Stereoselective Semi‐Hydrogenations of Alkynes by First‐Row (3d) Transition Metal Catalysts

Abstract

The chemo‐ and stereoselective semi‐hydrogenation of alkynes to alkenes is a fundamental transformation in synthetic chemistry, for which the use of precious 4d or 5d metal catalysts is well‐established. In mankind‘s unwavering quest for sustainability, research focus has considerably veered towards the 3d metals. Given their high abundancy and availability as well as lower toxicity and noxiousness, they are undoubtedly attractive from both an economic and an environmental perspective. Herein, we wish to present noteworthy and groundbreaking examples for the use of 3d metal catalysts for diastereoselective alkyne semi‐hydrogenation as we embark on a journey through the first‐row transition metals.

Alkyne semi‐hydrogenation: This review explores the fundamentally important transformation of alkynes to alkenes in a diastereoselective manner by means of 3d metal catalysis. Several examples ranging from scandium to zinc are presented, featuring E‐ or Z‐selective homogeneous or heterogeneous hydrogenations using either H₂ gas or alternative hydrogen donors.

1. Introduction

Reductive chemistry plays a pivotal role in the industrial preparation of numerous bulk chemicals. Hydrogenation reactions such as the Haber‐Bosch process, ^[1] the hydrogenative saturation of fats ^[2] or the production of methanol fuel from CO or CO₂ ^[3] are noteworthy examples, owing to their considerable impact on modern life. In general, the hydrogenation of alkenes or alkynes is of vital importance for the synthesis of myriad fine chemicals such as pharmaceuticals, fragrances and natural products (Scheme 1).[ ⁴ , ⁵ , ⁶ ] However, a partial hydrogenation of alkynes to alkenes remains a challenging task in catalysis for two main reasons: Firstly, the obtained alkenes may undergo facile onward hydrogenation to alkanes under similar catalytic conditions, which diminishes alkene yields and hampers product separation. High degrees of chemoselectivity between alkyne and alkene are crucial criteria for competent catalysts. While a plethora of active catalysts for the full hydrogenation of alkynes to alkanes have been developed, selective (3d) semi‐hydrogenation catalysts have remained more limited. Secondly, a partial alkyne hydrogenation may produce two diastereomeric products: the (E)‐alkene and the (Z)‐alkene (occasionally referred to as trans‐ and cis‐alkenes). Typically, only a single isomer is desired. Separation of E/Z product mixtures is notoriously problematic due to similar polarity and size, calling for high stereoselectivities in the catalytic step to avoid the undesired isomer altogether.

Scheme 1.

Dihydrogen as a sustainable reductant and applications of selective semi‐hydrogenations of alkynes in complex molecule syntheses.[ ⁴ , ⁵ ]

Ever since the pioneering works of Lindlar ^[7] for the Z‐selective semi‐hydrogenation of alkynes using a heterogenous Pb‐poisoned Pd catalyst, a multitude of potential catalytic systems has been developed, ^[8] a large portion of which are based on the later 4d or 5d transition metals. This is problematic from an economic as well as from an environmental perspective: The heavy 4d and 5d metals show a noticeably low abundancy in the earth's crust and are difficult to mine and process, making them costly due to limited supply. Furthermore, due to their scarcity, they are foreign to most terrestrial organisms and thus generally show a high degree of toxicity and potential environmental damage. An uprising field in green chemistry is to pursue new catalytic systems based on less toxic and environmentally benign elements such as 3d metals. ^[9] While few industrial processes are based on these metals, nature has been utilizing them for eons – indeed, most 3d metals can be found as an active center in metalloenzymes, proving their inherent potential as catalysts. ^[10] This review focuses on the diastereoselective semi‐hydrogenation of alkynes to either the (E)‐ or (Z)‐alkenes using H₂ gas and other hydrogen donors (transfer hydrogenations) and highlights current examples for 3d‐metal‐based catalysts. Herein, we provide an overview of the 3d elements starting with Sc, Ti, V and Cr, which have rarely been applied in semi‐hydrogenation catalysis save a few examples, followed by Mn, often employed in transfer hydrogenation reactions. A large portion of this review is dedicated to Fe, Co, Ni and Cu as the most commonly utilized 3d metals in alkyne semi‐hydrogenations, whereas Zn has thus far only been employed as an additive in the presence of another transition metal catalyst. The use of alternative methods of formal hydrogenations, i. e. via sequential hydrosilylation/hydroboration and protodesilylation/protodeboration is beyond the scope of this review. Topical reviews as well as most relevant recent works of such hydrofunctionalization methods will be referred to in the title of each chapter.

The main advantage of using molecular hydrogen (over alternative methods using boranes/silanes and protolytic workup) is the avoidance of by‐products and therefore the minimalization of waste. Common metrics of chemical sustainability are the atom economy, the environmental factor (E‐factor), as well as hazard potentials, which all display beneficial magnitudes for the usage of H₂ gas as reductant. ^[9] Additionally, an economic cost‐factor should be taken into consideration: While various silanes or boranes are rather inexpensive, the use of H₂ as the most widely available technical hydride source clearly should be preferred. Additionally, the growing production of “green hydrogen”, derived from large‐scale electrolysis of water using exclusively renewable energies, paves the way for fully sustainable and climate‐neutral protocols for both industrial as well as research applications.

2. Scandium[ ¹¹ , ¹² ]

As early transition metals are not intensively investigated as hydrogenation catalysts, only a single example for a selective semi‐hydrogenation using dihydrogen was reported for scandium in 2017. ^[13] A scandium complex featuring a PNP‐pincer ligand was applied as the catalyst for the hydrogenation of 3‐hexyne under mild conditions (4 bar H₂, 23 °C, 5 mol% catalyst). The (Z)‐alkene was obtained in high selectivity and yield (Scheme 2).

Scheme 2.

