Recent Progress in Heterogeneous Asymmetric ... - ACS Publications

Recent Progress in Heterogeneous Asymmetric...

0 downloads 94 Views 30MB Size

Review pubs.acs.org/CR

Recent Progress in Heterogeneous Asymmetric Hydrogenation of CO and CC Bonds on Supported Noble Metal Catalysts Fabian Meemken and Alfons Baiker* Institute for Chemical and Biochemical Engineering, Department of Chemistry and Applied Biosciences, ETH-Zurich, Hönggerberg, HCI, Vladimir Prelog Weg 1, CH-8093 Zurich, Switzerland ABSTRACT: The ease of separation, simple regeneration, and the usually high stability of solid catalysts facilitating continuous production processes have stimulated the development of heterogeneous asymmetric hydrogenation catalysis. The simplest and so far most promising strategy to induce enantioselectivity to solid metal catalysts is their modiﬁcation by chiral organic compounds, as most prominently represented by the cinchona-modiﬁed Pt and Pd catalysts for the asymmetric hydrogenation of activated C O and CC bonds. In this Review, we provide a systematic account of the research accomplished in the past decade on noble metal-based heterogeneous asymmetric hydrogenation of prochiral CO and CC bonds, including all important facets of these catalytic systems. The advances made are critically analyzed, and future research challenges are identiﬁed.

CONTENTS 1. Introduction 2. Hydrogenation of CO Bonds 2.1. Catalysts 2.1.1. Platinum 2.1.2. Other Noble Metals 2.2. Chiral Modiﬁers 2.2.1. Cinchona Alkaloids and Derivatives 2.2.2. Synthetic Modiﬁers 2.2.3. Additives and Comodiﬁers 2.3. Adsorption of Modiﬁers 2.3.1. Cinchonidine Adsorbed on Pt 2.3.2. Cinchona Derivatives Adsorbed on Pt and Other Noble Metals 2.3.3. Synthetic Modiﬁers 2.4. Substrates 2.5. Solvent Eﬀects 2.6. Reaction Pathway and Kinetics 2.6.1. Reaction Pathway 2.6.2. Enantiodiﬀerentiating Surface Complex 2.6.3. Kinetics and Rate Enhancement 2.7. Nonlinear Behavior and Inversion of Enantioselectivity 2.8. Fundamental Understanding from Surface Science Studies 2.8.1. Ex Situ Versus In Situ Studies 2.8.2. Studies on Pt Model Surfaces 2.8.3. Studies on Other Noble Metal Model Surfaces 3. Hydrogenation of CC Bonds 3.1. Catalysts 3.1.1. Chirally Modiﬁed Pd 3.1.2. Asymmetry Induced by (S)-Proline © 2017 American Chemical Society

3.2. Chiral Modiﬁers 3.2.1. Cinchona Alkaloids and Derivatives 3.2.2. Inﬂuence of Chiral Modiﬁer Concentration 3.2.3. Amine Additives 3.2.4. Proline-Based Chiral Modiﬁers 3.3. Substrates 3.3.1. Phenyl Cinnamic Acid and Derivatives 3.3.2. New Substrate Classes 3.4. Mechanistic Aspects 3.4.1. Cinchona-Modiﬁed Pd 3.4.2. Proline-Mediated Asymmetry on Pd 4. Conclusions and Outlook Author Information Corresponding Author ORCID Notes Biographies Acknowledgments Abbreviations References

11522 11523 11523 11524 11528 11529 11529 11531 11531 11532 11532 11533 11534 11535 11535 11536 11536 11537 11539 11540

11550 11550 11551 11552 11554 11555 11555 11556 11557 11557 11558 11561 11562 11562 11562 11562 11562 11562 11562 11563

1. INTRODUCTION Chirality plays an important role in many products such as pharmaceuticals, agrochemicals, ﬂavors, and fragrances. The synthesis of such compounds often includes catalytic asymmetric hydrogenation as a crucial step. In principle, there are three diﬀerent catalytic strategies for asymmetric hydrogenation: homogeneous, heterogeneous, and enzymatic

11541 11541 11542 11545 11547 11547 11547 11550

Received: May 12, 2017 Published: September 5, 2017 11522

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 1. Most Eﬃcient Asymmetric Hydrogenations of Activated Ketones on Chirally-Modiﬁed Pt and the required Reaction Conditions;11−18 Chiral Modiﬁer Structures Are Shown in Scheme 2

to be considered for the design of suitable catalytic systems for the asymmetric hydrogenation based on chirally modiﬁed noble metals, such as the molecular structures of the prochiral substrate (Scheme 1) and the chiral modiﬁer (Scheme 2), the properties of the heterogeneous catalyst, and the solvent used.10 In Scheme 1, some of the most eﬃcient asymmetric ketone hydrogenations and the respective reaction conditions for the chirally modiﬁed Pt catalyst are summarized. In this section, recent progress in the noble metal-catalyzed asymmetric ketone hydrogenation, including various aspects such as the development of new catalysts and chiral modiﬁers, the structural characterization of adsorbed modiﬁers, the role of additives and solvents, and the advancements in the mechanistic understanding of chirally modiﬁed metal catalysts, is reviewed

catalysis. Each of these strategies has its own advantages and limitations as concerns activity, selectivity, stability, handling, and regeneration of the catalyst. The ease of separation, simple regeneration, and the usually high stability of solid catalysts as well as the direct access to continuous production processes have spurred the development of heterogeneous asymmetric hydrogenation catalysis. Unfortunately, only very few intrinsically chiral solid materials exist that are active in hydrogenation and induce enantioselectivity. This limitation has triggered the development of various methods to impart chirality to catalytically active solid surfaces, embracing tethering and grafting of suitable metal complexes with chiral ligands as well as their encapsulation in a solid matrix. Another, practically much simpler strategy is the chiral modiﬁcation of metal catalysts, which is the focus of this Review. While various earlier reviews cover synthetic as well as mechanistic aspects of this ﬁeld,1−9 there is no comprehensive review that covers research done in the past decade. The aim of this Review is to provide a critical survey of the progress made after 2006. As a consequence, reference to earlier work is only made when necessary in the context of discussion of the recent developments. For earlier work the reader should consult the previous reviews.

2.1. Catalysts

Research done since 2006 deals mainly with the structural optimization of Pt-, Rh-, and Ir-based catalysts. Optimization of the size and shape of these supported noble metal nanoparticles (NPs) as well as the support material were recently in focus. Studies on the eﬀect of the acidity and basicity of the support showed that tuning of these properties can signiﬁcantly inﬂuence the catalytic performance. While most of the research was directed toward structural optimization of the wellestablished alumina-supported Pt catalyst, new Pt-based catalysts supported on various materials were developed that show interesting catalytic behavior.

2. HYDROGENATION OF CO BONDS Asymmetric hydrogenation of activated ketones on cinchonamodiﬁed Pt catalyst is probably the most studied and best understood chiral surface transformation. Several factors have 11523

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 2. Overview of Various Cinchona Alkaloids and the Most Eﬃcient Synthetic Chiral Modiﬁersa

a All eﬃcient chiral modiﬁers have three features in common: an aromatic moiety for adsorptive anchoring on the catalyst, stereogenic region(s), and an interacting amine moiety to guide the prochiral substrate.17,19,20 For the structure of the indicated substrate, refer to Scheme 1.

2.1.1. Platinum. Earlier studies of the enantioselective hydrogenation of ethyl pyruvate (EP) on supported Pt catalysts showed a signiﬁcant Pt particle-size dependence of both activity and enantioselectivity, indicating that Pt NPs of ∼3 nm show optimal catalytic performance.21 However, studies addressing the eﬀect of the morphology of the Pt NPs were missing, which prompted Schmidt et al.22 to study the structure sensitivity of EP as well as ketopantolactone (KPL) hydrogenation on chirally modiﬁed Pt NPs of 10 nm average size but diﬀerent morphologies (cubic, cubooctahedral, and octahedral), exposing diﬀerent fractions of (100) and (111) faces at the surface. The structural features together with the catalytic data shown in Figure 1 corroborate that the asymmetric hydrogenation is structure-sensitive, whereas the corresponding racemic hydrogenation seems to be structure-insensitive. Catalytic and theoretical investigations indicated that the likely cause for the particle-shape dependency of the enantioselective hydrogenation is the structure sensitivity of the adsorption of the chiral modiﬁer cinchonidine (CD) . The authors proposed that an ideal catalyst for the hydrogenation of activated ketones should contain predominantly Pt (111)

terraces, because this crystallographic face was found to be most active and enantioselective. Mechanistic insight into the structure sensitivity of the asymmetric catalytic hydrogenation of EP was provided by Guan et al., employing single-crystal Pt electrodes coupled with Raman spectroscopic analysis.23 The μ2(C,O) EP adsorbate prevailed at pristine Pt(111) and Pt(100), while the halfhydrogenation state (HHS) of the ketone was only observed after the introduction of surface defects by electrochemical roughening, indicating the requirement for low coordination sites for observing the HHS (also compare section 2.6.2 for the role of the HHS in the enantiodiﬀerentiating surface complex). Other studies14,24 focused on the inﬂuence of acidic and basic support properties on the performance of Pt-based catalysts. The performance of the standard Pt/Al2O3 was compared to Pt supported on various mixed oxides with enhanced acidic and basic properties, respectively. Hoxha et al.14 studied the enantioselective hydrogenation of methyl benzoylformate (MBF) and KPL on CD-modiﬁed ﬂame-made Pt/Al2O3−SiO2 and Pt/Al2O3−Cs2O catalysts. The acidic and basic properties of the support were systematically varied by addition of 5−80 wt % SiO2 or 0.25−10 wt % Cs2O to the 11524

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

shown that the acid−base properties of the Pt catalyst have a signiﬁcant eﬀect on the adsorption and hydrogenation of the CD modiﬁer.25 As illustrated in Figure 3, the diastereoselectivity during hydrogenation of the heteroaromatic ring of the quinoline moiety of CD depends on the catalyst used, and it was evidenced that the pro-(S) adsorption mode dominates on Pt/Al2O3 and Pt/Al2O3−SiO2, whereas pro-(R) prevails on Pt/ Al2O3−Cs2O. This switch of the adsorption mode of CD depending on the acid−base properties of the support had so far not been taken into account in studies on the mechanism of the Pt-catalyzed enantioselective hydrogenation and is likely to play a role when explaining the observed enantiodiﬀerentiation on these supported Pt systems. Li et al.26 prepared Pt catalysts supported on composite materials consisting of carbon incorporated with diﬀerent alumina contents by a chelate-assisted coassembly method. Carbon incorporated with 10−15 wt % alumina was favorable for achieving high Pt dispersion and retained the mesostructure of carbon. Pt particles supported on 15 wt % alumina−carbon composite modiﬁed by CD aﬀorded the highest ee (85%) in the asymmetric hydrogenation of ethyl 2-oxo-4-phenylbutyrate in acetic acid. Particularly interesting is the development of some new types of catalysts. Li and co-workers27 encapsulated Pt NPs in carbon nanotubes (CNTs) and tested these catalysts in the enantioselective hydrogenation of EP. Using CD as the chiral modiﬁer, these catalysts showed outstanding catalytic performance and a turnover frequency (TOF) exceeding 100.000 h−1 was reported at 60 bar H2 pressure and 293 K. Interestingly, when the Pt NPs were deposited onto the outer surface of the CNTs (Pt/CNTs(out)), TOF and ee were signiﬁcantly lower, as schematically shown in Figure 4. The high activity and enantioselectivity of the encapsulated Pt NPs was attributed to the unique properties of the nanochannels of CNTs, which showed higher adsorption uptake of CD compared to Pt/ CNTs(out) and a commercial carbon-supported Pt catalyst. It will be interesting to see whether this type of catalyst also performs as well in the enantioselective hydrogenation of other activated ketones. Later the same group described an unexpected eﬀect of water on the asymmetric hydrogenation of α-ketoesters on Pt NPs conﬁned in CNTs.28 By employing an acetic acid/water mixture as solvent, they achieved up to 95% ee in the asymmetric hydrogenation of EP, whereas in anhydrous acetic acid, the enantioselectivity was only 86%. The authors attributed the higher ee in acetic acid/water mixture to an enrichment of the chiral modiﬁer CD in the channels and also an enhancement of the ability of the channels to discriminate between the reactant and product molecules. More recently, the group investigated the eﬀect of the chemical state of Pt NPs deposited in or outside of the channels of CNTs in the enantioselective hydrogenation of EP.29 They showed that highly oxidized Pt species can be stabilized inside the channels of CNTs when Na+ ions are present. The high activity of this catalyst was attributed to the more electrophilic Pt species, which promote the interaction between the chiral modiﬁer and the reactant with the Pt and lead to an enrichment of hydrogen in the channels. They concluded that the function of CNTs is not only to enrich the concentration of reactant and modiﬁer molecules in the channels but also to stabilize Pt species in higher oxidation state. Sharma and Sharma30,31 synthesized Pt catalysts using Tween 20 as a facet-directing agent. The Pt(111)-rich NPs were deposited on various carbon supports including multiwall

Figure 1. Eﬀect of particle shape on catalytic performance of 10 nm sized Pt particles in the enantioselective hydrogenation of KPL. Adapted with permission from ref 22. Copyright 2009 American Chemical Society.

alumina support. As shown in Figure 2, the enantioselectivity was improved by enhancing the acidity via silica addition to the alumina support, reaching a maximum at ∼30 wt % silica content in the alumina−silica mixed oxide. By proper tuning of the acidity of the alumina−silica support, the highest enantioselectivity reported hitherto for the industrially important hydrogenation of KPL to (R)-pantolactone (enantiomeric excess (ee) of 94%) was achieved. Admixing of Cs2O to the alumina had an adverse eﬀect, indicating that basic supports are detrimental for enantioselectivity in the investigated enantioselective hydrogenations. The study revealed a critical inﬂuence of the support ionicity on the electronic properties of the Pt NPs, which was corroborated by a good correlation between the enantioselectivity and the ratio of CO adsorbed in bridged-to-linear (B/L) geometry. In a subsequent study24 the same group investigated the inﬂuence of acidic and basic supports on the enantioselective hydrogenation of acetophenone and some ring-substituted derivatives using the most acidic and basic catalysts of their former work and compared their performance to those of the unmodiﬁed and a commercial reference Pt/Al2O3 catalyst. The ee was improved with the acidic support, while it was diminished with the basic one. Interestingly, opposite behavior was observed for the reaction rates: the acidic support lowered the reaction rate, whereas the basic one enhanced it. It was also 11525

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 2. Hydrogenation of KPL over CD-modiﬁed Pt/Al2O3−SiO2 catalysts. Enantioselectivity determined at full conversion is plotted as a function of SiO2 content in the support (A), Pt particle size (B), and “normalized acidity” of the catalysts (C) . Variation of the ee with the electronic properties of Pt reﬂected by the B/L ratio of CO adsorption (D) . Filled symbols, KPL hydrogenation; open symbols, MBF hydrogenation. Adapted with permission from ref 14. Copyright 2010 Elsevier.

Figure 4. Asymmetric hydrogenation of α-ketoesters on the Pt nanoparticles encapsulated within the CNTs (Pt/CNTs(in)) and adsorbed onto CNTs (Pt/CNTs(out)) with CD as a chiral modiﬁer. Reproduced with permission from ref 27. Copyright 2011 Wiley-VCH.

Figure 3. Two major adsorption modes of CD on Pt aﬀording (S)- or (R)-CDH6 upon hydrogenation. Adapted with permission from ref 25. Copyright 2010 Elsevier.

inconsistencies should be clariﬁed. For example, in the hydrogenation of methyl pyruvate (MP), practically “perfect” enantioselection was reported, which seems quite unrealistic as an enantioselectivity of ∼99.9% ee has not even been reported in enantioselective homogeneous asymmetric hydrogenations.32 Considering the structure sensitivity of chirally modiﬁed metals, the very high enantioselectivity looks surprising as the reported Pt particle size varied signiﬁcantly between 2 and 20

CNTs, and as-prepared catalysts were tested in the asymmetric hydrogenation of several pyruvates using cinchonine (CN) as chiral modiﬁer. While very high ee is generally obtained with the chosen substrates (Scheme 1), the reported catalytic data, shown in Table 1, appears rather doubtful, and several 11526

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Table 1. Asymmetric Hydrogenation of Activated αKetoesters on Pt-Functionalized Multiwall CNT Compositesa (Reproduced with Permission from Ref 31; Copyright The Royal Society of Chemistry)

substrate

yieldb (%)

eec (%)

ethyl 3-bromo-2-oxopropanoate ethyl 3-methyl-2-oxobutanoate ethyl 2-oxo-4-phenylacetate ethyl 2-oxo-2-phenylbutanoate methyl 2-oxopropanoate (MP)

80.2 90.4 92.7 94.4 99.8

93.8 98.7 99.6 99.0 ∼99.9

a

Proposed reaction conditions: reaction temperature, 298 K; Pt HNC/MWNT catalyst amount, 30 mg; reaction time, 12 h; ethyl derivative, 75 μL; CN, 4 mg; acetic acid, 1.5 mL. bIsolated by bulb-tobulb distillation. cDetermined by high-performance liquid chromatography (HPLC) analysis.

nm. Furthermore, at 10 bar H2 pressure the TOFs on the level of 10−30 h−1 (assuming a Pt dispersion between 30−100% as this value is missing in the publication) are very low and the reported activity of the catalyst indicates severe limitations of the catalytic reaction, rendering high enantioselectivity of the heterogeneous hydrogenation even more doubtful.33 For comparison, the Pt catalyst containing mainly (111) facets synthesized by Schmidt et al.22 provided a TOF of 1.350 h−1 using only 1 bar H2 pressure and otherwise very similar experimental conditions, e.g., the same solvent and similar molar ratios. The successful application of Pt nanowires with a diameter of ∼1 nm was also reported.34 In the asymmetric hydrogenation of EP in water (1% acetic acid addition), an ee of 78% was achieved, at 1 bar and below 0.1 bar H2 pressures, which is the best ee reported using water as the solvent. The unsupported Pt catalyst was stable under moisture and air and could be easily recycled via phase separation without signiﬁcant loss of catalytic performance. An entirely diﬀerent catalyst-design strategy was applied by Keilitz et al.,35 who prepared a Pt catalyst consisting of a polymeric dendritic core−shell architecture in which the Pt particles were homogeneously stabilized. The dendritic polymer architecture shown in Figure 5 was made up of hyperbranched polyglycerol as hydrophilic core, alkanedioic acid as the inner hydrophobic shell, and monomethylated poly(ethylene glycol) as the outer hydrophilic shell. As-prepared catalysts were chirally modiﬁed with CD and tested in the asymmetric hydrogenation of EP, aﬀording up to 75% ee. Recycling tests indicated that the catalysts are stable up to eight recycles, but show some loss of activity due to aggregation of the Pt particles in further recycles. Beier et al.36 presented another successful catalyst-design strategy for asymmetric hydrogenation, which is illustrated in Figure 6. Pt NPs with a mean size of 2−3 nm and narrow size distribution were stabilized with various water-insoluble ionic liquids (ILs) . The size of the NPs could be tuned by employing CD as a chiral costabilizer. As-prepared catalysts provided high ee and fast hydrogenation rate at 30 bar H2 pressure in the enantioselective hydrogenation of MBF, indicating that this

Figure 5. (A) Schematic representation of dendritic double-shell polymer and general structures of dendritic core−multishell architectures. (B) TEM micrographs and particle-size distributions of samples (a) after preparation of Pt NPs at Pt/polymer 0.5 g g−1 and (b) after addition of extra polymer to obtain 0.1 g g−1. Adapted with permission from ref 35. Copyright 2010 Wiley-VCH.

design strategy may open the door to catalyst systems with enhanced performance. Tethering the cinchona alkaloid modiﬁers to the Pt-based solid catalyst could be beneﬁcial, in case an intolerable amount of the cinchona alkaloid, requiring additional separation and puriﬁcation steps, may be contained in the product. However, earlier work37 pursuing this strategy only resulted in chiral catalysts, which showed relatively poor performance. Recently, Zaera and co-workers made new attempts in this direction.38−40 Figure 7 shows schematically the strategy the authors followed to tether CD adjacent to the silica-supported Pt NPs. The study indicated that a spatially controlled manner leads to better catalytic performance in the enantioselective hydrogenation of EP than random distribution of the cinchona alkaloid on the silica support. Campos et al.41 synthesized a series of chiral porous titanate nanotubes on which 11-trimethoxysilyl cinchonidine (TMSCD) was anchored via direct surface reaction. These hybrid solids served as supports and were loaded with 1 wt % Pt. The well-characterized catalysts were tested in the enantioselective hydrogenation of 1-phenyl-1,2-propanedione (PPD) . Maximum ee and TOF achieved under optimized conditions were 37% (at 15 wt % nominal content of TMS-CD) and 9900 h−1 (at 10 wt % nominal content of TMS-CD) . For the bestperforming catalyst, hydrogen pressure, solvent eﬀects, and 11527

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

dihydro-4′,5′-diphenyl-2-(6-cyanopyridyl)imidazoline (PyIm) immobilized on the silica support. These catalysts were also examined for the enantioselective hydrogenation of PPD and aﬀorded ee values up to 46%. The enantiodiﬀerentiation of the catalysts was speculated to originate from substrate−modiﬁer interactions involving hydrogen bonding between the ketocarbonyl O atom and the NH moiety of PyIm. Modiﬁer-surface concentration, hydrogen pressure, and solvent nature were identiﬁed as crucial properties for optimal catalytic performance. Catalyst recycling tests showed a signiﬁcant loss in ee with each recycle, which the authors ascribed to some leaching. Azmat et al.43 used another approach. They developed a single-unit catalyst system for the enantioselective hydrogenation of EP. The chiral modiﬁer CD was tethered directly without prior modiﬁcation over carboxylate-functionalized SBA-15 by the reaction of the vinyl group in CD with −COOH groups in functionalized SBA-15 through ester linkage. Subsequently Pt NPs were deposited on the CDtethered SBA-15. The highest ee achieved with this catalyst in the asymmetric hydrogenation of EP was 71%, and the catalyst could be recycled three times without signiﬁcant loss in ee. Further work on tethered catalyst systems seems to be necessary to ﬁnally assess their potential for asymmetric hydrogenation. Functionalization of Pt NPs with (S)-proline (Pr) as chiral source has also yielded encouraging results in the asymmetric hydrogenation of acetophenone and ethylacetoacetate with enantioselectivities of 14%44 and 33%,45 respectively. Better recyclability of the enantioselective heterogeneous catalyst could be a promising advantage of functionalizing Pt NPs with chiral ligands. However, while the chemoselectivity of the recycled catalyst was found to remain on the same level, the stereoselectivity was completely lost.44 Future work in this direction has to show whether the recyclability can be improved and, more importantly, whether the enantioselectivity can be controlled more strongly by adapting the molecular structure of the ligand. In chemoselective hydrogenation of unsaturated aldehydes, Medlin and co-workers have recently shown that self-assembled monolayers of thiols on a Pt catalyst are suitable to control the chemoselectivity toward unsaturated alcohols.46 In the hydrogenation of cinnamaldehyde, the molecular structure of the thiol was found to have a pronounced eﬀect on the selectivity toward the alcohol and noncovalent molecular interactions were proposed to be responsible for controlling the chemoselectivity.47 It will be interesting to see whether this approach can be adapted to induce enantioselectivity on noble metals. 2.1.2. Other Noble Metals. Studies of the inﬂuence of the metal particle size on the catalytic behavior of supported Rhand Ir-based catalysts provide new inside into the structure sensitivity on these noble metals. Hoxha et al.48 prepared a series of Rh/Al2O3 catalysts with relatively narrow size distribution by ﬂame synthesis and tested these catalysts in the enantioselective hydrogenation of EP and ethyl 3-methyl-2oxobutyrate using quinine (QN) as chiral modiﬁer. Both reactions were strongly dependent on the mean Rh particle size. In the range from 0.9 to 1.7 nm, the TOF as well as the ee strongly decreased with smaller particle size, showing a similar trend as observed with Pt catalysts. This behavior is shown in Figure 8 for the asymmetric hydrogenation of EP. Interestingly, the structure sensitivity was much stronger for hydrogenations performed at low pressure (1 bar) compared to those carried out at 10−100 bar. This phenomenon has been explained by

Figure 6. (A) Strategy for the synthesis of IL-supported Pt catalyst for asymmetric hydrogenation. (B) TEM images and size distributions of Pt@[BMIm][PF6] (a) and PtCD@[BMIm][PF6] (b). Modiﬁed with permission from ref 36. Copyright 2012 American Chemical Society.

Figure 7. Schematic representation of the approach reported for the spatially controlled tethering of cinchona alkaloids next to Pt NPs dispersed on a high-surface-area silica support. Reproduced with permission from ref 39. Copyright 2015 The Royal Society of Chemistry.

recycling were studied. The same researchers42 also synthesized Pt/silica catalysts with a new chiral modiﬁer (4′R,5′S)-4′,5′11528

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

The eﬀect of the iridium particle size and the inﬂuence of the modiﬁer concentration on the catalytic performance of Ir/SiO2 in the enantioselective hydrogenation of EP, PPD, and acetophenone has been studied by Marzialetti et al.,50 who reported that both conversion and ee increase with Ir particle size. However, the size range of Ir particles was rather small (1.7−3.1 nm according to transmission electron microscopy (TEM) analysis), and therefore, further studies may be necessary to ﬁrmly conclude the structure sensitivity. A volcano-type curve was found for the dependence of activity and enantioselectivity on CD concentration, which was attributed to the competitive adsorption of substrate and modiﬁer on the catalyst surface. In another study51 the same group studied the eﬀect of the support and of promoters on the catalytic behavior of Ir in the enantioselective hydrogenation of PPD. They showed that catalysts with Ir present as Irδ+ species give rise to an increase of the reaction rate, which was attributed to the polarization of the carbonyl bond. They propose that these species may be generated during high-temperature reduction, in the case of a partially reducible support, or by the addition of promoter oxides. The eﬀect of addition of promoters to Ir/silica catalysts was also addressed by Jiang et al.52 They prepared silica-supported Ir catalysts, which were stabilized by triphenylphosphate (PPh3). When modiﬁed with a chiral diamine, derived from cinchona alkaloids, this catalyst exhibited high activity and high enantioselectivity in the hydrogenation for ortho-substituted aromatic ketones. Considerably less is known about the potential of supported Ru catalysts for enantioselective hydrogenation. Ye et al.53 prepared a series of oxide-supported Ru catalysts modiﬁed with achiral PPh3 and chiral (1R,2R)-1,2-diphenylethylenediamine. In the asymmetric hydrogenation of aromatic ketones, the enantioselectivity depended on the oxidic support material and followed the order MgO > Al2O3 > CeO2 > ZnO > SiO2, which was correlated with their surface basicity. The best performing catalyst Ru/MgO did not show a signiﬁcant decrease in ee after several recycles. The study indicated that the catalytic performance is inﬂuenced by the acidic and basic sites of the oxide support as well as by the size of the Ru particles.

Figure 8. Structure sensitivity of Rh-based catalyst. Inﬂuence of Rh mean particle size on the enantioselectivity and reaction rate of the hydrogenation of EP to (R)-ethyl lactate over QN-modiﬁed Rh/Al2O3 catalysts in toluene at 1 bar; ﬁlled symbols, data at 9 ± 2% conversion; open symbols, data after 2 h reaction time. Adapted with permission from ref 48. Copyright 2009 Elsevier.

