Review pubs.acs.org/CR
Homogeneous Transition Metal Catalysis of Acceptorless Dehydrogenative Alcohol Oxidation: Applications in Hydrogen Storage and to Heterocycle Synthesis Robert H. Crabtree* Department of Chemistry & Energy Sciences Institute, Yale University, P.O. Box 208107, New Haven, Connecticut 06520-8107, United States ABSTRACT: The different types of acceptorless alcohol dehydrogenation (AAD) reactions are discussed, followed by the catalysts and mechanisms involved. Special emphasis is put on the common appearance in AAD of pincer ligands, of noninnocent ligands, and of outer sphere mechanisms. Early work emphasized precious metals, mainly Ru and Ir, but interest in nonprecious metal AAD catalysis is growing. Alcohol−amine combinations are discussed to the extent that net oxidation occurs by loss of H2. These reactions are of potential synthetic interest because they can lead to N heterocycles such as pyrroles and pyridines. AAD also has green chemistry credentials in that an oxidation occurs without the need for an oxidizing agent and hence without the waste formation that would result from its use.
α to the OH group. As an organic oxidation method, eq 1 also has green chemistry1 credentials because the absence of an oxidizing agent avoids both the hazards and the waste formation associated with the oxidant, as well as maximizing atom efficiency. Since the best known classical reagents in oxidation2 are toxic and produce waste, this is a welcome aspect. The alchemists wanted to convert base metals into precious metals; we now know this is impossible, at least by chemical means, but we are today still trying to do the next best thing to wrap base metals in ligands that give them the catalytic properties of precious metals. Indeed, precious metal catalysts are predominant in AAD and so this area provides a good test bed for trying to bring about this transition to base metals. The successful cases of AAD to be covered here involve a wide variety of organic pathways, discussed in section 2, as well as an unusually large number of novel types of ligand, as discussed in section 3. Beyond the usual chelates and pincers, we also find a number of rather exotic multidentate ligands not seen elsewhere in homogeneous catalysis. These are often multifunctional ligands3sometimes also called cooperative or noninnocent ligandsin which the ligand has functions that go beyond simple binding to the metal. In this way typical metal functions can be “outsourced” to the ligands, leading to metal−ligand cooperation (MLC) or even to cases where the metal appears to be a pure spectator. In AAD these extra functions include being able to undergo reversible protonation or being intrinsically redox-active. As we see in what follows, AAD is particularly well-suited to the
CONTENTS 1. Introduction 2. Substrates 2.1. Methanol 2.2. Higher Monohydric Alcohols 2.3. Polyols 2.4. Alcohol−Amine Combinations 3. Catalysts and Mechanisms 3.1. Osmium and Ruthenium Catalysts 3.2. Iridium and Rhodium Catalysts 3.3. Base Metal Catalysts 4. Conclusions Author Information Corresponding Author ORCID Notes Biography Acknowledgments References
9228 9230 9230 9231 9232 9233 9235 9236 9238 9240 9241 9242 9242 9242 9242 9242 9242 9242
1. INTRODUCTION Acceptorless alcohol dehydrogenation (AAD) by thermally activated homogeneous transition metal catalysis (e.g., eq 1) has much to teach us. As the best explored class of substrate dehydrogenation, it provides a template for extending acceptorless dehydrogenation to a much wider variety of other substrates. cata.
RCH 2OH ⎯⎯⎯→ RCHO + H 2
(1)
Special Issue: CH Activation
Equation 1 can be also considered as including a CH activation step, albeit of a particularly reactive CH bond since it is located © 2017 American Chemical Society
Received: August 17, 2016 Published: January 4, 2017 9228
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
MLC approach and the presence of cooperative ligands that can also help make nonprecious metals behave more like the precious metals in terms of catalytic activity. Another aspect of AAD involves energy chemistry: the present fossil fuel energy economy may become untenable at some point in the future ifsome would say “when”climate change effects become sufficiently severe. Of the alternatives that have been suggested, a proposed hydrogen economy has a long history. The colorful polymath J. B. S. Haldane (1892−1964) suggested such a thing in a speech from 1923 that was published in the following year.4 Personally, I think that four hundred years hence the power question in England may be solved somewhat as follows: The country will be covered with rows of metallic windmills working electric motors [generators] which in their turn supply current at a very high voltage to great electric mains [national electricity grid]. At suitable distances, there will be great power stations where during windy weather the surplus power will be used for the electrolytic decomposition of water into oxygen and hydrogen. These gases will be liquefied, and stored in vast vacuum jacketed reservoirs, probably sunk in the ground. [...] In times of calm, the gases will be recombined in explosion motors [internal combustion engines] working dynamos which produce electrical energy once more, or more probably in oxidation cells [fuel cells]. Substantial objections have been raised to this idea, however.5 Hydrogen is inconvenient to store, being a flammable permanent gas with a high tendency to escape containment. Of the numerous proposed possibilities,6 the method relevant to this review is catalytic hydrogenation/dehydrogenation to bring about covalent H2 storage/release from liquid organic hydrogen carriers, or LOHCs.7,8 LOHCs can be easily contained and transported, allowing much of the current transportation fuel infrastructure to be maintained.9 Among LOHCs, MeOH has been identified as one promising candidate because it is easily storable. If it were adopted, it could be used directly as a fuel alternatively, eq 1 provides an easy H2 release pathway that could prove broadly useful for supplying hydrogen fuel cells. Indeed, in a comprehensive 2006 book, Olah and coauthors10 have proposed a “methanol economy” where the methanol could come from any of a wide array of possible sources, including CO2 hydrogenation, as well as act as a source material for a wide variety of chemical intermediates and products. Homogeneous organometallic catalysts are best known for reductive or redox-neutral reactions but less commonly for oxidation. Because oxidation typically requires a primary oxidant, typical P- and C-donor ligands are often vulnerable to oxidative degradation by that oxidant under the reaction conditions. Organometallic complexes can still be precatalysts for reactions involving such primary oxidants, but they may then undergo loss of a C-donor ligand such as Cp*, CO, arene, or alkene to give a coordination complex as the resting state of the catalyst.11 Dehydrogenative oxidation such as AAD differs from classical oxidation in that there is no primary oxidant, so any ligands, particularly C-donor ligands, stand a better chance of surviving the catalytic process. Standard oxidation methods tend to give low selectivity between primary and secondary alcohols. When metalcatalyzed, AAD can show enhanced selectivity because primary alcohols are then often significantly more reactive. For example,12 RuCl2(PPh3)3 selectively oxidized a primary hydroxyl in the presence of a secondary hydroxyl in the conversion of
1,10-undecadiol to the aldehyde in 89% yield at 25 °C over 2 h (eq 2). This was only barely catalytic, however, since the “catalyst” loading was 75%; indeed, the Ru complex might itself have acted as hydrogen acceptor. Raising the temperature permitted a lower loading, but the selectivity was then degraded. One disadvantage of eq 1 is the much lower driving force available than in the case where an oxidant is present. This was originally countered by introduction of a hydrogen acceptor such as tBuCHCH2,13,14 but we then lose the green advantage of omitting the oxidant. Higher temperatures can drive the reaction in the absence of a hydrogen acceptor by enhancing both the thermodynamics and the kinetics at the same time. This not only speeds the reaction steps but it also enhances the entropic driving force via the −TΔS component of ΔG as a result of forming two products, one being a gas. The solubility of a perfect gas falls to zero at the solvent boiling point, so if the reaction is carried out at reflux the gas concentration in the solvent phase can be reduced to an arbitrarily low level, thus helping drive the reaction by continuously sweeping out the H2 product.15 This has proved useful in driving the “acceptorless” endothermic dehydrogenations of alkanes,16 and in one early case even using a nonprecious metal catalyst, W(triphos)H6 {triphos = PhP(CH2CH2PPh2)2}.17 Another very useful application of alcohol oxidation via eq 1 depends on the much higher reactivity of the aldehyde or ketone product versus the starting alcohol. This has provided a recently developed tandem route to a variety of derivatives formed by nucleophilic attack on the carbonyl compound followed by transfer of the two H atoms removed in the initial dehydrogenative oxidation step to the organic intermediate to form the final product. This “hydrogen borrowing” process is shown in Figure 1 for the case of amine alkylation, typically
Figure 1. Typical hydrogen borrowing cycle.
catalyzed by a Ru pincer catalyst. Here an endothermic initial step is driven by the exothermic second and third steps. Extensions of this concept to the synthesis of the pharmaceutically important benzodiazepines have appeared very recently. Figure 2 shows the two sequential hydrogen borrowing steps in the reaction, which proceeds at 160 °C for 16 h in 65% yield for the R = benzyl case with [Ru(p-cymene)Cl2]2/Xantphos catalyst.18 This hydrogen borrowing pathway has proved notably successful for the synthesis of heterocycles22 and has been extensively reviewed,19−21 so this net redox-neutral pathway is thus only mentioned briefly here. In what follows the term “alcohol dehydrogenation” will therefore imply the acceptorless version unless otherwise stated. Apart from the reviews already noted, a number of other important ones cover parts of the present topic or related areas. For example, Alberico and Nielsen cover methanol conversion 9229
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
Figure 2. One-pot, hydrogen borrowing benzodiazepine synthesis.18
molecule is released to form CH2O with a room temperature ΔG° of +15.2 kcal/mol, counting the formaldehyde as CH2O(g). In a solvent the formaldehyde will be stabilized to some extent, for example by hydrate formation, so this is an upper limit for the endoergicity. The high thermodynamic stability of CO makes eq 4 much less unfavorable with a ΔG° of +6.9 kcal/mol. Equation 5 is even more favorable thanks to the formation of CO2: ΔG° is now +2.2 kcal/mol.23 This third pathway has several other advantages. Not only is the final product, CO2, much less toxic, both to people and to catalysts, than either CO or CH2O, but also a part of the hydrogen comes from water.
to CO2 and CO2 to methanol,23 Nielsen looks at AAD of alcohols in general,24 Johnson et al.25 cover catalytic dehydrogenation of formic acid and monohydric alcohols, Trincado et al. discuss AAD of monohydric alcohols,26 Chamoun et al.27 review hydrogen storage materials in general, Chelucci et al. discuss 2-(aminomethyl)pyridine Ru and Os complexes as catalysts for a variety of processes including AAD,28 and Soloveichik looks at LOHCs in fuel cells.29 Wang et al.30 review CO2 hydrogenation to formate and methanol in connection with the storage step of the hydrogen storage problem. Kopylovich, Pombeiro, and co-workers have an extensive review on recent work in catalytic alcohol oxidation in general.31 Beller and co-workers have reviewed the role of pincer catalysts in transfer hydrogenation, hydrogen borrowing, and AAD reactions.32 A detailed review of computational mechanistic work in AAD chemistry comes from Li and Wang.33 All this indicates that the AAD and related fields have seen a recent sharp increase in activity with the main applications being in energy chemistry, green chemistry, and organic synthetic methods.
CH3OH → CH 2O + H 2
(3)
CH3OH → CO + 2H 2
(4)
CH3OH + H 2O → CO2 + 3H 2
(5)
In order for this third pathway to CO2 (eq 5) to operate, however, it is usually necessary to avoid forming CO via the pathway of eq 4, because CO can be a catalyst poison. To avoid CO formation, a water molecule can intervene at the CH2O stage to give the hydrate, CH2(OH)2, that can then go on to HCOOH, a molecule that is generally found34 to transform easily into CO2 and H2 with a wide variety of catalysts. If so, the resulting sequence is shown in eq 6. The mass percent of hydrogen released per unit mass of MeOH in eq 6 is 12% if we count the water or an impressive 18.8% if the water is available on-site and does not have to be transported along with the MeOH to the point of use, fuel transport costs being an important consideration in the global analysis.
2. SUBSTRATES This section emphasizes the organic aspectsthe alcohol substrates and their principal organic reactions under acceptorless dehydrogenative thermal homogeneous catalysis. This means the discussion of the catalysts themselves and their organometallic mechanisms is left to section 3. Alcohols are the focus of attention here because they are capable of undergoing fully reversible hydrogenation−dehydrogenation and thus can be considered true storage agents. Photocatalytic dehydrogenation is relevant to solar-to-fuel aspirations but not to hydrogen storage or organic synthesis; although it is not covered here, excellent reviews are available.23 Dehydrogenative net oxidation with H2 evolution is still in the early development stage for organic synthetic applications, however, but a few examples featuring complex organic molecules are described in this section. A note of caution applies to any discussion of rates of acceptorless dehydrogenation. Because the reaction can be reversible and the rate of departure of H2 from the solution may be very dependent on conditions, such as reflux rate and reactor configuration, significant variability is therefore to be expected between different laboratories and in experiment/theory comparisons, so AAD is not always kinetically well-behaved.