Application of a PNP‐scandium complex for Z‐selective semi‐hydrogenation of an internal alkyne. A dihydrido complex was proposed as active species. ^[13]

3. Titanium[ ¹¹ , ¹⁴ , ¹⁵ ]

Similar to scandium, titanium‐catalyzed (semi‐)hydrogenation reactions have remained a rarity. The first report of a direct titanium‐catalyzed semi‐hydrogenation was published by Sonogashira and Hagihara in 1965. ^[16a] The authors employed [Cp₂Ti(CO)₂] (Cp=cyclopentadienyl) as pre‐catalyst in the semi‐hydrogenation of just a handful of alkynes at 50 bar H₂ and 50–60 °C after reaction in benzene or heptane for 30 min (Scheme 3a). 1‐Pentyne, 1‐hexyne and phenylacetylene were converted to the alkenes with yields of 95 %, 90 % and 95 %, respectively. However, t‐butyl‐acetylene afforded a 60/40 mixture of alkane and alkene; diphenylacetylene gave bibenzyl in 90 % yield; 3‐heptyne showed no conversion. The authors speculated that a titanacyclopropene intermediate is operative, however, no mechanistic support was given. An optimized method was later reported by Dixneuf. ^[16b] Ligand substitution of the same pre‐catalyst with trimethylphosphine in the presence of diphenyl‐acetylene and subsequent treatment with H₂ afforded the labile Cp₂Ti(PMe₃) complex that displayed very high activity and selectivity in the semi‐hydrogenation of a few alkynes to the Z‐alkenes. Full conversion was already observed at 1 bar H₂ and room temperature (Scheme 3b). ^[16b]

Scheme 3.

a) Ti‐catalyzed semi‐hydrogenations by Sonogashira; ^[16a] and b) optimized conditions for low‐pressure hydrogenations by Dixneuf. ^[16b]

Application of a ternary Pd@Cu/TiO₂ mixture as hydrogen transfer catalyst for alkyne reductions under ultraviolet light irradiation was reported in 2020. ^[17] The authors proposed a mechanism based on a photocatalytic cycle: The excitation of the photocatalyst (TiO₂) in the UV generates an exciton. The positive holes are capable of oxidizing the hydrogen‐donating isopropanol, producing acetone and protons. The electrons in the conduction band in combination with the protons subsequently form the active hydrogen species at the Pd@Cu‐cocatalyst, which can selectively hydrogenate the alkyne to the (Z)‐alkene. [Cp₂Ti^IIICl] was reported to act as a mediator in a hydrogen transfer reaction from water. ^[18] A variety of well‐known catalysts based on Pd, Rh or Ir were used as the actual hydrogenation catalyst, while the Ti(III) complex was employed in stoichiometric amounts for water reduction. One reported catalyst was the Lindlar catalyst, which afforded high Z‐selectivities.

4. Vanadium[ ¹¹ , ¹⁹ ]

Few systems are known for the semi‐hydrogenation towards a single alkene isomer that directly involve vanadium. In 2011, a cationic vanadium complex was reported by Toste and co‐workers which was capable of a highly selective conversion of alkynes under mild conditions (1 bar H₂, 60 °C, 24–72 h, 20 mol% [V]) to yield exclusively (Z)‐alkenes. ^[20] Both internal and terminal alkynes were successfully converted using this method, however, only a small substrate scope of unfunctionalized substrates was presented, leaving the application of this method for functionalized substrates unclear. The authors provided a mechanistic proposal, involving a homogeneous bifunctional bis(imido)vanadium species (Scheme 4), which is capable of taking up one hydrogen molecule with contemporaneous formation of a monohydrido vanadium complex with one newly formed amido ligand. Cis‐Hydrovanadation of the alkyne, followed by a 1,2‐α‐NH elimination with the amido ligand, ^[21] leads to liberation of the (Z)‐alkene. Association of trimethylphosphine regenerates the initial catalyst species. This mechanistic proposal is based on DFT calculations, experiments with H₂/D₂‐mixtures and parahydrogen‐induced polarization (PHIP) NMR experiments. No scrambled HD‐hydrogenation products have been observed and the PHIP‐NMR experiment led to transfer of polarization; both of these results demonstrated that only one H₂ molecule is involved in the hydrogenation of one alkyne molecule.

Scheme 4.

Preparation of the cationic vanadium catalyst and proposed mechanism for the selective semi‐hydrogenation. ^[19]

Two similar Z‐selective vanadium‐catalyzed hydrogenations of alkynes were reported in 2014 ^[22] and 2017. ^[22] Both systems are based on organovanadium sites (V(Mes)(thf)) on either a catechol‐functionalized porous organic polymer (catPOP) ^[22] or on silica, SiO₂ (Scheme 5). ^[23] Both catalysts exhibit a similar structural motif. The former authors ^[22] reported 37 % conversion of diphenyl‐acetylene with exclusive formation of stilbene (Z/E 98/2, 1.2 mol% [(catPOP)V(Mes)(thf)], 14 bar H₂, 60 °C, 8 h, methylcyclohexane). Full conversion is observed with higher catalyst loading, albeit accompanied by increasingly significant amounts of over‐hydrogenation products. The latter authors ^[23] reported 88–91 % conversion of diphenylacetylene with 78–90 % formation of stilbene (Z/E not given, 0.4–0.5 mol% [(SiO₂)V(Mes)(thf)] with different SiO₂/V ratios, 14 bar H₂, 75 °C, 1–17 h, dodecane). Upon longer reaction times, the catalyst led to the further conversion of the alkene to the fully hydrogenated alkane.

Scheme 5.