2.2. Chiral Modiﬁers

The organic modiﬁer used to bestow chiral selection to the noble metal surface is of paramount importance for eﬃcient asymmetric hydrogenation. While the originally used cinchona alkaloids54,55 are still the most versatile chiral modiﬁers, in the past few decades there has been continuous striving to discover synthetic modiﬁers, which may allow better tailoring of their structure and adopting to a speciﬁc substrate. Thus, in Scheme 2 we distinguish between synthetic modiﬁers and cinchona alkaloids and derivatives thereof. Most of the recent research has focused on the exploration of the relationship between the structural properties of cinchona alkaloids and their eﬃciency in enantiodiﬀerentiation. 2.2.1. Cinchona Alkaloids and Derivatives. Previous work up to 2006, summarized in ref 3, shows that cinchona alkaloids and some of their derivatives can induce strong stereochemical control in hydrogenations on noble metal catalysts. It is generally accepted that cinchona alkaloids are excellent chiral modiﬁers as their molecular structure features three crucial functions: an aromatic anchoring unit, a stereogenic region(s), and an interacting moiety inclined to form hydrogen bonding with the substrate.3,56 It has been

the altered inﬂuence of the deactivation caused by the Al2O3 catalyzed aldol condensation, which is more relevant at 1 bar due to the lower reaction rate of the hydrogenation. These results corroborate that there is no advantage in using noble metal NPs below 2 nm, as also emerges from the previous studies of the enantioselective hydrogenation over supported Pt catalysts. Although Ir-based catalysts have earlier been shown to perform less well in the asymmetric hydrogenation of αketoesters, some studies have been devoted to further exploring their potential. For this purpose Ir/Al2O3 and Ir/SiO2 catalysts were prepared using two diﬀerent preparation methods, that is, incipient wetness impregnation and ﬂame spray pyrolysis.49 These catalysts were tested using the enantioselective hydrogenation of EP and CD as the chiral modiﬁer. Diﬀerent enantioselectivity to (R)-ethyl lactate was achieved by varying the support and the preparation method. On the basis of the structural characterization of the catalysts, the authors proposed that the diﬀerent catalytic behavior has to be attributed to the diﬀerent surface structure of the Ir NPs caused by the support interaction. 11529

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

the behavior of CD and various O-phenyl ether derivatives of CD, such as O-phenyl CD (PhOCDs), O-(3,5-dimethylphenyl)-CD (dMePhOCD), and O-[3,5-bis(triﬂuoromethyl)phenyl]-CD (tFPhOCD). Mondelli et al.66 investigated the enantiodiﬀerentiating behavior of novel C9 ﬂuorinated cinchona alkaloid derivatives in the Pt-catalyzed asymmetric hydrogenation of α-ketoesters. The enantioselectivity was close to those observed with the parent alkaloids, and a direct comparison with the conformationally labile deoxy-cinchonidine indicated that the C9 ﬂuorine atom is important for good performance. The inﬂuence of the conﬁguration at the C8 and C9 positions of cinchona alkaloids was also in the focus of the study of Schmidt et al.69 They compared the eﬃciency of CD, CN, and 9-epi-cinchonidine (ECD) as chiral modiﬁers in the hydrogenation of various ketones, including MBF, KPL, methylglyoxal dimethylacetal (PA), and triﬂuoroacetophenone (TFAP) on a Pt/Al2O3 catalyst. The catalytic results in Table 3

proposed that the conformation of cinchona alkaloids is also critical, because their conformation in solution or upon adsorption onto metal surfaces can directly inﬂuence their interactions with substrates and, hence, the stereochemical outcome of reactions. There are several factors that can aﬀect the conformational behavior of cinchona alkaloids, such as temperature,57 protonation of the quinuclidine moiety,58,59 and presence of solvents60 as well as peripheral groups in the molecular scaﬀold.61 More recent studies mainly focused on the performance of various ether derivatives of cinchona alkaloids61−66 as well as several rigid cinchona alkaloid O-ethers67−72 in the asymmetric hydrogenation of various substrates. The lesson learned from these studies is that bulky substituents in C9−O ethers generally lead to lower ee compared to small substituents and the parent cinchona alkaloids, as shown in Table 2 for the asymmetric hydroTable 2. Eﬃciency of Ether Derivatives of CD Used As Chiral Modiﬁers of Pt/Al2O3 in the Hydrogenation of Activated Ketones in THF at 1 bar (Reproduced with Permission from Ref 73; Copyright 2004 Elsevier)

Table 3. Enantioselective Hydrogenation of Various Activated Ketones Using CD, ECN, and CN as Chiral Modiﬁersa

substrate

CD

PhOCD

dMePhOCD

tFPhOCD

substrate

EP KPL

80 (R) 50 (R)

21 (S) 21 (S)

26 (S) 36 (S)

25 (S) 16 (S)

MBF

genations of EP and KPL in the solvent tetrahydrofuran (THF). Interestingly, some substituents even lead to the opposite sense of enantiodiﬀerentiation than that induced by the parent cinchona alkaloid. This behavior has been attributed to conformational and steric changes of the adsorbed modiﬁer, as demonstrated by Vargas et al.65 (Figure 9), who compared

modiﬁer

ee (%)

CD ECN CN

92 (R) 80 (R) 88 (S)

CD ECN CN

89 (R) 59 (R) 80 (S)

CD ECN CN

90 (R) 72 (R) 72 (S)

CD ECN CN

80 (R) 57 (R) 42 (S)

KPL

PA

TFAP

rate (mmol g−1 h−1) 29 196 15 191 628 7750 1150 4300 6.2 77 6.7 29 69 58 4.7 42

a

The reaction conditions diﬀer for each substrate; for the experimental conditions, refer to the original work. Reproduced with permission from ref 69. Copyright 2012 Elsevier.

show that the absolute conﬁguration of the main products was not aﬀected by changing the conﬁguration at C9 and that the enantiodiscrimination of the noncyclic α-ketoester (MBF) was the least sensitive to the conformational change in the chiral modiﬁer. In contrast, for the triﬂuoro-actived ketone TFAP a higher ee was observed with ECD (57% ee to the (R)-alcohol) compared to CN (42% ee to the (S)-alcohol) . Nevertheless, for all substrates the rates were signiﬁcantly lower for ECD. Theoretical calculations indicated that the lower reaction rates observed for ECD-modiﬁed surfaces cannot be attributed to the diﬀerent adsorption strength of this modiﬁer. Other cinchona alkaloid derivatives, which have gained considerable attention, are rigid cinchona alkaloid Oethers.70−72 Margitfalvi and Tálas72 investigated the enantiodiﬀerentiating behavior of rigid cinchona alkaloids, including αisocinchonine (ICN) and β-ICN in the asymmetric hydrogenation of EP in toluene at low modiﬁer concentration (Figure 10) . They reported that the ICNs showed signiﬁcantly

Figure 9. (Top) Chemical structures of the cinchona alkaloid modiﬁers. (Bottom) Two views of a stable conformer of PhOCD adsorbed on platinum; in addition to the torsional angles τ1 and τ2, the two degrees of freedom associated with the phenyl ring, torsional angles τ3 and τ4, are also indicated; Pt, gray; C, orange; H, white; N, blue; O, red. The carbon atoms of the quinuclidine (left) and of the phenyl moieties (right) have been darkened. Reproduced with permission from ref 65. Copyright 2007 Wiley VCH. 11530

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 10. Structure of various investigated alkaloids. Reproduced with permission from ref 72. Copyright 2006 Elsevier.

diﬀerent kinetic behavior compared to the ﬂexible cinchona alkaloids, CD and CN, characterized by the absence of rate acceleration, low ee values, and independence of ee on modiﬁer concentration. Other rigid modiﬁers, such as (−)-sparteine and (+)-dihydrovinpocetine (DHVIN), also did not show the typical rate-acceleration behavior of CD and CN. Bucsi et al.71 showed that β-ICN-modiﬁed Pt/Al2O3 also exhibits low ee in the enantioselective hydrogenation of EP. Balázsik et al.70 studied the enantioselective hydrogenation of EP, MBF, KPL, and PA on Pt/Al2O3 modiﬁed with αisoquinine (IQN) and compared the behavior of this rigid modiﬁer with that of QN. Under mild experimental conditions (room temperature (rt), 1 bar H2 pressure), the ee values were lower for IQN than for QN, corroborating the general trend observed in the earlier investigations with rigid modiﬁers. The inversion of enantioselectivity observed previously with β-ICN (which possesses C8(R) and C9(S) atoms) in toluene as solvent did not occur in the presence of IQN (which possesses C8(S) and C9(R) atoms). 2.2.2. Synthetic Modiﬁers. Most previously synthesized modiﬁers have been derived from NEA.3 More recently, Holland et al.74 synthesized a series of imidazolidinone and proline derivatives and tested them concerning their potential as chiral modiﬁers for the Pt-catalyzed asymmetric hydrogenation of KPL, MBF, and TFAP performed at low H2 pressures (1 and 10 bar) . No improved enantiodiﬀerentiation could be achieved compared to the parent cinchona alkaloids. However, the studies revealed a performance enhancement of (S)-Pr-derived modiﬁers, indicating some potential in improving catalyst performance by ﬂuorine introduction. Mondelli et al.75 synthesized various peptides in which the natural amino acid tryptophan (Trp) was used as the ﬁrst member of the peptidic chain. Trp was used because of its structural resemblance to cinchona alkaloids. The amino acid moiety of Trp was elongated via peptidic bond formation with other natural amino acids, and the as-synthesized peptides were tested as chiral modiﬁers in the asymmetric hydrogenation of KPL. Assorted results are depicted in Figure 11. The structure of the asymmetric environment created by the peptide adsorption on the Pt surface was investigated by density functional theory (DFT) calculations. Although the enantioselectivity achieved with these peptides was signiﬁcantly lower compared to that of cinchona alkaloids, peptides may bear an interesting potential for chiral modiﬁcation of noble metal surfaces due to their huge structural variability. Nevertheless, a

Figure 11. In the asymmetric hydrogenation of KPL on Pt catalyst, the amino acid Trp and some Trp-based peptides involving glycine (gly) provided the indicated ee in acetic acid. Adapted from ref 75. Copyright 2009 American Chemical Society.

combinatorial approach may be necessary to ﬁnd the most suitable peptide. Maris et al.76 applied mandelic acid and various derivatives thereof as chiral modiﬁers in the asymmetric hydrogenation of KPL. Systematic variation of the modiﬁer structure showed that among the tested mandelic acid derivatives 1-naphthylglycolic acid performed best. Attenuated total reﬂection-infrared (ATRIR) spectroscopy showed that the modiﬁers were present on the Pt catalyst surface as monomers with internal hydrogen bond between the carboxylic OH group and the carboxyl carbonyl group, as well as that the phenyl or naphthalene rings take on a tilted position relative to the surface. The relatively poor enantioselectivity achieved in KPL hydrogenation was attributed to the weak adsorption of the modiﬁers and to the weakness of the O−H−O type H bond between substrate and modiﬁers. 2.2.3. Additives and Comodiﬁers. In several studies77−80 the application of additives and comodiﬁers has been addressed. Szöri et al.81 showed that the addition of small amounts of triﬂuoroacetic acid (TFA) boosts the ee from 50 to 92% in the enantioselective hydrogenation of TFAP over CD-modiﬁed Pt. Molecular interactions between CD, the acid additive TFA, and the triﬂuoro-activated substrate TFAP on a Pt/Al2O3 catalyst were investigated in the presence of H2 and toluene using in situ ATR-IR spectroscopy combined with modulation excitation spectroscopy (MES).77 On the basis of this spectroscopic study, the involvement of TFA in the diastereomeric intermediate surface complex was proposed, in which the quinuclidine N atom of the adsorbed CD forms an N−H···O-type hydrogen-bonding interaction with the triﬂuoroactivated ketone, the corresponding alcohol, and the acid additive. As depicted in Scheme 3, two possible surface conﬁgurations of a monodentate acid−base adduct in which the carboxylate of TFA resides at the quinuclidine N atom of CD were proposed to improve the stereochemical control with the TFA additive. (More information on the structure of the diastereomeric surface complex is provided in section 2.6.2.) 11531

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 3. Enantiodiﬀerentiating Intermediate Surface Complexes between Coadsorbed Ketone TFAP and Chiral Modiﬁer CD in the Absence and Presence of TFAa

a The N−H···O type hydrogen-bonding interaction between CD and (R)-TFBA (product) is also depicted. For adsorbed modiﬁer conformation, open(4) is assumed. Reproduced with permission from ref 77. Copyright 2014 Wiley-VCH.

Sano et al.79 studied the Pt-catalyzed enantioselective hydrogenation of MBF in the presence of various ILs and three structurally related chiral modiﬁers: CD, PhOCD, and Onaphthyl-CD (NaphOCD). Addition of 1% IL to the reaction mixture improved the ee by up to 12−93% in polar organic solvents (alcohols). Only ILs possessing a heteroaromatic cation were found to interact strongly with Pt or the O-aryl function of PhOCD and NaphOCD. However, with some ILs the reaction rate strongly decreased. This phenomenon was ascribed to the strong interaction of the ILs with the Pt surface leading to active-site blocking. The eﬀect of poly(acrylic acid) (PAA) used as a stabilizer of Pt NPs has been investigated by Schmidt et al.78 They observed a strong inﬂuence of the stabilizer on the selectivity in the hydrogenation of CD over PAA-stabilized Pt NPs with welldeﬁned shape distributions. In the presence of PAA, the Pt NPs aﬀorded a diasteromeric excess (de) of (R)-5,6,7,8-hexahydroCD in the hydrogenation of the heteroaromatic ring of CD in toluene, which is also the major diastereomer in acetic acid. After oxidative removal of PAA, the corresponding (S)diastereomer was observed in excess, which increased with higher Pt(111)/Pt(110) ratio. Thus, the PAA stabilizer played a dual role: it controlled the size and shape of the nanoparticles and aﬀected the rate and diastereoselectivity of the hydrogenation of CD. How this scenario may aﬀect the asymmetric hydrogenation requires further studies. Tálas et al.80 showed that the addition of achiral tertiary and secondary amine additives to the enantioselective hydrogenations of EP and MBF in toluene on CD-modiﬁed Pt/ Al2O3 increases both the reaction rate and the ee at low CD concentration. Interestingly, a similar enhancement of the catalytic performance was not observed with primary amines as additives. On the basis of circular dichroism spectroscopy measurements, they showed that in the presence of both secondary and tertiary achiral amines the virtual concentration of CD increased due to a shift of the dimer−monomer equilibrium of CD. This increase of the virtual CD concentration was proposed to be an important factor contributing to the observed improvement of the enantioselectivity.

and theoretical studies65,84,85,90,91 provided evidence for the coexistence of diﬀerent adsorption modes, which are characterized by diﬀerent conformations and adsorption strengths. While considerable information is available for the adsorption mode of cinchona-based modiﬁers, there is a lack of information about the adsorption of most synthetic modiﬁers. Only the adsorption of 1-(1-naphthyl) ethylamine (NEA) and (R,S)-pantoylnaphthylethylamine (PNEA) has been investigated to some depth.92−94 2.3.1. Cinchonidine Adsorbed on Pt. Using ﬁrstprinciple methods Vargas and Baiker90 investigated the conformations CD adopts upon adsorption onto a Pt(111) surface. They found eight conformationally diﬀerent adsorption states corresponding to diﬀerent rotations around the τ1 and τ2 degrees of freedom (cf. Figure 9) . The study corroborated the previously proposed special role of the open(3) conformer as a chiral docking site, but the rich conformational ﬂexibility observed on the Pt surface invoked that other conformers of CD may also act as chiral enantiodiﬀerentiating sites. Hahn et al.91 addressed the inﬂuence of hydrogen coadsorption on the adsorption mode of CD and CN using the Pt(111) surface as a model by DFT. Figure 12 shows some of the DFT results, which indicate that hydrogen coadsorption has a strong inﬂuence on the adsorption geometry of the cinchona alkaloids. At all hydrogen coverages the cinchona alkaloids adsorbed via the quinoline moiety. Quinoline adsorbs nearly parallel to the surface in the absence of H2 and at low hydrogen coverage, but at higher hydrogen coverage it assumes tilted adsorption. The adsorbed cinchona alkaloid is destabilized at higher hydrogen coverage and partial hydrogenation of the quinoline moiety or hydrogen transfer to the vinyl group. In contrast, hydrogen transfers to the N atom of the quinoline and the quinuclidine moiety lead to stabilization. The study showed that, depending on the surface coverage of hydrogen, the adsorbed cinchona alkaloid may assume various tilted orientations with diﬀerent stabilities resulting in diﬀerently modiﬁed chiral sites. This study has recently been extended by considering the adsorption of CD and CN in the presence of oxygen, and the stability of the chiral modiﬁers under oxidative conditions was shown, which is a necessary prerequisite for extending the scope of chirally modiﬁed metals to asymmetric oxidations, such as the oxidative kinetic resolution of racemic secondary alcohols.95 More recently, Motobayashi et al.86 studied the inﬂuence of coadsorbed hydrogen on the orientation of CD on a Pt surface

2.3. Adsorption of Modiﬁers

In the past few years considerable progress has been made in resolving the adsorption mode of cinchona alkaloids and their derivatives on noble metal particles. Both spectroscopic65,82−89 11532

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

The crucial role of tilted CD surface species for enantioselection in the Pt-catalyzed asymmetric hydrogenation of KPL was evidenced in a recent ATR-IR operando spectroscopic study.87 The correlation between the surface coverage of the tilted modiﬁer species and the obtained ee in Figure 13B shows relatively poor ee at low coverage and the best ee at saturation of the Pt/C catalyst where tilted CD prevails. A similar behavior was also observed recently for the asymmetric hydrogenation of 4-methoxy-6-methyl-2-pyrone (MeoP) on CN-modiﬁed Pd.96 In this operando spectroscopic study, using ATR-IR combined with UV−vis spectroscopies, it was shown that the enantiodiﬀerentiation induced by the cinchona modiﬁer varies signiﬁcantly depending on the adsorption mode of the quinoline anchoring moiety of the cinchona alkaloid. The most stable, ﬂat adsorbed CN (parallel to the surface), predominant at low surface coverage, was much less eﬃcient in enantiodiﬀerentiation than the more weakly adsorbed tilted CN species, corroborating the observation made earlier in the Pt-catalyzed asymmetric hydrogenation of KPL. The operando spectroscopic evidence of the crucial role of tilted cinchona species runs counter to the often-made assumption that the most strongly adsorbed cinchona modiﬁer is most important for enantiodiﬀerentiation. Existing mechanistic models mainly focus on chiral modiﬁers adsorbed via a nearly parallel quinoline ring,3,77,97,98 which is the most stable adsorption mode indicated by DFT studies. Less strongly bound tilted species were considered to be less eﬃcient in enantiodiﬀerentiation.52,99,100 Hence, in future research special attention should be given to the critical role of tilted cinchona species for enantiodiﬀerentiation. Besides the surface coverage and the presence of hydrogen, coadsorbed substrates or additives aﬀect the adsorption mode of the cinchona modiﬁer. Alteration of the dominant surface rotamer of adsorbed CD by the presence of an activated ketone substrate has been shown by following the stereoselectivity in the competing hydrogenation of the quinoline anchor of CD. On Pt/Al2O3 the chiral modiﬁers CD and 10,11-dihydro-CD (HCD) predominantly adopt a pro-(S) adsorption geometry, while in the presence of substrates (MBF, KPL, or EP) the dominant adsorption conformation of CD is inverted to pro(R), as illustrated for MBF in Scheme 4. On top of the altered stereoselectivity, both the modiﬁer and substrate are also hydrogenated faster compared to the hydrogenation activity in the absence of their binding partner, revealing strong modiﬁer− substrate interaction as a decisive factor in the prevailing surface processes.101 2.3.2. Cinchona Derivatives Adsorbed on Pt and Other Noble Metals. The adsorption mode of CD, CN, and the rigid α-ICN modiﬁer from CH2Cl2 solution was also studied on Rh and Pt metal ﬁlms using ATR-IR spectroscopy and DFT calculations.85 The spectroscopic study revealed signiﬁcant diﬀerences in the adsorption mode of the modiﬁers, depending on the rotational ﬂexibility of the modiﬁer, the Pt group metal, and the presence of hydrogen. For the rigid αICN tilted adsorption prevailed on both metals in the absence as well as in the presence of hydrogen. In contrast, CD and CN were found to adsorb on Pt mainly with the quinoline ring parallel to the surface. On Rh only tilted species were observed in the presence of hydrogen, which was attributed to the fast hydrogenation of the aromatic quinoline anchor. The adsorption of CD has also been investigated on monometallic Au and bimetallic Pt−Au and Pd−Au thin

Figure 12. Relative binding energy and orientation of CD adsorbed on (hydrogen-covered) Pt. The relative energy is calculated with respect to the binding energy of CD on clean Pt(111) . Green lines show the results where no H transfer to the CD molecule takes place. Blue lines show the results for H transfer to the quinuclidine N′ atom, and red lines show the results for H transfer to the C11 atom. Conﬁgurations of CD are shown at θH = 0.5 ML with H atoms preadsorbed on active sites ((a) light green), at θH = 0.5 ML without interaction between CD and H (b), at θH = 0.5 ML with H transfer to C11 (c), at θH = 0.5 ML with H transfer to the quinoline N′ atom (d), at θH = 1 ML with H transfer to N′, C11, and C4′ (e), and at θH = 1 ML with H transfer to N′ and C4′ (f) . H atoms that are transferred to the molecule are shown in black. Reproduced with permission from ref 91. Copyright 2015 PCCP Owner Societies.

by in situ ATR-IR spectroscopy. As shown in Figure 13A, the quinoline moiety of CD was found to be reoriented from a πbound nearly parallel to an N-bound upright orientation in the presence of hydrogen, which indicates the same tendency as observed by the DFT calculations.91 In this experimental study, the reason for the reorientation was attributed to the repulsive interaction of a quinoline π-orbital and the negative surface charge induced by dissociative adsorption of H2 and to the hydrogenation of the CC double bond at the 10,11-position of the quinuclidine moiety. 11533

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 13. (A) Spectroscopic detection of quinoline tilting of CD adsorbed on Pt in the presence of chemisorbed hydrogen. Reproduced with permission from ref 86. Copyright 2015 Chemical Society of Japan (CSJ). (B) Operando spectroscopic characterization showing the correlation of signal intensity from tilted CD surface species and obtained ee in the asymmetric hydrogenation of KPL on CD-Pt/C. Reproduced with permission from ref 87. Copyright 2017 Elsevier.

orientation of the phenyl ring reshapes the chiral surface environment formed by the adsorbed modiﬁers, and consequently in some cases, e.g., in the hydrogenation of KPL, even opposite enantioselective control was achieved without altering the absolute conﬁguration of the modiﬁer. Furthermore, competitive adsorption experiments between modiﬁers allowed the assignment of a relative adsorption strength of the modiﬁers. 2.3.3. Synthetic Modiﬁers. Among the synthetic modiﬁers particularly NEA has been in the focus.72−74,81 Recently, Gordon and Zaera93 studied the uptake of NEA and structurally related compounds from CCl4 solutions onto a polycrystalline Pt surface (Figure 14) using in situ reﬂection absorption

Scheme 4. Adsorption and Hydrogenation of HCD (CDH2) in Toluene (left) and the Change of the Dominant Adsorption Mode in the Presence of MBF (right)a

a

For the sake of simplicity, only the hydrogenation of the heteroaromatic ring is shown here. Reproduced with permission from ref 101. Copyright 2010 Elsevier

ﬁlms prepared by physical vapor deposition.102 The ATR-IR spectroscopic studies supported by DFT calculations showed that the adsorbed CD is more stable against desorption on Au than on Pt, because intermolecular interactions dominate over metal−adsorbate interactions. The ﬁndings show that the formation of isolated chiral sites on the gold surface is suppressed due to the establishment of an adsorbed structure involving intermolecularly bound CD molecules (more results from surface science studies are provided in section 2.8.3) . Considerable eﬀort has also been made for gaining information about the adsorption behavior of cinchona-derived modiﬁers.65,103 Bonalumi et al.103 investigated the adsorption behavior of O-methyl cinchonidine (MeOCD) and Otrimethylsilyl derivatives of CD on an alumina-supported Pt ﬁlm by means of in situ ATR-IR spectroscopy and DFT calculations. Applying the same methods also the adsorption of O-phenyl derivatives of CD, including PhOCD, dMePhOCD, and tFPhOCD, was investigated, and their behavior was compared to that of the parent CD.65,84 Experiments and calculations supported the contention that the spatial

Figure 14. Reﬂection absorption infrared spectra for, from top to bottom, (S)-NEA (s-NEA), 1-naphthylmethylamine (NMA), (S)(−)-N,N-dimethyl-1-(1-naphthyl) ethylamine (s-DNE), 1-ethylnaphthalene (EtN), and quinoline (Q) adsorbed from 1 mM CCl4 solutions onto the surface of a polished platinum foil. Signiﬁcant uptake was observed only for (S)-NEA and NMA. Reproduced with permission from ref 93. Copyright 2013 Wiley-VCH.

infrared spectroscopy (RAIRS). On the basis of their comparative adsorption study employing closely related compounds with diﬀerent steric hindrance around the N atom, the authors concluded that chiral modiﬁers such as NEA are bound to the surface predominantly via their aminic N. This suggestion stands in contrast to previous ﬁndings, e.g., surface science studies by McBreen and co-workers have shown that πbound NEA is prevalent on Pt(111).104,105 However, this 11534

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

discrepancy may be explained by the vastly diﬀerent conditions of measurements, that is, ultrahigh-vacuum (UHV) conditions employed by McBreen and co-workers versus CCl4 solution employed by Zaera and co-workers. It is well-known that the solvent has a strong inﬂuence on the adsorption behavior of NEA as reﬂected by the very diﬀerent catalytic performances observed with various solvents.19 Nevertheless, it is surprising in the study of Gordon and Zaera that no adsorption of quinoline was detected on the Pt foil, which had previously been detected in an in situ spectroscopic study106 and indicated in theoretical studies.107 It shows that such in situ spectroscopic studies involving catalytic solid−liquid interfaces are very sensitive to the experimental conditions, not only to the choice of the solvent but also to the properties of polycrystalline noble metals, and due care has to be taken to compare such spectroscopic results and to transfer mechanistic implications to practical catalysis (also compare section 2.8.1, Ex Situ Versus In Situ Studies). Alteration of the preferred adsorption mode (πbound via aromatic ring versus N-bound) depending on the surface coverage of NEA could be a possible explanation for the contradicting observations, but conclusive experiments resolving this aspect are still lacking. Meemken et al.94 investigated the adsorptive anchoring and structural stability of (R)-NEA and (R,S)-PNEA on a technical 5 wt % Pt/Al2O3 catalyst in the presence of H2, the solvent toluene, and the acid additive TFA. In catalytic hydrogenation of KPL under such conditions, (R)-NEA and (R,S)-PNEA provide ee’s of 8% and 60%, respectively. The in situ spectroscopic study revealed that (R)-NEA is predominately anchored on Pt via a tilted naphthalene group, while (R,S)PNEA adsorbs via both the naphthalene and pantoyl moieties. The carbonyl group was found to be positioned nearly parallel to the catalyst surface, and the adsorption conﬁguration and an enantiodiﬀerentiating surface interaction forming preferentially the (R)-enantiomer, shown in Figure 15, were proposed. In addition to the depicted adsorption of PNEA on Pt, rotation of the pantoyl group by 180° around the C−N bond could also be envisaged as an adsorption conﬁguration, which would be in agreement with the surface IR spectrum. In such an adsorption conﬁguration, the geminal methyl groups could contribute to the shape of the pro-(R) chiral pocket for the substrate KPL. The authors also discussed various factors inﬂuencing the stability of NEA-based chiral modiﬁers.