+H 2 O
CH3OH ⎯⎯⎯→ CH 2O ⎯⎯⎯⎯⎯→ HCOOH ⎯⎯⎯→ CO2 −H 2
−H 2
−H 2
(6)
In an alternative energy context, assuming hydrogen becomes available from solar, wind, biomass, nuclear, or other nonfossil fuel sources, the MeOH LOHC could be formed by hydrogenation of CO2, reversing eq 6. Heterogeneous copper catalysts do this conveniently, for example, although high temperatures and pressures are required.35 A ruthenium PNP pincer catalytic system36 used by Olah and co-workers gives a CO2 hydrogenation to MeOH that is fully reversible with a mere pressure swing.37 A number of current hydrogen storage studies report irreversible release of H2 without too much concern for the key storage step, the regeneration of the hydrogenated form of the carrier. The term “hydrogen storage” may therefore best be restricted to cases such as (CO2 + 3H2)/ (MeOH + H2O) where both directions are possible. Even where the selectivity for eq 6 is complete, however, one drawback of MeOH as an LOHC is that the concomitant production of CO2 in the H2 release step. This requires a CO2
2.1. Methanol
Methanol is perhaps the most common alcohol substrate for hydrogen evolution catalysis, but it has several pathways by which it can release H2, depending on the conditions and the catalyst. The simplest case is shown in eq 3, where a single H2 9230
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
separation step to avoid carbon release into the atmosphere, but this would be hard to do in transport applications, for example. Alternatively, if MeOH were consumed in an internal combustion engine, the consequent CO2 release into the air would only conform to the green ideal if it had originally come from air. This carbon could then be considered as having been temporarily “borrowed” from the atmosphere. A variety of different embodiments of hydrogen storage are possible, and we cannot be sure which ones will be preferred in any of the different niches that may open up in a future energy economy. It is therefore prudent for us to develop a range of LOHCs and of catalysts so that a variety of potential applications can be supported by an adequate science base. Other pathways can also take over under certain conditions. For example, the CH2O or HCOOH oxidation products from MeOH can be trapped as the acetal (eq 7) or the ester (eq 8). These are also of historical interest because they were identified in Maitlis’s 1985 work with RuCl2(PPh3)3 which was one of the first reports to identify homogeneous catalysis of H2 evolution from MeOH.38
2.2. Higher Monohydric Alcohols
The thermodynamic data for a number of relevant reactions are shown in Table 1. The ΔG values are mostly positive at Table 1. Some Thermodynamic Data (kcal/mol)a
a
3CH3OH → CH 2(OMe)2 + H 2O + H 2
(7)
2CH3OH → CH3OOCH + 2H 2
(8)
reaction
ΔH°298
ΔG°298
CH3OH(l) = CH2O(g) + H2(g) CH3OH(l) = CO(g) + 2H2(g) CH3OH(l) + H2O(l) = CO2(g) + 3H2(g) EtOH(l) = MeCHO(l) + H2(g) i-PrOH(l) = Me2CO(l) + H2(g) PhCH2OH(l) = PhCHO(l) + H2(g)
31.0 30. 6 31.8 19.4 16.4 17.5
15.2 6.9 2.2 6.4 −0.8 1.5
NIST database.
298 K as a result of an unfavorable ΔH term, but this term is progressively compensated at higher temperature by a favorable −TΔS term that results from the entropy gain associated with the release of H2 and sometimes also of other gases. A part of the requirement for relatively high temperatures in these AAD reactions may therefore be thermodynamic. This analysis assumes an equilibrium exists, but this is of course far from the case in the real systems because these reactions are often run under reflux. Under such circumstances the product H2 is driven out of solution dynamically, in which case Le Chatelier’s principle provides an additional driving force from the product being actively removed from solution.14,15 The first example of the homogeneous transition metal catalyzed dehydrogenation of any alcohol to give free H2 was Murahashi’s 1981 Ru3(CO)12-catalyzed conversion of a range of higher monohydric alcohols into esters and diols into lactones (eq 9).48 A hydrogen acceptor was added for enhancing the
Another early report, also from 1985, came from Saito, who found formaldehyde production via eq 3 to be catalyzed by a variety of Ru(II) precatalysts, including [Ru(OAc)Cl(PEtPh2)3].39 Another useful feature of this report was the identification of the catalyst deactivation pathway, a topic that has received less academic attention than it deserves.40 This proved to be decarbonylation of the intermediate formaldehyde to produce inactive Ru(II) carbonyl complexes. Maitlis also saw carbonyl complexes at the end of the reaction in his case but did not specifically identify them as deactivation products. The MeOH dehydrogenation pathway to give CO (eq 4) is often undesired, not only because CO can poison homogeneous catalysts but also because CO contamination of the H2 stream can be a problem for downstream processes such as fuel cell operation. Although not an alcohol, formic acid deserves a brief mention since it is on the methanol pathway of eq 6 and is very often itself proposed as a hydrogen storage material. Its big advantage is that reversible catalytic interconversion of CO2−H2 mixtures into HCOOH is well-established.25,41 A disadvantage is that the mass percent H in HCOOH is a relatively low 4.3%. Coproduction of CO2 in the release step is a problem shared with MeOH, but MeOH does at least provide 3 equiv of H2, not just 1 equiv. Finally, a number of very recent prior reviews have covered formic acid as a hydrogen storage medium.25,34,42,43 Another intermediate on the MeOH pathway of eq 6 is formaldehyde, and the dehydrogenation of aqueous formaldehyde by a Ru catalyst has been discussed in the context of hydrogen generation.44 The relatively high toxicity and volatility of methanol and its oxidation products other than CO2 is another disadvantage common to all of them, considering that a future hydrogen economy based on any of them would require extensive distribution of these materials in the public sphere. A number of key early reports45 cover AAD of secondary alcohols with Grubbs’s catalyst (1, 5% loading, 110 °C, 24 h),40 Shvo’s catalyst46 (2, 2% loading, 145 °C, 15 h; the four Ph substituents on the C5 ring are omitted for clarity), and a combination of RuCl3 with a number of bulky phosphines (0.0315% loading, 90 °C, 6 h).47
yield, but the H2-evolution pathway was sufficiently favored to give 40−50% yields without the additive. Once again, the primary > secondary selectivity of eq 2 is maintained even though the latter is expected to be thermodynamically favored (Table 1). Of course, the conditions were very harsh180 °C for 24 hso the true catalyst may have been a derivative of the precatalyst. Blum and Shvo’s 1985 conversion of benzyl alcohol to benzyl benzoate, the catalyst precursor being a simple ruthenium carbonyl derivative, (η4-tetramethylcyclopentadienone)Ru(CO)3 may therefore have been the first truly homogeneous case.49 Ethanol and isopropanol are the most commonly encountered substrates in more recent work, however. These are advantageous in having a higher driving force for dehydrogenation than MeOH but carry a lower mass percentage of H2 (iPrOH, 3.3%; EtOH, 4.3%). iPrOH also has the advantage of being much more resistant to decarbonylation and therefore of any possible subsequent catalyst deactivation by the resulting CO which makes it the most common hydrogen donor solvent in transfer 9231
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
hydrogenation (TH).50 Isopropanol has the additional advantage in that its oxidation product, acetone, is more volatile than iPrOH itself and a transfer hydrogenation can thus be driven by distilling this component out of the reaction mixture. Homogeneous transition metal catalysis of transfer hydrogenation greatly predates hydrogen evolution catalysis, going back to Henbest’s 1964 IrHCl2(dmso)3 catalyzed reduction of cyclohexanones with iPrOH to give the thermodynamically less stable axial alcohol products.51 Organic synthetic applications of dehydrogenative oxidation are still rare, but a wide variety of higher alcohols has proved susceptible. For example, K. Fujita et al. have used a [Cp*IrCl2(2-hydroxypyridine)] catalyst to dehydrogenate a wide array of both primary and secondary alcohols to the corresponding aldehydes and ketones. The system showed considerable tolerance of functional groups, such as −OMe, −Cl, −Br, −COOMe, −CF3, and −NO2; conversions and yields often exceeded 90% under suitable conditions.52 Direct conversion of alcohols to carboxylic acids by dehydrogenative oxidation has also proved possible. For example, Milstein and co-workers have reported their Ru pincer catalyst, 3, is active at a mere 0.2% loading at the reflux temperature of a basic aqueous solution of the alcohol. The yields varied from 60% to quantitative over a wide range of substrates including conversion of 1,5-dihydroxypentane to glutaric acid (61% + 14% lactone) and of 2,6-hydroxymethylpyridine to dipicolinic acid (quantitative).53
the catalyst is only able to hydrogenate the CC bond, not the CO, H2 is released to form the coupled ketone. This is the case for the Cp*Ir dipyridonate complex, [Cp*Ir(2,2′bpyO)(H2O)] (4), where coupled ketones are formed in 70−90% yield with Cs2CO3 as base after 6 h reflux (Figure 3). A slight ambiguity in this report is the fact that the reflux was only carried out under air, so there is a remote possibility that aerial O2 was the true oxidant.55
2.3. Polyols
Zhao and Hartwig’s 2005 work56 on γ-butyrolactone formation from 1,4-butanediol has several points of interest. First, the reaction goes ahead without base, solvent, or hydrogen acceptor at 205 °C with a variety of Ru catalysts at 0.05% loading. The very high temperatures led the authors to favor alkylphosphines as ligands because they tend to be more thermally stable, although such temperatures are well beyond the range that is generally regarded as safe for the preservation of homogeneous catalysts from decomposition. A range of catalysts were examined and the deactivation products identified in some cases, for example with the formation of cis,mer-RuH2(CO)(PMe3)3 from RuH2(PMe3)4 via decarbonylation of the intermediate aldehyde formed in partial oxidation of the diol. Reversibility arguments were used to choose catalysts, the thought being that ones known to carry out the hydrogenation would be good candidates for the reverse process. The authors point out that this argument, although suggestive, is not conclusive because the conditions of the forward and backward processes are not the same. Lin et al. previously obtained butyrolactone from the diol but in lower yield with IrH5(PiPr3)2 at 1% loading at a much milder 75 °C.57 The polyol that has attracted the most attention in the context of alternative energy is glycerol. This is because it is a byproduct of the biofuel industry, being the main residue about 10% by weight after transesterification of fats to give biodiesel. Having a very low cost of a few cents per pound, minimal toxicity, and a high production rate, said to be 7 × 109 L/year in the United States alone,58 its upgrading to useful materials is of great interest. Numerous pathways have been suggested,59 but the ones relevant to the present discussion involve catalytic dehydrogenation to the monoaldehydes. In one case dihydroxyacetone was the product but only by a TH route, thus without H2 evolution.60 Conversion to lactic acid (LA) salts is also possible under basic conditions, this time with H2 evolution, for example with Cp*Ir (5),61 N-heterocyclic carbene (NHC) catalysts58 (6), and iron pincer precatalysts (7).62
Dehydrogenative cross-coupling of primary and secondary alcohols was also seen.54 Here the primary alcohol undergoes AAD to give the carboxylic acid, which is then trapped as the ester by the secondary alcohol, implying selectivity in favor of oxidation of the primary alcohol. In a different sequence starting from a mixture of primary and secondary alcohols, AAD produces the ketone and aldehyde, respectively, followed by an aldol reaction to give the α,β-unsaturated ketone. Under hydrogen borrowing conditions, this is hydrogenated to the saturated alcohol in a net redox-neutral reaction, but if
Figure 3. Dehydrogenation of primary and secondary alcohol mixtures can lead to a ketone/aldehyde mixture that can undergo an aldol reaction, followed by complete hydrogenation (hydrogen borrowing) or partial hydrogenation (AAD), depending on the catalyst.