Structural motifs of the [(SiO₂)V(Mes)(thf)]‐ as well as the catPOP‐vanadium catalyst. Structure proposals based on XAS and EPR.[ ²² , ²³ ]

5. Chromium ^[11]

The earliest report on chromium‐catalyzed selective semi‐hydrogenation of alkynes stems from Sodeoka and co‐workers, published in 1985. ^[24] They demonstrated the application of “piano stool” (arene)tricarbonylchromium(0) complexes as highly efficient catalysts (Scheme 6). 1‐Phenyl‐1‐propyne was successfully hydrogenated (20 mol% [Cr], 69 bar H₂, 120 °C, acetone) to (Z)‐1‐phenyl‐1‐propene in high yield, while no (E)‐isomer or fully hydrogenated alkane have been observed, even under prolonged reaction times. However, a limited substrate scope was reported and the harsh conditions left room for an improvement of this protocol. Akin to the catPOP‐vanadium system (vide supra), a similar chromium‐based system has been developed (Scheme 7). The initial report relied on [Cr(CH₂SiMe₃)₄] as the precursor, ^[22] which was later substituted by [CrPh₃(thf)₃] ^[25] due to improved protonolysis behavior and increased stability towards decomposition. These effects led to a more defined catalytic system, which was formed within 2 h instead of the prolonged reaction times or elevated temperatures otherwise necessary. Both catalytic systems show high efficiency towards (semi‐)hydrogenation, with the earlier report exhibiting a higher activity towards full hydrogenation of diphenylacetylene (81 % 1,2‐diphenylethane, 19 % (Z)‐stilbene) than the [CrPh₃(thf)₃]‐based catalyst, which predominantly resulted in the formation of (Z)‐stilbene (35 % (Z)‐stilbene, 1 % (E)‐stilbene, 6 % 1,2‐diphenylethane) under the same conditions (5 mol% [Cr], 14 bar H₂, 60 °C, 6 h, methylcyclohexane). As a mechanistic proposal for the observed activity and the function of the catecholate moiety, the authors suggested that, upon dissociation of a THF molecule, a free coordination site for the alkyne is provided on the Cr center. Additionally, free coordination sites for hydrogen are required. The polymeric catecholate ligand inhibits the formation of chromium dimers and consequent loss of catalytic activity due to the blockage of coordination sites. The authors verified this proposal by poisoning the active catalyst with bipyridine, resulting in a total loss of hydrogenation activity.

Scheme 6.

Z‐Selective semi‐hydrogenation of 1‐phenyl‐1‐propyne using a (methyl benzoate)tricarbonylchromium(0) complex. ^[24]

Scheme 7.

Precursors for the preparation of catPOP‐chromium catalysts. Structures were proposed based on XAS and EPR.^,[22][25]

A recent development in Cr‐catalyzed semi‐hydrogenation was the application of commercially available [Cr(acac)₃] in combination with the reductant diisobutylaluminium hydride (DIBAL−H) in THF. ^[26] While CrCl₂ and CrCl₃ (entries 1 and 2, table 1), showed low conversions towards the alkene, Cr(acac)₃ provided an efficient and simple catalytic process for the selective semi‐hydrogenation under mild conditions (5–10 mol% catalyst, <5 bar H₂, 25 °C, 3–24 h, THF). A variety of functional groups such as halides, ethers, esters, and ketones were tolerated under these conditions. The highest selectivity towards (Z)‐alkenes was achieved with dialkylacetylenes, compared to the lower selectivity for diarylacetylenes. Mechanistically, the acetylacetonate is initially reduced; this has been confirmed by the isolation of the reduced species (removal of carbonyl group, reduction to alcohols with subsequent elimination to alkene products). Additionally, the chromium is reduced, which results in an aggregation to Cr nanoparticles, which have been detected by TEM.

Table 1.

Chromium‐catalyzed semi‐hydrogenation of alkynes. ^[26]

Entry	Alkyne	Alkene [%][a]	Z/E [a]
1[b] 2[c] 3[d] 4[e]		19 28 51 95	17/1 16/1 12/1 6/1
5[f]		96	3/1
6[g]		97	10/1
7		99	>99/1

Conditions: 0.2 mmol alkyne, 0.4 mL THF. [a] Determined by ¹H‐NMR, GC‐FID. [b] 5 mol% CrCl₂, 10 mol% DIBAL−H. [c] 5 mol% CrCl₃, 15 mol% DIBAL−H. [d] 5 mol% [Cr(acac)₃], 15 mol% DIBAL−H. [e] 1.3 bar H₂, 9 h. [f] 3 bar H₂. [g] 12 h.

6. Manganese[ ¹¹ , ²⁷ , ²⁸ , ²⁹ , ³⁰ ]

While various examples for full hydrogenations using manganese catalysts are known, the semi‐hydrogenation of alkynes remains a challenging task. ^[31] The first reports were contemporaneously published by Zhou et al. ^[32] and Brzozowska et al. ^[33] in 2018. Both systems rely on hydrogen donors to carry out a transfer semi‐hydrogenation. Zhou and co‐workers applied an N‐heterocyclic silylene manganese complex (1 mol%) in THF at 55 °C for the E‐selective transfer hydrogenation of alkynes, using ammonia borane as the hydrogen source (Scheme 8). ^[32] A variety of functional groups such as ethers, halides, or esters were tolerated and gave the semi‐hydrogenation product with high yield and E‐selectivity. Mechanistic proposals for the observed reactivity are based on the formation of a molecular Mn‐hydrido complex by means of the ammonia borane. No conversion under H₂ atmosphere was observed in the absence of H₃N‐BH₃. Unfortunately, no proposals for the origin of the E‐selectivity were provided. Similarly, the system by Brzozowska et al. ^[33] utilized a PNP‐pincer ligand and ammonia borane in methanol as the hydrogen donor for the Z‐selective semi‐hydrogenation of alkynes in generally high yields (see Scheme 9). Analogously to the system of Zhou et al., it was hypothesized that the initial alkoxido‐Mn complex undergoes ligand substitution to a hydrido complex by means of ammonia borane. Subsequent cis‐hydrometallation and protonation of the intermediate vinyl complex by methanol regenerates the alkoxido species and releases the (Z)‐alkene (Scheme 10). The presence of small amounts of (E)‐alkene was explained by isomerization in a second catalytic cycle. However, the coordination of the alkyne to the Mn center is energetically preferred over the (Z)‐alkene coordination and thus the isomerization is limited.