Figure 15. Proposed adsorption of (R,S)-PNEA (A) and the structure of the pro-(R)-KPL-PNEA surface complex (B). Reproduced with permission from ref 94. Copyright 2015 The Royal Society of Chemistry. Note that the stereochemistry of PNEA has been revised,108 and the correct conﬁguration is (R,S)-PNEA.

role of the solvent have been relatively scarce in the past few years. Mechanistic understanding on the behavior of cinchona alkaloids depending on the properties of the solvent has been provided by Zaera and co-workers.59,61,88,89 The solubility of a cinchona alkaloid may vary by as much as 5−6 orders of magnitude and displays a volcano-type correlation with solvent polarity and dielectric constant, as depicted in Figure 16.89 In general, cinchona alkaloids with the C9(R)−C8(S) chirality, such as in CD, QN, and their 10,11-dihydro analogues, are more soluble than those with C9(S)−C8(R), such as in CN, quinidine (QD), and their 10,11-dihydro analogues. The inﬂuence of peripheral groups in these closely related cinchona alkaloids was also investigated, and the speciﬁc position of the peripheral substitution was used to explain their diﬀerent solubilities. On the basis of entropic and energetic considerations, these studies relate the adsorption geometry as well as the surface coverage of various cinchona alkaloids to their diﬀerent solubilities and provide a rationale for the partly signiﬁcant deviations in the overall catalytic performance of the cinchona-Pt catalytic system.61 The choice of solvent may even control the sense of enantioselection. Pereniguez et al.109 studied the eﬀect of the solvent in the enantioselective hydrogenation of TFAP on CNmodiﬁed Pt/Al2O3 as a model reaction using 10 diﬀerent solvents. Inversion of the sense of enantiodiﬀerentiation from (S)-alcohol to (R)-alcohol was observed in strongly basic solvents as well as at increasing hydrogen pressure or conversion, which was attributed to the formation of hemi-

2.4. Substrates

Scheme 1 gives an overview over the various substrates containing activated CO bond(s), which have been hydrogenated on chirally modiﬁed noble metals with excellent catalytic performances. Substrates yielding CN > QN. Note that the adsorption on Pd from isopropanol solutions of CD and CN shows similar adsorption strengths.136 Szöllösi et al.145 studied the enantioselective hydrogenation of TFAP in a continuous-ﬂow reactor system over Pt/Al2O3 modiﬁed by cinchona alkaloids using a solvent mixture consisting of toluene/AcOH (9/1) in the absence and presence of 0.1% volume/volume (v/v) TFA. Interestingly, Pt-CN and Pt-QD aﬀorded the (S)-product in a toluene/acetic acid mixture, whereas addition of 0.1% v/v TFA to the reaction mixture led to inversion of the enantioselectivity to the (R)product. The reason for the unprecedented inversion was discussed, focusing on the nucleophilic intermediate complex formation. The authors proposed that in the hydrogenation of TFAP the reaction route involves the equilibrium of electrophilic and nucleophilic intermediate complexes. However, the formation of the nucleophilic intermediate complex (N → C O) leading to the (S)-product in a toluene/acetic acid mixture, which supposedly is dominant in the absence of TFA, is 11541

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

systems and in situ spectroscopic averaged data often complement each other in achieving a better understanding of heterogeneous asymmetric catalysis. In the past decade, important fundamental insight into the functioning of chirally modiﬁed noble metal surfaces has emerged from surface science studies. Lessons learned from surface science studies on enantioselective chemistry on solid surfaces and perspectives in this ﬁeld have recently been discussed.2 The main focus was on improving the understanding of adsorption and intermolecular surface interactions of prochiral substrates and chiral molecules as well as the formation of enantiodiﬀerentiating modiﬁer−substrate complexes. 2.8.2. Studies on Pt Model Surfaces. Lavoie et al.151 studied the adsorption of the frequently used model substrate MP on Pt(111) using TPD, RAIRS, and STM. They showed that MP undergoes C−H bond scission at room temperature, leading to surface-mediated enol formation and assembly into H-bonded superstructures. The self-assembly process could be severely inhibited by either keeping the Pt sample below the temperature for C−H bond scission or applying a background pressure of 1.3 × 10−6 mbar H2. They proposed that the superstructure formation is due to easily reversible enol assembly. The superstructure formation in the absence of background H2 pressure was previously observed by Bonello et al.,152 but, in contrast, they attributed it to an irreversible polymerization reaction. The superstructure formation may explain the earlier made observation in the enantioselective hydrogenation of MP that contacting the Pt/Al2O3 catalysts with this substrate in the absence of hydrogen is detrimental for the catalytic performance.153,154 STM studies supported by DFT calculations of the adsorption of TFAP on Pt(111)155 showed that TFAP forms stable C−H···O bound dimers and trimers on the surface, as shown in Figure 22. These structures were proposed to mimic those formed in the cinchona−TFAP surface interaction, thereby accounting for the relatively small rate diﬀerences observed between racemic and enantioselective hydrogenation. The authors suggested that the formation of C−H···O bound TFAP prochiral assemblies on Pt(111) could be the reason why α-phenyl ketone substrates do not show strong rate enhancement in the Orito reaction. Nevertheless, note that two of the most frequently studied ketones containing an α-phenyl, MBF,69 and TFAP do show signiﬁcant rate enhancement. The phenomenon of an accelerated rate on CD-modiﬁed Pt catalyst in the hydrogenation of TFAP is more sensitive to the applied reaction conditions, but a signiﬁcant enhancement (r modified ≈ 4.2runmodified) is observed at high TFAP concentration, which contradicts the suggested role of such TFAP dimers on the hydrogenation rate and on the deterioration of ee.114 These surface science studies were later extended, including some structurally related compounds of TFAP, to elucidate the role of aryl-CH···O bonding in forming self-assembled lownuclearity structures and to compare aryl-CH···O bonding by ester and ketone carbonyl functions.156 The formation of homochiral dimers and trimers of TFAP has been corroborated, and DFT calculations revealed aryl-CH···O bonding as the driving force for dimer formation. Chemisorbed 2,2,2triﬂuorovinylbenzene and octaﬂuoroacetophenone did not form such self-assembled structures, because they lack carbonyl and aryl-CH groups, respectively. Structures involving only one aryl-CH···O group are formed by adsorbed methyl benzoate, as distinct from two aryl-CH··· O bonds with TFAP. The same

Figure 22. STM images of TFAP on Pt(111) at rt (bias, −1.0 V; tunnel current, 1.0 nA) . STM images show well-deﬁned groups of two (A1, A2, B1, B2), three (A3, B1, B3, B4), four (B1, B2), or six (B6) molecules over the entire surface. Reproduced with permission from ref 155. Copyright 2008 American Chemical Society.

group also studied the chemical transformation and subsequent self-assembly of prochiral ketones and their corresponding alcohol products on Pt(111) using STM and high-resolution electron energy loss spectroscopy (HREELS) .157 They observed that the product methyl lactate undergoes dehydrogenation, leading to the same adsorbed enol assemblies that are formed directly upon adsorption of the prochiral (R)-ketoester. Similar behavior was also observed for 1-phenylethanol, where the enol tautomer of acetophenone was formed via chemisorption interaction, inducing strong interadsorbate aryl-CH···O bonding between the α-phenyl ketone formed through oxidation of the parent α-phenyl alcohol. The authors proposed that the observed alcohol-to-ketone and alcohol-toenol transformations may aﬀect the rate and enantioselectivity of the asymmetric hydrogenation. The chiral alcohol product may react back to the corresponding prochiral ketone, which in turn reenters the catalytic cycle as a higher probability to undergo hydrogenation at nonmodiﬁed racemic sites, thereby lowering the enantioselectivity and conversion. These results suggest that enolization is relevant to the platinum-catalyzed hydrogenation performed under nonoptimal conditions, that is, when the surface coverage of hydrogen is too low. It is known that the addition of a small quantity of TFA increases the ee in the asymmetric hydrogenation of TFAP on CD-modiﬁed Pt/Al2O3.81 An STM study by Brunelle et al.158 11542

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 23. Classiﬁcation of 900 modiﬁer−substrate complexes formed by (R)-NEA and TFAP on Pt(111) at rt into distinct stereochemical and regiochemical arrangements. Each entry shows a 2 nm-by-2 nm STM image of an archetypal complex of a particular family, along with the percentage of the total complex population it represents. The remaining 7% mostly consists of complexes where TFAP binds to the naphthyl instead of the ethylamine group and irregularly shaped complexes likely caused by high local coverages. STM images were recorded at a 1.2 V bias and 0.3 nA current with the sample at rt. Reproduced with permission from ref 161. Copyright 2011 American Association for the Advancement of Science.

provides a feasible explanation for this behavior. The authors assumed that the earlier observed aryl-CH···O interactions increase the rate of hydrogenation in TFAP dimers and that this competing, racemic hydrogenation decreases the achievable ee. Either a high coverage of the modiﬁer133 or inhibition of the hydrogenation at nonmodiﬁed sites by disrupting dimerization of TFAP could counter this scenario. STM images of the coadsorption of TFAP and TFA showed that TFA inserts between TFAP molecules on Pt(111) at rt, thereby eﬀectively disrupting the noncovalent network of aryl-CH···O bonds that holds TFAP in dimer pairs. However, this interpretation of the lower eﬃciency of TFAP dimer stands in contrast to catalytic studies as higher ee and signiﬁcant rate enhancement were observed at higher TFAP concentration.114 The adsorption of cinchona alkaloids on single-crystal surface has also been investigated by STM. Time-lapsed STM studies on the adsorption of CD and CN on Pt(111) and Pd(111) revealed signiﬁcant surface mobility of the adsorbed modiﬁers, which was enhanced with increasing H 2 background pressure.159,160 It could be speculated that the high mobility of the adsorbed cinchona alkaloids is beneﬁcial for the enantiodiﬀerentiation, because the interaction with the adsorbed substrate is not spatially restricted, unlike chiral modiﬁers, which are immobilized on the surface by grafting, tethering, or encapsulation. Goubert et al.108 studied surface processes of the synthetic chiral modiﬁer PNEA on Pt(111), which is eﬃcient in the asymmetric hydrogenation of KPL.17 The STM studies revealed dissociative adsorption of a fraction of PNEA yielding two fragments, which the authors attributed to a process involving C−N bond scission. They also evidenced that C−N bond scission occurs under hydrogenation conditions on PNEA-modiﬁed Pt/Al2O3, forming the aminolactone, amino4,4-dimethyldihydrofuran-2-one (AF). On the basis of their STM studies of the coadsorption of (S)-AF with TFAP and KPL, which showed the formation of isolated 1:1 complexes with TFAP and ﬂuxional supramolecular assemblies with KPL, the authors speculated that such assemblies could contribute to the overall enantioselectivity observed of PNEA-modiﬁed Pt catalysts. However, later catalytic experiments with (S)-AF did not conﬁrm this contention.94

Particularly interesting are STM investigations, which embrace the formation of diastereomeric surface complexes between chiral modiﬁers and prochiral substrates, because these interactions are considered to be crucial for enantiodiﬀerentiation. Despite their superior catalytic performance, cinchona alkaloids are rather demanding chiral modiﬁers for STM studies on intermolecular interactions due to their more diﬃcult handling (solid, melting point 205 °C) and bulky molecular structure. Therefore, STM studies on intermolecular chiral interactions have been carried out almost exclusively with the simpler synthetic modiﬁer (R)-NEA. In their seminal STM work, Demers-Carpentier et al.161 studied the molecular preorganization of the prochiral substrate TFAP and the chiral modiﬁer (R)-NEA on Pt(111) using a combination of rt STM measurements and DFT calculations. The prochirality of >900 diastereomeric complexes was determined by visual inspection and compared to catalytic tests on (R)-NEA-modiﬁed Pt/Al2O3 in acetic acid. In a ﬁrst step, the adsorption of (R)-NEA and TFAP on Pt(111) was characterized separately. The STM images revealed two diﬀerent prochiral surface conformers of (R)-NEA, which were present in a 7:3 ratio. The prochiral substrate TFAP formed dimers upon adsorption. When both (R)-NEA and TFAP were coadsorbed, in addition to TFAP dimers, various surface complexes consisting of TFAP bound to (R)-NEA were observed, which are categorized in Figure 23 with respect to their prochirality. An important lesson learned from this surface science investigation is that the preorganization of modiﬁer and substrate plays an important role for the formation of enantiodiﬀerentiating complexes and that multiple diastereomeric complexes contribute to the overall stereochemical outcome of the hydrogenation, that is, the stereochemical outcome of the reaction cannot be attributed to a single structure of a (R)-NEA−TFAP complex. In a subsequent study,162 the authors carried out experiments with TFAP both on a clean surface and on a surface populated with chemisorbed (R)-NEA. On the nonmodiﬁed surface, introduction of H2 at a background pressure of 10−6 mbar resulted in a rapid breakup of TFAP dimer structures followed by the gradual removal of all TFAP and the formation of a half-hydrogenated 11543

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

inaccessible enol state, and such a double-bond migration facilitates the hydrogenation of TFAP on Pt(111) . Later the groups of McBreen and Hammer also investigated the structure and stability of bimolecular complexes involving the popular model substrate KPL.105 They showed that the two surface conformers of (R)-NEA interact in distinct ways with KPL on the Pt(111). The energetically most viable calculated (R)-NEA−KPL complex was stabilized by multiple noncovalent intermolecular interactions. Both carbonyl groups form NH···O bonds to the modiﬁer, and the keto-carbonyl group also forms an arene−CH···O interaction. In this complex, the oxygen of the keto-carbonyl forms a bond to a Pt atom positioned in a location where the oxygen can simultaneously form NH···O and CH···O bonds and participate in an η2(C,O) chemisorption interaction. This particular twocarbonyl contact arrangement stereodirects KPL into a pro-(R) conﬁguration and activates the carbonyl group. The importance of various features (chemisorption geometries of both modiﬁer and substrate, atomic structure of the metal, and multiple noncovalent interactions) in deﬁning a chiral pocket is nicely illustrated by this example. The overall pro-(R)/pro-(S) prochiral ratio observed for the surface complexes was found to be much lower than the enantiomeric ratio reported in previous studies on the hydrogenation of KPL on (R)-NEAmodiﬁed Pt/Al2O3 in acetic acid.19 This discrepancy is likely due to the vastly diﬀerent conditions of the catalytic and the UHV studies. Lemay et al.165 compared bimolecular complexes chemisorbed on Pt(111) formed by the prochiral substrate KPL and the structurally analogous chiral modiﬁers (R)-NEA, (R)-Nmethyl-1-(1-naphthyl) ethylamine (MNEA), and (R)-1-naphthyl-1,2-ethanediol (NED). The STM motifs of the formed complexes, which are categorized in Figure 25, permitted an estimation of the prochiral state of complexed KPL based on complexation geometries predicted by DFT calculations. Signiﬁcant diﬀerences in the distribution of induced prochiral states were observed irrespective of the relatively minor diﬀerence in the molecular structures between (R)-NEA and (R)-MNEA. Strikingly, the prochiral ratios measured for (R)NEA and (R)-NED were within the range of enantiomeric ratios reported in the literature for these modiﬁers under catalytic reaction conditions. This is remarkable in the light of the high parametric sensitivity of these reactions and the vastly diﬀerent conditions used in the STM measurements and inherent to the DFT calculations. Goubert and McBreen166 compared the structure of diastereomeric complexes formed by MBF and (R)-NEA on Pt(111) with previously published STM images and DFT calculated structures for MTFP-(R)-NEA124 and TFAP-(R)NEA161 complexes. They analyzed how the combination of a phenyl group and a ketoester group in MBF alters the complexation patterns with respect to TFAP (possesses a phenyl group but no ester carbonyl) and with MTFP (possesses an ester carbonyl but no phenyl group). Compared to MTFP, MBF was found to form signiﬁcantly diﬀerent types of complexes due to avoidance of phenyl-(R)-NEA steric interactions. Particularly, conﬁgurations involving keto-carbonyl hydrogen-bonding interactions favored for MTFP are replaced by ester carbonyl hydrogen-bonding interactions for MBF. The authors concluded that the longer ketoester group, as compared to the TFAP acetyl group, gives rise to a variety of binding conﬁgurations, which results in lower prochiral selectivity than those observed for TFAP complexes. Unfortunately, no

intermediate. Hydrogen addition to a mixture of (R)-NEA and TFAP on Pt(111) led to the removal of TFAP without any change in the population of the modiﬁer, a prerequisite for an eﬃcient chirally modiﬁed catalyst.162 The coexistence of multiple pro-(S)- and pro-(R)-complexes was also found, employing the prochiral ketone methyl 3,3,3-triﬂuoropyruvate (MTFP) .124 More recently, Goubert et al.163 showed that, at rt, frequently used in asymmetric hydrogenation, both the prochiral reactant and the chiral modiﬁer must be considered mobile species that can reorganize dynamically during the formation of the chiral complexes. Rasmussen et al. investigated the activation and hydrogenation of TFAP on Pt(111) by means of DFT.164 In the energy diagrams shown in Figure 24, the routes via formation of an alkoxy, a hydroxyl, and an enol surface intermediate species are included, while simultaneous addition of both H atoms was found to be energetically very unfavorable. Interestingly, the theoretical study revealed that the lowestenergy route involves the remote activation of TFAP by interaction of the phenyl ring with Pt, leading to an otherwise

Figure 24. Energy diagram of all the investigated routes of hydrogenation. Solid symbols represent transition states; open symbols denote minima. The reference energy is calculated from the separated components: a four-layer Pt(111) slab, an isolated TFAP molecule, and isolated H2. Zero-point energy corrections have been applied. (Insets) Geometries of adsorbed TFAP at diﬀerent stages of hydrogenation. In all cases, an appropriate number of hydrogen atoms are on the Pt surface (2 for the ketone, 1 for the three intermediates). The triangles (Δ/∇) represent the transition-state energy for the addition of a hydrogen atom to the bridged and apical carbons adjacent to the ipso carbon. They are colored blue and yellow to indicate if this addition was done from the hydroxy or enol state, respectively. In the case of the ketone, hydroxy, and alkoxy geometries, there are Pt atoms inserted into the carbonyl group bonds. Reproduced with permission from ref 164. Copyright 2014 American Chemical Society. 11544

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 25. (i−iii) Schematic categorization of STM motifs for ensembles of (R)-NEA-KPL, (R)-MNEA-KPL, and (R)-NED-KPL 1:1 complexes in terms of the location and prochirality of complexed KPL. The abundances of pro-(R) and pro-(S) complexes are indicated in red and blue, respectively. The indicated prochiralities are inferred from the directionality of KPL motifs. Nonassigned complexes (indicated in white) are those for which the directionality of KPL cannot be determined from a visual inspection of the STM motif. Motifs clearly not belonging to the illustrated subsets are termed others. Reproduced with permission from ref 165. Copyright 2016 Elsevier.

from this calorimetric study. First, the coadsorption of the two diﬀerent chiral compounds leads to vastly diﬀerent adsorption properties, indicating attractive 1:1 interaction of PO only to NEA. The second conclusion concerns the chiral preference of the modiﬁed Pt in PO titration. Interestingly, no diﬀerence in absolute adsorption amount or sticking coeﬃcient was observed using enantiopure (S)- and (R)-PO molecules, regardless of the NEA coverage, which stands in sharp contrast to other studies on the desorption from Pt(111).92 Deviations in the type of measurements, direct adsorption by SCAC versus indirect desorption by TPD, was oﬀered as a preliminary explanation for these strikingly diﬀerent results. 2.8.3. Studies on Other Noble Metal Model Surfaces. There is a lack of studies on well-deﬁned Pd surfaces, which could clarify some fundamental aspects important for the Pdcatalyzed asymmetric hydrogenation of CC bonds. Recently, Tysoe’s group studied the adsorption of (S)-NEA168 and MP169 on Pd(111). As on Pt(111),104 both prochiral surface conformers of this chiral amine were also found on Pd(111), but only with a minor excess in favor of the exo-NEA conformer.168 A recent surface science study also addressed the coadsorption of these molecules employing TPD, STM, and DFT calculations.134 The measured TPD proﬁles indicated a small enhancement of MP hydrogenation depending on the NEA coverage on Pd(111). Stable surface complexes involving the enol tautomer of MP and CO···H2N hydrogen-bonding interactions were indicated by the DFT calculations, and the formation of the more easily reducible CC bond in the diastereomeric surface complex is provided as a rationale for explaining the often observed rate enhancement in the Ptcinchona catalytic system (compare section 2.6.3, Kinetics and Rate Enhancement (speciﬁcally on Pt catalyst)) . Unfortunately, no catalytic data for this reaction employing a NEAmodiﬁed Pd catalyst is available. However, earlier catalytic studies of MP hydrogenation on cinchona-modiﬁed Pt and Pd

catalytic data for the hydrogenations of MBF and MTFP on NEA-modiﬁed Pt catalyst is available in the literature, but on cinchona-modiﬁed catalysts ketoester substrates generally yield higher ee compared to TFAP. While STM and DFT calculations were the principle working horses of most studies on model surfaces, there are several studies that were based on other analytical surface methods. The focus of several single-crystal studies was the adsorption of chiral modiﬁers, and mostly the simpler chiral modiﬁer NEA was used. Lee et al.92 investigated the adsorption of NEA on Pt(111) using RAIRS and TPD, both under UHV and from liquid CCl4 solutions. The experiments with NEA corroborated the formation of supramolecular chiral templates. Furthermore, complexation of individual NEA with the chiral probe molecule propylene oxide (PO) was evidenced. Titrations of NEAmodiﬁed Pt(111) using PO indicated a relative enhancement in the adsorption of one enantiomer over the other at intermediate NEA coverages, a behavior expected from the templating mechanism. This study also demonstrates that diﬀerent results are obtained depending on the conditions (vacuum versus solution). Gordon and Zaera93 studied the uptake of NEA and similar compounds from CCl4 solutions onto a surface of a polycrystalline Pt foil using in situ RAIRS (Figure 14) . Their data suggest that the adsorption of those compounds is not via π-bonding of the aromatic ring, as stated before, but rather by involving the amine moiety (compare section 2.3, Adsorption of Modiﬁers). Single-crystal adsorption calorimetry (SCAC) was introduced by Schauermann and co-workers as a new tool to study the energetics of chirally modiﬁed metal surfaces.167 The adsorption of two chiral compounds, (R)-NEA and (S)-2methylbutanoic acid, was characterized by SCAC using PO as the chiral probe molecule. Two main conclusions were drawn 11545

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 8. In Situ Preparation of the Active Chiral Modiﬁer in the Asymmetric Hydrogenation of EP Formed by Reductive Alkylation of the Condensate of the Modiﬁer Precursor (R)-NEA and the Substrate EP on Pt Catalyst (Adapted with Permission from Ref 171; Copyright 1996 Elsevier)

Scheme 9. Catalytic Performance (Rate and ee) of the Most Eﬃcient Asymmetric Oleﬁn Hydrogenations on CD- or CNModiﬁed Pd Catalysts; Reaction Conditions Are Also Provided176−179

11546

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

summarized by Wells and Wilkinson170 indicated that the reaction over Pd diﬀers from that over Pt in important parts. A striking dissimilarity is the absence of a rate enhancement over modiﬁed Pd catalysts, which contrasts the behavior observed over Pt catalysts. To our knowledge there is no reported data in the literature on rate enhancement on NEA-modiﬁed Pd catalysts, neither for the hydrogenation of MP nor for any other substrate. Pertinent catalytic studies may clarify to which extent the observations made on the model system in ultrahigh vacuum can be used for understanding the catalytic processes occurring under reaction conditions. A critical point is the choice of the substrate MP. As shown by Minder et al.171 and depicted in Scheme 8, unsubstituted pyruvates are prone to react with primary amines, such as NEA, to form the corresponding imine, which undergoes subsequent reduction on Pt to form the actual active chiral modiﬁer in this asymmetric hydrogenation.171 In a previous STM study by the groups of McBreen and Hammer, the corresponding ketone MTFP was chosen to avoid the obscuring side-reaction.124 It is not clear whether the reductive alkylation of NEA also occurs on Pd, but it seems probable. Thus, it remains to be proven whether the reductive alkylation can be ruled out under reaction conditions of the Pd-catalyzed asymmetric hydrogenation of alkyl pyruvates. Nevertheless, in addition to previous STM studies performed on Pt(111), the newly emerging work on Pd(111) will represent an important cornerstone for understanding the fundamental diﬀerences as well as similarities of chiral catalysis on Pt and Pd metal catalysts. On Au(111) surfaces, Behzadi et al.172 investigated the adsorption of CD by means of TPD, low-energy electron diﬀraction (LEED), and X-ray photoelectron spectroscopy (XPS). In the monolayer the alkaloid was bound via N-lone pair electrons of its quinoline part rather than via the π-system of this aromatic moiety. For the multilayers intact molecular desorption was observed, whereas in the ﬁrst monolayer decomposition occurred upon heating above 400 K. This behavior indicates a stronger interaction in the monolayer, most likely due to strong lateral intermolecular interaction, stabilizing the monolayer. LEED patterns did not indicate any long-range ordered structures. Experiments with quinoline also showed bonding via the N-lone pair, corroborating that CD interacts with Au(111) via the N atom of the quinoline group, resulting in an adsorption geometry in which the aromatic part is tilted or perpendicular with respect to the surface plane. The adsorption of CD was also studied on monometallic Au and bimetallic Pt−Au and Pd−Au thin ﬁlms using ATR-IR spectroscopy combined with DFT calculations.102 Interestingly, on Au the alkaloid formed an adsorbed layer that showed higher stability against desorption than the corresponding adsorption on Pt. This behavior was attributed to the dominance of intermolecular interactions over metal−adsorbate interactions because spectroscopic features due to strongly ﬂat adsorbed species were absent. DFT calculations indicated comparable adsorption energy on Au (7−10 kcal/mol) for ﬂat and tilted orientations of the quinoline ring but a signiﬁcantly lower one compared to adsorption on Pt (ca. 40 kcal/mol). The consequence is that the formation of intermolecularly bound CD molecules is energetically favored on Au compared to the creation of isolated chiral sites.

