group already several supramolecular approaches have been used in asymmetric hydrogenation, for example zinc porphyrin based,13 UREAphos14 and METAMORPhos.15 Also in other groups many supramolecular approaches have been used, for example by Takacs et al.16 The supramolecular approach used in this project is based on double hydrogen bonding. A library of rhodium complexes has already been reported by Breit et al, containing a supramolecular combination of two monodentate ligands (Figure 12), which are derivatives of thymine and adenine (Figure 13), with a chiral phosphine attached to it.17 They associate with a double hydrogen bond as proposed by Watson and Crick for the bonding of adenine and thymine in deoxyribonucleic acid.18

Figure 12 library of chiral aminopyridine and isoquinolone ligand metal complexes

Figure 13 base pairing of adenine and thymine

In this appendix a novel ligand system will be discussed, of which synthesis was attempted. These novel ligands are tautomers of each other (Figure 14), meaning only one component needs to be synthesised, which will tautomerise in solution and will form a complex with rhodium using self assembly. Because of the double supramolecular bond, a rigid backbone is expected. The stereogenic centre is the carbon atom next to the phosphine, which results in two stereogenic centres in this type of bidentate ligand, one for every tautomer. A single component has c1-symmetry and combined the bidentate ligand will have c2-symmetry, like BINAP by Noyori et al.19

Figure 14 supramolecular bidentate ligand rhodium complex

Ligand Synthesis

The synthesis of the ligands 12, 13 and 14 was attempted, but was unfortunately unsuccessful. In the case of ligand 12, synthesis failed because the secondary amine was phosphorylated as well as the β carbon (Figure 16). Perhaps the secondary amine can be protected to make it less nucleophilic, but that wasn’t further investigated.

7-azaindoline was prepared from the nickel catalyzed hydrogenation of 7-azaindole. Synthesis of 13 failed because it was not possible to selectively phosphorylate on the 2-position of the 7-azaindoline. Phosphorylation also occurred at the 7-position resulting in an aminophosphine.

For ligand 14, the double five member ring first needed to be synthesized according to literature, but unfortunately the purification was very tedious and the yield was only <11%, while in literature a yield of 36% was reached20. There was too little of the double ring to continue to the next step in synthesis.

Figure 15 bidentate ligands consisting of supramolecular combination of tautomers

Figure 16 Phosphorylation of secondary amine of 1


I would like to thank prof. dr. Joost Reek for giving me the opportunity to do this bachelor project. I would also like to thank the entire Homogeneous Catalysis group for their help and amiability during this project. Especially, I would like to thank drs. Jeroen Wassenaar for his daily guidance.
  1. References

1 J. G. de Vries, C. J. Elsevier, The Handbook of Homogeneous Hydrogenation, Wiley-VCH, Weinheim, 2007

2 P. Knightley, H. Evans, E. Potter, M. Wallace, Suffer The Children: The Story of Thalidomid , The Viking Press, New York, 1979

3 M. Iguchi, J. Chem. Soc. Japan, 1938, 60, 1287

4 F. Jardine, J. Osborn, G. Wilkinson, J. Young, Chem. Ind. (London), 1965, 560

5 J. Halpern, D. P. Riley, A. S. C. Chan, J. J. Pluth, J. Am. Chem. Soc., 1977, 98, 4450

6 J. Wassenaar, M. Kuil, J. N. H. Reek, adv. Synth. Catal., 2008, 350, 1610 - 1614

7 J. Wassenaar, unpublished manuscript, INDOLPhos: Readily available hybrid phosphine-phosphoramidite ligands for asymmetric catalysis

8 H. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth. Catal., 2002, 345, 103 - 151

9 K. M. Bertini Gross, Y. M. Jun, P. Beak, J. Org. Chem., 1997, 62, 7679 - 7689

10 X. Dong, C. Erkay, J. Mol. Catal. A: Chem, 2003, 211, 73 -81.

11 Y. Sun, R. N. Landau, J. Wang, C. LeBlond, D. G. Blackmond, J. Am. Chem. Soc., 1996, 118(6), 1348 - 1353

12 A. L. Minnaard, B. L. Feringa, L. Lefort, J. G. de Vries, , Acc. Chem. Res., 2007, 40 , 1267 – 1277

13 V. F. Slagt, P. W. N. M. van Leeuwen, J. N. H. Reek, Chem. Commun., 2003, 2474 -2475

14 A. J. Sandee, A. M. van der Brurg, J. N. H. Reek, Chem. Commun., 2006, 864 - 866

15 F. W. Patureau, M. Kuil, A. J. Sandee, J. N. H. Reek, Angew. Chem. Int. Ed., 2008, 47, 3180 - 3183

16 J. M. Takacs, D. S. Reddy, S. A. Moteki, D. Wu, H. Palencia, J. Am. Chem. Soc., 2004, 126, 4494 - 4495

17 M. Weis, C. Waloch, W. Seiche, B. Breit, J. Am. Chem. Soc., 2006, 128, 4188 - 4189

18 J. D. Watson, F. H. D. Crick, Nature, 1953, 171, 738 - 739

19 A. Miyashita, A. Yasuda, H. Takaya, K. Toriumi, T. Ito, T. Souchi, R. Noyori, J. Am. Chem. Soc., 1980, 102, 7932

20 T. Mas, R. M. Claramunt, M. D. Santa Maria, D. Sanz, S. H. Alarcon, M. Perez-Torralba, J. Elguero, ARKIVOC, 2002, 48 - 61

The database is protected by copyright ©hestories.info 2017
send message

    Main page