Scheme 8.

Silylene‐manganese catalyst for E‐selective semi‐hydrogenation. Conversions are given; E‐selectivity in parentheses. ^[32]

Scheme 9.

A PNP‐pincer manganese complex for the Z‐selective semi‐hydrogenation of alkynes applied to various substrates. ^[33] Isolated yields and Z/E ratios are given next to each substrate.

Scheme 10.

Proposed mechanism of PNP/manganese‐catalyzed hydrogenation (left) and isomerization (right); from DFT calculations; R= ⁱ Pr. ^[33]

The same group later reported a similar catalyst based on a [MnBr(CO)₂(PNP)] for Z‐selective semi‐hydrogenation without relying on ammonia borane as a hydride donor; both hydrogen atoms stem from methanol (Scheme 11). ^[34] After elimination of HBr by means of a base, methanol protonates the Mn=N bond and coordinates to the Mn center, resulting in a manganese methoxide intermediate. β‐Hydride elimination releases formaldehyde and generates a hydrido species, which further reacts with the alkyne under standard hydrometallation‐protonation steps to form the (Z)‐alkene. The system operates under forcing conditions: 150 °C, 2 equivalents of Cs₂CO₃ as the base and 14 h of reaction time in a methanol/toluene mixture are required. Nonetheless, the authors report a broad substrate scope with high yields and Z‐selectivities.

Scheme 11.

A PNP‐manganese complex for the semi‐hydrogenation using methanol as the hydrogen source. Both the hydrido ligand on manganese as well as the hydrogen atom on the nitrogen were proven to originate from methanol. ^[34]

In a recent example, Torres‐Calis and García reported an E‐selective semi‐hydrogenation of internal alkynes using isopropanol as the sole hydrogen source and [MnBr(CO)₃(dippe)] as pre‐catalyst (dippe=1,2‐bis(diisopropylphosphino)ethane) under basic conditions (4 mol% [Mn], 10 mol% NaOMe, ⁱ PrOH/THF, 100 °C, 4 h). ^[35] The authors proposed an initial displacement of the bromido by isopropanolate, followed by β‐hydride elimination to provide a hydridomanganese complex. Subsequent hydrometallation of the alkyne and proto‐demetallation by isopropanol gave the (Z)‐alkene. In a separate cycle, the (Z)‐alkene is isomerized, dictated by the thermodynamic equilibrium between E‐ and Z‐isomers. Diphenyl acetylene was converted to stilbene with 98 % yield and E/Z=97/3. Substituted diarylalkynes generally gave much lower yields and selectivities; 4‐(phenylethynyl)toluene afforded the alkene in 51 % yield with E/Z=52/48. Dialkylalkynes were unreactive.

The first two examples for a direct semi‐hydrogenation using gaseous H₂ as opposed to a transfer hydrogenation were reported nearly parallel by Zubar et al. ^[36] and by Garbe et al. ^[37] Both systems again operated with pincer‐manganese complexes, which form the active hydride species upon hydrogenation of the Mn=N double bond in the presence of a base. It was proposed[ ³⁶ , ³⁷ , ³⁸ ] that the alkyne hydrogenation step occurs via an outer‐sphere mechanism (Scheme 12). Similar reaction conditions were applied in both cases, employing 1–5 mol% [Mn], 2.5 equiv. KO ^t Bu with respect to [Mn], 30–60 °C, 20–30 bar H₂ and 16 h in a non‐polar solvent (heptane or toluene). High yields (>90 %) and high selectivities (>90/10) towards the (Z)‐alkenes were achieved.

Scheme 12.

Comparison between two Mn catalysts for the semi‐hydrogenation of alkynes using H₂. After initial formation of the hydrido‐Mn species, an outer sphere mechanism was proposed for the alkyne hydrogenation step.[ ³⁵ , ³⁶ , ³⁷ ] Top left: Mn catalyst of Zubar et al. ^[36] Top right: Mn catalyst of Garbe et al. ^[37] P and X are used as placeholders.

A recent example of an E‐selective semi‐hydrogenation was reported by Farrar‐Tobar and co‐workers employing a Mn hydrido complex as the catalytically active species. ^[39] The reaction proceeds both with gaseous H₂ (1 mol% [Mn], 30 bar H₂, 16 h, 60 °C, toluene) and H₂ generated in situ from the methanolysis of KBH₄ (1 mol% [Mn], 0.5 equiv. KBH₄, 20 h, 90 °C, MeOH). The authors performed mechanistic experiments and detailed DFT calculations that suggest a Z‐selective hydrogenation step, followed by an isomerization to the (E)‐alkene (Scheme 13). This protocol tolerated esters, aryl halides (F, Cl, Br), alcohols and phenols, amines and heteroarenes. The use of in situ generated hydrogen instead of 30 bar H₂ drastically improved the yields and selectivities of the reaction (Table 2). Dialkyl‐substituted alkynes showed high yields of alkene, but only low selectivities of 67–75 % for the in situ method.

Scheme 13.

Mechanistic proposal for the Z‐selective semi‐hydrogenation catalyzed by an Mn−H complex. ^[39] R= ⁱ Pr.

Table 2.

Use of an alkyl‐Mn pre‐catalyst for the E‐selective hydrogenation of internal alkynes using two different H sources. ^[39]

	Conditions A		Conditions B
Alkyne	Alkene [%]	E/Z	Alkene [%]	E/Z
	>99	>99/1	>99	>99/1
	90	>99/1	96	>99/1
	99	>99/1	44	98/2
	69[a]	>99/1	traces	/
	95	99/1	75	>99/1
	99[b]	98/2	54	73/27
	99	67/33	91	80/20

[a] 25 h, [b] 2 mol % [Mn].