3. HYDROGENATION OF CC BONDS Compared to noble metal-catalyzed asymmetric hydrogenation of activated CO bonds, heterogeneous asymmetric hydrogenation of CC bonds has been investigated less intensively, in spite of the high importance of asymmetric oleﬁn hydrogenation for organic synthesis and its widespread application for the production of pharmaceuticals.173,174 Nevertheless, several groups have contributed to a signiﬁcant progress in this research direction in the past decade. Enantioselectivity up to 96% has been reported for speciﬁc substrates under mild conditions (rt and low hydrogenation pressure). In the following, two diﬀerent catalytic systems are discussed for enantioselective hydrogenation of prochiral oleﬁns: chirally modiﬁed Pd and (S)-proline-induced asymmetry. The two most popular substrate classes for chirally modiﬁed metals are bulky aromatic alkenoic acids like phenyl cinnamic acid (PCA) and several of its derivatives as well as aliphatic alkene acids such as 2-methyl hexenoic acid (MHA), 2-methyl pentenoic acid (MPA), or 2-methyl butanoic acid (MBA). Catalytic studies involving these diﬀerent alkenoic acids indicate mechanistic diﬀerences regarding the structure of the diastereomeric surface complex responsible for enantiodiﬀerentiation. For aromatic and aliphatic acids, chiral modiﬁer−acid surface complexes with a ratio of 1:1, and 1:2 or even 1:3 are proposed, respectively. Very high enantioselectivity can also be obtained in asymmetric hydrogenations of 2-pyrones, particularly in the hydrogenation of 4-methoxy-6-methyl-2-pyrone (MeoP) on CN-modiﬁed Pd with an ee exceeding 90%.175 The most eﬃcient enantioselective hydrogenations on cinchonamodiﬁed Pd catalysts are summarized in Scheme 9. For asymmetric hydrogenation of α,β-unsaturated ketones, the (S)-proline-mediated hydrogenation of isophorone (IP) on Pd catalysts can yield good enantioselection. Several catalytic and mechanistic studies have shown the distinctly diﬀerent behavior of this catalytic system compared to hydrogenations on chirally modiﬁed metals, and even the classiﬁcation as heterogeneous asymmetric hydrogenation had been in dispute. Recently, the reaction mechanism was revisited, and comprehensive experimental evidence substantiated the earlier contention that the enantioselection takes place on the catalytic surface. Therefore, the asymmetry induced by (S)-proline (Pr) is discussed in this Review as an alternative strategy for heterogeneous asymmetric CC bond hydrogenation. Note that research on non-noble metal-catalyzed asymmetric CC bond hydrogenation, most prominently represented by the Ni-tartaric acid system, will not be covered here. The progress in this ﬁeld up to about 2011 has been summarized elsewhere.6 3.1. Catalysts

3.1.1. Chirally Modiﬁed Pd. A striking diﬀerence between the asymmetric hydrogenation of ketones and prochiral oleﬁns is that in the former reaction other noble metals beside the most frequently used Pt catalysts can provide good enantioselection (see section 2.1), whereas for prochiral oleﬁns almost exclusively Pd catalysts are applied. In the recent past, systematic variation of the Pd particles size and shape has revealed important insight into the structure−activity/enantioselectivity relationship of chirally modiﬁed Pd.180,181 In addition, the inﬂuence of the support material182−184 and the design of new catalysts185,186 were in focus. 11547

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

inantly (100) faces or 9 nm Pd nanospheres (S-9) containing mainly (111) facets, indicating that these hydrogenations are rather structure-insensitive with respect to the particle shape. However, evidence for two diﬀerent types of hydrogenation sites was found by varying the particle size. While a higher activity but a lower enantioselectivity were observed with small-sized Pd nanocubes containing a relatively large amount of edge sites, higher fraction of ﬂat planes in large nanocubes provided a higher ee but at lower rate. The authors suggested that the enantioselective hydrogenation is favored at plane surface sites, while edge sites are intrinsically active for racemic hydrogenation. On the basis of comparative tests, any signiﬁcant inﬂuence of particle stabilizer polyvinylpyrrolidone residuals from the catalyst synthesis was ruled out. Comparing the reported structure-sensitivity studies, in the hydrogenation of both aliphatic alkene acids180 and the 2pyrone MeoP,181 generally better enantioselectivity was observed with larger particles. Nevertheless, in the study by Strobel et al.,181 the particle size was varied in a signiﬁcantly narrower range. Furthermore, the opposing structure−activity relationship indicates mechanistic diﬀerences for these two substrate classes. For the asymmetric hydrogenation of aliphatic prochiral substrates, Szöri et al.187 also observed poorer enantioselection with smaller particles. They synthesized Pd NPs supported on nanoparticle−graphene (Gn) by coreduction of Pd and graphite oxide using NaBH4. The 5 wt % Pd/Gn catalyst with the lowest dispersion of 72% gave the best activity and enantioselectivity, e.g., an ee of 47% to the (S)-enantiomer was obtained in the asymmetric hydrogenation of MHA in toluene at 50 bar H2 pressure. In comparison, the other two graphene-supported catalysts prepared by precipitation with Na2CO3 had a narrower Pd particle-size distribution with 83% and 86% dispersion but only gave 31% and 29% ee, respectively. Kubota et al.183 examined the eﬀect of Pd metal loading employing TiO2- and CeO2-supported catalysts. They also optimized the reaction conditions for the asymmetric hydrogenation of PCA with respect to the reaction temperature, the concentrations of the substrate, and the chiral modiﬁer. In general, better catalytic performance was observed with TiO2supported compared to CeO2-supported catalysts. As reductive pretreatment of the catalysts at high temperatures decreased the ee, the authors speculated that strong metal−support interaction was unfavorable for the enantioselective hydrogenation. At higher Pd loading a better ee was achieved, reaching up to 90% under optimized conditions. The authors suggested that the enantioselective hydrogenation takes place preferably on Pd(111) faces. However, in a follow-up study the same authors observed a diﬀerent trend for Pd catalyst supported on carbon.182 Low catalytic activity was achieved with a 1 wt % Pd/C catalyst, which was attributed to inaccessibility of Pd by CD and PCA due to the deposition of NPs inside very small pores. A metal loading of 5 wt % Pd/ C, however, aﬀorded the best ee. On TiO2-supported catalyst a metal loading of 38 wt % Pd was necessary to achieve good ee. As no correlation between the dispersion of the diﬀerent Pd catalysts and the ee was found, the authors suggested that the properties of the support have a strong eﬀect, which, however, was not speciﬁed in more detail. The eﬀect of the support acidity was studied by Trung et al.,184 who synthesized Pd catalysts supported on activated carbon, graphene oxide (GO), or CNTs using a deposition− precipitation method. As the Pd/GO catalyst with high support

The inﬂuence of Pd dispersion on the ee was investigated using ﬂame-synthesized Pd/Al2O3 catalysts with various particle sizes (1−5 nm). In the asymmetric hydrogenation of the CC bond in MeoP, higher enantioselectivity and activity was observed with the larger Pd particles.181 Deeper insight into the structure−activity/enantioselectivity relationship of Pd nanoparticles was provided by an extensive study of Chen et al.180 Using a seeded-growth method, they synthesized a series of well-deﬁned Pd nanocubes and nanospheres with diﬀerent sizes and shapes, which are shown in Figure 26. The synthesized Pd

Figure 26. Particle-size distribution (top) and high-resolution TEM (HRTEM) (bottom) of 9 nm Pd cubes (left) and spheres (right). Reproduced with permission from ref 180. Copyright 2013 The Royal Society of Chemistry.

particles were immobilized on silica, and the Pd/SiO2 catalysts were tested in the enantioselective hydrogenations of the aliphatic substrates MPA and MBA. In contrast to well-deﬁned Pt/SiO2 particles for CO hydrogenation,22 the catalytic performance summarized in Table 4 is similar using either silica-supported 9 nm Pd nanocubes (C-9) exposing predomTable 4. Catalytic Performance of Pd Cubes (Designated As C-6, C-9, C-14, and C-19) and Pd Spheres (Designated As S9) in the Enantioselective Hydrogenation of MPA (Reproduced with Permission from Ref 180; Copyright 2013 The Royal Society of Chemistry)a catalyst

Pd (wt %)

Pd size (nm)

C-6 C-9 C-14 C-19 S-9

1.76 1.75 1.85 1.71 1.72

5.8 ± 1 9.6 ± 1 14.8 ± 2 19.3 ± 2 9.0 ± 2

Xedge/plane

Pd surface 2 −1 (m g )

TOF (h−1)

ee (%)

1.89 1.20 0.63 0.42 1.26

72 43 28 22 55

142 87 56 33 80

18 25 31 35 21

Reaction conditions: 16.5 μmol of Pd, 66 μmol of CD, 1 mmol of MPA, and 10 mL of ethanol, 303 K, 2 MPa H2 pressure. Note that the rather low ee values of these hydrogenation were obtained without amine additive. a

11548

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 27. TEM images of silica-free mesoporous Pd network catalysts. (a, b) Pd-308-KBH4, (c, d) Pd-353-KBH4, (e) skeletal model of a fragment of single gyroid with I4132 symmetry, (f) Pd-373-KBH4, (g) TEM image simulation of Pd-373-KBH4, (h, i) Pd-403-KBH4, (j) Pd-373-H2, (k) Pd373-N2H4·H2O, and (l) skeletal model of a fragment of double gyroid with Ia3d symmetry. (Insets) Corresponding SAED patterns and Fourier diﬀractogram. In the designations Pd-x-y, x refers to the hydrothermal treatment temperature (K) of the original KIT-6 template and y refers to the reducing agent. Reproduced with permission from ref 198. Copyright 2014 Elsevier.

had to be dried beforehand.189 As the same positive eﬀect was observed by admixing of benzene, but not by addition of hexane, the droplets of toluene containing the catalyst were suggested to shield water from the catalyst surface. As an extension to the application of CNTs for the asymmetric hydrogenation of ketoesters on cinchona-modiﬁed Pt,27 Guan et al.186 synthesized Pd catalysts supported on CNTs and compared their performance to a standard Pd catalyst supported on active carbon. While the particle sizes of Pd deposited on the inside (4.9 nm) and outside (5.2 nm) of the tubes were similar, higher activity and enantioselectivity were observed for catalysts with Pd particles deposited inside the nanotubes, designated as Pd/CNT(in), for the asymmetric hydrogenation of PCA and several of its derivatives. In the asymmetric hydrogenation of the substrate α,β-para-dimethoxy PCA in wet 1,4-dioxane as solvent, an ee of 80% was achieved

acidity provided very low ee in the hydrogenation of PCA, the authors proposed acidic support materials to be detrimental to the enantioselection due to competitive adsorption of the chiral modiﬁer on such support sites. Another property of the catalyst that was proposed to be detrimental to the catalytic hydrogenation is residual water.188 After pretreatment of commercial Pd/C catalysts at 80 °C, good enantioselectivity is readily achieved. As PCA and CD have diﬀerent ionic properties (polarity), the authors argued that removal of surface water is crucial to enable better access of the substrate and the chiral modiﬁer to the same active sites in a concerted manner. While the hydrogenation in water generally yielded poor ee, Sugimura and co-workers reported that admixing of “droplet amounts” of toluene leads to similarly high ee obtained in the asymmetric hydrogenation of PCA as in the usually best solvent, wet 1,4-dioxane, but the Pd/C catalyst 11549

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

3.2. Chiral Modiﬁers

for Pd/CNT(out), while that of Pd/CNT(in) was 92%. The authors speculated that the enhanced catalytic performance of Pd/CNT(in) is due to an enrichment of chiral modiﬁer, substrate, and achiral amine additive inside the tube. However, it should be noted that high enantioselectivity of 92% has also been reported by others190 using a conventional Pd catalyst. A Pd catalyst supported on TiO2-coated multiwalled CNTs did not show enhanced catalytic performance.191 A novel catalyst design based on metallo-organic hybrid material synthesis has been reported by Durán Pachón et al.185 They showed that chiral induction to a metal can be performed by ﬁrst doping and then extracting the chiral dopant. Palladium salt solutions were reduced in the presence of the four pseudoenantiomers of cinchona alkaloids. While Pd retained its chiral properties after extraction of the chiral modiﬁer, poor enantioselectivity could only be achieved in the hydrogenation of IP by addition of the modiﬁer to the reaction mixture. Thus, the metallo-organic hybrid material synthesis was rather ineﬃcient in bestowing enantioselectivity to heterogeneous hydrogenations. 3.1.2. Asymmetry Induced by (S)-Proline. Catalytic studies on the performance of Pd catalysts in the asymmetric hydrogenation of IP in the presence of (S)-Pr mainly focused on the kinetic resolution of the corresponding cyclohexane product 3,3,5-trimethylcyclohexanone (TMCH), which consumes the (R)-enantiomer and provides the (S)-enantiomer in excess192,193 (see also section 3.4.2 for recent mechanistic insight into the origin of enantioselection) . The size and shape of Pd particles and the properties of the support material were reported to strongly inﬂuence the kinetic resolution.192,194−196 Mhadgut et al.192 studied the eﬀect of the catalyst support, employing a series of alkaline earth metal carbonate-supported Pd catalysts, eight diﬀerent solvents, and (S)- and (R)-Pr. Under the chosen reaction conditions (no premixing of the reaction solution), the left-handed and right-handed chirality in Pr always induced the same enantioselection in the TMCH product. The best performance was found with a Pd/CaCO3 catalyst in the solvent ethanol, and it was claimed that the properties of this support enhance the modiﬁer adsorption; therefore, both chiral catalysis and kinetic resolution can contribute to the ee. In follow-up research they extended the range of modiﬁers by including various derivatives of (S)-Pr.194 Better catalyst performance with basic support materials was also observed by Li et al.,195 who attributed the beneﬁcial eﬀect to improved adsorption of (S)-Pr and more eﬃcient kinetic resolution. In another study, Pd/MgO catalysts with Pd particle size ranging from 1.8 to 20.6 nm were employed.196 With increasing particle size an inversion of enantioselectivity was reported, and with large particles (≥11.9 nm) the (R)-product was observed in excess. However, as the authors simply specify that the reaction solution was “pre-mixed in a quartz tube”,196 the extent of condensation of (S)-Pr and IP in the used reaction solution is unclear, which is a decisive reaction parameter for the sense of enantioselection in this hydrogenation.197 Wang et al.198 investigated the adsorption of (S)-Pr on various free-standing mesoporous Pd catalysts, depicted in Figure 27, and reported a conﬁnement eﬀect, that is, the enantioselectivity could be enhanced by >10% ee by the topology, pore size, and lattice structure of the Pd catalyst. While the hydrogenation of acetophenone was used as a test reaction in this study, similar structure−performance eﬀects are likely to be relevant also for CC bond hydrogenation.

The inﬂuence of the molecular structure of the chiral modiﬁer on the enantioselection in asymmetric hydrogenations on cinchona-modiﬁed Pd catalyst has received relatively little attention in the past.199 The importance of an intact quinoline moiety as the anchoring unit has been reported.200 The alcohol function at the C9 has been suggested to play a more important role compared to asymmetric hydrogenations of activated ketones on Pt,201 but a thorough investigation is missing. Despite the molecular structure of the chiral modiﬁer, the addition of amine additives, such as benzylamine (BA), is an established strategy to improve the catalytic activity and the enantioselectivity in the asymmetric hydrogenation of prochiral alkenoic acids. For best performances, addition in roughly equivalent amounts (0.6−2 equiv) with respect to the substrate is suggested. Recent research has focused on the elucidation of structural requirements of the chiral modiﬁer using cinchonabased derivatives 69,202−205 and on the role of amine additives.177,203,206−209 3.2.1. Cinchona Alkaloids and Derivatives. In asymmetric hydrogenations of α-ketoesters on cinchona-modiﬁed Pt, switching the chiral modiﬁer from CD to the pseudoenantiomer CN generally leads to the formation of the opposite product alcohol enantiomer with only minor deviations in the observed ee. However, for other activated ketone substrates, such as TFAP (compare Table 3), the observed absolute ee values deviate more signiﬁcantly, indicating very high speciﬁcity of cinchona alkaloid diastereomers in imparting chirality to noble metal surfaces. On cinchona-modiﬁed Pd the dependence of enantiodiscrimination on the absolute conﬁguration of the chiral modiﬁer appears to be even more pronounced. For example, in the asymmetric hydrogenation of PCA, chiral modiﬁcation with CD (C8(S), C9(R)) is always more eﬃcient compared to the modiﬁcation with CN (C8(R), C9(S)), and the ee is higher by at least 20% irrespective of the catalytic conditions.202 As CD and CN possess similar adsorption strength on Pd,135,203 it was concluded that the shape of the chiral pocket and the enantiodiﬀerentiating interaction with the prochiral substrate are diﬀerent for the two pseudoenantiomers on Pd catalysts. The very high speciﬁcity of the chiral sites imparted by CD and CN on Pd is corroborated by the observation that in the enantioselective hydrogenation of the CC bond in 2-pyrones the eﬃciency of the two pseudoenantiomers is inverted, and for this substrate class chiral modiﬁcation with CN provides roughly 5−10% higher ee.175 The inﬂuence of the conﬁguration at the C8 and C9 atoms of the cinchona modiﬁer was evaluated in a catalytic study by Schmidt et al.69 employing the diastereomer ECD with the C8(S)−C9(S) conﬁguration as chiral modiﬁer. From the catalytic data in Table 5, it emerges that enantioselection and catalytic activity strongly depend on the chiral conﬁguration of the modiﬁer in the hydrogenations of MHA, PCA, and MeoP. The results of quantum chemical calculations shown in Figure 28 indicate a similar adsorption strength of ECD, CD, and CN, and inadequate anchoring on the catalyst surface was considered unlikely to explain the worsened catalytic performance of ECD. Instead, the results revealed the crucial importance of the stereochemical arrangement at the C8−C9 alcohol linker for the enantiodiﬀerentiating ability of cinchona chiral modiﬁers for asymmetric CC bond reduction. The vinyl bond of cinchona alkaloids at C10−C11 is generally not considered to play an important role for the 11550

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

enantioselectivity at low conversion in the asymmetric hydrogenation of PCA, which was not observed when the 10,11-dihydro derivative HCD was used. Nevertheless, for both modiﬁers the ﬁnal enantioselection was the same, and from a catalysis point of view, this induction period is negligible. However, substitution of the vinyl group with an ethyl alcohol group decreased the enantioselectivity.205 The cinchona modiﬁer−prochiral acid interaction in solution was investigated by Szöllösi et al.203 using NMR spectroscopy with the focus on the role of the alcohol function at the C9 atom. The study included cinchona methyl ethers and various PCA derivatives bearing diﬀerent substituents at the aromatic rings. The liquid-phase study conﬁrmed a bidentate interaction between CD and the acid involving the quinuclidine N and the C9-OH of the chiral modiﬁer, as presumed previously.201 According to the NMR data, the less-acidic β-phenyl-paraOCH3 substrate acid (83% ee) formed a stronger +N−H···O− CO bond and a weaker C9−O−H···OC−O bond, compared to the unsubstituted acid (70% ee). However, the relation between the enantioselective control and the acid−base strength was not unambiguous, because the α-phenyl-orthoOCH3 substrate acid, which gave an ee of 76%, formed an even stronger +N−H···O−CO bond and a weaker C9−O−H··· OC−O bond. The deviation of this relation based on measurements in liquid solution indicates the importance of additional interactions on the heterogeneous catalyst, which are sensitive to the enantioselective process. Compared to the parent cinchona alkaloids, the methyl ethers gave the opposite enantiomer with low ee. As illustrated in Scheme 10, the authors explained the inversion of the sense of enantioselection by steric eﬀects originating from the O-methyl group reshaping the chiral site and by the increased ﬂexibility of cinchona ether−substrate complexes (monodentate) compared to the complex involving the parent cinchona alkaloid (bidentate). Szöllösi et al.205 extended their study on C9-ethers and conﬁrmed the steric inﬂuence of the ether substituent on the sense of enantioselection. Increasing the bulkiness of the ether group gradually changed the sense of enantioselection in the hydrogenation of (E)-α-(ortho-methoxyphenyl)-β-(para-ﬂuorophenyl) propenoic acid, that is, from the (S)-enantiomer with an ee of 86% obtained with CD to the (R)-enantiomer with an ee of up to 50% with the bulkiest trimethylsilylether of CD, Si(CH3)2C(CH3) 3-CD. Furthermore, it was shown that either the lack of asymmetry at the C9 atom due to epimerization or a partial hydrogenation of quinoline at the benzene ring results in a severe loss of ee (7% and 14%, respectively) . 3.2.2. Inﬂuence of Chiral Modiﬁer Concentration. An often-observed unfavorable behavior of cinchona-modiﬁed Pd is the rate deceleration of the hydrogenation on modiﬁed catalyst relative to that on unmodiﬁed catalyst, which is assumed to originate from competitive adsorption of the chiral modiﬁer and the prochiral oleﬁn. The group of Sugimura investigated the inﬂuence of the chiral-modiﬁer concentration on the hydrogenation activity and enantioselectivity. The catalytic data in Figure 29 show the ee and relative rate increase as a function of liquid CD concentration. They observed an increasing hydrogenation rate with increasing CD concentration, and the maximum ee was obtained at the highest relative increase of the rate at a modiﬁer−Pd−substrate ratio of 3:10:500.177 As they observed a 30% faster hydrogenation relative to the racemic hydrogenation in the absence of CD, they also use the term “ligand acceleration”,177,202 which, however, might be

Table 5. Enantioselective Hydrogenation of Various Prochiral Oleﬁns Using CD, ECN, and CN As Chiral Modiﬁersa substrate

modiﬁer

ee (%)

CD ECN CN

92 (S) 80 (S) 88 (R)

CD ECN CN

76 (S) 8 (S) 51 (R)

CD ECN CN

70 (S) 1 (R) 74 (R)

MHA

PCA

MeoP

rate (mmol g−1 h−1) 136 55 0.8 51 36 23 2.9 17 27 4.7 1.9 6.1

a

The reaction conditions diﬀer for each substrate; for the experimental conditions, refer to the original work. Reproduced with permission from ref 70. Copyright 2012 Elsevier.

Figure 28. Conformations of CD, CN, and ECD adsorbed on Pt (a− c) and Pd (d−f) surfaces. The pictures show only the ﬁrst metal layer (white spheres) to simplify the representation. C atoms are colored green, O are red, N are blue, and H are white. At the top-right corner of each picture, the name and the adsorption energy of the represented conformation are reported. Reproduced with permission from ref 69. Copyright 2012 Elsevier.

enantioselection on cinchona-modiﬁed Pt or Pd. Mameda et al. 204 reported an induction period for the obtained 11551

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 10. Proposed Structures of the Modiﬁer−Acid Complexes in Liquid Phase with the Parent Cinchona Alkaloids (S1) and with Their Methyl Ethers (S2) (Reproduced with Permission from Ref 203; Copyright 2010 Elsevier)

Figure 29. Product ee (left) and relative initial rate (=observed initial rate/unmodiﬁed initial rate, right) for the hydrogenation of PCA, DMPCA, or FMPCA with CD-modiﬁed Pd/C. Reproduced from ref 177. Copyright 2013 The Chemical Society of Japan.

misleading in this context. The term “ligand acceleration” or “rate enhancement” has been coined in the ﬁeld of chirally modiﬁed metals to express the faster hydrogenation over the chirally modiﬁed catalyst compared to the unmodiﬁed catalyst.210 In their study employing cinchona-modiﬁed Pd catalysts, all the hydrogenations (modiﬁed and unmodiﬁed) were performed in the presence of 1 equiv of the amine additive BA, which leads to better enantioselection (see section 3.2.3) . However, amine addition to unmodiﬁed hydrogenations also decelerates the rate by a factor of roughly 20%.206 Therefore, rather than “ligand acceleration”, a relative rate increase using low chiral modiﬁer concentrations was observed in the presence of BA. More importantly, though, the reported dependence of the hydrogenation activity and enantioselectivity revealed crucial aspects of this catalytic system. First of all, only a relatively small CD-to-Pd ratio of 0.3:1 (often ratios of 6:1 or even higher are used in reported studies) is necessary to obtain the maximum enantioselectivity, and the excess of CD interferes with the eﬃcient hydrogenation. At the highest relative rate increase, the contribution of racemic hydrogenation on unmodiﬁed surface is minimized, leading to the maximum attainable ee. These studies also included various substituted PCAs.211 As the highest relative rate increases of 370% and 270% were observed in the most-eﬃcient asymmetric hydrogenations of α,β-para-dimethoxy PCA (DMPCA, 91% ee) and α-ortho-ﬂuoro-β-para-methoxy PCA (FMPCA, 92% ee), respectively, it was concluded that the paramethoxy or para-ﬂuorine substituents on the β-phenyl ring improve the turnover rate at the modiﬁed site and thus the ee. The ortho-methoxy substitution at the α-phenyl ring, which improves the ee in case the para-electron-withdrawing substituent on the β-phenyl ring is also present, was proposed

to lead to high ee due to improved intrinsic enantioselectivity at the chirally modiﬁed site. The synergy of these two eﬀects can lead to an ee of 93% in the asymmetric hydrogenation of αortho-methoxy-β-para-methoxy PCA (see section 3.3 for additional information on the substrate structure). 3.2.3. Amine Additives. The origin of the beneﬁcial eﬀect of the amine addition has been debated in the literature. There is mutual consensus that the amine accelerates product desorption, thereby facilitating an eﬃcient catalytic turnover at the chirally modiﬁed site. On the basis of an in situ spectroscopic study,207 it has also been suggested that the coadsorption of BA increases the fraction of π-bound CD, which is the surface species generally assumed to be responsible for the enantioselective control on the Pd catalyst (for the characterization of adsorbed chiral modiﬁers and discussion about active species on Pt, refer to section 2.3) . In this in situ spectroscopic study, the involvement of BA in the enantiodiﬀerentiating surface complex involving CD as chiral modiﬁer and PCA as prochiral substrate was ruled out. In contrast, on the basis of liquid-phase NMR investigations, the involvement of BA in the enantiodiﬀerentiating surface complex has also been proposed.203,209 Kim and Sugimura206 investigated the eﬀect of amine addition in more detail and found a strong solvent dependence of the BA eﬀect in the hydrogenation of PCA, tiglic acid, and some of their derivatives. The choice of the solvent was reported to have a more pronounced eﬀect on the enantioselectivity in the absence of amine. In the hydrogenation of PCA in wet 1,4-dioxane, the enantioselectivity was enhanced from 57% to 83% by the addition of the amine, while its presence had only a very minor eﬀect on the hydrogenation in aprotic toluene (78% instead of 75%). A similar trend was 11552

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

observed for the other tested derivatives of PCA and tiglic acid. There is also a lack of information on the inﬂuence of the residual chiral modiﬁer concentration remaining in the liquid reaction solution after saturation of the catalyst surface, because there is essentially no accurate quantitative information on the chiral modiﬁer surface coverage during the hydrogenation. Nakai et al.208 attempted to close this gap and investigated the concentration eﬀect of several amine bases on the catalytic performance, including the amine chiral modiﬁer CD itself. They discovered that even a catalytic amount of the strong amine base 1,8-diazabicyclo[5.4.0]undec-7-ene is suﬃcient to reach high ee and that the beneﬁcial amine eﬀect on the enantioselection is also brought about by an excess amount of the cinchona alkaloid dissolved in the liquid phase. While a high liquid-phase concentration of the chiral modiﬁer diminishes the reaction rate, the enantioselection is improved similarly as with other amine bases. An important conclusion of their study is that the required amount of amine is rather dependent on the amount of added chiral modiﬁer and the used quantity of Pd catalyst, instead of the amounts of PCA or solvent. Another suggestion for the beneﬁcial eﬀect of amine addition is the involvement of BA in the enantiodiﬀerentiating surface complex. In an NMR spectroscopic study in wet DMF, Szöllösi et al.203 found evidence for the formation of some trimolecular complexes in solution, such as PCA-BA-CD as well as PCA-BAMeOCD complexes, shown in Scheme 11.

Figure 30. Time-resolved 2D surface plot of ATR-IR spectra during adsorption and desorption of (a, b) PCA and (c, d) DPPA over CDmodiﬁed Pd at 323 K in (a, c) absence and (b, d) presence of BA. The unit of absorbance (color bar) is 10−3. Reproduced with permission from ref 207. Copyright 2012 American Chemical Society.