In these cases, dehydrogenation is followed by dehydration and then reincorporation of the lost H2O in an internal Cannizzaro rearrangement to give LA (Figure 4). In effect, the 9232
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
2.4. Alcohol−Amine Combinations
The hydrogen borrowing outcome of alcohol−amine substrate combinations with a reversible dehydrogenation catalyst was illustrated in Figure 1. Since no hydrogen is released, this is a net redox-neutral process and does not fall within the scope of this review. This type of reaction arises when the catalyst securely binds the two H atoms released by the alcohol so that they can be reincorporated into the substrate-derived imine in a later step of the reaction. In contrast, Milstein’s70 catalyst 9 brings about a net oxidative process (eq 11) that relies on the
Figure 4. Glycerol conversion to lactate via AAD.
dehydration step is followed by a rehydration in the Cannizzaro step to form LA. The iridium catalyst is also efficient for conversion of authentic biofuel plant residues, which are brown in color and obviously very impure. Some unusual catalyst deactivation products, [Ir6(IMe)8(CO)2H14]2+, [Ir4(IMe)7(CO)H10]2+, and [Ir4(IMe)8H9]3+, were also identified63,64 (IMe = N,N′-dimethylimidazol-2-ylidene) from these catalysts. Sugar alcohols are also of interest in this connection. Along with glycerol, the C4−C6 analogues erythritol, xylitol, and sorbitol are considered potentially useful biomassderived feedstocks (Figure 5) since they can be derived from
two H atoms released by the alcohol being sufficiently weakly held by the metal so as to be lost as H2, a net oxidation thus becoming possible. The same alcohol/amine combination can now undergo a four-electron oxidation with elimination of 2 equiv of H2. The main significance of the work, though, was that this led to a green amide synthesis from nonstandard RCH2OH/R′NH2 precursors, avoiding the waste generated by the standard RCOOH/R′NH2 coupling when it is promoted by stoichiometric carbodiimides. In view of the importance of amides in many fields, such as pharmaceuticals, this procedure should find multiple applications. The organic part of the pathway adopted is shown in eq 11: catalytic AAD provides the aldehyde; attack by the amine leads to a hemiaminal, normally a precursor to an imine but here intercepted to give the amide by catalytic dehydrogenation with a Ru PNN pincer complex at 0.1% loading. This route was later extended to synthesis of peptides and pyrazines,71 and tertiary amides were also formed from secondary amines under very similar conditions.72
Figure 5. Some sugar alcohols that undergo AAD.
agricultural or lumber residues.65 The same iridium catalysts useful for glycerol also convert these higher polyols into lactic acid; although the selectivity remains high (75−90%) the yield is much reduced (6−40%). Presumably the dehydrogenation is now followed by a retro-aldol step that cleaves a C−C bond, but only with low selectivity. Any C3 fragment produced in this way can go to LA, which is stable under the conditions, but a C1 fragment may be converted to formate and then on to CO2. Ethylene glycol was a significant C2 byproduct in this case.66 Along with glycerol, ethylene glycol was also a substrate for hydrogen generation from a pincer ruthenium system.67 In contrast with the methanol series of eq 6, polyols higher than ethylene glycol tend to have low toxicity,68 a factor that enhances their suitability as energy vectors likely to become widely distributed in the public sphere. Of more synthetic interest, Ngo et al. have reported on the interconversion of glucocorticoids with an iridium catalyst (8). Equation 10 illustrates the transformations, all of which use the same catalyst but under different conditions.69
The two pathways, to amide (eq 11) and to the secondary amine (Figure 1), are finely balanced as shown by the behavior of one ruthenium catalyst that leads to a nearly equimolar mixture of the two products (eq 12).73 Mechanistic work suggests that if the hemiaminal remains metal-bound, it undergoes oxidation to form the amide, but if the hemiaminal is released into solution, it dehydrates to the imine that then goes on to give the secondary amine by a hydrogen borrowing pathway.
9233
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
This general reaction type has also been extended to aromatic heterocycles, such as a pyrrole synthesis from a diol and an amine with a Ru catalyst (eq 13, R = Ph, PhCH2).
In the proposed pathway, catalytic dehydrogenation of the diol to the dione is followed by attack of the amine in what then becomes a Paal−Knorr pyrrole synthesis from dione and amine. Secondary alcohols were preferred, giving yields of 45−50%.74,75This example is unusual in that four H atoms of the substrate are removed in the overall process. Another pyrrole synthesis, this time using catalyst 10, exploits the availability of amino alcohols which are dehydrogenated to amino aldehydes in situ. These normally highly labile intermediates are then immediately trapped by an aromatic ketone, followed by an aldol ring closing step to give 2,5-disubstituted pyrroles in good to moderate yields. An excess equivalent of the ketone also acts as hydrogen acceptor, so in this case H2 is not evolved. Many examples were given including that of the core pyrrole of Lipitor (Figure 6), although in this case the yield was only 33%.76
γ-aminoalcohols has been carried out with a Ru catalyst under basic conditions. The pathway shown in eq 15 is postulated (solid arrows), but an alternative ring closing by a basecatalyzed, Robinson-type annulation to give the dihydropyridine, followed by metal catalyzed dehydrogenation to the quinoline (dotted arrows), seems attractive. In any case, a wide variety of related structures proved to be accessible in moderate to excellent yields.78 Michlik and Kempe79,80 have used an unusual iridium PNP pincer catalyst (11) for similar transformations to give a wide variety of differently functionalized pyridines and pyrroles in good to excellent yields at 90−130 °C over 24 h with NaOtBu as base. The authors also point out that many of the required reactants can be obtained from renewable lignocellulose.
The earliest quinoline synthesis via AAD comes from Ishii and co-workers,81 who used simple IrCl3 or [Ir(cod)Cl]2 catalysts with KOH under solvent-free conditions to convert 2-aminobenzyl alcohol and acetophenone to 2-phenyl quinoline in good to excellent yields. In the AAD step, the aminoalcohol was converted to the aminoaldehyde, with the remaining steps following a Friedländer pathway. A pyrazole synthesis from a 1,3-diol and an alkyl hydrazine proceeds in a similar way via AAD in fair to moderate yield at 110 °C with Rh(CO)H2(PPh3)3/Xantphos at 3% loading.82 The catalyst resting state was not reported, and it seems likely that the yield and loading could be improved by moving to some of the more modern catalysts. A wide variety of pyridines and quinolines were synthesized in 24−80% yield by Milstein and co-workers via AAD of γ-aminoalcohol/secondary alcohol mixtures using their PNN Ru pincer catalyst (9, 0.5% loading, KOtBu, reflux, 24 h).83 A similar benzimidazole synthesis by AAD from an alcohol and benzene-1,2-diamine was extended to a quinoxaline synthesis by moving to a 1,2-diol using the Kempe PNP Ir pincer catalyst (11) and KOtBu in diglyme or THF (eq 16).84 Similar principles apply to a multicomponent synthesis by Kempe and co-workers85 of a wide range of pyrimidines in
Figure 6. Synthesis of the core pyrrole of Lipitor by dehydrogenative oxidation. Ar = p-FC6H4.
A regioselective three-component pyrrole synthesis that proceeds in good to excellent yield via AAD has been reported by Beller and co-workers (eq 14). This involves a ketone, a
1,2-diol, and a primary amine at 130 °C for 16 h, with KOtBu as base and [Ru(p-cymene)Cl2]2/Xantphos as catalyst.77 A dehydrogenative synthesis of pyridines and related heterocycles from an monohydric alcohol and a variety of 9234
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
instead of the nitrile, and the imine can then also undergo nucleophilic attack by the starting amine to give secondary amines and imines, in most cases leading to the “wrong” products or to a complex mixture. In an interesting development, a particularly bulky NNN pincer complex of ruthenium has recently been shown to bring about double amine dehydrogenation at 130 °C with good selectivity (eq 17).90 The preference
good to excellent yields from an amidine, a primary alcohol, and a secondary alcohol catalyzed by their Ir PNP pincer catalyst. A quinazoline synthesis has been reported with [Cp*IrCl2]2 as catalyst that combines nitrile hydrolysis with acceptorless dehydrogenation activity to convert readily accessible precursors to the target compounds. As shown in Figure 7, the nitrile is
for double dehydrogenation over other pathways was attributed to the imine intermediate having a high binding constant to the catalyst as well as fast kinetics for its subsequent dehydrogenation. As in the case of eq 12, retention of the imine intermediate by the metal leads to net oxidation; in contrast, release often leads to a redox-neutral hydrogen borrowing pathway.
Figure 7. Quinazoline synthesis that includes a dehydrogenative oxidation step catalyzed by [Cp*IrCl2]2.
hydrolyzed to the amide by the oxime cosubstrate, and the aldehyde cosubstrate then reacts with the amine group α to the newly formed amide to give a dihydroquinazolinone that is subsequently dehydrogenated to the final product.86 This group has also reported similar chemistry elsewhere.87 Since secondary alcohols go to ketones under AAD conditions, it is not surprising to find that the pathway can be redirected to form hydrazones by addition of a suitable hydrazine. This proved possible at 130 °C with [Cp*IrCl2]2 (0.5 mol %) as catalyst and KOH as base.88 A related approach has led to a synthesis of ureas in good to excellent yields with a PNP Ru pincer catalyst closely related to 7 (0.5 mol %) under base-free conditions at 140 °C for 24 h (Figure 8). In this case methanol provides the carbonyl group
3. CATALYSTS AND MECHANISMS This section emphasizes the inorganic aspectscatalyst structures and mechanism. Literature reports indicate that the historically important AAD catalysts were often based on the precious metals Ru and Ir but nonprecious metal catalysts are now appearing in greater numbers. Particularly for nonprecious metals it is common to find pincer or related multidentate ligands, which have a number of favorable characteristics. Their multidentate character is expected to enhance the thermal stability of the catalyst, a useful property in view of the relatively high temperatures and agressively basic media often involved in the AAD reaction. Many of the ligands are noninnocent, typically undergoing deprotonation, a process that generates a ligand-centered basic site that is in principle able to abstract a proton from the alcohol. The resulting alkoxide may bind to the metal or else directly transfer a hydride from the α-CH position to the metal. In the former case, a β elimination can ensue (eq 18) or, in the latter, a transfer of (H+ + H−) to the complex (eq 19). Thus, we can have either inner or outer sphere
Figure 8. Synthesis of unsymmetrical ureas by dehydrogenative oxidation of MeOH.
with one or more amines providing the base and coreactant. With only one type of amine present, symmetrical ureas can be obtained, often in good to high yield. For unsymmetrical ureas to be favored, the reaction is run with just one type of amine until the formamide stage is reached. The second type of amine is then added to complete the synthesis, which can often give good to excellent yields. No hydrogen acceptor is present, so 3 equiv of H2 are released in the process.89 Although acceptorless amine dehydrogenation goes beyond the strict bounds of the present review, a brief mention is needed. One potentially attractive feature of amines of the type RCH2NH2 is the possibility of double dehydrogenation to the nitrile, RCN, a process that would approximately double the mass percent H storable over the corresponding alcohol. Unfortunately, many catalysts produce imines RCHNH
mechanisms depending on the details of the system. By avoiding the requirement for oxidative addition, these steps make it easier for nonprecious metals to participate since these metals often prefer one-electron to two-electron redox steps. Indeed, in several of the cycles to be discussed in this section there is no change at all in the metal oxidation state, a feature that in principle opens the way to employment of nontransition main group elements. 9235
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
3.1. Osmium and Ruthenium Catalysts
Work in methanol dehydrogenation began with RuCl3·3H2O, which needed a high catalyst loading and a big excess of strong base, typically NaOMe, but the resulting catalysts were relatively inefficient, giving HCOONa as product at 80 °C.39,91 By introducing ancillary ligands, both the catalyst loading and the amount of base could be reduced. For example, Maitlis92 preferred [RuCl2(PPh3)3] as precatalyst, a choice which avoided the need for a base and permitted a lower catalyst loading but required 150 °C. In this work, MeOOCH and CH2(OMe)2 were both formed from MeOH. The influential Shvo catalyst, [(Ph4C4CO)Ru(CO)3] (2, eq 20; Ph groups omitted;
the mechanisms involved in the wide variety of reactions of these catalysts by reviewing existing density functional theory (DFT) work and adding their own contributions; they conclude that the deprotonation−dearomatization step of eq 21 is not always necessary, in which case a conventional inner sphere mechanism can take over. The full dehydrogenation of aqueous methanol to CO2 by eq 6, also termed “methanol reforming”, has proved possible with catalysts of types 12 (M = Ru or Fe; R = Ph or iPr; X = Cl or BH4),102 13 (R = tBu; X = Cl),103 and 14104 (Scheme 1). Scheme 1. Some AAD Catalysts
open box = vacant site), was also shown to dehydrogenate alcohols.93 The tricarbonyl first loses CO and is reduced to generate the dinuclear resting state shown in eq 20.46 This in turn dissociates to give the catalytically active monomer. The cyclopentadienone oxygen then provides a pendant basic site that acts in concert with the electrophilic metal to abstract 2H from the substrate in the form of (H+ + H−); the resulting reduced catalyst then liberates H2. This result was important in that it introduced metal−ligand cooperation into the AAD field for the first time. In another early report, Cole-Hamilton saw H2 release from C1−C4 alcohols with NaOH and [RuH2(N2)PPh3)3] as catalyst, again at 150 °C.94 The authors argue that the fact that these complexes have a high tendency to form dihydrogen complexes leads to easy H2 loss and thus net alcohol oxidation. Perhaps as a result of chelation, ethylene glycol was the best substrate with a turnover frequency (TOF) of 1185 h−1. The high reaction temperature of 150 °C is sufficient to raise concerns about homogeneous catalyst stability, however. Since reductive elimination of H2 from the catalyst is required for dehydrogenative oxidation, the presence of high trans effect, electron-withdrawing ligands on the metal might be helpful. The NHCs mentioned above partly fulfill these criteria, since they are indeed high trans effect ligands. Likewise, the trihalostannyl group, not only a high trans effect ligand but also a strong π acceptor, found early use as a ligand in the catalysts [Ru(SnCl3 )5 (PPh3)]3−,95 [CpRu(SnF3 )(PPh3) 2],96,97 and [Ru(SnCl3)2(P{OMe}3)3],98 where the phosphite added to the electron withdrawing character of the ligand set. An important advance was the introduction of pincers as a ligand class by Shaw99 and as a ligand in catalysts for AAD reactions by Milstein and co-workers.100 The initial study described the alcohol to ester conversion of eq 8, which occurs in the range of 115−157 °C over 2.5−24 h with nearquantitative yields in many cases. Deprotonation of one of the benzylic positions (eq 21) creates an internal base that permits proton abstraction from the substrate, a key part of any AAD mechanism. Hall and co-worker101 have helped clarify some of
This pathway has the great advantage of producing not just 1 or 2, but 3 equiv of H2 per substrate molecule. An additional constraint on such a catalyst is the need for water stability, a feature no doubt enhanced by the presence of PNP (12) and PNN (13) pincers or even of tetradentate ligands (14), rather than monodentate ones. Indeed, the iPr complex of type 12 proved to be extremely robust, giving a turnover number (TON) of 3.5 × 105. A clue to the mechanism in the case of 12 is the fact that the NH of the pincer needed to be deprotonated to give the active catalyst. This step provides a basic N adjacent to the electrophilic metal, suggesting the possibility of a Noyoritype, outer sphere mechanism (eq 18.)105,106 in which MeOH, CH2(OH)2, and HCOOH successively transfer H− from the substrate CH to the metal and H+ from the substrate OH to the adjacent nitrogen. Free formate can be detected as an intermediate, but no formaldehyde is seen, suggesting that formate dehydrogenation may be turnover limiting as DFT work also suggests.107 Catalyst 12 (X = H) also efficiently dehydrogenates alcohols such as iPrOH by the outer sphere Noyori mechanism with a TOF of >8000 h−1 over 2 h.36 The Milstein catalyst, 13, is noninnocent in that the benzylic position of the ligand is easily deprotonated. In the key step of the mechanism for dehydrogenative oxidation of alcohols, this proton is believed to be transferred to the adjacent Ru hydride, where it forms an H2 complex leading to H2 dissociation. This proton loss generates an internal base site that participates in the mechanism by deprotonating a water molecule prior to nucleophilic attack by OH− on the intermediate aldehyde complex.53 9236
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
fully as effective in stabilizing both the Ru(0) and the Ru(II) states in the proposed cycle and devising similar disparate combinations could thus become a general strategy for catalyst design. The ligand conformation and consequent absence of cis vacant sites may inhibit the classic β elimination of an alkoxide intermediate that would normally be the product-forming step with a more conventional catalyst. Instead we have direct intervention of the ligand by a transfer of the β CH hydride to the carbon of the ligand CN bond (steps c and e in Figure 9) in close analogy with an Oppenauer oxidation2 in organic chemistry (Figure 10). Likewise, instead of the more
Closely related catalysts of structure [RuH(BH4)(pincer)] (pincer = PNP or PNN type)108 convert primary and secondary alcohols to esters (eq 8) and ketones, respectively, as well as converting diols to lactones (eq 9) with high yields and turnover numbers (∼1000). The PNN catalyst could also reverse the AAD process by ester hydrogenation. Thus, in THF solution with butyl butyrate under H2 (10 atm) at 110 °C for 12 h, this catalyst gave 97% 1-butanol. Catalyst 14104 operates at 90 °C with 0.5% catalyst loading (TON = 540 after 10 h) by a mechanism that is still under discussion but in which the noninnocent character of the ligand plays a clear role. Figure 9 shows a possible catalytic cycle,
Figure 10. Key mechanistic step of the Oppenauer oxidation2 (left) compared with that in step c of Figure 9 (right), where the N-donor ligand of 3 becomes the hydride acceptor but the substrate alcohol remains the hydride donor.