7. Iron[ ¹¹ , ²⁷ , ⁴⁰ , ⁴¹ ]

Some of the earliest reports for iron‐catalyzed semi‐hydrogenations of alkynes stem from Thompson and co‐workers as far back as the 1940 s ^[42] and from Y. Urushibara, eponymous for the Urushibara catalyst, in the 1960s. ^[43] Thompson used an iron‐aluminium alloy, however, only very few aliphatic alkynes were successfully hydrogenated to (Z)‐alkenes at 70–100 bar H₂ and 100–150 °C. ^[42] A similar Fe−Al alloy catalyst was later described by Armbrüster et al. ^[44]

The Urushibara catalyst ^[43] is prepared by reducing an iron precursor (hydrates of FeCl₂ or FeCl₃) with zinc dust in water. After the exothermic reaction has ceased, the heterogeneous catalyst is washed with water and digested with acetic acid to obtain a powder which was subsequently applied as a catalyst for the Z‐selective hydrogenation of alkynes (2‐butyne‐1,4‐diol). However, high catalyst loadings (>50 wt%, due to a low catalyst surface area), and temperatures (80–100 °C, due to hydrogen uptake only starting at 80 °C) as well as an elevated H₂ pressure (>50 bar) were required. With these harsh reaction conditions, alternative iron‐catalyzed processes were highly desirable.

The [CpFe(μ₃‐CO)]₄ cluster has been employed for the semi‐hydrogenation of terminal and internal alkynes at elevated pressure (7–70 bar H₂) and temperatures (100–140 °C) by Pittman and co‐workers. ^[45] Internal alkynes reacted more sluggishly compared to terminal alkynes, the latter of which underwent slow isomerization to internal alkenes if alkyl‐substituted. No full hydrogenation to alkanes was observed.

Bianchini et al. ^[46] reported a semi‐hydrogenation of terminal alkynes using the PPPP‐pincer‐Fe hydride complex ([FeH(P(CH₂CH₂PPh₂)₃)(X₂)]BPh₄, X=H, N). Terminal alkynes were hydrogenated at 1 bar H₂ and 20–60 °C (Scheme 14).

Scheme 14.

Left: Fe‐tetraphos‐catalyst of the semi‐hydrogenation of terminal alkynes by Bianchini et al. ^[46] Right: Related transfer‐hydrogenation mechanism by Wienhöfer et al. using formic acid (4P=tetraphos=(P(CH₂CH₂PPh₂)₃). ^[47]

A similar catalyst was applied to the transfer hydrogenation of terminal alkynes by Wienhöfer et al. in 2012. ^[47] Fe(BF₄)₂ ⋅ 6 H₂O was employed in the presence of the tetraphos ligand (P(CH₂CH₂PPh₂)₃) with formic acid acting as the hydrogen donor (Scheme 14). Under optimized conditions (0.75–2.5 mol% [Fe], 2–3 equiv. formic acid, 40 °C, 5 h, THF), numerous terminal alkynes were converted quantitatively to alkenes while no full hydrogenation to alkanes was observed. Functional groups such as halides, esters, or alcohols were well tolerated. An iron hydride similar to the Bianchini complex ^[46] was proposed to form by decarboxylation of the intermediate formate complex. After coordination of the alkyne, the two H atoms are successively added to the alkyne by means of hydrometallation and reductive elimination. Deuteration experiments using DCO₂H for the hydrogenation of 4‐acetylphenylacetylene showed ∼50 % deuteration of the α‐position of the styrene, verifying the presence of an initial [Fe(HD)] species. Moreover, the β‐position is non‐selectively deuterated to yield a 1 : 1 E/Z‐mixture.

The iron catalysts of Bart et al. ^[48] and Gieshoff et al., ^[49] albeit developed for full hydrogenation to alkanes, initially form Z‐alkenes by semi‐hydrogenation of several internal alkynes, which can be obtained at shorter reaction times.

Another interesting approach was presented by the latter group already in 2014, which relied on a biphasic solvent/ionic liquids system containing iron nanoparticles, generated in situ by reduction of FeCl₃ with ethylmagnesium chloride (Scheme 15). ^[50] Z‐Selective semi‐hydrogenation was accomplished by attaching a nitrile moiety to the cation to serve as a selective catalyst poison to impede the otherwise occurring full hydrogenation, which is also observed in the absence of the ionic liquid altogether. Structural separation of the nitrile and ionic liquid functions by using a combination of acetonitrile and the commercially available ionic liquid [BMMIm]NTf₂ gave superior catalyst activities and stereoselectivities. The main advantage of this system lies in the fact that the ionic liquid may be reused several times (up to 8 reaction cycles) after phase separation, during which the IL‐stabilized iron nanoparticles remain in the IL phase, whereas the hydrogenation products remain in the organic phase. Disadvantageously, harsh reaction conditions proved necessary for this biphasic system, due to the sluggish interphase hydrogen transfer and the high viscosity of the ionic liquid at lower temperatures. Thus, the same group opted for a method with milder reaction conditions. ^[51] The reaction of Fe(acac)₂ with DIBAL−H in THF yielded a heterotopic catalyst capable of Z‐selective semi‐hydrogenation of alkynes under very mild conditions (1–5 bar H₂, 30 °C; Scheme 16). The observed reactivity may be explained by a “buffering” effect of the acetylacetonate ligand due to an initial ligand‐centered reduction, followed by a delayed, intramolecular reduction of the iron center. This effect prevents an otherwise rapid total reduction of the Fe(II) species and consequent agglomeration to larger particles (observed with TEM). This study additionally investigated the reduction step via XANES and EXAFS spectroscopy and compared it to the reduction of Fe(hmds)₂ (hmds = 1,1,1,3,3,3‐hexamethyldisilazane), the latter having been applied as an efficient catalyst for full hydrogenation to alkanes under comparable reaction conditions (Scheme 16). ^[52] The reduction of Fe(hmds)₂ by DIBAL−H initially results in a cluster structure, which subsequently undergoes further agglomeration in the presence of additional reductant.