Monitoring of catalytic solid−liquid interface revealed restructuring of surface acid−base adducts involving the product but not the substrate, contrasting the proposed involvement of BA in the enantiodiﬀerentiating surface complex. On the basis of the in situ spectroscopic data, the authors proposed the catalytic cycle shown in Scheme 12 for

Scheme 11. (A) Liquid-Phase Modiﬁer−Acid Complexes in the Presence of BA with the Parent Cinchona Alkaloids (S3−S5) and with Their Methyl Ethers (S6) (Reproduced with Permission from Ref 203; Copyright 2010 Elsevier)

Scheme 12. Proposed Catalytic Cycle in the Presence of BAa

a

Indices (l) and (ad) represent liquid-phase and adsorbed species, respectively. Reproduced from ref 207. Copyright 2012 American Chemical Society.

the asymmetric hydrogenation of PCA on CD-modiﬁed Pd catalyst in the presence of BA. The time-resolved ATR-IR spectroscopic data reported for the coadsorption of CD and BA also indicated an altered surface conﬁguration of CD from Nlone pair bound CD to π-bound CD, which was proposed to lead to a better stereochemically controlled catalyst. However, the experimental conditions in this study on the transient adsorption of the chiral modiﬁer were poorly chosen with respect to the concentration of CD and the adsorption kinetics of the modiﬁer, and overlapping signal contributions from dissolved species were not taken into account. Therefore, the proposed dynamic change in the adsorption conﬁguration by the presence of BA should be treated with care. In the asymmetric hydrogenation of aliphatic prochiral acids, addition of BA also leads to an enhanced enantioselection on cinchona-modiﬁed Pd catalyst. A variety of primary amines were screened in the asymmetric hydrogenation of MBA,213 and basic and steric requirements of the amine additive were identiﬁed. Except for naphthyl methylamine, which probably

In another study involving CD-O-ethers, which generally invert the sense of enantioselection, the same authors205 also noted that addition of BA always increased the amount of enantiomer produced by the parent cinchona alkaloid, and the amine facilitated desorption of partially hydrogenated cinchona alkaloids. These observations were interpreted as an indication for the participation of BA in the enantiodiﬀerentiating surface step during the asymmetric hydrogenation of PCA or PCA derivatives. In an in situ spectroscopic study on the asymmetric hydrogenation of PCA, Meemken et al.207 conﬁrmed the more-eﬃcient removal of product acids from the chirally modiﬁed catalyst in the presence of BA, as proposed initially by Nitta and co-workers.212 The ATR-IR spectra in Figure 30 show the adsorption and desorption of PCA and the product acid diphenyl propanoic acid (DPPA) in the absence and presence of BA on Pd/TiO2 catalyst in methanol. 11553

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

adsorbs competitively to CD, all tested amines improved the enantioselection and decreased the hydrogenation rate. The best ee of 64% was achieved with (R)-2-phenylethylamine (62% with BA) at 293 K, but a chiral preference of the amine additive with the (R)-conﬁguration cannot be inferred from the slightly lower ee of 61% with (S)-2-phenylethylamine considering a reasonable error margin. More basic aliphatic amines also increased the ee signiﬁcantly, and addition of 1dodecaneamine as the amine with the longest hydrocarbon improved the ee to nearly the same level (60% ee) as obtained with BA. The inﬂuence of the molecular structure of the amine additive was further investigated by the same authors209 in the asymmetric hydrogenation of MHA, including diﬀerent H2 pressures and reaction temperatures. They found that secondary amines are similarly eﬃcient and concluded that the right basicity coupled with steric requirements are crucial for enhancing the enantioselection. The ee could be further improved from 68% with BA to 71% using N-methyl BA at a reaction temperature of 273 K. The study was later extended,214 and the eﬀect of BA and N-methyl-BA was investigated in various solvents on Pd catalyst modiﬁed with CD and some CD derivatives. Enantioselection was reported to be generally good in apolar and polar aprotic solvents (toluene, ethyl acetate, and acetone) and poor in strong H-bond donor solvents such as water or acetic acid. While the ee was generally low in protic methanol, in aprotic toluene the ee improved at higher initial acid concentration, which should favor acid-CD dimer formation in the liquid phase. Regarding the structure of the enantiodiﬀerentiating surface complex, the observed dilution eﬀect was interpreted as an indication for better enantiodiﬀerentiation of surface complexes involving two or multiple substrate acid molecules (2:1 or 3:1 acid−chiral modiﬁer complexes) . The observed concentration dependence seems to indicate a mechanistic distinction between the asymmetric hydrogenation of aliphatic and aromatic acids. For the aromatic acids lower substrate concentration is favorable for the enantioselection, suggesting better enantioselective control of 1:1 complexes. However, other properties of the substrate such as the solubility or the adsorption strength can also signiﬁcantly inﬂuence the outcome of catalytic reactions, and in situ spectroscopic studies might be necessary to resolve the structure of the active diastereomeric complexes. Regarding the beneﬁcial eﬀect of amine addition in the asymmetric hydrogenation of aliphatic prochiral acids, liquidphase molecular complexes have been proposed as shown in Scheme 13, and the involvement of BA in the enantiodiﬀerentiating surface complex was proposed based on three observations.214 First, the participation of BA in the ratedetermining step was suggested, because the hydrogenation rate increased less signiﬁcantly in the presence of BA with increasing solvent polarity. Second, the amine eﬀect is more pronounced using CD or CN compared to C9−O-methyl ether derivatives or QN (C6−O-methyl substitution), which was explained with steric eﬀects hindering the participation of BA. Third, using chiral-modiﬁer mixtures, an increased deviation from the nonlinear behavior is observed in the presence of amine, which is suggested to originate from an increased adsorption strength of the two used chiral modiﬁers in the presence of amine. Compared to the asymmetric hydrogenations of aromatic and aliphatic prochiral monocarboxylic alkene acids, the dicarboxylic acid itaconic acid shows additional peculiarities with respect to the reaction conditions. While for the

Scheme 13. Possible Structures of the Intermediates of Aliphatic Acid Hydrogenation in the Presence of Amine Additives in Aprotic Solvents (Reproduced with Permission from Ref 214; Copyright 2014 Elsevier)

asymmetric hydrogenation of PCA addition of BA prior to substrate addition led to a considerable decrease of ee,215 for the dicarboxylic acid the best result was obtained when the amine was added to the reaction solution prior to the chiral modiﬁer and the substrate.216 By optimizing the conditions with respect to the hydrogen pressure, the order of adding the reaction components, the choice of the solvent, the reaction temperature, the amine-to-substrate ratio, and the amine additive structure, an ee of up to 58% could be achieved. 3.2.4. Proline-Based Chiral Modiﬁers. In contrast to asymmetric hydrogenation of activated ketones on Pt catalysts, asymmetric hydrogenation of CC bonds on Pd catalysts is generally reported to require high chiral modiﬁer concentration to yield good ee.180 While it was in fact shown that lower concentration than generally assumed is beneﬁcial,177 stronger adsorption of the modiﬁer could be desirable to counter competitive adsorption with the prochiral substrate. In addition, the susceptibility of aromatic CC bond to reduction on Pd is proposed to deter the stability of the adsorbed quinoline anchor of cinchona alkaloids, leading to a loss of enantioselection.200 A possible strategy to prevent the undesired side-reaction is to implement a diﬀerent anchoring moiety in the chiral modiﬁer molecule, which is more resistant against hydrogenation on Pd. As a ﬁrst attempt Watson et al.217 employed surface-tethered chiral modiﬁers by covalent sulfur anchoring and tested the proline-derived ligands depicted in Figure 31 in the asymmetric hydrogenation of IP. While enantioselectivity could be bestowed to the Pd catalyst, the observed catalytic activity and ee were in general on a very poor level. The best sulfur-containing chiral modiﬁer achieved an ee of 14%. The presence of the thio-ether strongly poisoned the Pd catalyst, and compared to the unmodiﬁed hydrogenation, the rate was reduced by roughly 3 orders of magnitude. Nevertheless, this study provided important insight into the design of chiral 11554

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 31. Molecular structures of (S)-Pr containing derivatives tested in the asymmetric hydrogenation of IP (top). Dependence of TMCH product ee on initial modiﬁer ligand concentration. Each data point corresponds to running the reaction up to a conversion of ∼60%. A ligand concentration of 0.47 mmol dm−3 corresponds to 0.1 mol % ligand with respect to IP (middle). Dependence of activity (reactant conversion after a ﬁxed time (168 h)) on ligand concentration for (S)-2-(tert-butylthiomethyl)pyrrolidine and (S)-2-(phenylthiomethyl)pyrrolidine (bottom). Reproduced with permission from ref 217. Copyright 2009 American Chemical Society.

application range. In addition, there have been several discoveries of eﬃcient asymmetric hydrogenations of additional substrates in the past,179,223,224 which not only extend the application range of chirally modiﬁed metals but also shed light on the structural requirements of the prochiral substrate for cinchona-modiﬁed metal catalyst. 3.3.1. Phenyl Cinnamic Acid and Derivatives. While PCA had long been considered the most suitable prochiral alkenoic acid substrate for CD-modiﬁed Pd, Sugimura et al.215 revealed the beneﬁcial eﬀect of proper substitution at the phenyl rings to enhance the ee from 72% for PCA to 82% using DMPCA. As the optimized reaction conditions were the same for the two hydrogenations, the authors concluded that the number of CD-modiﬁed sites should be the same, and the methoxy-substituted acid experiences a higher intrinsic enantioselectivity. Deeper insight into the structure−enantioselectivity relationship with respect to the molecular structure of the prochiral acid was revealed by the catalytic study of Sugimura et al.218 employing various α- and β-substituted propenoic acids. While the aryl substituent in β-position is crucial to yield high product

modiﬁers covalently bound to the catalyst by sulfur. The size of the alkyl group at the sulﬁde was shown to have a decisive eﬀect on the observed ee, which was linked to the varying degree of ligand dispersion on the surface. In catalytic experiments using varying sulfur ligand-to-Pd ratios, shown in Figure 31, the inﬂuence of the molecular structure of the thiol-ether chiral modiﬁers were investigated. The authors proposed that small substituents lead to island formation on the surface, while big substituents favor dispersion as separated ligand molecules. 3.3. Substrates

The asymmetric hydrogenations of PCA and aryl-substituted derivatives of PCA shown in Scheme 9 are excellent model reactions to study cinchona-modiﬁed Pd catalyst, reaching ee values up to 96%.176,178 Despite the encouraging strong stereochemical control of speciﬁc hydrogenations, the small substrate pool represents one of the major limitations of cinchona-modiﬁed Pd. In the past decade numerous catalytic studies involving various structurally related alkenoic acids have provided important insight into the structure−enantioselectivity relationship,176,218−222 which will help to broaden the 11555

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

ee, prochiral acids carrying α-alkyl substituents can also be hydrogenated with good enantioselectivity. The best ee of 86% was reported for α-isopropyl-β-para-methoxyphenyl propenoic acid, which was proposed to lead to excellent enantioselectivity due to the proper bulkiness of the α-substituent. Bulkier substituents such as cyclohexyl or t-butyl (no reaction observed) decreased the enantioselective control. The positive eﬀect of the para-methoxy substitution at the βphenyl ring on enantioselection was further investigated by Szöllösi et al.219 The systematic variation of the substituents at the two phenyl rings revealed electronic as well as steric eﬀects of the methoxy group, inﬂuencing the hydrogenation activity as well as the enantioselectivity. The inﬂuence of the aromatic ring substituent in ortho position was explained by a steric eﬀect, and ortho substitution at the α- and the β-phenyl increases and decreases the catalytic performance (activity and ee), respectively. Due to proper substitution in the substrate, (E)α-(ortho-methoxyphenyl)-β-(para-methoxyphenyl) propenoic acid, favorable steric and electronic eﬀects could be combined, providing an ee of 92%. The same authors176 further studied the eﬀect of the substrate structure employing ﬂuorinesubstituted PCA derivatives. They obtained the best results with PCA derivatives carrying an ortho-substituent at the αphenyl and a para-substituent at the β-phenyl ring. As they observed the same enhancement with the small ﬂuorine atom in (E)-α-(ortho-ﬂuoro)propenoic acid as with the respective methoxy-substituted acid, steric hindrance should only play a minor role, and rather an additional interaction of this substituent with the modiﬁer was suggested to lead to better enantioselective control. An electron-releasing substitution at the β-phenyl ring, which decreases the acidity of the substrate, was proposed to enhance the stereochemically controlling interaction between the chiral modiﬁer and the substrate (electronic eﬀect) and to decrease the adsorption strength of the acid at the chirally modiﬁed site. In the hydrogenation of (E)-α-(ortho-ﬂuoro)-β-(para-ﬂuoro)propenoic acid, an ee of 96% was reported at a reaction temperature of 273 K (91% at 294 K) . On the basis of their extensive study involving 42 diﬀerently mono- and disubstituted propenoic acids, Sugimura et al.220 proposed the general substrate structure−enantioselectivity relationship shown in Figure 32. In contrast to the beneﬁcial ﬂuorine substitution at the phenyl rings of PCA, asymmetric hydrogenation of ﬂuorinated

unsaturated carboxylic acids was rather ineﬃcient.225 Among the three tested triﬂuoromethyl α,β-unsaturated acids, the hydrogenation of (E)-4,4,4-triﬂuoro-3-phenyl-2-butenoic acid aﬀorded the best ee of 43%. The catalytic performance of heteroaromatic derivatives of PCA was also assessed.221 A 2furyl group in β-position has neither a positive nor a negative eﬀect on the ee, but simultaneous reduction of the furyl ring and the oleﬁnic CC bond was observed, indicating the coplanarity of these two groups of the adsorbed substrate. A pyridyl group at α-position lowers the ee, supposedly due to competitive interaction involving the N atom of the pyridyl ring. Further information on the adsorption conﬁguration of PCA on unmodiﬁed and CD-modiﬁed Pd catalyst was revealed by the study of Szöllösi et al.,222 employing chlorine-substituted PCAs. While the electron-releasing chloro substitution at the βphenyl ring also improved the enantioselection, fast dehalogenolysis of the para-Cl was observed, indicating the parallel arrangement of this phenyl ring with respect to the surface. In contrast, a tilted arrangement of the α-phenyl ring as illustrated in Figure 33 was inferred from the signiﬁcantly slower scission

Figure 33. Schematic illustration of the arrangement on the Pd surface of (E)-2,3-diphenylpropenoic acids interacting with the modiﬁer. Reproduced with permission from ref 222. Copyright 2010 Elsevier.

of the C−Cl, either in ortho- or para-position. Disubstitution in ortho,ortho-position diminished the hydrogenation rate, which was also found in the hydrogenation of the α-phenyl diﬂuoro acid, and the adsorption of the CC bond of such substrates must be sterically hindered. Following the established design principles with respect to the molecular prochiral acid structure, an ee of 95% could be obtained at 295 K in the asymmetric hydrogenation of (E)-α-(ortho-methoxy)-β-(meta,paradiﬂuoro)propenoic acid, which is the currently best-reported enantioselection obtained with CD-modiﬁed Pd catalyst for asymmetric CC bond hydrogenation at 295 K (Scheme 9) .176 3.3.2. New Substrate Classes. Szöllösi et al.223 reported on the asymmetric hydrogenation of N-methyl-3,4-dehydronipecotic acid, achieving an ee of 60% of the (S)-enantiomer using CN as chiral modiﬁer. In contrast to the asymmetric hydrogenation of PCA, CD provided poorer enantioselection with only 40% (R). Regarding the molecular structure of the substrate, it was noticed that the free carboxylic acid in alphaposition to the CC bond and the alkylation of the N atom are key features. On the same Pd/Al2O3 catalyst, no conversion of the secondary amine guvacine was observed at 1 and 5 bar H2 pressure. Another interesting substrate class, which shows a peculiar behavior with respect to the used hydrogen pressure, is Nacetyldehydroamino acids. Using CD-modiﬁed Pd/TiO2, (R)N-acetylphenylalanine is formed in the hydrogenation of 2acetamidocinnamic acid with an ee of 36% at 1 bar H2 pressure. Addition of BA or an increase in the hydrogen pressure to 50 bar provided the other (S)-enantiomer with a low ee of 5−6%. In the asymmetric hydrogenation of 2-acetamidoacrylic acid,

Figure 32. (A) Newman projections of the substrate looking from the chiral amine adsorbed on the Pd surface to give (a) a major enantiomer and (b−d) a minor enantiomer, the latter of which is caused by (b) misidentiﬁcation of α-aryl as the oleﬁn part, (c) loss of the conformational planarity, or (d) lack of the stereorecognition of the carboxyl group. Reproduced with permission from ref 220. Copyright 2009 Elsevier. 11556

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

modiﬁer feed concentration, and stopping or decreasing of the CD concentration led to a sudden drop in ee. The authors interpreted the high sensitivity of the enantioselection on the liquid-phase concentration of CD by a preferred chiral modiﬁer−substrate interaction in the liquid phase and claimed that such a “salt-like complex” adsorbs stronger on the catalyst than the genuine chiral modiﬁer or substrate alone. Nevertheless, it should be mentioned that fast saturation of the quinoline ring anchor of CD is favored at these applied high hydrogen pressures, and replenishment of partially hydrogenated CD is required to maintain good enantioselection.200 Furthermore, the relevance of the proposed interaction in liquid-phase contrasts the ﬁndings of Sugimura and co-workers at atmospheric pressure, who showed that the best catalytic performance is obtained at relatively low chiral modiﬁer concentrations.177,211 A ﬁrst attempt to model the kinetics of chirally modiﬁed Pd catalyst was reported by Murzin and Szöllösi,230 focusing on the kinetic description of the nonlinear phenomenon of chiral modiﬁer mixtures, which is an important topic regarding the (optical) purity of the used chiral modiﬁer. The model allows one to simulate experimentally observed maxima of ee depending on the composition of binary modiﬁer mixtures, which was achieved by considering the dependence of one modiﬁer on the rate constant of a step involving the other modiﬁer. However, considering the lack of mechanistic understanding of the underlying surface processes, the application of kinetic models is in a very early stage. The authors themselves pointed out the overparametrization of the system and that the model is not able to determine values of their parameters with statistical signiﬁcance. The theoretical study by Meier et al.231 provides insight into the molecular structure of diastereomeric acid−base complexes in liquid phase pertinent to asymmetric hydrogenations on CDmodiﬁed Pd. NMR and IR spectroscopic data of the solution phase were interpreted with the aid of quantum chemical calculations, and the structure of acid−base complexes involving the aliphatic acid MHA and CD was elucidated. The study revealed that MHA and CD do not tend to form 1:1 acid−base adducts in solution in the absence of the Pd catalyst, and rather 3:1 complexes, as shown in Figure 34, exist abundantly in toluene. The catalytic study on the asymmetric hydrogenation of aliphatic acids by Makra and Szöllösi214 also indicated the importance of acid−modiﬁer complexes involving more than one acid molecule, because better enantioselection was achieved in aprotic solvents, including toluene, at higher substrate concentration. However, catalytic studies employing structurally altered substrates (section 3.3) and various catalysts (section 3.1) convey the crucial role of the heterogeneous catalyst, as well as the high sensitivity of the surface processes pertinent to this heterogeneous catalytic system. Nevertheless, theoretical studies such as the study by Meier et al. will be necessary to understand the enantioselection and should be regarded as a reference for the analysis of surface spectra. Very recently, the adsorption of the cinchona modiﬁer CN was characterized under working conditions at the catalytic solid− liquid interface involving a supported polycrystalline Pd catalyst and the solvent isopropanol.96 As depicted in Figure 35, four diﬀerently bound cinchona surface species were identiﬁed and the enantiodiﬀerentiation was shown to vary signiﬁcantly depending on the adsorption mode of the quinoline-anchoring moiety of the cinchona

only the (S)-enantiomer was obtained under the investigated reaction conditions, and the best ee of 58% was achieved at 50 bar H2 pressure, 275 K, and by the addition of an equivalent amount of BA.224 With an ee of up to 89% of the cockroach attractant (+)-tetrahydro-2H-pyran-3-carboxylate, the newly discovered substrate 5,6-dihydro-2H-pyran-3-carboxylic acid, also depicted in the overview in Scheme 9, shows potential for synthetic application of CD-modiﬁed Pd.179 The authors also noted that they had not found any report in the literature for this hydrogenation using a homogeneous catalyst. Comparison with structurally similar substrates revealed the crucial importance of the cyclic structure, the free carboxylic acid group, and the position of the ring oxygen atom relative to the prochiral CC group, which permits exploitation of the beneﬁcial eﬀect of the achiral amine additive and consequently leads to obtain the high ee. Maximum enantioselection of 89% was found at 50 bar H2 pressure, while 1, 10, and 100 bar led to ee’s of 42%, 82%, and 70%, respectively. Application of CD-modiﬁed Pd, Ir, and Ru catalysts was unsuccessful in the hydrogenation of the 6,7-dimethoxy-3,4dihydroisoquinoline providing only 3−4% ee.226 In contrast, using triphenylphosphane-stabilized catalyst modiﬁed with diphenylenediamine gave an enantioselection of 70−80% and a Ru(II)-aminophosphane complex even gave an ee of 97% in the hydrogenation of such N-heterocyclic compounds. However, the catalyst suﬀered from leaching of the chiral complexes to the liquid phase. 3.4. Mechanistic Aspects

3.4.1. Cinchona-Modiﬁed Pd. Most of the current mechanistic understanding of chirally modiﬁed Pd catalyst for asymmetric hydrogenation of prochiral oleﬁns is derived from general catalytic studies, and only a few kinetic and in situ spectroscopic studies exist. The group of Williams reported two kinetic studies. In the hydrogenation of the aliphatic substrate MHA on Pd/Al2O3, among the three investigated solvents the highest catalytic activity was obtained in methanol, while the rates of hydrogenation in dioxane were lower by a factor of 3−7, and the use of CH2Cl2 strongly poisoned the catalyst.227 The highest reported ee for this substrate was 30%. In the hydrogenation of the aromatic substrate 2-methyl cinnamic acid, a strong solvent dependence of the activity and enantioselectivity was observed, too.228 As a zero reaction order in the hydrogen pressure (in a range from 1 to 30 bar) and a ﬁrst-order substrate acid concentration dependence was found, and no eﬀect on the rate was observed by addition of product, the authors concluded that substrate adsorption and activation has to be the rate-determining step. An isotopic investigation using D2 could conﬁrm that hydrogen is not involved in the rate-determining step. Regarding the chiral modiﬁcation, they reported a strong inhibiting eﬀect by the addition of CD. However, the modiﬁer-to-Pd ratio was >7 to 1 very high in these studies, which might have obscured important kinetic information. Molar ratios in the range of 0.3−1 provide the best catalytic performance (compare section 3.2.2.) .177,211 Hermán et al.229 studied the asymmetric hydrogenations of PCA, MBA, MHA, and itaconic acid in a continuous ﬁxed-bed reactor at hydrogen pressures between 10 and 50 bar and obtained similar performances as usually obtained in a slurry reactor. The enantioselection was sensitive to the chiral 11557

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

3.4.2. Proline-Mediated Asymmetry on Pd. The (S)-Prmediated asymmetric hydrogenation of IP has been the focus of several mechanistic studies.192,193,196,197,233,234 The understanding of the reaction mechanism has recently been reviewed.6,235 Nevertheless, the origin of enantioselection of this catalytic system has been the subject of a controversial debate, and even the classiﬁcation as heterogeneous asymmetric hydrogenation had been in dispute. In contrast to the small amount of chiral modiﬁer generally employed for chirally modiﬁed metals (molar ratios of chiral modiﬁer-to-Pd < 1 are used; see section 3.2), early catalytic results indicated that equivalent amounts of (S)-Pr with respect to the prochiral substrate are required to obtain good enantioselection, which can reach up to 99% ee of (S)-TMCH. While Tungler and co-workers originally proposed enantioselectivity to result from stereoselective hydrogenation of an IP-(S)-Pr condensate on the heterogeneous catalyst,236,237 the discovery of the preferential condensation of (R)-TMCH with (S)-Pr in methanol solution192 prompted the questioning of the classiﬁcation as heterogeneous asymmetric hydrogenation.193 To stress the importance of kinetic resolution for the origin of enantioselection, the solution phase of this catalytic system was monitored by UV−vis spectroscopy,193 competitive adsorption processes were investigated on a Pt electrode in an electrochemical study,234 and displacement measurements of the prochiral substrate, the amino acid, and the chiral product were performed under catalytic conditions. Even though the heterogeneous catalytic system had only been monitored ex situ in the solution phase or under model conditions (the hydrogenation has to be performed on supported Pd catalyst in methanol, and other noble metals and solvents lead to totally diﬀerent results), Lambert and coworkers193 claimed that all earlier mechanistic hypotheses were wrong and enantioselectivity is merely the result of kinetic resolution of the cyclohexanone product in the liquid phase, which does not include the heterogeneous catalyst. However, the displacement experiments using a Pd/C catalyst were not properly designed, as neither the simultaneous addition of (S)Pr and IP (premixing time was not speciﬁed) nor the subsequent addition of IP to (S)-Pr premodiﬁed catalyst were able to focus on the crucial role of the IP-(S)-Pr condensate. Recently, it was shown that the molar ratio of IP-(S)-Pr condensate to Pd catalyst is a decisive reaction parameter for the observed ee, and the enantioselectivity-controlling steps do take place on the heterogeneous catalyst.197 Depending on the surface coverage of the Pd catalyst, the reaction is controlled either by kinetic resolution ((S)-pathway) or by chiral catalysis ((R)-pathway), and the reaction mechanism in Scheme 14 can explain the complex behavior of this catalytic system. Regarding the surface processes occurring on the Pd catalyst, weak adsorption of (S)-Pr is frequently suggested, making the amino acid a poor chiral surface modiﬁer. Wang et al.198 investigated a few adsorption conﬁgurations of (S)-Pr on a Pd slab by means of DFT. They found the most stable adsorption in the H- and O-adsorption illustrated in Figure 36 with rather low adsorption energies of −10.4 and −14.1 kcal mol−1, respectively. Formation of the zwitterion in the O-adsorption mode is in agreement with XPS results238 and in situ IR spectroscopic characterization.197 In situ spectroscopic monitoring of Pd/C revealed that both (S)-Pr and IP interact rather weakly with the Pd catalyst in the presence of methanol. They also do not adsorb competitively on polycrystalline Pd. Instead, the condensation products control the adsorption on the metal surface and, therefore, the

Figure 34. O−H bond length trajectories of MHA interacting with the quinuclidine N of CD during the ab initio molecular dynamics (18.4 ps). The initial complex structure was the 1:3 complex shown as C. The two diﬀerent oxygen atoms in MHA are noted as O and O′. Reproduced with permission from ref 231. Copyright 2008 American Chemical Society.

alkaloid. According to the operando spectroscopic characterization, the weaker bound CN surface species III with the quinoline ring anchor tilted along both axes provides the most eﬃcient enantioselection, providing up to 87% ee in the asymmetric hydrogenation of MeoP (compare also section 2.3, Adsorption of Modiﬁers) . Sun and Williams232 studied the interactions between CD and prochiral alkenoic acids spectroscopically at the catalytic solid−liquid interface involving a Pd/Al2O3 powder catalyst. The substrate 2-methyl cinnamic acid was found to adsorb only molecularly on the alumina support as a dimer in the solvent CH2Cl2. On the Pd catalyst, the acid adsorbed predominately as a bidentate bridging carboxylate. Comparing the adsorption of CD in methanol and in CH2Cl2, the similar band positions of the adsorbed alkaloid indicated no signiﬁcant diﬀerence in the adsorption orientation of the chiral modiﬁer in the protic and aprotic solvents.150 The substrate MPA was also included in the spectroscopic study. In the presence of CD, the position of the absorption bands induced by the aliphatic acid were more ﬁxed, which was interpreted as a preferred adsorption conﬁguration on CD-modiﬁed surface compared to unmodiﬁed surface. On the basis of the position of the carboxylate bands, a 1:1 CD− MPA diastereomeric surface complex was suggested, while 2:1 or even higher acid−modiﬁer complexes are proposed to play a more important role based on observations of catalytic studies (higher substrate concentration leads to better enantioselection) . Another in situ spectroscopic study207 employing a CDmodiﬁed Pd/TiO2 catalyst showed acid−base interaction between the chiral modiﬁer and PCA on the catalyst surface. The study also revealed restructuring of surface acid−base adducts in the presence of the amine additive BA. While the carboxylate band position of the product acid DPPA changed upon introduction of an equivalent amount of BA, no signiﬁcant spectral change was observed for the substrate acid PCA (see section 3.2.3) . 11558

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Figure 35. Correlation of CN surface coverage to the catalytic performance (TOF, ee) revealing four regimes (I−IV) each predominately covered by one of the identiﬁed CN surface species. Accordingly, the tilted CN surface species III is the active species, which is most relevant for the enantioselection. Reproduced with permission from ref 96. Copyright 2017 American Chemical Society.

underlying catalysis. If the molar ratio of the IP-(S)-Pr condensate to Pd catalyst is suﬃciently high (exceeding roughly 700 for the investigated Pd/C catalyst),197 this chiral condensate is the most abundant surface species and asymmetric catalysis occurs, that is, (S)-Pr is not consumed but mediates the enantioselective hydrogenation to (R)-TMCH according to the reaction mechanism in Scheme 14. If the concentration of the cyclohexenone condensate in the reaction mixture is too low to cover all available catalytic sites, hydrogenation of the cyclohexanone condensate occurs, which consumes TMCH and (S)-Pr. As the diastereomer (R)-TMCH-(S)-Pr is signiﬁcantly faster hydrogenated than the (S)-TMCH-(S)-Pr condensate, (S)-TMCH is produced by kinetic resolution. As for the chiral catalysis (R)-pathway, the enantioselectivity-determining reaction step of the kinetic resolution ((S)-pathway) does not take place in the liquid

solution but on the heterogeneous catalyst. Even though McIntosh et al.193 observed a lower ee of (S)-TMCH with increasing premixing time (increasing concentration of IP-(S)Pr condensate in the reaction mixture), they falsely interpreted the role of the cyclohexenone product as spectator species, not considering competing enantioselective pathways forming the opposite enantiomer. It had also been overlooked that the enantioselectivity of the kinetic resolution in the liquid solution is very poor, rapidly reaching the equilibrium with a diastereomeric ratio of 150:133 in favor of (R)-TMCH-(S)Pr. The reason for the very eﬃcient kinetic resolution is the heterogeneous hydrogenation of the reaction mixture containing both diastereomers in equilibrium, because the (R)-TMCH(S)-Pr condensate is hydrogenated signiﬁcantly faster, leading to the observed ee of the (S)-enantiomer.197 11559

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Scheme 14. Enantioselectively Competing Reaction Pathways Involving Kinetic Resolution ((S)-Pathway) and Chiral Catalysis ((R)-Pathway) for the (S)-Pr-Mediated Asymmetric Hydrogenation of IP on Supported Pd Catalyst (Reproduced with Permission from Ref 197; Copyright 2015 American Chemical Society)

Figure 36. Adsorption of (S)-Pr on the Pd slab. (a) Adsorption through the H atoms in the pyrrolidine ring; (b) adsorption through the O atoms in the carboxylate group. Reproduced with permission from ref 198. Copyright 2014 Elsevier.