conventional oxidative addition step to install the substrate, we have a intramolecular deprotonation of the substrate −OH by the N-donor atom (step b in Figure 9). Another unusual step is step f in Figure 9, also repeated in step a. As a speculative suggestion, this may involve base promoted β-elimination, followed by protonation of the hydride by the conjugate acid of the base with release of H2. A very recent mechanistic study by Li and Hall110 confirms the outer sphere character of the reaction and the noninnocence of the ligand. The new mechanism differs from the prior proposal, however, in having the MeOH transfer two of its hydrogen atoms to a ligand N atom and to a C atom of the diazadiene unit of the ligand which makes the metal much more of a spectator in the mechanism. Gauvin and co-workers111 have examined a series of Ru PNP pincer precatalysts for conversion of primary alcohols into the corresponding carboxylates. For example, NaOH-promoted reaction of 1-butanol yields butyrate at 130 °C over 3 h with TONs of 500−4000. Air neither inhibited nor promoted the reaction and recyclability was excellent over five runs. A DFT study112 compared outer sphere versus inner sphere AAD mechanisms for two different Ru complexes, one with the NNN pincer of eq 17 that goes by an entirely metal-based inner sphere mechanism and the other with the PNN pincer 13 (R = But) that requires metal−ligand cooperation (MLC). The difference in behavior was correlated with the ease of tautomerization of eq 21, favored for the PNN pincer but not accessible for the NNN case. The MLC mechanism of course requires the tautomeric product forms of eq 21 to be easily available because these forms are required to permit substrate deprotonation by the ligand. As might be expected, an inner sphere strategy was also favored where two vacant or at least labile cis sites were available at the metal, making the NNN system suitable for β elimination of an alkoxide ligand; in contrast, an MLC mechanism only requires one such site at the metal. NHC ligands are still relatively rare in the field, but their high trans effect should help promote the departure of H2 from the metal. Indeed, Peris, Albrecht, and co-worker113 have looked at complex 15, an unusual example of an “abnormal” N-heterocyclic carbene (NHC). This gives dehydrogenative oxidation of benzyl
Figure 9. Possible catalytic cycle for complex 14, adapted from the reported one104 and covering just the CH3OH to HCOOH part of the total sequence.
adapted from the reported one, going from MeOH to HCOOH. Catalyst 14 also very efficiently decomposes HCOOH to H2 and CO2, completing the process. Complex 14 is particularly interesting in several respects. The ligand choice is very unusual in this type of homogeneous catalysis, where P-donors and unsaturated N-donors are most often encountered, but almost never an alkene as in 14. Alkenes can of course be placeholder ligands that depart under the catalytic conditions, such as cyclooctadiene in the [(cod)IrLL′]BF4 hydrogenation catalyst series,109 but such lability does not seem to be the case here. Alkenes can also be substrates in a catalytic reaction, as in the Wacker process, but then they are actor, not simply spectator, ligands, again unlike the present case. Classic P-donor ligands are often seen, no doubt because they have intermediate hard−soft character that stabilizes both the reduced state and the oxidized state of the metal in a typical catalytic cycle. In 14, in contrast, we have a pair of much softer alkenes combined with a pair of much harder N-donors. The N-donors being trans to the alkenes may help stabilize both metal−ligand bonds by a push−pull “antisymbiotic” effect. This ligand combination seems to be 9237
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
alcohol at 70 °C in the absence of base with improved performance over the “normal” IBu NHC analogue. Because the normally firmly bound Cp* is lost from Cp*Ir NHC catalysts under similar catalytic conditions,114 as discussed in section 3.2, loss of the arene seems plausible in this case. The same catalyst at 5% loading is also effective for dehydrogenative oxidative coupling of hexylamine and PhCH2OH to form the corresponding amide in >95% yield after 20 h at reflux.
3.2. Iridium and Rhodium Catalysts
The price rise in rhodium which kept it above $2000/oz. for most of the time period from mid-2005 to mid-2011 discouraged work in the area, and in those years the comparatively much less expensive iridium came to be dominant over rhodium in homogeneous catalysis in general and in AAD in particular. Because [Cp*IrCl2]2 is commercially available, it is perhaps the most easily accessible AAD catalyst. It was first used for hydrogen borrowing conversions but more recently for AAD, as noted in section 2.88 Unlike the ruthenium case, the ligands typically chosen for iridium, such as P-donors, NHCs, or Cp*, are more often relatively conventional. Conventional they may be, but they can do some very unconventional things under catalytic conditions: in one recent case, Cp* loss from Cp*Ir(IMe)2Cl (IMe = N,N′-dimethylimidazolylidene) was identified as the precatalyst activation step.114 Since Cp and Cp* are currently considered as reliable spectator ligands, at least in the absence of harsh primary oxidants,117 this work provides a clue hinting that such ligand loss may be much more common than currently believed. Such a loss is expected to open up multiple labile sites at the metal that could favor catalysis, as long as some ligands remain to stabilize the complex and thus prevent metal deposition or other deactivation pathway. In this case the NHCs were shown to resist detachment from the metal under the catalytic conditions. Among all the transition metals, IrIII−L bonds are among the most stable both thermodynamically and kinetically, so if Cp*−Ir bonds are labile under AAD conditions, the same lability must be considered as a possibility for the other metals as well. The reversible deprotonation of a 2-pyridinol ligand was exploited by Fujita, Yamaguchi, and co-worker for AAD of secondary alcohols.118 2-Pyridinol is of interest both from the presence of an organometallic derivative in an Fe hydrogenase enzyme active site119 and for its ability to hydrogen bond to adjacent functionality, such as a hydride.120,121 The ligand has also proved widely useful beyond AAD chemistry.122 Catalyst 18, shown nearby, gave the best AAD results (TON = 2120) compared to the analogous complexes of 3-pyridinol and 4-pyridinol. The proposed cycle is shown in Figure 11.
In the hands of Bruneau, Fischmeister, and co-worker, the IMes (IMes = N,N′-dimesitylimidazol-2-ylidene) analogue of 4 gave acceptorless dehydrogenation of alcohols with K3PO4 as base.115 Schley et al.74 looked at catalyst 16 for dehydrogenative lactam and pyrrole (eq 12) formation at 125 °C with a flow of N2 to continuously remove the H2 formed. Computational data indicated the intermediacy of an H2 complex favorable to dihydrogen loss and thus to net oxidation. An unexpected feature of the intermediate hemiaminalate complex came from DFT calculations on a model system. This identified an intraligand hydrogen bond between the hemiaminal nitrogen and either the aminomethyl NH group or the dihydrogen ligand (eq 22) with the former having a much lower barrier for the key H2 loss step. Once again, the tendency to prefer dihydrogen over dihydride complexation in the intermediate seems to favor loss of H2 and therefore also net oxidation.
An unusual tridentate ligand was employed by Sun and co-workers78 as a precatalyst (17) for the quinoline synthesis previously mentioned (eq 15). A very high efficiency was found, with a Ru loading of just 0.025 mol % giving very satisfactory results in this multistep dehydrogenative pathway. This pincer is unusual in preferring a fac configuration, validated by an X-ray crystal structure. Once again, we see an unusual ligand type appearing in the dehydrogenative oxidation literature, and in this case one which is chiral, thus providing additional possibilities for future applications in asymmetric catalysis.
Figure 11. Catalytic cycle proposed for the Fujita−Yamaguchi catalyst, 18 ([Ir] = Cp*ClIr).
Mechanistic work on the role of the cooperative ligand in this system has come from Rauchfuss and co-workers.123 They also saw [Cp*2Ir2H2L]+ and [Cp*IrHCl(LH)] (LH = 2-hydroxypyridine) in the reaction medium and considered these to be intermediates. A related and improved ligand with several successful applications in the field is illustrated in eq 23 along with its
Osmium is much more rarely employed in homogeneous catalysis than ruthenium, but where the two can be directly compared for AAD, Os does comparably well.116 Dehydrogenative catalysis with Os has also very recently been reviewed.28 9238
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
by a Diels−Alder cycloaddition, this framework provides the usual PCP pincer unit but with two novel points. The C-donor is not sp2 type as is usually formed by arene cyclometalation but rather an sp3 type formed by alkyl group cyclometalation. The presence of this sp3 carbon must enhance both the trans effect and the donor power of the central ligand unit. Second, the pendant hydroxymethyl group appears to be able to intervene directly in the mechanism by binding of the alkoxide form to the metal, the cycle of Figure 12 being proposed. The catalyst is
reversible deprotonation behavior, although this characteristic was not thought to be involved in the mechanism.124−126 The monoanion is a good catalyst for the complete dehydrogenation
of aqueous MeOH/NaOH via eq 5; reflux over 20 h gave up to 84% yield of H2. As expected from eq 5, formaldehyde and formic acid were also viable substrates. A long-term study over 150 h produced a TON of over 104, showing the robust character of the ligand, which should be considered as analogous not to an oxidation-sensitive phenol but to an oxidation-stable lactam. Complete departure of Cp* may again be occurring because otherwise only one labile site is present, in which case it would be hard for the proposed [Cp*(L)IrOMe]− intermediate to β-eliminate without any cis vacancy being present. A similar situation exists for other Cp*Ir(chelate)L catalysts where, if complete departure is not considered, Cp* slip to an η1 form has to be proposed to generate enough sites for β-elimination.127 Historically, pincer ligands take pride of place in the iridium area, thanks to work from Jensen and from Goldman, who showed how these ligands produce very robust catalysts, particularly applicable to alkane dehydrogenation but also active for AAD. According to Werkmeister et al.32 more papers featuring pincers have appeared for iridium than for any other metal. The pincer series, illustrated in diagram 19, has another advantagethere are numerous points at which the structure can be usefully varied, as detailed in a recent review.13 The PR2 group is mainly involved in steric protection of the active site; the X group, typically C or N, is involved in electronic effects, notably trans influence (C > N); the Y group, typically CH2, NH, or O, modifies the electronics both of the neighboring P donor and of the aryl ring; and the Z substituent has some effect on the electron density of the aryl ring and can also enhance solubility. So far the great majority of alcohol dehydrogenation reactions in the pincer series involve the presence of a hydrogen acceptor, and so do not fall into the AAD area.