Scheme 15.

An Fe‐nanoparticle‐catalyzed semi‐hydrogenation of alkynes in a biphasic organic solvent/ionic liquid medium. ^[50] Reaction conditions with IL‐2: 5 mol% [Fe], 1 equiv. MeCN, 60 bar H₂, 80 °C, 18 h, heptane/[BMMIm]NTf₂ (3/1). Yields (GC, ¹H‐NMR) and Z/E ratios are given below each substrate. [a] reaction in IL‐1, reaction time 48 h.

Scheme 16.

Iron‐catalyzed Z‐selective semi‐hydrogenation of alkynes under mild conditions (top). Role of the anionic ligand on the chemoselectivity of the catalyst (X=acac vs. N(SiMe₃)₂).[ ⁵¹ , ⁵² ]

In 2017, Tejeda‐Serrano and co‐workers reported a highly Z‐selective semi‐hydrogenation by means of iron(II,III) oxide nanoparticles immobilized onto an acidic metal oxide (ZnO, TiO₂, ZrO₂). ^[53] High conversions and selectivities were observed for most substrates at 10 bar H₂ and 100 °C. Additionally, this method offers a remarkable approach for the industrial removal of residual acetylene from a stream of ethylene gas, where the presence of alkynes may pose a problem for subsequent catalytic conversions of the alkene (e. g. polymerization). It was hypothesized that surface hydroxyl groups on the metal oxide solid support assist in the dissociation of coordinated hydrogen.

In 2019, Gorgas et al. ^[54] reported a Z‐selective semi‐hydrogenation using a bench‐stable cationic hydridoiron(II) complex, featuring a PNP‐pincer as well as an aminoborane ligand. The active hydrido‐dihydrogen complex is generated upon addition of hydrogen. Whilst terminal alkynes showed complete hydrogenation to the corresponding alkanes, this method delivered exclusively (Z)‐alkenes when internal alkynes, 1,3‐diynes, or 1,3‐enynes were hydrogenated, whereas no (E)‐alkenes or alkanes were observed at 5 bar H₂ after 1 h at room temperature. DFT calculations were performed to provide mechanistic insights (Scheme 17). ^[54]

Scheme 17.

PNP−Fe‐complex‐catalyzed semi‐hydrogenation of internal alkynes. (P=P ⁱ Pr₂, positive charges of the Fe complexes omitted for clarity). ^[54]

Beside these examples for a Z‐selective semi‐hydrogenation, a handful of protocols have been reported for E‐selective hydrogenations by means of iron catalysis. The first report by Srimani and co‐workers ^[55] used a PNP‐pincer‐Fe complex for an E‐selective semi‐hydrogenation at 90 °C and low hydrogen pressure (Scheme 18). Various functionalized alkynes were converted with high yield and selectivity. For some substrates, the fully hydrogenated alkane products could be observed, necessitating defined reaction conditions for every substrate by variation of reaction time and pressure. Isomerization experiments with (Z)‐stilbene verified the E‐selectivity of the overall reaction: After initial Z‐selective semi‐hydrogenation, the (Z)‐alkene undergoes rapid isomerization to its (E)‐isomer. A bimetallic catalyst for E‐selective semi‐hydrogenation was reported by Karunananda and Mankad. ^[56] Whereas iron could be included as part of the catalyst, the study mainly focused on ruthenium due to higher conversion and improved selectivity. The mechanistic proposal, based on DFT calculations of the transition state, implies that the hydrogenation step operates Z‐selectively at the “second” metal (copper or silver) and iron or ruthenium are responsible for the hydrogen activation. ^[56] A subsequent isomerization step leads to the formation of the (E)‐alkene. In 2019, Espinal‐Viguri and co‐workers reported a (nacnac)Fe‐catalyzed E‐selective transfer‐hydrogenation using a borane as the hydride source and an amine as the proton source. ^[57] This method was mainly applied for the hydrogenation of alkenes, however, a single internal and a handful of terminal alkynes were converted to the corresponding alkenes when one equivalent of amine and borane each is employed.

Scheme 18.

A PNP‐iron complex for the E‐selective semi‐hydrogenation of internal alkynes. GC yields of alkene, GC yields of alkane (in parentheses, if applicable), E/Z ratios, and reaction conditions are given next to each substrate. ^[55]

In addition to the methods presented here, a multitude of different systems have been reported for which a semi‐hydrogenation of alkynes could be observed upon shorter reaction times, although the intermediate alkenes were generally further hydrogenated as the reaction progressed. ^[58]

8. Cobalt[ ¹¹ , ²⁷ , ⁵⁹ , ⁶⁰ ]

An early example of cobalt‐catalyzed semi‐hydrogenation utilized the hydridocobalt complexes ([CoH(CO)₂(P ⁿ Bu₃)₂] and [CoH(CO)(P ⁿ Bu₃)₃]). ^[61] Both catalysts enabled selective hydrogenations of 1‐pentyne to 1‐pentene (30 bar H₂, 40–60 °C, 45 min), whereas nearly no isomerization to internal alkenes was observed. Upon longer reaction times (4 h), the hydrogenation of internal alkynes to (Z)‐alkenes proved feasible, however, a hydrogenation of the newly generated alkenes to alkanes was observed at even longer reaction times.