Figure 37. Asymmetric heterogeneous hydrogenation of IP in the presence of covalently bound (S)-2-(tert-butylthiomethyl)pyrrolidine. (a) Diﬀerence in steric inhibition for reactant IP−modiﬁer conﬁgurations of the iminium intermediate that leads to (R)- and (S)-TMCH. (b) Enhanced steric interaction between the geminal dimethyl group (Me*) of IP and the tert-butyl group of the modiﬁer upon tilting IP from ﬂat to ∼42°. At 42°, it is clear that these two groups would sterically interfere in the orientations depicted when following the direction of approach indicated by blue arrows. Reproduced with permission from ref 239. Copyright 2010 American Chemical Society.

A rationale for the diastereoselective hydrogenation of the imine (or enamine) condensate on the Pd surface was provided by Beaumont et al.239 based on steric and geometric eﬀects. They proposed steric hindrance between the geminal dimethyl group on IP and the tert-butyl group in the proline-derived modiﬁer to lead to the preferential formation of the pro-(S) conformer of the surface condensate, as depicted in Figure 37. In the catalytic hydrogenation of IP in the presence of (S)-2(tert-butylthiomethyl)pyrrolidine, formation of (S)-TMCH was found with an ee of 14%.217 However, it appears that subtle changes in the molecular structure may even alter the sense of stereochemical control, because opposite diastereoselectivity is observed for the hydrogenation of the (S)-Pr-IP condensate (right-handed

product chirality)197 and the surface condensate involving (S)-2-(tert-butylthiomethyl)pyrrolidine (left-handed product chirality).217 It is not even clear whether (S)-Pr and its derivatives form the imine or the enamine structure with the α,β-unsaturated ketone IP, and even an oxazolidione 11560

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

intermediate has been suggested.240 In homogeneous catalysis, evidence for exclusive formation of the enamine in acetonitrile was recently published, although involving an aldehyde and not the ketone IP.241 To better understand the origin of the diastereoselective hydrogenation of such chiral condensates on Pd, it is crucial to resolve the molecular structure of the chiral reactive intermediate. As pointed out by Beaumont et al.,239 care has to be taken when comparison to chemistry in homogeneous solution is made and analogies to surface chemistry are drawn. On the basis of the recent ﬁndings, the (S)-Pr-mediated asymmetric hydrogenation of IP is certainly a heterogeneous asymmetric hydrogenation. The discovered pathway of chiral catalysis could evolve as a new strategy for heterogeneous asymmetric hydrogenation. In contrast to chirally modiﬁed metals, in the (S)-Pr-mediated asymmetric hydrogenation the chiral environment is not induced at the active metal surface, but the chiral information is transferred to the prochiral substrate to achieve chiral catalysis. Future research has to show whether this approach can be extended to other α,βunsaturated ketones and if structure−enantioselectivity relationships can be derived for this catalytic system based on simpler liquid-phase investigations.

tryptophan, due to the similarity of the crucial functional parts to cinchona alkaloids, the availability of functional groups suitable for facile adaption of the molecular structure to a speciﬁc substrate, and relatively low cost. A promising strategy to cope with the complexity of the catalytic system and to guide the design process in a more rational way is the application of operando spectroscopy, which ideally has to be designed for tracing measured enantioselectivity back to molecular processes at the catalytic solid−liquid interface. As the parametric sensitivity of chirally modiﬁed metals is very high and subtle changes, e.g., changes of the hydrogen pressure or solvent, often result in a signiﬁcant decrease or complete loss of enantioselection, decisive information for the design of an enantioselective catalyst might be lost by characterization under nonoperating ex situ conditions. From a synthetic point of view, future research has to show whether the improving mechanistic understanding is fruitful for the development of enantioselective surface reactions and how emerging structure−enantioselectivity relationships contribute to a more rational design. The predictive power of quantum chemical calculations might also beneﬁt from the continuous progress in (enantioselective) catalyst characterization, and computational methods could soon become a relatively inexpensive and powerful tool to support the experimental approach. From a mechanistic point of view, one of the most important open questions in the ﬁeld of chirally modiﬁed metal catalysts concerns information on the actual surface coverage of chiral modiﬁers (number of enantioselective sites) and their intrinsic enantioselectivity (performance of a single modiﬁed site) during the reaction (Scheme 5). As both the actual number and the intrinsic catalytic performance of chirally modiﬁed sites are not extractable with current methods, the contribution of nonmodiﬁed catalyst (remaining racemic sites) to the overall observed catalytic performance can only be assumed based on macroscopic observations. A methodology for describing the intrinsic performance of a chirally modiﬁed site for a speciﬁc substrate (rate and enantioselectivity) would be ideal for a rational catalyst design. In general, deeper quantitative information about the surface processes occurring on an operating chirally modiﬁed metal catalyst is needed as it would guide future research eﬀort decisively by revealing whether the catalyst design has to aim at improving the stereochemical control at the modiﬁed site, enhancing the stability of chiral sites, or eliminating racemic active sites on the heterogeneous catalyst. Regarding the latter aim, today there is still no selective blocking strategy available, which only aﬀects racemic sites. Eﬀective strategies for blocking racemic sites could revolutionize the development of chirally modiﬁed noble metal catalysts for asymmetric catalysis. A diﬀerent strategy to induce enantioselection to heterogeneous hydrogenations is to transfer the chiral information to the prochiral substrate. It was recently uncovered that the Pdcatalyzed heterogeneous asymmetric hydrogenation of IP in the presence of (S)-Pr is based on such a mechanism. However, the inﬂuence of crucial design parameters such as the molecular structures of the substrate and the chiral auxiliary have not been explored yet, and a ﬁnal assessment of the potential of this alternative strategy requires further insight. Future research has to show whether this approach can be extended to other α,βunsaturated ketones and if structure−enantioselectivity relationships can be derived for this catalytic system based on simpler liquid-phase investigations.

4. CONCLUSIONS AND OUTLOOK Progress has been made in several aspects relevant for heterogeneous asymmetric hydrogenation, including the development of new catalysts and chiral modiﬁers, extension of the scope of substrates, structural characterization of adsorbed modiﬁers, understanding of the role of additives and solvents, and a deeper insight into the reaction mechanisms. Despite the promising, recent extension of the range of substrates of cinchona-modiﬁed Pd, the general knowledge about the asymmetric hydrogenation of activated ketones on chirally modiﬁed Pt catalysts is still more advanced. Both noble metals show only a few common features; the most prominent is that enantiodiscrimination occurs on the chirally modiﬁed noble metal and is controlled by hydrogen-bonding interaction(s) between coadsorbed chiral modiﬁer and prochiral substrate. However, the observed structure sensitivity with respect to the molecular structures of the chiral modiﬁer and the prochiral substrate, the solvent eﬀects, and other reaction parameters indicate major diﬀerences and unifying principles of the enantioselective surface processes on Pt and Pd are barely evident. On Pt, highly enantiospeciﬁc interactions have been identiﬁed, and the enantiodiﬀerentiating interaction may even diﬀer for two substrates from the same substrate class, e.g., the ester-activated ketone KPL forms two cooperative hydrogen-bonding interactions to CD for very eﬃcient enantiocontrol, while the triﬂuoro-activated ketone TFAP requires the acid additive TFA to reside at the quinuclidine N atom of CD for strong enantiocontrol. While the underlying surface processes of asymmetric hydrogenation on chirally modiﬁed noble metal catalysts have been signiﬁcantly better characterized, the current information about the enantiodiﬀerentiating step and, in particular, about the structure of the enantiodiﬀerentiating surface complex is still incomplete. Present progress in ﬁnding suitable modiﬁers and substrates still relies on an empirical and at best semiempirical approach, which could be made more eﬃcient by applying combinatorial methods. Regarding the design of modiﬁers, an interesting road to follow might be the application of structurally modiﬁed amino acids such as derivatives of 11561

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

ATR-IR

attenuated total reﬂection infrared (spectroscopy) BA benzylamine B/L bridged to linear [BMIm][PF6] 1 - b u t y l - 3 - m e t h y l i m i d a z o l i u m t r i s (pentaﬂuoroethyl) triﬂuorophosphate CNTs carbon nanotubes CD cinchonidine CN cinchonine de diastereomeric excess DFT density functional theory DHVIN dihydrovinpocetine DMF dimethylformamide dMePhOCD O-(3,5-dimethylphenyl)cinchonidine DMPCA α,β-para-dimethoxy phenyl cinnamic acid DPPA diphenyl propanoic acid ECD 9-epi-cinchonidine ee enantiomeric excess (%) = 100·|(R) − (S)|/ [(R) + (S)] EP ethyl pyruvate FMPCA α-ortho-ﬂuoro-β-para-methoxy phenyl cinnamic acid Gly glycine Gn nanoparticle graphene GO graphene oxide HCD 10,11-dihydrocinchonidine HHS half-hydrogenation state HREELS high-resolution electron energy loss spectroscopy ICN isocinchonine ILs ionic liquids IP isophorone IQN isoquinine KPL ketopantolactone LEED low-energy electron diﬀraction MBA 2-methyl butanoic acid MBF methyl benzoylformate MeoP 4-methoxy-6-methyl-2-pyrone MP methyl pyruvate MeOCD O-methyl cinchonidine MHA 2-methyl hexenoic acid MNEA N-methyl-1-(1-naphthyl) ethylamine MPA 2-methyl pentenoic acid MTFP methyl 3,3,3-triﬂuoropyruvate MWNT multiwall nanotube NaphOCD O-naphthyl cinchonidine NPs nanoparticles NEA naphthylethylamine NED 1-naphthyl-1,2-ethanediol PA methylglyoxal dimethylacetal PAA poly(acrylic acid) PCA phenyl cinnamic acid PNEA pantoylnaphthylethylamine PO propylene oxide PhOCD O-phenyl cinchonidine PPD 1-phenyl-1,2-propanodione PPh3 triphenylphosphine Pr proline PTFE 1-phenyl-2,2,2-triﬂuoroethanol PyIm (4′R,5′S)-4′,5′-dihydro-4′,5′-diphenyl-2-(6-cyanopyridyl) imidazoline QD quinidine QN quinine

Further important challenges are the extension of the application of heterogeneous asymmetric hydrogenation catalysis to CN bond hydrogenation, which has so far been rather unsuccessful,242 and ultimately to other asymmetric reactions. First steps in this direction have been reported for allylic substitutions243−245 and the hydroformylation of oleﬁns,246 but the development of such heterogeneous asymmetric reactions is still in its infancy. Broadening the scope of reactions will require new heterogeneous catalytic systems and deeper insight into the molecular surface processes occurring on chiral catalysts. In spite of the signiﬁcant diﬀerences between heterogeneous and homogeneous asymmetric catalysis, some aspects of the latter, such as the more profound understanding of molecular mechanisms and the extensive pool of suitable chiral ligands, could further help in fertilizing the future development of heterogeneous asymmetric catalysis.

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. ORCID

Fabian Meemken: 0000-0001-9983-0394 Alfons Baiker: 0000-0003-1408-464X Notes

The authors declare no competing ﬁnancial interest. Biographies Fabian Meemken received an M.Sc. in Chemical Engineering from the Georgia Institute of Technology, Atlanta, Georgia, U.S.A., in 2010, and a Diploma in Industrial Engineering from the Technical University Berlin, Germany, in 2011. After a six-month internship at Lonza AG, he started his doctoral studies at ETH Zürich. In 2014, he obtained his Ph.D. and assumed his current position as senior scientist at the Chemistry Department at ETH Zürich. His main research interests comprise heterogeneous catalysis in liquid phase, chemoselective and enantioselective hydrogenation, and the development of in situ and operando spectroscopy of catalytic solid−liquid interfaces. Alfons Baiker studied Chemical Engineering at ETH Zurich and earned his Ph.D. degree in 1974. After several postdoctoral stays at various universities, he ﬁnished his habilitation at Stanford University (California) and returned to ETH in 1980, where he started his own research group focusing on heterogeneous catalysis and reaction engineering. He moved up to the ranks to become Full Professor in 1990. His research interests, documented in over 900 publications in refereed journals and numerous patents, are centered on catalyst design and novel catalytic materials, mechanisms and kinetics of catalytic surface processes, asymmetric hydrogenation, selective oxidation, environmental catalysis, chiral surfaces, in situ and operando spectroscopy, and the application of supercritical ﬂuids and ionic liquids in catalysis.

ACKNOWLEDGMENTS Financial support by the Swiss National Science Foundation and the Foundation Claude & Giuliana is kindly acknowledged. A.B. thanks former and present co-workers for their invaluable contributions to this research. F.M. thanks Professor Konrad Hungerbühler (ETH Zürich) for supporting his research. ABBREVIATIONS AF amino-4,4-dimethyldihydrofuran-2-one 11562

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews RAIRS rt SCAC SO(4) STM TEM TFA TFBA TFAP THF tFPhOCD TMCH TMS-CD TOF TPD Trp v/v wt % UHV XPS

Review

(15) von Arx, M.; Mallat, T.; Baiker, A. Platinum-Catalyzed Enantioselective Hydrogenation of Aryl-Substituted Trifluoroacetophenones. Tetrahedron: Asymmetry 2001, 12, 3089−3094. (16) Kü n zle, N.; Szabo, A.; Schü r ch, M.; Wang, G. Z. Enantioselective Hydrogenation of a Cyclic Imidoketone over Chirally Modified Pt/Al2O3. Chem. Commun. 1998, 1377−1378. (17) Diezi, S.; Hess, M.; Orglmeister, E.; Mallat, T.; Baiker, A. An Efficient Synthetic Chiral Modifier for Platinum. Catal. Lett. 2005, 102, 121−125. (18) von Arx, M.; Mallat, T.; Baiker, A. Highly Efficient PlatinumCatalyzed Enantioselective Hydrogenation of Trifluoroacetoacetates in Acidic Solvents. Catal. Lett. 2002, 78, 267−271. (19) Orglmeister, E.; Mallat, T.; Baiker, A. Synthetic Modifiers for Platinum in the Enantioselective Hydrogenation of Ketopantolactone: A Test for the Mechanistic Models of Ketone Hydrogenation. Adv. Synth. Catal. 2005, 347, 78−86. (20) Schürch, M.; Heinz, T.; Aeschimann, R.; Mallat, T.; Pfaltz, A.; Baiker, A. Design of New Modifiers for the Enantioselective Hydrogenation of Ethyl Pyruvate. J. Catal. 1998, 173, 187−195. (21) Wehrli, J. T.; Baiker, A.; Monti, D. M.; Blaser, H. U.; Jalett, H. P. Enantioselective Hydrogenation of α-Ketoesters: Influence of Reaction Medium and Conversion. J. Mol. Catal. 1989, 57, 245−257. (22) Schmidt, E.; Vargas, A.; Mallat, T.; Baiker, A. Shape-Selective Enantioselective Hydrogenation on Pt Nanoparticles. J. Am. Chem. Soc. 2009, 131, 12358−12367. (23) Guan, S. L.; Donovan-Sheppard, O.; Reece, C.; Willock, D. J.; Wain, A. J.; Attard, G. A. Structure Sensitivity in Catalytic Hydrogenation at Platinum Surfaces Measured by Shell-Isolated Nanoparticle Enhanced Raman Spectroscopy (SHINERS). ACS Catal. 2016, 6, 1822−1832. (24) Hoxha, F.; Schmidt, E.; Mallat, T.; Schimmoeller, B.; Pratsinis, S. E.; Baiker, A. Hydrogenation of Acetophenone Derivatives: Tuning the Enantioselectivity via the Metal-Support Interaction. J. Catal. 2011, 278, 94−101. (25) Schmidt, E.; Hoxha, F.; Mallat, T.; Baiker, A. Metal-Support Interaction in Pt/Alumina: Inversion of Diastereoselectivity by Tuning the Acid-Base Properties of the Support. J. Catal. 2010, 274, 117−120. (26) Li, Q.; Zhang, X. Q.; Xiao, M. T.; Liu, Y. J. Alumina Incorporated with Mesoporous Carbon as a Novel Support of Pt Catalyst for Asymmetric Hydrogenation. Catal. Commun. 2013, 42, 68−72. (27) Chen, Z.; Guan, Z.; Li, M.; Yang, Q.; Li, C. Enhancement of the Performance of a Platinum Nanocatalyst Confined within Carbon Nanotubes for Asymmetric Hydrogenation. Angew. Chem., Int. Ed. 2011, 50, 4913−4917. (28) Guan, Z.; Lu, S.; Chen, Z.; Li, C. An Unexpected Effect of Water on the Asymmetric Hydrogenation of alpha-Ketoesters on Platinum Nanoparticles Confined in Carbon Nanotubes. J. Catal. 2013, 305, 19−26. (29) Guan, Z. H.; Lu, S. M.; Li, C. Highly Oxidized Pt Species Stabilized Inside Carbon Nanotubes for Asymmetric Hydrogenation. Chin. J. Catal. 2015, 36, 1535−1542. (30) Sharma, P.; Sharma, R. K. Platinum Functionalized Multiwall Carbon Nanotube Composites as Recyclable Catalyst for Highly Efficient Asymmetric Hydrogenation of Methyl Pyruvate. RSC Adv. 2015, 5, 102481−102487. (31) Sharma, P.; Sharma, R. K. Asymmetric Hydrogenation of αKetoesters on the Pt(111) Surface. New J. Chem. 2016, 40, 9038− 9041. (32) Noyori, R.; Ohkuma, T. Asymmetric Catalysis by Architectural and Functional Molecular Engineering: Practical Chemo- and Stereoselective Hydrogenation of Ketones. Angew. Chem., Int. Ed. 2001, 40, 40−73. (33) Garland, M.; Jalett, H. P.; Blaser, H. U. Mass Transfer Considerations for the Enantioselective Hydrogenation of αKetoesters Catalyzed by Cinchona Modified Pt/Al2O3. Stud. Surf. Sci. Catal. 1991, 59, 177−184.

reﬂection absorption infrared spectroscopy room temperature single-crystal adsorption calorimetry surface open(4) scanning tunneling microscopy transmission electron microscopy triﬂuoroacetic acid α,α,α-(triﬂuoromethyl) benzyl alcohol α,α,α-triﬂuoroacetophenone tetrahydrofuran O-[3,5-bis(triﬂuoromethyl)phenyl] cinchonidine 3,3,5-trimethylcyclohexanone 11-trimethoxysilyl cinchonidine turnover frequency temperature-programmed desorption tryptophan volume/volume weight percentage ultrahigh vacuum X-ray photoelectron spectroscopy

REFERENCES (1) Blaser, H. U.; Pugin, B.; Spindler, F.; Thommen, M. From a Chiral Switch to a Ligand Portfolio for Asymmetric Catalysis. Acc. Chem. Res. 2007, 40, 1240−1250. (2) Gellman, A. J.; Tysoe, W. T.; Zaera, F. Surface Chemistry for Enantioselective Catalysis. Catal. Lett. 2015, 145, 220−232. (3) Mallat, T.; Orglmeister, E.; Baiker, A. Asymmetric Catalysis at Chiral Metal Surfaces. Chem. Rev. 2007, 107, 4863−4890. (4) Heitbaum, M.; Glorius, F.; Escher, I. Asymmetric Heterogeneous Catalysis. Angew. Chem., Int. Ed. 2006, 45, 4732−4762. (5) Zaera, F. Chiral Modification of Solid Surfaces: A Molecular View. J. Phys. Chem. C 2008, 112, 16196−16203. (6) Kyriakou, G.; Beaumont, S. K.; Lambert, R. M. Aspects of Heterogeneous Enantioselective Catalysis by Metals. Langmuir 2011, 27, 9687−9695. (7) Bartok, M. Heterogeneous Catalytic Enantioselective Hydrogenation of Activated Ketones. Curr. Org. Chem. 2006, 10, 1533−1567. (8) Murzin, D. Y.; Maki-Arvela, P.; Toukoniitty, E.; Salmi, T. Asymmetric Heterogeneous Catalysis: Science and Engineering. Catal. Rev.: Sci. Eng. 2005, 47, 175−256. (9) Zhan, E. S.; Chen, C. H.; Li, Y.; Shen, W. J. Heterogeneous Asymmetric Hydrogenation over Chiral Molecule-Modified Metal Particles. Catal. Sci. Technol. 2015, 5, 650−659. (10) Baiker, A. Crucial Aspects in the Design of Chirally Modified Noble Metal Catalysts for Asymmetric Hydrogenation of Activated Ketones. Chem. Soc. Rev. 2015, 44, 7449−7464. (11) Sutyinszki, M.; Szö ri, K.; Felfö ldi, K.; Bartók, M. 98% Enantioselectivity in the Asymmetric Synthesis of a Useful Chiral Building Block by Heterogeneous Method: Enantioselective Hydrogenation of Ethyl-benzoylformate over Cinchona Modified Pt/Al2O3 Catalysts in the Acetic Acid. Catal. Commun. 2002, 3, 125−127. (12) Huck, W. M.; Mallat, T.; Baiker, A. Non-Linear Effect of Modifier Composition on Enantioselectivity in Asymmetric Hydrogenation over Platinum Metals. Adv. Synth. Catal. 2003, 345, 255−260. (13) Török, B.; Balázsik, K.; Bartók, M.; Felföldi, K.; Bartók, M. New Synthesis of a Useful C3 Chiral Building Block by a Heterogeneous Method: Enantioselective Hydrogenation of Pyruvaldehyde Dimethyl Acetal over Cinchona Modified Pt/Al2O3 Catalysts. Chem. Commun. 1999, 1725−1726. (14) Hoxha, F.; Schimmoeller, B.; Cakl, Z.; Urakawa, A.; Mallat, T.; Pratsinis, S. E.; Baiker, A. Influence of Support Acid-Base Properties on the Platinum-Catalyzed Enantioselective Hydrogenation of Activated Ketones. J. Catal. 2010, 271, 115−124. 11563

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

(34) Erathodiyil, N.; Gu, H.; Shao, H.; Jiang, J.; Ying, J. Y. Enantioselective Hydrogenation of α-Ketoesters over AlkaloidModified Platinum Nanowires. Green Chem. 2011, 13, 3070−3014. (35) Keilitz, J.; Nowag, S.; Marty, J. D.; Haag, R. Chirally Modified Platinum Nanoparticles Stabilized by Dendritic Core-Multishell Architectures for the Asymmetric Hydrogenation of Ethyl Pyruvate. Adv. Synth. Catal. 2010, 352, 1503−1511. (36) Beier, M. J.; Andanson, J. M.; Mallat, T.; Krumeich, F.; Baiker, A. Ionic Liquid-Supported Pt Nanoparticles as Catalysts for Enantioselective Hydrogenation. ACS Catal. 2012, 2, 337−340. (37) Blaser, H. U.; Jalett, H. P.; Muller, M.; Studer, M. Enantioselective Hydrogenation of α-Ketoesters using Cinchona Modified Platinum Catalysts and Related Systems: A Review. Catal. Today 1997, 37, 441−463. (38) Hong, J.; Lee, I.; Zaera, F. Cinchona Alkaloids Tethered on Porous Silica as Enantioselective Heterogeneous Catalysts. Top. Catal. 2011, 54, 1340−1347. (39) Hong, J.; Lee, I.; Zaera, F. Correlated Bifunctionality in Heterogeneous Catalysts: Selective Tethering of Cinchonidine Next to Supported Pt Nanoparticles. Catal. Sci. Technol. 2015, 5, 680−689. (40) Hong, J. H.; Zaera, F. Interference of the Surface of the Solid on the Performance of Tethered Molecular Catalysts. J. Am. Chem. Soc. 2012, 134, 13056−13065. (41) Campos, C. H.; Torres, C. C.; Osorio-Vargas, P.; Mella, C.; Belmar, J.; Ruiz, D.; Fierro, J. L. G.; Reyes, P. Immobilised Chiral Inducer on Pt-Based Mesoporous Titanate Nanotubes as Heterogeneous Catalysts for Enantioselective Hydrogenation. J. Mol. Catal. A: Chem. 2015, 398, 190−202. (42) Campos, C. H.; Torres, C. C.; Leyton, A.; Belmar, J.; Mella, C.; Osorio-Vargas, P.; Ruiz, D.; Fierro, J. L. G.; Reyes, P. A New NonCinchona Chiral Modifier Immobilized on Pt/SiO2 Catalysts for Enantioselective Heterogeneous Hydrogenation. Appl. Catal., A 2015, 498, 76−87. (43) Azmat, M. U.; Guo, Y.; Guo, Y.; Wang, Y. Q.; Lu, G. Z. An Easy and Effective Approach towards Heterogeneous Pt/SiO2-Cinchonidine Catalyst System for Enantioselective Hydrogenation of Ethyl Pyruvate. J. Mol. Catal. A: Chem. 2011, 336, 42. (44) Schrader, I.; Warneke, J.; Backenkohler, J.; Kunz, S. Functionalization of Platinum Nanoparticles with L-Proline: Simultaneous Enhancements of Catalytic Activity and Selectivity. J. Am. Chem. Soc. 2015, 137, 905−912. (45) Schrader, I.; Neumann, S.; Himstedt, R.; Zana, A.; Warneke, J.; Kunz, S. The Effect of Particle Size and Ligand Configuration on the Asymmetric Catalytic Properties of Proline-Functionalized Pt-Nanoparticles. Chem. Commun. 2015, 51, 16221−16224. (46) Schoenbaum, C. A.; Schwartz, D. K.; Medlin, J. W. Controlling the Surface Environment of Heterogeneous Catalysts Using SelfAssembled Monolayers. Acc. Chem. Res. 2014, 47, 1438−1445. (47) Kahsar, K. R.; Schwartz, D. K.; Medlin, J. W. Control of Metal Catalyst Selectivity through Specific Noncovalent Molecular Interactions. J. Am. Chem. Soc. 2014, 136, 520−526. (48) Hoxha, F.; van Vegten, N.; Urakawa, A.; Krumeich, F.; Mallat, T.; Baiker, A. Remarkable Particle Size Effect in Rh-Catalyzed Enantioselective Hydrogenations. J. Catal. 2009, 261, 224−231. (49) Dongil, A. B.; Bachiller-Baeza, B.; Rodriguez-Ramos, I.; Guerrero-Ruiz, A.; Mondelli, C.; Baiker, A. Structural Properties of Alumina- and Silica-Supported Iridium Catalysts and their Behavior in the Enantioselective Hydrogenation of Ethyl Pyruvate. Appl. Catal., A 2013, 451, 14−20. (50) Marzialetti, T.; Oportus, M.; Ruiz, D.; Fierro, J. L. G.; Reyes, P. Enantio- selective Hydrogenation of 1-Phenyl-1,2-propanedione, Ethyl Pyruvate and Acetophenone on Ir/SiO2 Catalysts. Effect of Iridium Loading. Catal. Today 2008, 133-135, 711−719. (51) Marzialetti, T.; Fierro, J. L. G.; Reyes, P. Iridium-Supported Catalyst for Enantioselective Hydrogenation of 1-Phenyl-1,2-propanedione: The Effects of the Addition of Promoter and the Modifier Concentration. Catal. Today 2005, 107−108, 235−243. (52) Jiang, H. Y.; Yang, C. F.; Li, C.; Fu, H. Y.; Chen, H.; Li, R. X.; Li, X. J. Heterogeneous Enantioselective Hydrogenation of Aromatic