Figure 12. Proposed mechanism of action of the Gelman catalyst.
also well set up to permit dihydrogen bonding (21) between the ligand OH and the adjacent hydride, a structure known to facilitate proton transfer to give a dihydrogen complex that can subsequently lead to loss of H2.130 The catalyst was effective for AAD in refluxing p-xylene at 0.1% loading for a wide variety of primary and secondary alcohols either with or without base. Dehydrogenative reaction of primary and secondary alcohols to give the cross-coupled ketone was also seen.131 Taddei and coworkers saw the same type of net oxidative cross-coupling with [Ru(cod)Cl2]n/PTA (PTA = 1,3,5-triaza-7-phosphaadamantane) catalyst.132
Among NHC-based Ir catalysts, Albrecht and co-workers have looked at an unusual binucleating ligand which forms the dinuclear abnormal NHC precatalyst, 22. This dehydrogenates benzyl alcohol at 150 °C,133 but given prior work,114 Cp* loss cannot be entirely excluded.
A particularly unusual NHC catalyst comes from Tu and coworkers,134 who describe an iridium coordination polymer linker by biface bis-NHC linkers of type 23 (R = Me or nBu) that is particularly active for the glycerol to lactate conversion in the presence of KOH at only 0.0003% loading. Because of its insolubility, it could be easily recycled for more than 30 separate runs. This provides an excellent example of the advantages that can sometimes accrue from heterogenization of a molecular
A series of PCP pincer complexes of type 19 (e.g., X = C; Y = O; R = tBu; Z = H; M = IrHCl) in only 0.025% loading is also effective for AAD of a variety of secondary alcohols with RONa as base and giving TONs up to 3500, but primary alcohols undergo stoichiometric decarbonylation instead.128 A pincer series of a very unusual type (20, [Ir] = IrHCl) has been reported by Gelman and co-workers.129 Easily constructed 9239
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
of H2. The activity of the NMe pincer was similar, so no special role was assigned to the NH group of 26.
catalyst and highlights the possibilities of polymeric catalysts where strongly held biface ligands act as linkers.
Although Rh is rarely seen in the context of AAD, [Cp*RhCl2]2/2,2′:6′,2″-terpyridine has been reported by Wang, Xiao, and co-workers135 as a catalyst for the dehydrogenative cross-coupling of a series of benzylic alcohols or aldehydes with a series of aliphatic alcohols. In each case the benzylic alcohol component provides the carboxyl group of the resulting ester and the alcohol provides the alkoxy group (eq 24). A series of benzylic alcohols was also successfully cross-coupled with the same series of aliphatic alcohols, again with H2 release. The phenyl group of the substrate could be successfully replaced by polycyclic aryls such as 2-pyryl and heterocycles such as 2-furyl, 2-thiophenyl, 4-pyridinyl, 2-quinolinyl, or 2-indolyl. As we saw above with a Cp*Ir system,114 the Cp* is lost in the activation step, although this aspect was not particularly emphasized by the authors. The activation step results in a dirhodium(II) complex (24, L = terpyridine) that is considered to be the active catalyst. This is not the only dinuclear AAD catalyst that has emerged from recent work: Bera and co-workers136 have employed a diruthenium(II) catalyst (25) in which a naphthyridine-diimine links the two centers. Like the diruthenium system, this is selective for benzylic alcohols, but the product in this case is the aldehyde.
The effectiveness32 of the PNP ligand set is emphasized by the fact that a number of iron pincer catalysts of type 12 (M = Fe, R = Pri or C6H11, X = BH4) effect AAD with reasonably good efficiency (TON = 9.1 × 103),139,140 so if one criterion for sustainability106 is adoptedturnovers per unit catalyst costthe iron catalyst clearly still has the advantage. W. D. Jones and co-workers141 reported base-free acceptorless heterocycle dehydrogenation along with the alcohol case, including diol conversion to the corresponding lactones.142 Contrary to some of the prior cases mentioned, secondary alcohols were selectively converted in the presence of primary ones. Reversibility was also achieved with hydrogenation occurring under mild conditions (1 atm, 20 °C).143 The proposed mechanism of the dehydrogenation shown in Figure 13 has the metal maintaining the Fe(II) state
Figure 13. Proposed catalytic cycle for AAD with the pincer catalyst shown (R = Pri or C6H11). The metal does not change oxidation state as a result of metal−ligand cooperation.
throughout, thanks to the metal−ligand cooperation involved in the case of 27 (Figure 13). This is relevant to the problem of replacing precious metals by cheap ones because catalytic cycles invoked for the precious metals typically involve two electron changes in oxidation state during their ubiquitous oxidative addition and reductive elimination steps. Such oxidation state changes are nonstandard for cheap metals such as Fe, where the one electron change from Fe(II) to Fe(III) is more typical. If a catalytic cycle can be constructed that does not involve any change in oxidation state, cheap metals clearly become much more plausible choices relative to their precious metal rivals. In fact, even main group elements should be viable. An example of this behavior144 comes from the Berben group, where an Al(III) complex of an NNN pincer has been shown to catalyze the dehydrogenative coupling of PhCH2NH2 to give PhCHNCH2Ph, H2, and NH3. Indeed, main group chemistry is currently on the rise145 and element−ligand cooperation could be a useful strategy to promote catalytic activity in these elements, thus extending the range of
3.3. Base Metal Catalysts
Some success has been achieved in base metal catalysis of AAD. Among ligands, PNP pincers are the most commonly encountered, but these are early days and many of the more sophisticated ligands incorporating metal−ligand cooperation as described in earlier sections are yet to be applied to the base metal problem. Since net oxidation requires dihydrogen loss from the metal, first-row metals with their generally lower M−L bond strengths seem good candidates, so much more can be expected from this area in the future. The first report on AAD with a cobalt catalyst comes from Hanson and co-workers, whose PNP pincer catalyst (26) is active for the dehydrogenation of secondary alcohols to ketones.137,138 The ancillary ketone ligand in 26 originated from a cyclometalation of a Co(I) precursor with the oxidation product from 1-phenylethanol dehydrogenation, acetophenone. A Co(I/III) cycle was suggested with a conventional β-elimination of an alkoxide followed by a reductive elimination 9240
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
homogeneous catalysis to a class of elements relatively neglected for catalytic applications. One of the longest known iron organometallics, CpFe(CO)2Cl, proved to be a catalyst for AAD, but the substrates are limited to 2-hydroxymethylpyridine and its derivatives and the very strong base NaH was required. The 3-hydroxymethylpyridine isomer did not undergo AAD, however, so the pyridyl group is believed to be involved in allowing the substrate to bind to the catalyst with displacement of CO giving an N,O chelate, CO being a ligand that is relatively hard to displace. The activity is excellent, with TON values up to 6.7 × 104. A conventional β-elimination mechanism was suggested, going via the chelated alkoxide and “CpFe(CO)H”.146 Going against the trend in favor of exotic ligands discussed above, very simple ligand sets have also proved somewhat effective: Hong and co-workers reported that Fe(III) acetylacetonate (8.5 mol %), 1,10-phenanthroline, and K2CO3 were effective in refluxing toluene over 48 h for the AAD of 1-phenylethanol,147 although, even here, a carbonate ligand has the potential to act as an internal base, and so may not be entirely innocent.148 Bernskoetter, Hazari, Holthausen, and co-workers149 have reported that the PNP pincer Fe complex 27 benefits from the presence of a cocatalytic amount of a Lewis acid activator such as LiBF4 for base-free, aqueous phase methanol dehydrogenation to give very impressive turnover numbers of up to 51 000. At least for this case, acid activation is superior to the more common base activation because the activator loading is lower and the reaction conditions are milder (EtOAc reflux) than in the standard base case. Kinetic data and DFT calculations are consistent with a four-step mechanism. Initial dehydrogenation of MeOH to formaldehyde is followed by hydration to CH2(OH)2. Dehydrogenation of this diol then yields H2 and an iron formate that decarboxylates to provide the second equivalent of H2. The Lewis acid speeds the reaction by facilitating dissociation of the formate, a step that is required for decarboxylation to occur because the formate CH bond, not the carboxylate oxygen, must interact with the metal for this step to occur. Nickel in the form of catalyst 28 has also proved a competent AAD catalyst, mostly in excellent conversions and yields, as shown by W. D. Jones and co-workers.150 Instead of the classical Cp ligand, a Cp analogue, tris(3,5-dimethylpyrazolyl)borate (=Tp′), proved successful in this case, as did a series of 2-hydroxyquinolines in the role of ancillary ligand. In addition, full reversibility of the dehydrogenation was achieved and the reactions were remarkably tolerant of a variety of substituents. Deprotonation of the parent 2-OH group of the hydroxyquinoline precursor provides a potentially pendant base which acts in concert with the metal to abstract (H+ + H−) from the substrate, probably in a concerted way. Dehydrogenation of cholesterol to cholesterone also proved possible without CC bond migration as would probably have occurred if an inner sphere mechanism had been adopted with a precious metal. Metal−ligand cooperation may therefore have selectivity advantages in avoiding this common side reaction in precious metal homogeneous catalysis.
Other than in oxidation driven by a primary oxidant, manganese is an unusual element in homogeneous catalysis, but the Milstein group151 now reports that their PNP pincer system permits alcohol−amine coupling in this case, too. Instead of giving the usual secondary amine by hydrogen borrowing, however, net oxidation takes place to give the imine in moderate to excellent yield, although the reaction is slow, requiring 60 h at ∼135 °C. As before, deprotonation of the pincer is needed to access the active form (29, R = tBu). This form provides a ligand-based site for transfer of a proton from the substrate alcohol while a hydride from the alcohol α-CH is transferred to the metal, all occurring in an outer sphere process. This once again allows the whole cycle of Figure 14 to proceed without change of oxidation state.
Figure 14. Proposed cycle for the Milstein Mn catalyst showing the direct intervention of the ligand, permitting the metal to remain Mn(I) throughout the cycle.