In 2016, Tokmic and Fout reported an E‐selective semi‐hydrogenation of internal alkynes using an NHC‐cobalt pre‐catalyst, which is converted into the active dihydrogen species in situ, and H₂. ^[62] The origin of the selectivity was explained by the dual role of the catalyst for initial Z‐selective hydrogenation (Scheme 19, left) as well as subsequent isomerization to the (E)‐isomer (Scheme 19, right). The semi‐hydrogenation has been tested for various substrates, for which generally high E/Z‐ratios were achieved under mild conditions (1–5 mol% [Co], 4 bar H₂, 30 °C, 17 h) in moderate to excellent yields (59–96 %, Scheme 20).

Scheme 19.

Proposed reaction mechanism for the cobalt‐catalyzed E‐selective semi‐hydrogenation, supported by (PHIP)‐NMR spectroscopy. ^[62] Left: Z‐selective semi‐hydrogenation; right: Z‐to‐E‐isomerization of alkenes.

Scheme 20.

Substrate scope for the cobalt‐catalyzed E‐selective semi‐hydrogenation by Tokmic and Fout. ^[62] Isolated yields and E/Z ratios are given below each substrate.

Similarly, Fu et al. ^[63] used a PNP‐pincer cobalt complex for a selective transfer semi‐hydrogenation using ammonia borane. With a variation of the PNP‐pincer ligand, both a Z‐ and an E‐selective hydrogenation were possible (Scheme 21). In the initial mechanistic proposal, steric effects due to a difference in bulkiness of the ligands ( ⁱ Pr vs. ^t Bu) played an important role. Later detailed mechanistic investigations ^[63b] revealed that the ^t Bu ligand could, in fact, promote the isomerization of the (Z)‐alkene to the (E)‐alkene as well. However, the isomerization process is hindered due to catalyst deactivation after ligand dissociation of one P‐ligand arm.

Scheme 21.

PNP‐pincer cobalt complexes as catalyst precursors for the selective E‐ or Z‐selective semi‐hydrogenation of alkynes. ^[63a]

A 2019 report from Gramigna and co‐workers documented that a bimetallic Zr−Co complex is active in the semi‐hydrogenation of alkynes. ^[64] Diphenylacetylene was converted primarily to stilbene after 30 h at 23 °C and 1 bar H₂, using 10 mol% of the Zr/Co‐complex (85 % conversion, 93/7 alkene/alkane, 79/21 Z/E). The reaction time could be decreased at elevated temperature, which in turn led to a near complete loss of Z‐selectivity (60 °C: 3 h, 57/43; 75 °C: 1 h, 51/49). Remarkably, a mechanism for the direct formation of the (E)‐alkene based on the bimetallic complex was proposed instead of a hydrogenation‐isomerization pathway.

A user‐friendly protocol for a Z‐selective semi‐hydrogenation has been developed by Chen et al., using a mixture of cobalt(II) acetate tetrahydrate, sodium borohydride and ethylenediamine in an ethanol/THF/water mixture. ^[65] In the absence of ethylenediamine, full hydrogenation to the alkane was observed. An extensive scope of functionalized substrates was tested with virtually full conversion and high Z/E ratios (Scheme 22). This method is especially attractive for practical use, as no prior ligand or complex syntheses is required, paired with mild reaction conditions. Unfortunately, no further mechanistic studies have been carried out.

Scheme 22.

Substrate scope for a semi‐hydrogenation protocol based on cobalt catalysis. ^[65] Conversions and Z/E/alkane ratios given below substrates.

A phosphine‐free NNN‐pincer ligand was employed for Z‐selective transfer semi‐hydrogenation of alkynes by Landge and co‐workers, using ammonia borane in methanol as the hydrogen source (Scheme 23). ^[66] In the presence of the cobalt complex, ammonia borane is converted into B(OMe)₃, releasing hydrogen, which is subsequently incorporated into the alkyne. The solvent methanol was revealed to exhibit a mediating effect on the protonation step, due to monodeuterium incorporation into the alkyne upon usage of CD₃OD. Numerous internal alkynes were selectively converted to the corresponding (Z)‐alkenes, while practically no formation of (E)‐alkenes or alkanes was observed in most cases. Moreover, terminal alkynes were successfully converted to terminal alkenes, which could then be further hydrogenated to the alkane upon increasing the reaction temperature and addition of additional ammonia‐borane and catalyst.

Scheme 23.

An NNN‐pincer cobalt complex for the semi‐hydrogenation of alkynes. ^[66] Isolated yields and Z/E ratios given below each substrate. [a] Yield of alkanes after addition of 4 mol% [Co] and additional ammonia‐borane at elevated temperature and longer reaction time.

In a 2016 report by the RajanBabu group, a Co(II) complex bearing a tridentate 2,6‐bis(arylimino)pyridine ligand was employed in the reduction of terminal alkynes and alkenes using (EtO)₂Si(Me)H as hydrogen source and a catalytic amount of NaBHEt₃ (Scheme 24). ^[67] While primarily developed for the reduction of alkenes to alkanes, the reduction of alkynes can be controlled toward selective alkene formation by the silane stoichiometry. The catalytically active Co species is presumably generated by addition of NaNHEt₃ at −78 °C, the silane is then introduced and the mixture is gradually warmed. Mechanistic investigations strongly suggested that both H atoms originate from the silane, as the use of deuterated solvents or the presence of D₂O did not effect deuterium incorporation. Additionally, the product yields roughly corresponded to the equivalents of silane when sub‐stoichiometric amounts were used. The authors proposed initial formation of a hydridocobalt(I) species by NaBHEt₃, followed by the insertion of the alkyne into Co−H. Reaction with the silane affords the alkene and a Co−Si species, which regenerates the Co(I) hydride and forms polymeric siloxanes. Only four terminal aliphatic alkynes were reported; functional groups such as bromides, alcohols, amines, carbonyls and epoxides were tolerated under the conditions; yields were consistently excellent (generally >90 %).

Scheme 24.