Ketones Catalyzed by Cinchona- and Phosphine-Modified Iridium Catalysts. Angew. Chem., Int. Ed. 2008, 47, 9240−9244. (53) Ye, L. M.; Lin, H. Q.; Zhou, H. C.; Yuan, Y. Z. Support and Size Effects of Ruthenium Catalysts with a Chiral Modifier for Asymmetric Hydrogenation of Aromatic Ketones. J. Phys. Chem. C 2010, 114, 19752−19760. (54) Orito, Y.; Imai, S.; Niwa, S.; Hung, N.-G. Asymmetric Hydrogenation of Methyl Benzoylformate using Platinum-Carbon Catalysts Modified with Cinchonidine. Yuki Gosei Kagaku Kyokaishi 1979, 37, 173−174. (55) Orito, Y.; Imai, S.; Niwa, S. Asymmetric Hydrogenation of Methyl Pyruvate using Platinum-Carbon Catalyst Modified with Cinchonidine. Nippon Kagaku Kaishi 1979, 1118−1120. (56) Studer, M.; Blaser, H. U.; Exner, C. Enantioselective Hydrogenation using Heterogeneous Modified Catalysts: An Update. Adv. Synth. Catal. 2003, 345, 45−65. (57) Urakawa, A.; Meier, D. M.; Rü egger, H.; Baiker, A. Conformational Behavior of Cinchonidine Revisited: A Combined Theoretical and Experimental Study. J. Phys. Chem. A 2008, 112, 7250−7255. (58) Ferri, D.; Bürgi, T.; Baiker, A. Molecular Interaction between Cinchonidine and Acetic Acid Studied by NMR, FTIR and Ab Initio Methods. J. Chem. Soc., Perkin Trans. 2 1999, 7, 1305−131. (59) Olsen, R. A.; Borchardt, D.; Mink, L.; Agarwal, A.; Mueller, L. J.; Zaera, F. Effect of Protonation on the Conformation of Cinchonidine. J. Am. Chem. Soc. 2006, 128, 15594−15595. (60) Bürgi, T.; Baiker, A. Conformational Behavior of Cinchonidine in Different Solvents: A Combined NMR and Ab Initio Investigation. J. Am. Chem. Soc. 1998, 120, 12920−12926. (61) Lai, J.; Ma, Z.; Mink, L.; Mueller, L. J.; Zaera, F. Influence of Peripheral Groups on the Physical and Chemical Behavior of Cinchona Alkaloids. J. Phys. Chem. B 2009, 113, 11696−11701. (62) Cserényi, S.; Felföldi, K.; Balázsik, K.; Szöllösi, G.; Bucsi, I.; Bartók, M. C9-O-Substituted Derivatives of Cinchona Alkaloids as Chiral Modifiers in the Orito Reaction: Effects of Structure of Modifiers on Sense of Enantioselectivity. J. Mol. Catal. A: Chem. 2006, 247, 108−115. (63) Bucsi, I.; Cserényi, S.; Felföldi, K.; Bartók, M. New Chiral Catalysts: Synthesis and Fragmentation Pattern of C9-O-Silanized Cinchonidines. React. Kinet. Catal. Lett. 2006, 87, 281−289. (64) Balázsik, K.; Bucsi, I.; Cserényi, S.; Szöllösi, G.; Bartók, M. Methylethers of Cinchona Alkaloids in Pt-Catalyzed Hydrogenation of Ethyl Pyruvate and Ketopantolactone: Effect of Stereochemical Factors on the Enantioselectivity. J. Mol. Catal. A: Chem. 2008, 280, 87−95. (65) Vargas, A.; Ferri, D.; Bonalumi, N.; Mallat, T.; Baiker, A. Controlling the Sense of Enantioselection on Surfaces by Conformational Changes of Adsorbed Modifiers. Angew. Chem., Int. Ed. 2007, 46, 3905−3908. (66) Mondelli, C.; Bucher, C.; Baiker, A.; Gilmour, R. A Novel Class of Fluorinated Cinchona Alkaloids as Surface Modifiers for the Enantioselective Heterogeneous Hydrogenation of α-Ketoesters. J. Mol. Catal. A: Chem. 2010, 327, 87−91. (67) Busygin, I.; Warna, J.; Toukoniitty, E.; Murzin, D. Y.; Leino, R. Hydrogenation of 1-Phenyl-1,2-propanedione over Pt Catalysts Modified by Cinchona Alkaloid O-Ethers and the Kinetic Resolution of the 1-Hydroxyketones Generated. J. Catal. 2008, 254, 339−348. (68) Busygin, I.; Nieminen, V.; Taskinen, A.; Sinkkonen, J.; Toukoniitty, E.; Sillanpaa, R.; Murzin, D. Y.; Leino, R. A Combined NMR, DFT, and X-ray Investigation of Some Cinchona Alkaloid OEthers. J. Org. Chem. 2008, 73, 6559−6569. (69) Schmidt, E.; Bucher, C.; Santarossa, G.; Mallat, T.; Gilmour, R.; Baiker, A. Fundamental Insights into the Enantioselectivity of Hydrogenations on Cinchona-Modified Platinum and Palladium. J. Catal. 2012, 289, 238−248. (70) Balázsik, K.; Martinek, T. A.; Bucsi, I.; Szöllösi, G.; Fogassy, G.; Bartók, M.; Olah, G. A. A New Rigid Cinchona-Modified (α-IQ) Platinum Catalyst for the Enantioselective Hydrogenation of Activated 11564

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Ketones: Data to the Origin of Enantioselection. J. Mol. Catal. A: Chem. 2007, 272, 265−274. (71) Bucsi, I.; Sutyinszki, M.; Felföldi, K.; Bartók, M. Hydrogenation of β-Isocinchonidine in Mild Conditions on Pt and Pd Catalysts using HPLC-ESI-Ion-trap MS: New Results on the Role of Structure of Cinchona Alkaloids in the Orito Reaction. Catal. Commun. 2006, 7, 104−108. (72) Margitfalvi, J. L.; Tálas, E. Anomalous Behavior of Rigid Cinchona Alkaloids in the Enantioselective Hydrogenation of Ethyl Pyruvate in an Aprotic Solvent. Appl. Catal., A 2006, 301, 187−195. (73) Diezi, S.; Mallat, T.; Szabo, A.; Baiker, A. Fine Tuning the “Chiral Sites” on Solid Enantioselective Catalysts. J. Catal. 2004, 228, 162−173. (74) Holland, M. C.; Meemken, F.; Baiker, A.; Gilmour, R. Chiral Imidazolidinone and Proline-Derived Surface Modifiers for the PtCatalysed Asymmetric Hydrogenation of Activated Ketones. J. Mol. Catal. A: Chem. 2015, 396, 335−345. (75) Mondelli, C.; Vargas, A.; Santarossa, G.; Baiker, A. Fundamental Aspects of the Chiral Modification of Platinum with Peptides: Asymmetric Induction in Hydrogenation of Activated Ketones. J. Phys. Chem. C 2009, 113, 15246−15259. (76) Maris, M.; Ferri, D.; Konigsmann, L.; Mallat, T.; Baiker, A. Why are α-Hydroxycarboxylic Acids Poor Chiral Modifiers for Pt in the Hydrogenation of Ketones? J. Catal. 2006, 237, 230−236. (77) Meemken, F.; Baiker, A.; Schenker, S.; Hungerbühler, K. Chiral Modification of Platinum by Co-Adsorbed Cinchonidine and Trifluoroacetic Acid: Origin of Enhanced Stereocontrol in the Hydrogenation of Trifluoroacetophenone. Chem. - Eur. J. 2014, 20, 1298−1309. (78) Schmidt, E.; Kleist, W.; Krumeich, F.; Mallat, T.; Baiker, A. Platinum Nanoparticles: The Crucial Role of Crystal Face and Colloid Stabilizer in the Diastereoselective Hydrogenation of Cinchonidine. Chem. - Eur. J. 2010, 16, 2181−2192. (79) Sano, S.; Beier, M. J.; Mallat, T.; Baiker, A. Potential of Ionic Liquids as Co-Modifiers in Asymmetric Hydrogenation on Platinum. J. Mol. Catal. A: Chem. 2012, 357, 117−124. (80) Tálas, E.; Zsila, F.; Szabó, P.; Margitfalvi, J. L. Influence of Achiral Amine Additives on the Orito’s Reaction. J. Mol. Catal. A: Chem. 2012, 357, 87−94. (81) Szöri, K.; Balázsik, K.; Cserényi, S.; Szöllösi, G.; Bartók, M. Inversion of Enantioselectivity in the 2,2,2-Trifluoroacetophenone Hydrogenation over Pt-Alumina Catalyst Modified by Cinchona Alkaloids. Appl. Catal., A 2009, 362, 178−184. (82) Meier, D. M.; Mallat, T.; Ferri, D.; Baiker, A. Competition of Chiral Modifiers on Platinum: A Transient Catalytic and in situ ATRIR Study in Continuous Reactors. J. Catal. 2006, 244, 260−263. (83) Meier, D. M.; Ferri, D.; Mallat, T.; Baiker, A. Molecular Insight into the Dynamics of Chiral Modification of Pt/Alumina. J. Catal. 2007, 248, 68−76. (84) Bonalumi, N.; Vargas, A.; Ferri, D.; Baiker, A. Catalytic Chiral Metal Surfaces Generated by Adsorption of O-Phenyl Derivatives of Cinchonidine. J. Phys. Chem. C 2007, 111, 9349−9358. (85) Schmidt, E.; Ferri, D.; Vargas, A.; Baiker, A. Chiral Modification of Rh and Pt Surfaces: Effect of Rotational Flexibility of CinchonaType Modifiers on their Adsorption Behavior. J. Phys. Chem. C 2008, 112, 3866−3874. (86) Motobayashi, K.; Tomioka, R.; Uchida, T.; Osawa, M. Effect of Hydrogen on the Orientation of Cinchonidine Adsorbed on Platinum: An ATR-SEIRAS Study. Chem. Lett. 2015, 44, 770−772. (87) Rodriguez-Garcia, L.; Hungerbuhler, K.; Baiker, A.; Meemken, F. Discrimination of Active Species in Liquid-Phase Hydrogenation on Supported Noble Metal Catalyst: An Operando Spectroscopic Study on the Asymmetric Hydrogenation of Ketopantolactone on Pt/Al2O3 and Pt/C Modified by Cinchonidine. Catal. Today 2017, 283, 66−73. (88) Ma, Z.; Lee, I.; Zaera, F. Factors Controlling Adsorption Equilibria from Solution onto Solid Surfaces: The Uptake of Cinchona Alkaloids on Platinum Surfaces. J. Am. Chem. Soc. 2007, 129, 16083− 1690.

(89) Ma, Z.; Zaera, F. Role of the Solvent in the AdsorptionDesorption Equilibrium of Cinchona Alkaloids between Solution and a Platinum Surface: Correlations Among Solvent Polarity, Cinchona Solubility, and Catalytic Performance. J. Phys. Chem. B 2005, 109, 406−414. (90) Vargas, A.; Baiker, A. First Principles Study of the Conformations of Cinchonidine on a Pt(111) Surface. J. Catal. 2006, 239, 220−226. (91) Hahn, K. R.; Seitsonen, A. P.; Baiker, A. Chiral Modification of Platinum: Ab Initio Study of the Effect of Hydrogen Coadsorption on Stability and Geometry of Adsorbed Cinchona Alkaloids. Phys. Chem. Chem. Phys. 2015, 17, 27615−27629. (92) Lee, I.; Ma, Z.; Kaneko, S.; Zaera, F. 1-(1-Naphthyl) Ethylamine Adsorption on Platinum Surfaces: On the Mechanism of Chiral Modification in Catalysis. J. Am. Chem. Soc. 2008, 130, 14597−14604. (93) Gordon, A. D.; Zaera, F. Adsorption of 1-(1-Naphthyl) ethylamine from Solution onto Platinum Surfaces: Implications for the Chiral Modification of Heterogeneous Catalysts. Angew. Chem., Int. Ed. 2013, 52, 3453−3456. (94) Meemken, F.; Steiger, T.; Holland, M. C.; Gilmour, R.; Hungerbuhler, K.; Baiker, A. Adsorption and Stability of Chiral Modifiers Based on 1-(1-Naphthyl)-ethylamine for Pt Catalysed Heterogeneous Asymmetric Hydrogenations. Catal. Sci. Technol. 2015, 5, 705−715. (95) Hahn, K. R.; Baiker, A. Coadsorption of Cinchona Alkaloids and Oxygen on Pt(111): A Theoretical Study. J. Phys. Chem. C 2016, 120, 20170−20180. (96) Rodríguez-García, L.; Hungerbühler, K.; Baiker, A.; Meemken, F. The Critical Role of Tilted Cinchona Surface Species for Enantioselective Hydrogenation. ACS Catal. 2017, 7, 3799−3809. (97) Meemken, F.; Maeda, N.; Hungerbuhler, K.; Baiker, A. Platinum-Catalyzed Asymmetric Hydrogenation: Spectroscopic Evidence for an O-H-O Hydrogen-Bond Interaction between Substrate and Modifier. Angew. Chem., Int. Ed. 2012, 51, 8212−8216. (98) Lavoie, S.; Laliberté, M.-A.; Temprano, I.; McBreen, P. H. A Generalized Two-point H-Bonding Model for Catalytic Stereoselective Hydrogenation of Activated Ketones on Chirally Modified Platinum. J. Am. Chem. Soc. 2006, 128, 7588−7593. (99) Maeda, N.; Hungerbuhler, K.; Baiker, A. Asymmetric Hydrogenation on Chirally Modified Pt: Origin of Hydrogen in the N−H−O Interaction between Cinchonidine and Ketone. J. Am. Chem. Soc. 2011, 133, 19567−19569. (100) Kubota, J.; Zaera, F. Adsorption Geometry of Modifiers as Key in Imparting Chirality to Platinum Catalysts. J. Am. Chem. Soc. 2001, 123, 11115−11116. (101) Schmidt, E.; Mallat, T.; Baiker, A. Substrate-Controlled Adsorption of Cinchonidine during Enantioselective Hydrogenation on Platinum. J. Catal. 2010, 272, 140−150. (102) Behzadi, B.; Vargas, A.; Ferri, D.; Ernst, K. H.; Baiker, A. Cinchonidine Adsorption on Gold and Gold-Containing Bimetallic Platinum Metal Surfaces: An Attenuated Total Reflection Infrared and Density Functional Theory Study. J. Phys. Chem. B 2006, 110, 17082− 17089. (103) Bonalumi, N.; Vargas, A.; Ferri, D.; Baiker, A. Chirally Modified Platinum Generated by Adsorption of Cinchonidine Ether Derivatives: Towards Uncovering the Chiral Sites. Chem. - Eur. J. 2007, 13, 9236−9244. (104) Demers-Carpentier, V.; Goubert, G.; Masini, F.; LafleurLambert, R.; Dong, Y.; Lavoie, S.; Mahieu, G.; Boukouvalas, J.; Gao, H. L.; Rasmussen, A. M. H.; et al. Direct Observation of Molecular Preorganization for Chirality Transfer on a Catalyst Surface. Science 2011, 334, 776−780. (105) Svane, K.; Dong, Y.; Groves, M. N.; Demers-Carpentier, V.; Lemay, J.-C.; Ouellet, M.; Hammer, B.; McBreen, P. H. Single-ChiralCatalytic-Surface-Sites: STM and DFT Study of Stereodirecting Complexes Formed between (R)-1-(1-Naphthyl)ethylamine and Ketopantolactone on Pt(111). Catal. Sci. Technol. 2015, 5, 743−753. (106) Ferri, D.; Bürgi, T. An in Situ Attenuated Total Reflection Infrared Study of a Chiral Catalytic Solid−Liquid Interface: 11565

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Cinchonidine Adsorption on Pt. J. Am. Chem. Soc. 2001, 123, 12074− 12084. (107) Santarossa, G.; Iannuzzi, M.; Vargas, A.; Baiker, A. Adsorption of Naphthalene and Quinoline on Pt, Pd and Rh: A DFT Study. ChemPhysChem 2008, 9, 401−413. (108) Goubert, G.; Demers-Carpentier, V.; Loach, R. P.; LafleurLambert, R.; Lemay, J.-C.; Boukouvalas, J.; McBreen, P. H. Aminolactone Chiral Modifiers for Heterogeneous Asymmetric Hydrogenation: Corrected Structure of Pantoyl-Naphthylethylamine, In-Situ Hydrogenolysis, and Scanning Tunneling Microscopy Observation of Supramolecular Aminolactone/Substrate Assemblies on Pt(111). ACS Catal. 2013, 3, 2677−2683. (109) Pereniguez, R.; Santarossa, G.; Mallat, T.; Baiker, A. Hydrogenation of 2,2,2-Trifluoroacetophenone: Molecular Insight into the Role of Solvent in Enantioselection. J. Mol. Catal. A: Chem. 2012, 365, 39−49. (110) Martin, G.; Maki-Arvela, P.; Murzin, D. Y.; Salmi, T. Solvent Effects in the Enantioselective Hydrogenation of Ethyl Benzoylformate. Catal. Lett. 2013, 143, 1051−1060. (111) Wang, J.; LeBlond, C.; Orella, C. F.; Sun, Y.; Bradley, J. S.; Blackmond, D. G. Modelling of Kinetically Coupled Selective Hydrogenation: Kinetic Rationalization of Pressure Effects on Enantioselectivity. Stud. Surf. Sci. Catal. 1997, 108, 183−190. (112) Meemken, F.; Hungerbühler, K.; Baiker, A. Monitoring Surface Processes During Heterogeneous Asymmetric Hydrogenation of Ketones on a Chirally Modified Platinum Catalyst by Operando Spectroscopy. Angew. Chem., Int. Ed. 2014, 53, 8640−8644. (113) Cakl, Z.; Reimann, S.; Schmidt, E.; Moreno, A.; Mallat, T.; Baiker, A. Multiple Cycle Reaction Mechanism in the Enantioselective Hydrogenation of α,α,α-Trifluoromethyl Ketones. J. Catal. 2011, 280, 104−115. (114) Meemken, F.; Baiker, A.; Dupré, J.; Hungerbühler, K. Asymmetric Catalysis on Cinchonidine-Modified Pt/Al2O3: Kinetics and Isotope Effect in the Hydrogenation of Trifluoroacetophenone. ACS Catal. 2014, 4, 344−354. (115) Baiker, A. Progress in Asymmetric Heterogeneous Catalysis: Design of Novel Chirally Modified Platinum Metal Catalysts. J. Mol. Catal. A: Chem. 1997, 115, 473−493. (116) Vayner, G.; Houk, K. N.; Sun, Y. K. Origins of Enantioselectivity in Reductions of Ketones on Cinchona Alkaloid Modified Platinum. J. Am. Chem. Soc. 2004, 126, 199−203. (117) Felföldi, K.; Varga, T.; Forgó, P.; Bartók, M. Enantioselective Hydrogenation of Trifluoromethylcyclohexyl Ketone on Cinchona Alkaloid Modified Pt/Alumina Catalyst. Catal. Lett. 2004, 97, 65−70. (118) Bonalumi, N.; Bürgi, T.; Baiker, A. Interaction between Ketopantolactone and Chirally Modified Pt Investigated by Attenuated Total Reflection IR Concentration Modulation Spectroscopy. J. Am. Chem. Soc. 2003, 125, 13342−13343. (119) Orglmeister, E.; Mallat, T.; Baiker, A. Zwitterion Formation: A Feasible Mechanism for the Pt-Catalyzed Enantioselective Hydrogenation of Ketones? J. Catal. 2005, 234, 242−246. (120) Martinek, T. A.; Varga, T.; Fülöp, F.; Bartók, M. NMR Spectroscopic and Theoretical Evidence of Cinchona AlkaloidKetopantolactone Complex Formation in Aprotic Solvents: Implications for the Mechanism of Pt-Catalyzed Enantioselective Hydrogenation of Activated Ketones. J. Catal. 2007, 246, 266−276. (121) Martinek, T. A.; Varga, T.; Balázsik, K.; Szöllösi, G.; Fülöp, F.; Bartók, M. Enantioselective Hydrogenation of Ketopantolactone using Pt−β-ICN Chiral Catalyst: Correlation between the Solution-State Concentration of a Nucleophilic β-Isocinchonine−Ketopantolactone Complex and Enantioselectivity. J. Catal. 2008, 255, 296−303. (122) Vargas, A.; Santarossa, G.; Iannuzzi, M.; Baiker, A. Chiral Recognition on Catalytic Surfaces: Theoretical Insight in a Biomimetic Heterogeneous Catalytic System. J. Phys. Chem. C 2008, 112, 10200− 10208. (123) Groves, M. N.; Goubert, G.; Rasmussen, A. M. H.; Dong, Y.; Lemay, J. C.; Demers-Carpentier, V.; McBreen, P. H.; Hammer, B. Structure Determination of Chemisorbed Chirality Transfer Com-

plexes: Accelerated STM Analysis and Exchange-Correlation Functional Sensitivity. Surf. Sci. 2014, 629, 48−56. (124) Demers-Carpentier, V.; Rasmussen, A. M. H.; Goubert, G.; Ferrighi, L.; Dong, Y.; Lemay, J. C.; Masini, F.; Zeng, Y.; Hammer, B.; McBreen, P. H. Stereodirection of an α-Ketoester at Submolecular Sites on Chirally Modified Pt(111): Heterogeneous Asymmetric Catalysis. J. Am. Chem. Soc. 2013, 135, 9999−10002. (125) Szöllösi, G.; Cserényi, S.; Fülöp, F.; Bartók, M. New Data to the Origin of Rate Enhancement on the Pt-Cinchona Catalyzed Enantioselective Hydrogenation of Activated Ketones using Continuous Flow Fixed-Bed Reactor System. J. Catal. 2008, 260, 245−253. (126) Tálas, E.; Margitfalvi, J. L.; Egyed, O. Additional Data to the Origin of Rate Enhancement in the Enantioselective Hydrogenation of Activated Ketones over Cinchonidine Modified Platinum Catalyst. J. Catal. 2009, 266, 191−198. (127) Szöllösi, G.; Cserényi, S.; Balázsik, K.; Fülöp, F.; Bartók, M. New Data in the Enantioselective Hydrogenation of Ethyl Pyruvate on Pt-Cinchon Origin of Rate Enhancement. J. Mol. Catal. A: Chem. 2009, 305, 155−160. (128) Szöllösi, G.; Cserényi, S.; Bucsi, I.; Bartók, T.; Fülöp, F.; Bartók, M. Origin of the Rate Enhancement and Enantiodifferentiation in the Heterogeneous Enantioselective Hydrogenation of 2,2,2Trifluoroacetophenone over Pt/Alumina Studied in Continuous Flow Fixed-Bed Reactor System. Appl. Catal., A 2010, 382, 263−271. (129) Murzin, D. Y. Quantification of Rate Acceleration in Asymmetric Catalytic Hydrogenation. J. Mol. Catal. A: Chem. 2008, 289, 91−94. (130) Toukoniitty, B.; Warna, J.; Mikkola, J. P.; Helle, M.; Saxen, H.; Murzin, D. Y.; Salmi, T. Modelling of Enantioselective and Racemic Hydrogenation of Ethyl Pyruvate on a Pt/Al2O3 Catalyst in the Presence of Microwave Irradiation. Chem. Eng. Process. 2009, 48, 837− 845. (131) Toukoniitty, E.; Warna, J.; Murzin, D. Y.; Salmi, T. Modelling of Transient Kinetics in Catalytic Three-Phase Reactors: Enantioselective Hydrogenation. Chem. Eng. Sci. 2010, 65, 1076−1087. (132) Balázsik, K.; Szö ri, K.; Szö llö si, G.; Bartók, M. New Phenomenon in Competitive Hydrogenation of Binary Mixtures of Activated Ketones over Unmodified and Cinchonidine-Modified Pt/ Alumina Catalyst. Catal. Commun. 2011, 12, 1410−1414. (133) Varga, T.; Felföldi, K.; Forgó, P.; Bartók, M. Heterogeneous Asymmetric Reactions: Part 38. Enantioselective Hydrogenation of Fluoroketones on Pt−Alumina Catalyst. J. Mol. Catal. A: Chem. 2004, 216, 181−187. (134) Mahapatra, M.; Burkholder, L.; Garvey, M.; Bai, Y.; Saldin, D. K.; Tysoe, W. T. Enhanced Hydrogenation Activity and Diastereomeric Interactions of Methyl Pyruvate Coadsorbed with R-1-(1Naphthyl) Ethylamine on Pd(111). Nat. Commun. 2016, 7, No. 12380. (135) Huck, W. R.; Bürgi, T.; Mallat, T.; Baiker, A. Asymmetric Hydrogenation on Platinum: Nonlinear Effect of Coadsorbed Cinchona Alkaloids on Enantiodifferentiation. J. Catal. 2003, 216, 276−287. (136) Huck, W.-R.; Mallat, T.; Baiker, A. Coadsorption of Cinchona Alkaloids on Supported Palladium: Nonlinear Effects in Asymmetric Hydrogenation and Resistance of Alkaloids Against Hydrogenation. Catal. Lett. 2003, 87, 241−247. (137) Diezi, S.; Szabo, A.; Mallat, T.; Baiker, A. Inversion of Enantioselectivity in the Hydrogenation of Ketopantolactone on Platinum Modified by Ether Derivatives of Cinchonidine. Tetrahedron: Asymmetry 2003, 14, 2573−2577. (138) Bonalumi, N.; Vargas, A.; Ferri, D.; Bürgi, T.; Mallat, T.; Baiker, A. Competition at Chiral Metal Surfaces: Fundamental Aspects of the Inversion of Enantioselectivity in Hydrogenations on Platinum. J. Am. Chem. Soc. 2005, 127, 8467−8477. (139) Balazs, L.; Mallat, T.; Baiker, A. Nonlinear Phenomenon in Heterogeneous Enantioselective Catalysis. J. Catal. 2005, 233, 327− 332. (140) Hoxha, F.; Mallat, T.; Baiker, A. Hydrogenation of αKetoesters and Ketoplantolactone on Rhodium Modified by Cinchona and Isocinchona Alkaloids. J. Catal. 2007, 248, 11−19. 11566

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

(141) Balázsik, K.; Szöllösi, G.; Bartók, M. New Data of Nonlinear Phenomenon in the Heterogeneous Enantioselective Hydrogenation of Activated Ketones. Catal. Lett. 2008, 124, 46−51. (142) Balázsik, K.; Cserényi, S.; Szöllösi, G.; Fülöp, F.; Bartók, M. New Data on the Orito Reaction: Effect of Substrate Structure on Nonlinear Phenomenon. Catal. Lett. 2008, 125, 401−407. (143) Bartók, M.; Balázsik, K.; Bucsi, I.; Szöllösi, G. Inversion of Enantioselectivity in the Hydrogenation of Ketopantolactone on a Ptβ-ICN Chiral Catalyst. J. Catal. 2006, 239, 74−82. (144) Szöllösi, G.; Balázsik, K.; Bucsi, I.; Bartók, T.; Bartók, M. Modifier−Substrate Interactions of Various Types in the Orito Reaction: Reversal of the Enantioselection in the Hydrogenation of Ketopantolactone on Pt Modified by β-isocinchonine and Ophenylcinchonidine. Catal. Commun. 2013, 32, 81−85. (145) Szöllösi, G.; Cserényi, S.; Bartók, M. Novel Evidence on the Role of the Nucleophilic Intermediate Complex in the Orito-Reaction: Unexpected Inversion in the Enantioselective Hydrogenation of 2,2,2Trifluoroacetophenone on Pt-Cinchona Chiral Catalyst Using Continuous-Flow Fixed-Bed Reactor. Catal. Lett. 2010, 134, 264−269. (146) Bartók, M. Unexpected Inversions in Asymmetric Reactions: Reactions with Chiral Metal Complexes, Chiral Organocatalysts, and Heterogeneous Chiral Catalysts. Chem. Rev. 2010, 110, 1663−1705. (147) Rodriguez, J. A.; Hanson, J. C.; Chupas, P. J. In-situ Characterization of Heterogeneous Catalysts; Wiley: Hoboken, NJ, 2013. (148) Ferri, D.; Bürgi, T.; Baiker, A. In situ ATR-IR Study of the Adsorption of Cinchonidine on Pd/Al2O3: Differences and Similarities with Adsorption on Pt/Al2O3. J. Catal. 2002, 210, 160−170. (149) Tan, S.; Sun, X.; Williams, C. T. In situ ATR-IR Study of Prochiral 2-Methyl-2-pentenoic Acid Adsorption on Al2O3 and Pd/ Al2O3. Phys. Chem. Chem. Phys. 2011, 13, 19573−19579. (150) Tan, S.; Williams, C. T. An In Situ Spectroscopic Study of Prochiral Reactant−Chiral Modifier Interactions on Palladium Catalyst: Case of Alkenoic Acid and Cinchonidine in Various Solvents. J. Phys. Chem. C 2013, 117, 18043−18052. (151) Lavoie, S.; Laliberté, M.-A.; Mahieu, G.; Demers-Carpentier, V.; McBreen, P. Keto-Enol Driven Assembly of Methyl Pyruvate on Pt(111). J. Am. Chem. Soc. 2007, 129, 11668−11669. (152) Bonello, J. M.; Williams, F. J.; Santra, A. K.; Lambert, R. M. Fundamental Aspects of Enantioselective Heterogeneous Catalysis: The Surface Chemistry of Methyl Pyruvate on Pt{111}. J. Phys. Chem. B 2000, 104, 9696−9703. (153) Margitfalvi, J. L.; Minder, B.; Tálas, E.; Botz, L.; Baiker, A. Enantioselective Hydrogenation of α-Keto Esters over Pt/Al2O3 Catalyst: Kinetic Aspects of the Rate Acceleration Effect Induced by Addition of Cinchonidine. Stud. Surf. Sci. Catal. 1993, 75, 2471−2474. (154) Liu, Z. M.; Li, X. H.; Ying, P. L.; Feng, Z. C.; Li, C. Fourier Transform Infrared Spectroscopic Study on the Adsorption of Ethyl Pyruvate on Pt/Al2O3: Side Reactions Suppressed by Adsorbed Hydrogen and Cinchonidine. J. Phys. Chem. C 2007, 111, 823−829. (155) Laliberté, M. A.; Lavoie, S.; Hammer, B.; Mahieu, G.; McBreen, P. H. Activation in Prochiral Reaction Assemblies on Pt(111). J. Am. Chem. Soc. 2008, 130, 5386−5387. (156) Demers-Carpentier, V.; Laliberté, M. A.; Pan, Y. X.; Mahieu, G.; Lavoie, S.; Goubert, G.; Hammer, B.; McBreen, P. H. Tuning Aryl−CH···O Intermolecular Interactions on Pt(111). J. Phys. Chem. C 2011, 115, 1355−1360. (157) Demers-Carpentier, V.; Laliberté, M.-A.; Lavoie, S.; Mahieu, G.; McBreen, P. H. Two-Dimensional Self-Assembly and Catalytic Function: Conversion of Chiral Alcohols into Self-Assembled Enols on Pt(111). J. Phys. Chem. C 2010, 114, 7291−7298. (158) Brunelle, J.; Demers-Carpentier, V.; Lafleur-Lambert, R.; Mahieu, G.; Goubert, G.; Lavoie, S.; McBreen, P. H. Disrupting ArylCH···O Interactions on Pt(111) Through the Coadsorption of Trifluoroacetic Acid and 2,2,2-Trifluoroacetophenone (TFAP): Inhibition of Competing Processes in Heterogeneous Asymmetric Catalysis. Top. Catal. 2011, 54, 1334−1339. (159) Wahl, M.; von Arx, M.; Jung, T. A.; Baiker, A. Time-Lapse STM Studies of Diastereomeric Cinchona Alkaloids on Platinum Metals. J. Phys. Chem. B 2006, 110, 21777−21782.