4. CONCLUSIONS From an organic chemistry perspective, AAD offers promise as a green oxidation procedure, minimizing waste formation and providing a source of potentially useful H2. Extension of these ideas beyond alcohols to amines and to dehydrogenative synthesis of aromatics, specially heteroaromatics, seems fully realistic. Asymmetric catalysis, by which only one enantiomer of a chiral substrate is oxidized, is an entirely unexplored aspect of the field. From the point of view of homogeneous catalysis, AAD provides a good test bed for two modern trends in the field. How can we best use multifunctional ligands to enhance catalytic activity, and can they help in transitioning from precious metal catalysis, that has dominated the field up to now, toward cheap metals, such as the first row of the transition metals? Since some redox active ligands permit the metal to remain in the same oxidation state throughout the catalytic cycle, first row, and even main group, elements seem plausible candidates. Another point of interest in AAD is the common occurrence of outer sphere catalysis, where the substrate is never directly bound to the metal but instead transfers (H+ + H−) to the ligand and metal, respectively. Noninnocent ligands are of rising interest because they can usefully add to the functionality of the complex. This useful combination of organocatalysis with transition metal catalysis is only at an early stage of development but is likely to be a fruitful concept for future development.3,152 Electrocatalytic oxidation is a neglected area that may gain more attention9 in future. In such a system (H+ + 2e) could be 9241
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
(9) Crabtree, R. H. Hydrogen Storage in Liquid Organic Heterocycles. Energy Environ. Sci. 2008, 1, 134−138. (10) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol Economy; Wiley-VCH: Weinheim, 2006. (11) Crabtree, R. H. The Stability of Organometallic Ligands in Oxidation Catalysis. J. Organomet. Chem. 2014, 751, 174−180. (12) Tomioka, H.; Takai, K.; Oshima, K.; Nozaki, H. Selective Oxidation of a Primary Hydroxyl in the Presence of Secondary One. Tetrahedron Lett. 1981, 22, 1605−1608. (13) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A. S. Dehydrogenation and Related Reactions Catalyzed by Iridium Pincer Complexes. Chem. Rev. 2011, 111, 1761−1779. (14) Crabtree, R. H. The Organometallic Chemistry of Alkanes. Chem. Rev. 1985, 85, 245−269. (15) Fujii, T.; Saito, Y. Thermocatalytic formation of MolecularHydrogen and Cyclo-octene from Cyclo-octane by Rhodium Complexes. J. Chem. Soc., Chem. Commun. 1990, 757−758. (16) Yukawa, K.; Fujii, T.; Saito, Y. Dehydrogeno-Aromatization of Cyclohexanes with Suspended Noble-Metal Catalysts. J. Chem. Soc., Chem. Commun. 1991, 1548−1549. (17) Aoki, T.; Crabtree, R. H. Homogeneous Tungsten, Rhenium, and Iridium Catalysts in Alkane Dehydrogenation Driven by Reflux of Substrate or of Cosolvent or by Inert Gas Flow. Organometallics 1993, 12, 294−298. (18) Jumde, V. R.; Cini, E.; Porcheddu, A.; Taddei, M. MetalCatalyzed Tandem 1,4-Benzodiazepine Synthesis Based on Two Hydrogen-Transfer Reactions. Eur. J. Org. Chem. 2015, 2015, 1068− 1074. (19) Nixon, T. D.; Whittlesey, M. K.; Williams, J. M. J. Transition Metal Catalysed Reactions of alcohols using Borrowing Hydrogen Methodology. Dalton Trans. 2009, 753−762. (20) Dobereiner, G.; Crabtree, R. H. Dehydrogenation as a Substrate-Activating Strategy in Homogeneous Transition Metal Catalysis. Chem. Rev. 2010, 110, 681−703. (21) Huang, F.; Liu, Z.; Yu, Z. C-Alkylation of Ketones and Related Compounds by Alcohols: Transition-Metal-Catalyzed Dehydrogenation. Angew. Chem., Int. Ed. 2016, 55, 862−875. (22) Nandakumar, A.; Midya, S. P.; Landge, V. G.; Balaraman, E. Transition-Metal-Catalyzed Hydrogen-Transfer Annulations: Access to Heterocyclic Scaffolds. Angew. Chem., Int. Ed. 2015, 54, 11022− 11034. (23) Alberico, E.; Nielsen, M. Towards a Methanol Economy based on Homogeneous Catalysis: Methanol to H2 and CO2 to Methanol. Chem. Commun. 2015, 51, 6714−6725. (24) Nielsen, M. Hydrogen Production by Homogeneous Catalysis: Alcohol Acceptorless Dehydrogenation. In Hydrogen Production and Remediation of Carbon and Pollutants; Lichtfouse, E., Schwarzbauer, J., Robert, D., Eds.; Springer: Heidelberg, 2015. (25) Johnson, T. C.; Morris, D. J.; Wills, M. Hydrogen Generation from Formic Acid and Alcohols using Homogeneous Catalysts. Chem. Soc. Rev. 2010, 39, 81−88. (26) Trincado, M.; Banerjee, D.; Grützmacher, H. Molecular Catalysts for Hydrogen Production from Alcohols. Energy Environ. Sci. 2014, 7, 2464−2503. (27) Chamoun, R.; Demirci, U. B.; Miele, P. Cyclic Dehydrogenation−(Re)Hydrogenation with Hydrogen-Storage Materials. Energy Technol. 2015, 3, 100−117. (28) Chelucci, G.; Baldino, S.; Baratta, W. Ruthenium and Osmium Complexes containing 2-(aminomethyl)pyridine (Ampy)-based ligands in Catalysis. Coord. Chem. Rev. 2015, 300, 29−85. (29) Soloveichik, G. L. Liquid Fuel Cells. Beilstein J. Nanotechnol. 2014, 5, 1399−1418. (30) Wang, W.-H.; Himeda, Y.; Muckerman, J. T.; Manbeck, G. F.; Fujita, E. CO2 Hydrogenation to Formate and Methanol as an Alternative to Photo- and Electrochemical CO2 Reduction. Chem. Rev. 2015, 115, 12936−12973. (31) Kopylovich, M. N.; Ribeiro, A. P. C.; Alegria, E. C. B. A.; Martins, N. M. R.; Martins, L.M.D.R.S.; Pombeiro, A. J. L. Catalytic
released rather than H2, thus directly interconverting electrical and chemical energy. In a possible future biomass-based energy economy, chemical commodities that are now made from fossil fuels will still need to be produced, but in a new way. The conversion of glycerol to lactic acid is just one example of sourcing one of these materials from a biodiesel residue. Bioderived alcohols could also be an easily storable form of H2, if hydrogen begins to play a more important role in the energy economy. In looking for improved catalysts for AAD, the ones known to be effective in hydrogen borrowing provide a plausible starting point. Destabilizing the hydride intermediates by ligand modifications discussed above should lead to H2 loss and net oxidation of the substrate. Finally, workers in the area have developed some very unusual ligand designs that incorporate functionality that go far beyond simple metal-binding and should provide inspiration for future exploration.
AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected]. ORCID
Robert H. Crabtree: 0000-0002-6639-8707 Notes
The author declares no competing financial interest. Biography Bob Crabtree, educated at New College, Oxford, with Malcolm Green, did his Ph.D. with Joseph Chatt at Sussex and spent four years in Paris with Hugh Felkin at the CNRS, Gif. At Yale since 1977, he is now a Whitehead Professor. He has been an ACS and RSC organometallic chemistry awardee, Baylor Medallist, Mond lecturer, Kosolapoff awardee, and Stauffer Lecturer, has chaired the ACS Inorganic Division, and is the author of an organometallic textbook, now in its sixth edition. Early work on catalytic alkane C−H activation and functionalization was followed by work on H2 complexes, dihydrogen bonding, and catalysis for water oxidation and alcohol dehydrogenation. He is a Fellow of the ACS, RSC, IUPAC, and the American Academy of Arts and Sciences.
ACKNOWLEDGMENTS I thank the TomKat Charitable Trust for a generous gift and the U.S. Department of Energy, Basic Energy Sciences, for Grants DE-FG02-07ER15909 and DE-SC0001059. REFERENCES (1) Sheldon, R. A. Green Chemistry and Resource Efficiency: towards a Green Economy. Green Chem. 2016, 18, 3180−3183. (2) Ley, S. V.; Madin, A. Comprehensive Organic Synthesis; Pergamon: Oxford, 1991; Vol. 7, Chapter 2.7−2.8, pp 348−389. (3) Crabtree, R. H. Multifunctional Ligands in Transition Metal Catalysis. New J. Chem. 2011, 35, 18−23. (4) Haldane, J. B. S. Daedalus; or, Science and the Future; Paul, Trench, Trubner & Co.: London, 1924. (5) Bossel, U. Does a Hydrogen Economy Make Sense? Proc. IEEE 2006, 94, 1826−1837. (6) Zhu, Q.-L.; Xu, Q. Liquid organic and Inorganic Chemical Hydrides for High-Capacity Hydrogen Storage. Energy Environ. Sci. 2015, 8, 478−512. (7) Schlapbach, L.; Zuttel, A. Hydrogen-Storage Materials for Mobile Applications. Nature 2001, 414, 353−358. (8) Eberle, U.; Felderhoff, M.; Schueth, F. Chemical and Physical Solutions for Hydrogen. Angew. Chem., Int. Ed. 2009, 48, 6608−6630. 9242
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
Oxidation of Alcohols:Recent Advances. Adv. Organomet. Chem. 2015, 63, 91−174. (32) Werkmeister, S.; Neumann, J.; Junge, K.; Beller, M. Pincer-Type Complexes for Catalytic (De)Hydrogenation and Transfer (De)Hydrogenation Reactions: Recent Progress. Chem. - Eur. J. 2015, 21, 12226−12250. (33) Li, H. X.; Wang, H. X. Computational Mechanistic Studies of Acceptorless Dehydrogenation Reactions Catalyzed by Transition Metal Complexes. Sci. China: Chem. 2012, 55, 1991−2008. (34) Grasemann, M.; Laurenczy, G. Formic acid as a hydrogen source. Energy Environ. Sci. 2012, 5, 8171−81. (35) Palo, D. R.; Dagle, R. A.; Holladay, J. D. Methanol Steam Reforming for Hydrogen Production. Chem. Rev. 2007, 107, 3992− 4021. (36) Nielsen, M.; Kammer, A.; Cozzula, D.; Junge, H.; Gladiali, S.; Beller, M. Efficient Hydrogen Production from Alcohols under Mild Reaction Conditions. Angew. Chem., Int. Ed. 2011, 50, 9593−9597. (37) Kothandaraman, J.; Czaun, M.; Goeppert, A.; Haiges, R.; Jones, J.-P.; May, R. B.; Prakash, G. K. S.; Olah, G. A. Amine-Free Reversible Hydrogen Storage in Formate Salts Catalyzed by Ruthenium Pincer Complex without pH Control or Solvent. ChemSusChem 2015, 8, 1442−1451. (38) Smith, T. A.; Aplin, R. P.; Maitlis, P. M. The RutheniumCatalysed Conversion of Methanol into Methyl Formate. J. Organomet. Chem. 1985, 291, c13−c14. (39) Shinoda, S.; Itagaki, H.; Saito, Y. Dehydrogenation of Methanol in the Liquid Phase with a Homogeneous Ruthenium Complex Catalyst. J. Chem. Soc., Chem. Commun. 1985, 860−861. (40) Crabtree, R. H. Deactivation in Homogeneous Transition Metal Catalysis: Causes, Avoidance and Cure. Chem. Rev. 2015, 115, 127− 150. (41) Joó, F. A Breakthrough in Sustainable Production of Formate Salts: Combined Catalytic Methanol Dehydrogenation and Bicarbonate Hydrogenation. ChemCatChem 2014, 6, 3306−3308. (42) Yoo, J. S.; Christensen, R.; Vegge, T.; Norskov, J. K.; Studt, F. Theoretical Insight into the Trends that Guide the Electrochemical Reduction of Carbon Dioxide to Formic Acid. ChemSusChem 2016, 9, 358−363. (43) Singh, A. K.; Singh, S.; Kumar, A. Hydrogen Energy Future with Formic Acid: a Renewable Chemical Hydrogen Storage System. Catal. Sci. Technol. 2016, 6, 12−40. (44) Heim, L. E.; Schlörer, N. E.; Choi, J.-H.; Prechtl, M. H. G. Selective and Mild Hydrogen Production using Water and Formaldehyde. Nat. Commun. 2014, 5, 3621. (45) Adair, G. R. A.; Williams, J. M. J. Oxidant-free Oxidation: Ruthenium Catalysed Dehydrogenation of Alcohols. Tetrahedron Lett. 2005, 46, 8233−35. (46) Conley, B. L.; Pennington-Boggio, M. K.; Boz, E.; Williams, T. J. Discovery, Applications, and Catalytic Mechanisms of Shvo’s Catalyst. Chem. Rev. 2010, 110, 2294−2312. (47) Junge, H.; Beller, M. Ruthenium-Catalyzed Generation of Hydrogen from iso-Propanol. Tetrahedron Lett. 2005, 46, 1031−1034. (48) Murahashi, S.-i.; Ito, K.-i.; Naota, T.; Maeda, Y. Ruthenium Catalyzed Transformation of Alcohols to Esters and Lactones. Tetrahedron Lett. 1981, 22, 5327−5330. (49) Blum, Y.; Shvo, Y. Catalytically reactive (tetracyclone) (CO)2(H)2Ru and Related Complexes in Dehydrogenation of Alcohols to Esters. J. Organomet. Chem. 1985, 282, C7−10. (50) Wang, D.; Astruc, D. The Golden Age of Transfer Hydrogenation. Chem. Rev. 2015, 115, 6621−6686. (51) Haddad, Y. M. Y.; Henbest, H. B.; Husbands, J.; Mitchell, T. R. B. Reduction of Cyclohexanones to Axial Alcohols via Iridium Containing Catalysts. Proc. Chem. Soc. (London) 1964, 361−365. (52) Fujita, K.; Yoshida, T.; Imori, Y.; Yamaguchi, R. Dehydrogenative Oxidation of Primary and Secondary Alcohols Catalyzed by a Cp*Ir Complex Having a Functional C,N-Chelate Ligand. Org. Lett. 2011, 13, 2278−2281.