(160) von Arx, M.; Wahl, M.; Jung, T. A.; Baiker, A. Adsorption and Surface Mobility of Cinchonidine on Pt(111) Studied by STM. Phys. Chem. Chem. Phys. 2005, 7, 273−277. (161) Demers-Carpentier, V.; Goubert, G.; Masini, F.; LafleurLambert, R.; Dong, Y.; Lavoie, S.; Mahieu, G.; Boukouvalas, J.; Gao, H.; Rasmussen, A. M. H.; et al. Direct Observation of Molecular Preorganization for Chirality Transfer on a Catalyst Surface. Science 2011, 334, 776−780. (162) Demers-Carpentier, V.; Goubert, G.; Masini, F.; Dong, Y.; Rasmussen, A. M. H.; Hammer, B.; McBreen, P. H. Scanning Tunneling Microscopy Measurements of the Full Cycle of a Heterogeneous Asymmetric Hydrogenation Reaction on Chirally Modified Pt(111). J. Phys. Chem. Lett. 2012, 3, 92−96. (163) Goubert, G.; Rasmussen, A. M. H.; Dong, Y.; Groves, M. N.; McBreen, P. H.; Hammer, B. Walking-Like Diffusion of Two-Footed Asymmetric Aromatic Adsorbates on Pt(111). Surf. Sci. 2014, 629, 123−131. (164) Rasmussen, A. M. H.; Groves, M. N.; Hammer, B. Remote Activation of Chemical Bonds in Heterogeneous Catalysis. ACS Catal. 2014, 4, 1182−1188. (165) Lemay, J. C.; Dong, Y.; Groves, M. N.; Demers-Carpentier, V.; Goubert, G.; Lafleur-Lambert, R.; Boukouvalas, J.; Hammer, B.; McBreen, P. H. A Comparative Study of Diastereomeric Complexes Formed by a Prochiral Substrate and Three Structurally Analogous Chiral Molecules on Pt(111). Surf. Sci. 2016, 646, 13−18. (166) Goubert, G.; McBreen, P. H. Surface Diastereomeric Complexes Formed by Methyl Benzoylformate and (R)-1-(1Naphthyl)ethylamine on Pt(111). ACS Catal. 2014, 4, 847−854. (167) Dementyev, P.; Peter, M.; Adamovsky, S.; Schauermann, S. Chirally-Modified Metal Surfaces: Energetics of Interaction with Chiral Molecules. Phys. Chem. Chem. Phys. 2015, 17, 22726−22735. (168) Boscoboinik, J. A.; Bai, Y.; Burkholder, L.; Tysoe, W. T. Structure and Distribution of (S)-α-(1-Naphthyl)-ethylamine on Pd(111). J. Phys. Chem. C 2011, 115, 16488−16494. (169) Garvey, M.; Bai, Y.; Boscoboinik, J. A.; Burkholder, L.; Sorensen, T. E.; Tysoe, W. T. Identifying Molecular Species on Surfaces by Scanning Tunneling Microscopy: Methyl Pyruvate on Pd(111). J. Phys. Chem. C 2013, 117, 4505−4514. (170) Wells, P. B.; Wilkinson, A. G. Platinum Group Metals as Heterogeneous Enantioselective Catalysts. Top. Catal. 1998, 5, 39−50. (171) Minder, B.; Schürch, M.; Mallat, T.; Baiker, A.; Heinz, T.; Pfaltz, A. Enantioselective Hydrogenation of Ethyl Pyruvate over Pt/ Alumina Modified by (R)-1-(1-Naphthyl)ethylamine Derivatives. J. Catal. 1996, 160, 261−268. (172) Behzadi, B.; Ferri, D.; Baiker, A.; Ernst, K. H. Adsorption Mode of the Chiral Modifier Cinchonidine on Au(111). Appl. Surf. Sci. 2007, 253, 3480−3484. (173) Knowles, W. S. Asymmetric Hydrogenations (Nobel lecture). Angew. Chem., Int. Ed. 2002, 41, 1998−2007. (174) Lennon, I. C.; Pilkington, C. J. The Application of Asymmetric Hydrogenation for the Manufacture of Pharmaceutical Intermediates: The Need for Catalyst Diversity. Synthesis 2003, 2003, 1639−1642. (175) Huck, W. R.; Mallat, T.; Baiker, A. Heterogeneous Enantioselective Hydrogenation of 2-Pyrones over Cinchona-Modified Palladium. New J. Chem. 2002, 26, 6−8. (176) Szöllösi, G.; Hermán, B.; Felföldi, K.; Fülöp, F.; Bartók, M. Up to 96% Enantioselectivities in the Hydrogenation of Fluorine Substituted (E)-2,3-Diphenylpropenoic Acids over CinchonidineModified Palladium Catalyst. Adv. Synth. Catal. 2008, 350, 2804− 2814. (177) Ogawa, H.; Mameda, T.; Misaki, T.; Okamoto, Y.; Sugimura, T. Anomalous Ligand Acceleration on Cinchona-Modified Pd/C during Asymmetric Hydrogenation of Properly Substituted Phenylcinnamic Acid. Chem. Lett. 2013, 42, 813−815. (178) Huck, W. R.; Mallat, T.; Baiker, A. Enantioselective Hydrogenation of a Methoxypyrone with Cinchona-Modified Palladium. Catal. Lett. 2002, 80, 87−92. (179) Szöri, K.; Szöllösi, G.; Bartók, M. The Enantioselective Hydrogenation of 5,6-Dihydro-2H-pyran-3-carboxylic Acid over a 11567

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Cinchona Alkaloid-Modified Palladium Catalyst: Asymmetric Synthesis of a Cockroach Attractant. New J. Chem. 2008, 32, 1354−1358. (180) Chen, C. H.; Zhan, E. S.; Ta, N.; Li, Y.; Shen, W. J. Enantioselective Hydrogenation of α,β-Unsaturated Carboxylic Acids on Pd Nanocubes. Catal. Sci. Technol. 2013, 3, 2620−2626. (181) Strobel, R.; Stark, W. J.; Mädler, L.; Pratsinis, S. E.; Baiker, A. Flame-Made Platinum/Alumina: Structural Properties and Catalytic Behaviour in Enantioselective Hydrogenation. J. Catal. 2003, 213, 296−304. (182) Kubota, T.; Ogawa, H.; Okamoto, Y.; Misaki, T.; Sugimura, T. Preparation of Pd/C Designed for Chiral Modified Catalyst: Comparison with Pd/TiO2 in Enantioselective Hydrogenation of αPhenylcinnamic Acid. Appl. Catal., A 2012, 437-438, 18−23. (183) Kubota, T.; Kubota, H.; Kubota, T.; Moriyasu, E.; Uchida, T.; Nitta, Y.; Sugimura, T.; Okamoto, Y. Enantioselective Hydrogenation of (E)-α-Phenylcinnamic Acid over Cinchonidine-modified Pd Catalysts Supported on TiO2 and CeO2. Catal. Lett. 2009, 129, 387−393. (184) Trung, T. S. B.; Kim, Y.; Kang, S.; Kim, S.; Lee, H. Carbonaceous Materials as Catalyst Supports for the Enantioselective Hydrogenation of (E)-α-Phenylcinnamic Acid: Effect of the Support Acidity. Appl. Catal., A 2015, 505, 319−325. (185) Durán Pachón, L. D.; Yosef, I.; Markus, T. Z.; Naaman, R.; Avnir, D.; Rothenberg, G. Chiral Imprinting of Palladium with Cinchona Alkaloids. Nat. Chem. 2009, 1, 160−164. (186) Guan, Z.; Lu, S.; Li, C. Enantioselective Hydrogenation of α,βUnsaturated Carboxylic Acid over Cinchonidine-Modified Pd Nanoparticles Confined in Carbon Nanotubes. J. Catal. 2014, 311, 1−5. (187) Szöri, K.; Puskás, R.; Szöllösi, G.; Bertóti, I.; Szépvölgyi, J.; Bartók, M. Palladium Nanoparticle-Graphene Catalysts for Asymmetric Hydrogenation. Catal. Lett. 2013, 143, 539−546. (188) Kim, T. Y.; Yokota, M.; Uchida, T.; Sugimura, T. Conditioning of Palladium-on-Carbon (Pd/C) for Uniform Cinchonidine Modification to Perform Highly Enantioselective Hydrogenation. Catal. Lett. 2009, 131, 279−284. (189) Sugimura, T.; Kim, T. Y. Enantioselective Hydrogenation in Water Over Chiral Modified Heterogeneous Catalyst Admixed With Organic Solvent. Catal. Lett. 2009, 130, 564−567. (190) Nitta, Y.; Watanabe, J.; Okuyama, T.; Sugimura, T. ActivationTemperature Dependence in Enantioselective Hydrogenation of Unsaturated Carboxylic Acids over Cinchonidine-Modified Pd/C Catalysts. J. Catal. 2005, 236, 164−167. (191) Szöllösi, G.; Németh, Z.; Hernádi, K.; Bartók, M. Preparation and Characterization of TiO2 Coated Multi-Walled Carbon NanotubeSupported Pd and its Catalytic Performance in the Asymmetric Hydrogenation of α,β-Unsaturated Carboxylic Acids. Catal. Lett. 2009, 132, 370−376. (192) Mhadgut, S. C.; Török, M.; Esquibel, J.; Török, B. Highly Asymmetric Heterogeneous Catalytic Hydrogenation of Isophorone on Proline Modified Base-Supported Palladium Catalysts. J. Catal. 2006, 238, 441−448. (193) McIntosh, A. I.; Watson, D. J.; Burton, J. W.; Lambert, R. M. Heterogeneously Catalyzed Asymmetric CC Hydrogenation: Origin of Enantioselectivity in the Proline-Directed Pd/Isophorone System. J. Am. Chem. Soc. 2006, 128, 7329−7334. (194) Mhadgut, S. C.; Török, M.; Dasgupta, S.; Török, B. Nature of Proline-Induced Enantiodifferentiation in Asymmetric Pd Catalyzed Hydrogenations: Is the Catalyst Really Indifferent? Catal. Lett. 2008, 123, 156−163. (195) Li, S.; Zhan, E. S.; Li, Y.; Xu, Y. D.; Shen, W. J. Enantioselective Hydrogenation of Isophorone and Kinetic Resolution of 3,3,5Trimethylcyclohexanone over Pd Catalysts in the Presence of (S)Proline. Catal. Today 2008, 131, 347−352. (196) Li, S.; Chen, C. H.; Zhan, E. S.; Liu, S. B.; Shen, W. J. Chirality Inversion in Enantioselective Hydrogenation of Isophorone over Pd/ MgO Catalysts in the Presence of (S)-Proline: Effect of Pd Particle Size. J. Mol. Catal. A: Chem. 2009, 304, 88−94. (197) Rodriguez-Garcia, L.; Hungerbuhler, K.; Baiker, A.; Meemken, F. Enantioselection on Heterogeneous Noble Metal Catalyst: Proline-

Induced Asymmetry in the Hydrogenation of Isophorone on Pd Catalyst. J. Am. Chem. Soc. 2015, 137, 12121−12130. (198) Wang, Y.; Su, N.; Ye, L.; Ren, Y.; Chen, X.; Du, Y.; Li, Z.; Yue, B.; Tsang, S. C. E.; Chen, Q.; et al. Tuning Enantioselectivity in Asymmetric Hydrogenation of Acetophenone and its Derivatives via Confinement Effect over Free-Standing Mesoporous Palladium Network Catalysts. J. Catal. 2014, 313, 113−126. (199) Sipos, E.; Tungler, A.; Fogassy, G. New Substrates and Modifiers in the Enantioselective Heterogeneous Catalytic Hydrogenation of the CC Double Bond. J. Mol. Catal. A: Chem. 2004, 216, 171−180. (200) Huck, W. R.; Mallat, T.; Baiker, A. Potential and Limitations of Palladium−Cinchona Catalyst for the Enantioselective Hydrogenation of a Hydroxymethylpyrone. J. Catal. 2000, 193, 1−4. (201) Nitta, Y.; Shibata, A. Enantioselective Hydrogenation of (E)-αPhenylcinnamic Acid on Pd/TiO2 Catalyst Modified by Cinchona Alkaloids: Effect of Modifier Structure. Chem. Lett. 1998, 27, 161−162. (202) Sugimura, T.; Ogawa, H. Acceleration of Enantioselective Hydrogenation of Olefins over Pd/C by Cinchonidine as a Chiral Modifier. Comparison with Cinchonine, Pseudoenantiomer. Chem. Lett. 2010, 39, 232−233. (203) Szöllösi, G.; Hermán, B.; Fülöp, F.; Bartók, M. Cinchona Methyl Ethers as Modifiers in the Enantioselective Hydrogenation of (E)-2,3-Diphenylpropenoic Acids over Pd Catalyst. J. Catal. 2010, 276, 259−267. (204) Mameda, T.; Misaki, T.; Okamoto, Y.; Sugimura, T. Induction Period for the Enantioselective Hydrogenation of α,β-Unsaturated Acids over a Cinchonidine-Modified Palladium Catalyst. React. Kinet., Mech. Catal. 2015, 114, 55−61. (205) Szöllösi, G.; Busygin, I.; Hermán, B.; Leino, R.; Bucsi, I.; Murzin, D. Y.; Fülöp, F.; Bartók, M. Inversion of the Enantioselectivity in the Hydrogenation of (E)-2,3-Diphenylpropenoic Acids over Pd Modified by Cinchonidine Silyl Ethers. ACS Catal. 2011, 1, 1316− 1326. (206) Kim, T. Y.; Sugimura, T. Is Amine Addition Vital for Highly Enantioselective Hydrogenation of α,β-Unsaturated Carboxylic Acid over Cinchonidine-Modified Palladium? J. Mol. Catal. A: Chem. 2010, 327, 58−61. (207) Meemken, F.; Maeda, N.; Hungerbuhler, K.; Baiker, A. Heterogeneous Asymmetric Hydrogenation of Prochiral Alkenoic Acid: Origin of Rate and Enantioselectivity Enhancement by Amine Addition. ACS Catal. 2012, 2, 464−467. (208) Nakai, K.; Misaki, T.; Okamoto, Y.; Sugimura, T. Kinetic and Stereochemical Analyses of Amine-Additive Effects on the Hydrogenation of α-Phenylcinnamic Acid over Cinchonidine-Modified Pd/ C. Bull. Chem. Soc. Jpn. 2015, 88, 300−302. (209) Makra, Z.; Szöllösi, G.; Bartók, M. Achiral Amine Additives in the Enantioselective Hydrogenation of Aliphatic α,β-Unsaturated Acids over Cinchonidine-Modified Pd/Al2O3 Catalyst. Catal. Today 2012, 181, 56−61. (210) Garland, M.; Blaser, H. U. A Ligand-Accelerated Reaction: Enantioselective Hydrogenation of Ethyl Pyruvate Catalyzed by Cinchona-Modified Pt/Al2O3 Catalysts. J. Am. Chem. Soc. 1990, 112, 7048−7050. (211) Mameda, T.; Nakai, K.; Misaki, T.; Okamoto, Y.; Sugimura, T. Substrate Dependent Ligand Acceleration in Enantioselective Hydrogenation of (E)-2,3-Diarylpropenoic Acid on Cinchonidine-Modified Pd/C. Catal. Today 2015, 245, 129−133. (212) Nitta, Y. Importance of Product Desorption in Enantioselective Hydrogenation of (E)-α-Phenylcinnamic Acid with a CinchonidineModified Pd/TiO2 Catalyst: Effect of Additives. Chem. Lett. 1999, 28, 635−636. (213) Szöllösi, G.; Makra, Z.; Bartók, M. Enantioselective Hydrogenation of (E)-2-Methyl-2-butenoic Acid over Cinchonidine Modified Pd Catalyst. Effect of the Structure of Achiral Amine Additives. React. Kinet. Catal. Lett. 2009, 96, 319−325. (214) Makra, Z.; Szöllösi, G. Hydrogenation of (E)-2-Methyl-2butenoic acid over Cinchona-Modified Pd Catalyst in the Presence of 11568

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Chemical Reviews

Review

Achiral Amines: Solvent and Modifier Effect. Catal. Commun. 2014, 46, 113−117. (215) Sugimura, T.; Watanabe, J.; Okuyama, T.; Nitta, Y. The Effect of the Substrate Structure on the Stereoselectivity in an Asymmetric Hydrogenation of Unsaturated Carboxylic Acids over CinchonidineModified Palladium Catalyst. Tetrahedron: Asymmetry 2005, 16, 1573− 1575. (216) Szö llö si, G.; Balázsik, K.; Bartók, M. Enantioselective Hydrogenation of Itaconic Acid over Cinchona Alkaloid Modified Supported Palladium Catalyst. Appl. Catal., A 2007, 319, 193−201. (217) Watson, D. J.; John Jesudason, R. J. B. R.; Beaumont, S. K.; Kyriakou, G.; Burton, J. W.; Lambert, R. M. Heterogeneously Catalyzed Asymmetric Hydrogenation of CC Bonds Directed by Surface-Tethered Chiral Modifiers. J. Am. Chem. Soc. 2009, 131, 14584−14589. (218) Sugimura, T.; Watanabe, J.; Uchida, T.; Nitta, Y.; Okuyama, T. Highly Enantioselective Hydrogenation of α-Alkyl-β-Arylpropenoic Acids over Cinchonidine-Modified Palladium Catalyst. Catal. Lett. 2006, 112, 27−30. (219) Szöllösi, G.; Hermán, B.; Felföldi, K.; Fülöp, F.; Bartók, M. Effect of the Substituent Position on the Enantioselective Hydrogenation of Methoxy-Substituted 2,3-Diphenylpropenoic Acids over Palladium Catalyst. J. Mol. Catal. A: Chem. 2008, 290, 54−59. (220) Sugimura, T.; Uchida, T.; Watanabe, J.; Kubota, T.; Okamoto, Y.; Misaki, T.; Okuyama, T. Structural Requirements for Substrate in Highly Enantioselective Hydrogenation over the CinchonidineModified Pd/C. J. Catal. 2009, 262, 57−64. (221) Hermán, B.; Szöllösi, G.; Felföldi, K.; Fülöp, F.; Bartók, M. Enantioselective Hydrogenation of Propenoic Acids Bearing Heteroaromatic Substituents over Cinchonidine Modified Pd/Alumina. Catal. Commun. 2009, 10, 1107−1110. (222) Szöllösi, G.; Hermán, B.; Szabados, E.; Fülöp, F.; Bartók, M. Reactions of Chlorine Substituted (E)-2,3-Diphenylpropenoic Acids over Cinchonidine-Modified Pd: Enantioselective Hydrogenation versus Hydrodechlorination. J. Mol. Catal. A: Chem. 2010, 333, 28−36. (223) Szöllösi, G.; Szöri, K.; Bartók, M. Enantioselective Hydrogenation of Arecaidine over Cinchona Alkaloid-Modified Palladium Catalyst: A Novel Route to Enantio-Enriched Nipecotic Acid Derivatives. J. Catal. 2008, 256, 349−352. (224) Szöllösi, G.; Szabó, E.; Bartók, M. Enantioselective Hydrogenation of N-Acetyldehydroamino Acids over Supported Palladium Catalysts. Adv. Synth. Catal. 2007, 349, 405−410. (225) Szöllösi, G.; Varga, T.; Felföldi, K.; Cserényi, S.; Bartók, M. Enantioselective Hydrogenation of Fluorinated Unsaturated Carboxylic Acids over Cinchona Alkaloid Modified Palladium Catalysts. Catal. Commun. 2008, 9, 421−424. (226) Balázsik, K.; Szöllösi, G.; Berkesi, O.; Szalontai, G.; Fülöp, F.; Bartók, M. Heterogeneous Asymmetric Hydrogenation of NHeterocyclic Compounds: Hydrogenation of Tetrahydroisoquinoline Derivatives. Top. Catal. 2012, 55, 880−888. (227) Tan, S.; Monnier, J. R.; Williams, C. T. Kinetic Study of Asymmetric Hydrogenation of α,β-Unsaturated Carboxylic Acid over Cinchona-Modified Pd/Al2O3 Catalyst. Top. Catal. 2012, 55, 512− 517. (228) Sun, X.; Monnier, J. R.; Williams, C. T. Enantioselective Hydrogenation of α-Methylcinnamic Acid over Pt/Al2O3P: A Kinetic Study of Solvent, Temperature and Pressure Effects. Catal. Lett. 2013, 143, 881−886. (229) Hermán, B.; Szöllösi, G.; Fülöp, F.; Bartók, M. Enantioselective Hydrogenation of α,β-Unsaturated Carboxylic Acids in Fixed Bed Reactor. Appl. Catal., A 2007, 331, 39−43. (230) Murzin, D. Y.; Szöllösi, G. Unusual Behavior of Modifier Mixtures in Heterogeneous Enantioselective Catalysis: Beyond Nonlinear Phenomena. React. Kinet., Mech. Catal. 2011, 103, 1−9. (231) Meier, D. M.; Urakawa, A.; Turra, N.; Ruegger, H.; Baiker, A. Hydrogen-Bonding Interactions in Cinchonidine-2-methyl-2-hexenoic Acid Complexes: A Combined Spectroscopic and Theoretical Study. J. Phys. Chem. A 2008, 112, 6150−6158.

(232) Sun, X.; Williams, C. T. In-situ ATR-IR Investigation of Methylcinnamic Acid Adsorption and Hydrogenation on Pd/Al2O3. Catal. Commun. 2012, 17, 13−17. (233) Györffy, N.; Tungler, A.; Fodor, M. Stereodifferentiation in Heterogeneous Catalytic Hydrogenation. Kinetic Resolution and Asymmetric Hydrogenation in the Presence of (S)-Proline: CatalystDependent Processes. J. Catal. 2010, 270, 2−8. (234) McIntosh, A. I.; Watson, D. J.; Lambert, R. M. Mechanistic Insights into the Proline-Directed Enantioselective Heterogeneous Hydrogenation of Isophorone. Langmuir 2007, 23, 6113−6118. (235) Chen, C. H.; Zhan, E. S.; Li, Y.; Shen, W. J. Enantioselective Hydrogenation of α-Dehydroamino Acids on a CinchonidineModified Palladium Catalyst. J. Mol. Catal. A: Chem. 2013, 379, 117−121. (236) Tungler, A.; Kajtar, M.; Mathe, T.; Toth, G.; Fogassy, E.; Petro, J. Enantioselective Hydrogenation of α,β-Unsaturated Ketones. Catal. Today 1989, 5, 159−171. (237) Tungler, A.; Mathe, T.; Petro, J.; Tarnai, T. Enantioselective Hydrogenation of Isophorone. J. Mol. Catal. 1990, 61, 259−267. (238) Gao, F.; Wang, Y.; Burkholder, L.; Tysoe, W. T. Chemistry of L-Proline on Pd(111): Temperature-Programmed Desorption and Xray Photoelectron Spectroscopic Study. Surf. Sci. 2007, 601, 3579− 3588. (239) Beaumont, S. K.; Kyriakou, G.; Watson, D. J.; Vaughan, O. P. H.; Papageorgiou, A. C.; Lambert, R. M. Influence of Adsorption Geometry in the Heterogeneous Enantioselective Catalytic Hydrogenation of a Prototypical Enone. J. Phys. Chem. C 2010, 114, 15075− 15077. (240) Fodor, M.; Tungler, A.; Vida, L. Asymmetric Hydrogenation of Isophorone in the Presence of (S)-Proline: Revival of a 20 Years Old Reaction. Catal. Today 2009, 140, 58−63. (241) Ashley, M. A.; Hirschi, J. S.; Izzo, J. A.; Vetticatt, M. J. Isotope Effects Reveal the Mechanism of Enamine Formation in L-ProlineCatalyzed α-Amination of Aldehydes. J. Am. Chem. Soc. 2016, 138, 1756−1759. (242) Tungler, A.; Fogassy, G. Catalysis with Supported Palladium Metal, Selectivity in the Hydrogenation of CC, CO and CN Bonds, from Chemo- to Enantioselectivity. J. Mol. Catal. A: Chem. 2001, 173, 231−247. (243) Yasukawa, T.; Miyamura, H.; Kobayashi, S. Chiral LigandModified Metal Nanoparticles as Unique Catalysts for Asymmetric C− C Bond-Forming Reactions: How Are Active Species Generated? ACS Catal. 2016, 6, 7979−7988. (244) Reimann, S.; Mallat, T.; Baiker, A. Enantioselective Allylic Substitution on Pt/Al2O3 Modified by Chiral Diphosphines. J. Catal. 2007, 252, 30−38. (245) Reimann, S.; Mallat, T.; Baiker, A. A New, Efficient Heterogeneous Pd Catalyst for Enantioselective Allylic Substitution. J. Catal. 2008, 254, 79−83. (246) Han, D. F.; Li, X. H.; Zhang, H. D.; Liu, Z. M.; Hu, G. S.; Li, C. Asymmetric Hydroformylation of Olefins Catalyzed by Rhodium Nanoparticles Chirally Stabilized with (R)-BINAP Ligand. J. Mol. Catal. A: Chem. 2008, 283, 15−22.

11569

DOI: 10.1021/acs.chemrev.7b00272 Chem. Rev. 2017, 117, 11522−11569

Recent Progress in Heterogeneous Asymmetric ... - ACS Publications

Recommend Documents