(53) Balaraman, E.; Khaskin, E.; Leitus, G.; Milstein, D. Catalytic Transformation of Alcohols to Carboxylic Acid Salts and H2 using Water as the Oxygen Atom Source. Nat. Chem. 2013, 5, 122−125. (54) Srimani, D.; Ben-David, Y.; Milstein, D. Direct Synthesis of Pyrroles by Dehydrogenative Coupling of Aminoalcohols with Secondary Alcohols Catalyzed by Ruthenium Pincer Complexes. Angew. Chem., Int. Ed. 2013, 52, 4012−15. (55) Wang, R.; Ma, J.; Li, F. Synthesis of alpha-Alkylated Ketones via Tandem Acceptorless Dehydrogenation/alpha-Alkylation from Secondary and Primary Alcohols Catalyzed by Metal−Ligand Bifunctional Iridium Complex Cp*Ir(2,2-bpyO)(H2O). J. Org. Chem. 2015, 80, 10769−10776. (56) Zhao, J.; Hartwig, J. F. Acceptorless, Neat, Ruthenium-Catalyzed Dehydrogenative Cyclization of Diols to Lactones. Organometallics 2005, 24, 2441−2446. (57) Lin, Y. R.; Zhu, X. C.; Zhou, Y. F. J. A convenient Lactonization of Diols to Gamma-Lactones and Delta-Lactones Catalyzed by Transition-Metal Polyhydrides. J. Organomet. Chem. 1992, 429, 269− 274. (58) Lu, Z.; Demianets, I.; Hamze, R.; Terrile, N. J.; Williams, T. J. A Prolific Catalyst for Selective Conversion of Neat Glycerol to Lactic Acid. ACS Catal. 2016, 6, 2014−2017. (59) Pagliaro, M.; Ciriminna, R.; Kimura, H.; Rossi, M.; Della Pina, C. From Glycerol to Value-Added Products. Angew. Chem., Int. Ed. 2007, 46, 4434−4440. (60) Crotti, C.; Kaspar, J.; Farnetti, E. Dehydrogenation of Glycerol to Dihydroxyacetone catalyzed by Iridium Complexes with P−N Ligands. Green Chem. 2010, 12, 1295−1300. (61) Sharninghausen, L. S.; Campos, J.; Manas, M. G.; Crabtree, R. H. Efficient, Selective and Atom Economic Catalytic Conversion of Glycerol to Lactic Acid. Nat. Commun. 2014, 5, 5084. (62) Sharninghausen, L. S.; Mercado, B. Q.; Crabtree, R. H.; Hazari, N. Selective conversion of Glycerol to Lactic Acid with Iron Pincer Precatalysts. Chem. Commun. 2015, 51, 16201−16204. (63) Campos, J.; Sharninghausen, L. S.; Crabtree, R. H.; Balcells, D. A Carbene-Rich But Carbonyl-Poor Ir6(IMe)8(CO)2(H)142+ Polyhydride Cluster as a Deactivation Product From Catalytic Glycerol Dehydrogenation. Angew. Chem., Int. Ed. 2014, 53, 12808−12811. (64) Sharninghausen, L. S.; Mercado, B. Q.; Crabtree, R. H.; Balcells, D.; Campos, J. Gel-assisted Crystallization of Ir4(IMe)7(CO)H102+ and Ir4(IMe)8H93+ Clusters derived from Catalytic Glycerol Dehydrogenation. Dalton Trans 2015, 44, 18403−10. (65) Palkovits, R.; Tajvidi, K.; Procelewska, J.; Rinaldi, R.; Ruppert, A. Hydrogenolysis of Cellulose Combining Mineral Acids and Hydrogenation Catalysts. Green Chem. 2010, 12, 972−978. (66) Manas, M. G.; Campos, J.; Sharninghausen, L. S.; Lin, E.; Crabtree, R. H. Selective Catalytic Oxidation of Sugar Alcohols to Lactic Acid. Green Chem. 2015, 17, 594−600. (67) Li, Y.; Nielsen, M.; Li, B.; Dixneuf, P. H.; Junge, H.; Beller, M. Ruthenium-Catalyzed Hydrogen Generation from Glycerol and Selective Synthesis Of Lactic Acid. Green Chem. 2015, 17, 193−8. (68) Ellis, F. W.; Krantz, J. C. Sugar alcohols XXII. Metabolism and Toxicity Studies with Mannitol and Sorbitol in Man and Animals. J. Biol. Chem. 1941, 141, 147−154. (69) Ngo, A. H.; Adams, M. J.; Do, L. H. Selective Acceptorless Dehydrogenation and Hydrogenation by Iridium Catalysts Enabling Facile lnterconversion of Glucocorticoids. Organometallics 2014, 33, 6742−6745. (70) Gunanathan, C.; Ben-David, Y.; Milstein, D. Direct Synthesis of Amides from Alcohols and Amines with Liberation of H2. Science 2007, 317, 790−792. (71) Gnanaprakasam, B.; Balaraman, E.; Ben-David, Y.; Milstein, D. Synthesis of Peptides and Pyrazines from β-amino Alcohols through Extrusion of H2 catalyzed by Ruthenium Pincer Complexes. Angew. Chem., Int. Ed. 2011, 50, 12240−12244. (72) Srimani, D.; Balaraman, E.; Hu, P.; Ben-David, Y.; Milstein, D. Formation of Tertiary Amides and Dihydrogen by Dehydrogenative Coupling of Primary Alcohols with Secondary Amines Catalyzed by 9243
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
Ruthenium Bipyridine-Based Pincer Complexes. Adv. Synth. Catal. 2013, 355, 2525−2530. (73) Nova, A.; Balcells, D.; Schley, N. D.; Dobereiner, G. E.; Crabtree, R. H.; Eisenstein, O. An Experimental-Theoretical Study of the Factors that Affect the Switch between Ruthenium-Catalyzed Dehydrogenative Amide Formation versus Amine Alkylation. Organometallics 2010, 29, 6548−6558. (74) Schley, N.; Dobereiner, G.; Crabtree, R. H. Oxidative Synthesis of Amides and Pyrroles via Dehydrogenative Alcohol Oxidation by Ruthenium Diphosphine Diamine Complexes. Organometallics 2011, 30, 4174−4179. (75) Srimani, D.; Ben-David, Y.; Milstein, D. Direct Synthesis of Pyrroles by Dehydrogenative Coupling of Aminoalcohols with Secondary Alcohols Catalyzed by Ruthenium Pincer Complexes. Angew. Chem., Int. Ed. 2013, 52, 4012−15. (76) Iida, K.; Miura, T.; Ando, J.; Saito, S. The Dual Role of Ruthenium and Alkali Base Catalysts in Enabling a Conceptually New Shortcut to N-Unsubstituted Pyrroles through Unmasked alphaAmino Aldehydes. Org. Lett. 2013, 15, 1436−1439. (77) Zhang, M.; Fang, X.; Neumann, H.; Beller, M. General and Regioselective Synthesis of Pyrroles via Ruthenium-Catalyzed Multicomponent Reactions. J. Am. Chem. Soc. 2013, 135, 11384−11388. (78) Pan, B.; Liu, B.; Yue, E. L.; Liu, Q. B.; Yang, X. Z.; Wang, Z.; Sun, W. H. A Ruthenium Catalyst with Unprecedented Effectiveness for the Coupling Cyclization of gamma-Amino Alcohols and Secondary Alcohols. ACS Catal. 2016, 6, 1247−1253. (79) Michlik, S.; Kempe, R. Regioselectively Functionalized Pyridines from Sustainable Resources. Angew. Chem., Int. Ed. 2013, 52, 6326− 6329. (80) Michlik, S.; Kempe, R. A. Sustainable Catalytic Pyrrole Synthesis. Nat. Chem. 2013, 5, 140−144. (81) Taguchi, K.; Sakaguchi, S.; Ishii, Y. Synthesis of Quinolines from Amino Alcohol and Ketones catalyzed by Ir(cod)Cl2 or IrCl3 under Solvent-Free Conditions. Tetrahedron Lett. 2005, 46, 4539−4542. (82) Schmitt, D. C.; Taylor, A. P.; Flick, A. C.; Kyne, R. E., Jr Synthesis of Pyrazoles from 1,3-Diols via Hydrogen Transfer Catalysis. Org. Lett. 2015, 17, 1405−1408. (83) Srimani, D.; Ben-David, Y.; Milstein, D. Direct Synthesis of Pyridines and Quinolines by Coupling of gamma -Amino-Alcohols with Secondary Alcohols. Chem. Commun. 2013, 49, 6632−4. (84) Hille, H.; Irrgang, T.; Kempe, R. The Synthesis of Benzimidazoles and Quinoxalines from Aromatic Diamines and Alcohols by Iridium-Catalyzed Acceptorless Dehydrogenative Alkylation. Chem. - Eur. J. 2014, 20, 5569−5572. (85) Deibl, N.; Ament, K.; Kempe, R. A Sustainable Multicomponent Pyrimidine Synthesis. J. Am. Chem. Soc. 2015, 137, 12804−7. (86) Zhao, W.; Liu, P.; Li, F. Quinazolinones from o-Aminobenzonitriles by One-Pot Sequential Selective Hydration/Condensation/Acceptorless Dehydrogenation Catalyzed by an Iridium Complex. ChemCatChem 2016, 8, 1523−1530. (87) Li, F.; Lu, L.; Liu, P. Acceptorless Dehydrogenative Coupling of o-Aminobenzamides with the Activation of Methanol as a C1 Source for the Construction of Quinazolinones. Org. Lett. 2016, 18, 2580− 2583. (88) Li, F.; Sun, C.; Wang, N. Catalytic Acceptorless Dehydrogenative Coupling of Arylhydrazines and Alcohols for the Synthesis of Arylhydrazones. J. Org. Chem. 2014, 79, 8031−8039. (89) Kim, S. H.; Hong, S. H. Ruthenium-Catalyzed Urea Synthesis using Methanol as the C1 Source. Org. Lett. 2016, 18, 212−215. (90) Hale, L. V. A.; Malakar, T.; Tseng, K.-N. T.; Zimmerman, P. M.; Paul, A.; Szymczak, N. K. The Mechanism of Acceptorless Amine Double Dehydrogenation by N,N,N-Amide Ruthenium(II) Hydrides: A Combined Experimental and Computational Study. ACS Catal. 2016, 6, 4799−4813. (91) Fujii, T.; Saito, Y. Catalytic Dehydrogenation of Methanol with Ruthenium Complexes. J. Mol. Catal. 1991, 67, 185−190. (92) Smith, T. A.; Aplin, R. P.; Maitlis, P. M. The RutheniumCatalysed Conversion of Methanol into Methyl Formate. J. Organomet. Chem. 1985, 291, c13−c14.
(93) Blum, Y.; Shvo, Y.; Chodosh, D. F. Structure of (Ph4C4CO) (CO)3Ru - a Catalyst Precursor in H-Transfer and Dehydrogenation Reactions of Alcohols. Inorg. Chim. Acta 1985, 97, L25−L26. (94) Morton, D.; Cole-Hamilton, D. J. Molecular Hydrogen Complexes in Catalysis: Highly Efficient Hydrogen Production from Alcoholic Substrates Catalysed by Ruthenium Complexes. J. Chem. Soc., Chem. Commun. 1988, 1154−1156. (95) Shinoda, S.; Yamakawa, T. One-step Formation of Methyl Acetate with Methanol used as the sole source and Catalysis by Ru(II)-Sn(II) Cluster Complexes. J. Chem. Soc., Chem. Commun. 1990, 1511−12. (96) Einaga, H.; Yamakawa, T.; Shinoda, S. Synthesis of Acetic-Acid from Methanol alone by Homogeneous Metal-Complex Catalyst. J. Mol. Catal. A: Chem. 1995, 97, 35−40. (97) Robles-Dutenhefner, P. A.; Moura, E. M.; Gama, G. J.; Siebald, H. G. L.; Gusevskaya, E. V. Synthesis of Methyl Acetate from Methanol catalyzed by Cp(phosphine)2RuX and Cp(phosphine)2Ru(SnX3) (X = F, Cl, Br): Ligand Effect. J. Mol. Catal. A: Chem. 2000, 164, 39−47. (98) Yamakawa, T.; Hiroi, M.; Shinoda, S. Catalytic Reaction of Methanol with a series of Ru(II) Complexes and the Mechanism of the Formation of Acetic-Acid from Methanol alone. J. Chem. Soc., Dalton Trans. 1994, 2265−69. (99) Moulton, C. J.; Shaw, B. L. Transition Metal-Carbon Bonds 0.42. Complexes of Nickel, Palladium, Platinum, Rhodium And Iridium with Tridentate Ligand 2,6-bis(di-t-butylphosphino)methylphenyl. J. Chem. Soc., Dalton Trans. 1976, 1020−1024. (100) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Facile Conversion of Alcohols into Esters and Dihydrogen Catalyzed by New Ruthenium Complexes. J. Am. Chem. Soc. 2005, 127, 10840−10841. (101) Li, H.; Hall, M. B. Computational Mechanistic Studies on Reactions of Transition Metal Complexes with Noninnocent Pincer Ligands: Aromatization−Dearomatization or Not. ACS Catal. 2015, 5, 1895−1913. (102) Nielsen, M.; Alberico, E.; Baumann, W.; Drexler, H.-J.; Junge, H.; Gladiali, S.; Beller, M. Low-Temperature Aqueous-Phase Methanol Dehydrogenation to Hydrogen and Carbon Dioxide. Nature 2013, 495, 85−89. (103) Hu, P.; Diskin-Posner, Y.; Ben-David, Y.; Milstein, D. Reusable Homogeneous Catalytic System for Hydrogen Production from Methanol and Water. ACS Catal. 2014, 4, 2649−52. (104) Rodriguez-Lugo, R. E.; Trincado, M.; Vogt, M.; Tewes, F.; Santiso-Quinones, G.; Grützmacher, H. A Homogeneous Transition Metal Complex for clean Hydrogen Production from Methanol− Water Mixtures. Nat. Chem. 2013, 5, 342−347. (105) Noyori, R.; Hashiguchi, S. Asymmetric Transfer Hydrogenation Catalyzed by Chiral Ruthenium Complexes. Acc. Chem. Res. 1997, 30, 97−102. (106) Eisenstein, O.; Crabtree, R. H. Outer Sphere Hydrogenation Catalysis. New J. Chem. 2013, 37, 21−27. (107) Lei, M.; Pan, Y.; Ma, X. The Nature of Hydrogen Production from Aqueous-Phase Methanol Dehydrogenation with Ruthenium Pincer Complexes Under Mild Conditions. Eur. J. Inorg. Chem. 2015, 2015, 794−803. (108) Zhang, J.; Balaraman, E.; Leitus, G.; Milstein, D. Electron-rich PNP- and PNN-type ruthenium(II) hydrido borohydride pincer complexes. Synthesis, structure, and catalytic dehydrogenation of alcohols and hydrogenation of esters. Organometallics 2011, 30, 5716− 5724. (109) Crabtree, R. H. Iridium Compounds in Catalysis. Acc. Chem. Res. 1979, 12, 331−338. (110) Li, H. X.; Hall, M. B. Role of the Chemically Non-Innocent Ligand in the Catalytic Formation of Hydrogen and Carbon Dioxide from Methanol and Water with the Metal as the Spectator. J. Am. Chem. Soc. 2015, 137, 12330−12342. (111) Zhang, L.; Nguyen, D. H.; Raffa, G.; Trivelli, X.; Capet, F.; Desset, S.; Paul, S.; Dumeignil, F.; Gauvin, R. M. Catalytic Conversion of Alcohols into Carboxylic Acid Salts in Water: Scope, Recycling, and Mechanistic Insights. ChemSusChem 2016, 9, 1413. 9244
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
(112) Hou, C.; Zhang, Z. H.; Zhao, C. Y.; Ke, Z. F. DFT Study of Acceptorless Alcohol Dehydrogenation Mediated by Ruthenium Pincer Complexes: Ligand Tautomerization Governing Metal Ligand Cooperation. Inorg. Chem. 2016, 55, 6539−6551. (113) Prades, A.; Peris, E.; Albrecht, M. Oxidations and Oxidative Couplings catalyzed by Triazolylidene Ruthenium Complexes. Organometallics 2011, 30, 1162−1167. (114) Campos, J.; Hintermair, U.; Brewster, T. P.; Takase, M. K.; Crabtree, R. H. Catalyst Activation by Loss of Cyclopentadienyl Ligands in Hydrogen Transfer Catalysis with Cp*IrIII Complexes. ACS Catal. 2014, 4, 973−985. (115) Shahane, S.; Fischmeister, C.; Bruneau, C. Acceptorless Ruthenium Catalyzed Dehydrogenation of Alcohols to Ketones and Esters. Catal. Sci. Technol. 2012, 2, 1425−1428. (116) Baratta, W.; Bossi, G.; Putignano, E.; Rigo, P. Pincer and Diamine Ru and Os Diphosphane Complexes as Efficient Catalysts for the Dehydrogenation of Alcohols to Ketones. Chem. - Eur. J. 2011, 17, 3474−3481. (117) Hintermair, U.; Sheehan, S.; Parent, A.; Ess, D.; Richens, D.; Vaccaro, P.; Brudvig, G. W.; Crabtree, R. H. Precursor Transformation during Molecular Oxidation Catalysis with Organometallic Iridium Complexes. J. Am. Chem. Soc. 2013, 135, 10837−10851. (118) Fujita, K.; Tanino, N.; Yamaguchi, R. Ligand-Promoted Dehydrogenation of Alcohols Catalyzed by Cp*Ir Complexes. A New Catalytic System for Oxidant-Free Oxidation of Alcohols. Org. Lett. 2007, 9, 109−111. (119) Shima, S.; Lyon, E. J.; Sordel-Klippert, M.; Kauss, M.; Kahnt, J.; Thauer, R. K.; Steinbach, K.; Xie, X.; Verdier, L.; Griesinger, C. The Cofactor of the Iron-Sulfur Cluster Free Hydrogenase Hmd: Structure of the Light-Inactivation Product. Angew. Chem., Int. Ed. 2004, 43, 2547−2551. (120) Peris, E.; Lee, J. C., Jr.; Rambo, J. R.; Eisenstein, O.; Crabtree, R. H. Factors affecting the Strength of N-H···H-M Hydrogen Bonds. J. Am. Chem. Soc. 1995, 117, 3485−3491. (121) Park, S. H.; Lough, A. J.; Yap, G. P. A.; Morris, R. H. The Effect of Ancillary Ligands on Intramolecular Proton-Hydride (N-H··· H-Ir) Bonding in Complexes of Iridium(III). J. Organomet. Chem. 2000, 609, 110−122. (122) Moore, C. M.; Dahl, E. W.; Szymczak, N. K. Beyond H2: exploiting 2-Hydroxypyridine as a Design Element from Fe-hydrogenase for Energy-Relevant Catalysis. Curr. Opin. Chem. Biol. 2015, 25, 9−17. (123) Royer, A. M.; Rauchfuss, T. B.; Gray, D. L. Organoiridium Pyridonates and Their Role in the Dehydrogenation of Alcohols. Organometallics 2010, 29, 6763−6768. (124) Fujita, K.; Kawahara, R.; Aikawa, T.; Yamaguchi, R. Hydrogen Production from a Methanol−Water Solution Catalyzed by an Anionic Iridium Complex Bearing a Functional Bipyridonate Ligand under Weakly Basic Conditions. Angew. Chem., Int. Ed. 2015, 54, 9057−60. (125) Zeng, G.; Sakaki, S.; Fujita, K.; Sano, H.; Yamaguchi, R. Efficient Catalyst for Acceptorless Alcohol Dehydrogenation. ACS Catal. 2014, 4, 1010−1020. (126) Kawahara, R.; Fujita, K.; Yamaguchi, R. Cooperative Catalysis by Iridium Complexes with a Bipyridonate Ligand. Angew. Chem., Int. Ed. 2012, 51, 12790−12794. (127) Zhang, D.-D.; Chen, X.-K.; Liu, H.-L.; Huang, X.-R. A Computational Mechanistic Study of pH-Dependent Alcohol Dehydrogenation catalyzed by a novel C,N or C,C cyclometalated Cp*Ir Complex in Aqueous Solution. New J. Chem. 2014, 38, 3862−73. (128) Polukeev, A. V.; Petrovskii, P. V.; Peregudov, A. S.; Ezernitskaya, M. G.; Koridze, A. A. Dehydrogenation of Alcohols by Bis(phosphinite) Benzene Based and Bis(phosphine) Ruthenocene Based Iridium Pincer Complexes. Organometallics 2013, 32, 1000− 1015. (129) Musa, S.; Shaposhnikov, I.; Cohen, S.; Gelman, D. Ligand− Metal Cooperation in PCP Pincer Complexes: Rational Design and Catalytic Activity in Acceptorless Dehydrogenation of Alcohols. Angew. Chem., Int. Ed. 2011, 50, 3533−3537.
(130) Crabtree, R. H.; Siegbahn, P. E. M.; Eisenstein, O.; Rheingold, A. L.; Koetzle, T. F. A New Intermolecular Interaction: Unconventional Hydrogen Bonds with Element-Hydride Bonds as Proton Acceptor. Acc. Chem. Res. 1996, 29, 348−54. (131) Musa, S.; Ackermann, L.; Gelman, D. Dehydrogenative CrossCoupling of Primary and Secondary Alcohols. Adv. Synth. Catal. 2013, 355, 3077−3080. (132) Jumde, V. R.; Gonsalvi, L.; Guerriero, A.; Peruzzini, M.; Taddei, M. A Ruthenium-Based Catalytic System for a Mild Borrowing-Hydrogen Process. Eur. J. Org. Chem. 2015, 2015, 1829− 1833. (133) Valencia, M.; Muller-Bunz, H.; Gossage, R. A.; Albrecht, M. Enhanced Product Selectivity promoted by Remote Metal Coordination in Acceptor-Free Alcohol Dehydrogenation Catalysis. Chem. Commun. 2016, 52, 3344−3347. (134) Sun, Z.; Liu, Y.; Chen, J.; Huang, C.; Tu, T. Robust Iridium Coordination Polymers: Highly Selective, Efficient, and Recyclable Catalysts for Oxidative Conversion of Glycerol to Potassium Lactate with Dihydrogen Liberation. ACS Catal. 2015, 5, 6573−6578. (135) Cheng, J. J.; Zhu, M. J.; Wang, C.; Li, J. J.; Jiang, X.; Wei, Y. W.; Tang, W. J.; Xue, D.; Xiao, J. L. Chemoselective dehydrogenative Esterification of Aldehydes and Alcohols with a Dimeric Rhodium(II) Catalyst. Chem. Sci. 2016, 7, 4428−4434. (136) Dutta, I.; Sarbajna, A.; Pandey, P.; Rahaman, S. M. W.; Singh, K.; Bera, J. K. Acceptorless Dehydrogenation of Alcohols on a Diruthenium(II,II) Platform. Organometallics 2016, 35, 1505−1513. (137) Zhang, G.; Hanson, S. K. Cobalt-Catalyzed Alcohol Hydrogenation and Dehydrogenation Reactions. Org. Lett. 2013, 15, 650− 653. (138) Zhang, G.; Vasudevan, K. V.; Scott, B. L.; Hanson, S. K. J. Understanding the Mechanisms of Cobalt-Catalyzed Hydrogenation and Dehydrogenation Reactions. J. Am. Chem. Soc. 2013, 135, 8668− 8681. (139) Alberico, E.; Sponholz, P.; Cordes, C.; Nielsen, M.; Drexler, H.-J.; Baumann, W.; Junge, H.; Beller, M. Selective Hydrogen Production from Methanol with a Defined Iron Pincer Catalyst under Mild Conditions. Angew. Chem., Int. Ed. 2013, 52, 14162−66. (140) Bielinski, E. A.; Lagaditis, P. O.; Zhang, Y.; Mercado, B. Q.; Würtele, C.; Bernskoetter, W. H.; Hazari, N.; Schneider, S. Lewis AcidAssisted Formic Acid Dehydrogenation Using a Pincer-Supported Iron Catalyst. J. Am. Chem. Soc. 2014, 136, 10234−10237. (141) Chakraborty, S.; Brennessel, W. W.; Jones, W. D. A Molecular Iron Catalyst for the Acceptorless Dehydrogenation and Hydrogenation of N-Heterocycles. J. Am. Chem. Soc. 2014, 136, 8564−8567. (142) Chakraborty, S.; Lagaditis, P. O.; Förster, M.; Bielinski, E. A.; Hazari, N.; Holthausen, M. C.; Jones, W. D.; Schneider, S. WellDefined Iron Catalysts for the Acceptorless Reversible Dehydrogenation-Hydrogenation of Alcohols and Ketones. ACS Catal. 2014, 4, 3994−4003. (143) Bonitatibus, P. J., Jr.; Chakraborty, S.; Doherty, M. D.; Siclovan, O.; Jones, W. D.; Soloveichik, G. L. Reversible Catalytic Dehydrogenation of Alcohols for Energy Storage. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 1687−1692. (144) Berben, L. A. Catalysis by Aluminum(III) Complexes of NonInnocent Ligands. Chem. - Eur. J. 2015, 21, 2734−2742. (145) Power, P. P. Main-group Elements as Transition Metals. Nature 2010, 463, 171−177. (146) Kamitani, M.; Ito, M.; Itazaki, M.; Nakazawa, H. Effective Dehydrogenation of 2-pyridylmethanol Derivatives catalyzed by an Iron Complex. Chem. Commun. 2014, 50, 7941−44. (147) Song, H.; Kang, B.; Hong, S. H. Fe-Catalyzed Acceptorless Dehydrogenation of Secondary Benzylic Alcohols. ACS Catal. 2014, 4, 2889−2895. (148) Balcells, D.; Nova, A.; Clot, E.; Gnanamgari, D.; Crabtree, R. H.; Eisenstein, O. Mechanism of Homogeneous Iridium-Catalyzed Alkylation of Amines with Alcohols from a DFT Study. Organometallics 2008, 27, 2529−2535. (149) Bielinski, E. A.; Förster, M.; Zhang, Y.; Bernskoetter, W. H.; Hazari, N.; Holthausen, M. C. Base-Free Methanol Dehydrogenation 9245
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
Chemical Reviews
Review
Using a Pincer-Supported Iron Compound and Lewis Acid Cocatalyst. ACS Catal. 2015, 5, 2404−2415. (150) Chakraborty, S.; Piszel, P. E.; Brennessel, W. W.; Jones, W. D. A Single Nickel Catalyst for the Acceptorless Dehydrogenation of Alcohols and Hydrogenation of Carbonyl Compounds. Organometallics 2015, 34, 5203−5206. (151) Mukherjee, A.; Nerush, A.; Leitus, G.; Shimon, L. J. W.; Ben David, Y.; Espinosa Jalapa, N. A.; Milstein, D. Manganese-Catalyzed Environmentally Benign Dehydrogenative Coupling of Alcohols and Amines to Form Aldimines and H2: A Catalytic and Mechanistic Study. J. Am. Chem. Soc. 2016, 138, 4298−4301. (152) Luca, O. R.; Crabtree, R. H. Redox Active Ligands in Catalysis. Chem. Soc. Rev. 2013, 42, 1440−1459.
9246
DOI: 10.1021/acs.chemrev.6b00556 Chem. Rev. 2017, 117, 9228−9246