Cobalt‐catalyzed reduction of terminal alkynes by RajanBabu and coworkers. ^[67]

In 2017, a heterogeneous cobalt catalyst capable of a Z‐selective semi‐hydrogenation of both internal and terminal alkynes using hydrogen gas was developed by Chen et al. ^[68] The cobalt catalyst was prepared via pyrolysis of a Co(OAc)₂/phenanthroline mixture in the presence of SiO₂ as solid support. The thus synthesized cobalt nanoparticles were encapsulated in an N‐doped graphene layer and were used to hydrogenate various different internal alkynes (Table 3, top). A similar report from the same year also utilized N‐graphitic modified (NGR) cobalt nanoparticles. ^[69] In this process, ammonia borane was used as the hydrogen source, allowing for milder reaction conditions and thus a minimization of potential isomerization reactions occurring at elevated temperatures. Again, internal and terminal alkynes were selectively converted. Additionally, it was demonstrated that the heterogeneous Co catalyst could be reused after filtration without loss of activity or selectivity for up to nine cycles (Table 3, bottom).

Table 3.

Heterogeneous N‐graphitic cobalt catalysts for the hydrogenation of internal and terminal alkynes. Entries 1–6; ^[68] entries 7–10. ^[69]

Entry	Alkyne	Cond.	Alkene [%]	Z/E
1[a]		A	98 (96)[b]	9/1 (9/1)[b]
2[c]		A	95	6/1
3[a]		A	99	1.5/1
4[d]		A	63	mostly Z + isomers
5[a]		A	73	4 % alkane
6[e]		A	66	13 % alkane
7[f]		B	93	100/0
8[f]		B	90	100/0
9[g]		B	99	–
10[g]		B	99	–

Conditions: [a] 1 mol% [Co], 30 bar H₂, 15 h; [b] after three runs; [c] 3 mol% [Co], 40 bar H₂, 8 h; [d] 4 mol% [Co], 40 bar H₂, 10 h; [e] 1 mol% [Co], 30 bar H₂, 8 h; 100 °C; [f] 24 h; [g] 18 h.

In 2019, a cobalt‐catalyzed transfer semi‐hydrogenation using water as the hydrogen source was reported by Li et al. ^[70] The combination of cobalt(II) iodide with 3 equiv. Zn metal in the presence of 10 equiv. H₂O in methanol led to the formation of predominantly (Z)‐alkenes at 60 °C in generally high yields and selectivities. A noteworthy modification of this system included the use of the ligand dppe (1,2‐bis(diphenylphosphino)ethane) and the change of the solvent from methanol to acetonitrile. Using these modifications, the diastereoselectivity switched from (Z)‐ to (E)‐alkenes (Scheme 25). Mechanistic proposals for both systems feature an initial reduction of the Co(II) center (in CoI₂) to a Co(I) species by Zn, to which the alkyne can subsequently coordinate in preparation for the hydrogenation step. The E‐selectivity was hypothesized to result from steric hindrance of the ligand and from direct isomerization during the hydrogenation step. No isomerization was observed from free (Z)‐alkenes to (E)‐alkenes in control experiments.

Scheme 25.

Transfer semi‐hydrogenation of internal alkynes to (Z)‐alkenes (left) and (E)‐alkenes (right). ¹H‐NMR yields vs. 1,3‐benzodioxole and Z/E ratios are given. ^[70]

Bimetallic nanoparticles featuring cobalt and aluminium have been reported to fully hydrogenate terminal alkynes by Schmolke et al. ^[71] However, in the presence of coordinating groups para to the alkynyl substituent, semi‐hydrogenated products are obtained, likewise when the reaction time is shortened. One internal alkyne has been hydrogenated under the same reaction conditions, leading to the (Z)‐alkene as the major product.

A recent study by the group of Petit ^[72] used the commercial Co(I) precursor [CoCl(PPh₃)₃] to generate cobalt nanoparticles via disproportionation in THF. Rapid agglomeration and subsequent deactivation of the catalyst was inhibited by the addition of fatty alcohols. The nanoparticles were probed for the semi‐hydrogenation of internal alkynes, leading to mainly (Z)‐alkenes. Alkynes featuring coordinating groups such as alcohols instead delivered mainly the (E)‐alkenes and considerably higher amounts of fully hydrogenated alkanes. The coordinating groups allow for strong binding to the Co‐NP surface, resulting in a longer contact time with the catalytically active site and thus in the full hydrogenation of the (Z)‐alkenes to alkanes, while the (E)‐alkenes were mostly preserved. A reduction of H₂ pressure evaded the formation of alkanes and retained the Z‐selectivity, albeit with lower overall conversion.

An innovative visible‐light‐driven transfer semi‐hydrogenation was reported in 2020 by Tian and co‐workers. ^[73] The photo‐catalytic system revolves around an Ir(III) photocatalyst, which can oxidize the formal hydride source diisopropylethyl‐amine upon photoexcitation to generate an Ir(II) species and an iminium ion after formal elimination of a hydrogen radical. The Co(II) from the cobalt source is subsequently reduced by Ir(II), closing the photocatalytic cycle and leading to Co(I). Protonation results in a Co(III) hydride, which can Z‐selectively hydrometallate the alkyne; protodemetallation and two‐fold reduction of the Co(III) species in two separate photocatalytic cycles re‐forms the Co(I) species while the (Z)‐alkene is isomerized to the (E)‐alkene under photocatalytic conditions (Scheme 26). Cobalt(II) bromide paired with tri‐n‐butylphosphine was used as the cobalt source, while acetic acid was employed as the proton source. After 14 h at room temperature, several terminal and internal alkynes were converted to the alkene products in high yields. However, due to the photocatalytic E/Z‐isomerization being an equilibrium reaction, the Z/E‐ratios obtained by this method were low (approx. 70/30 in most cases).

Scheme 26.

The hypothesized catalytic cycles involved in the photochemical transfer semi‐hydrogenation of alkynes. ^[72]

A potential industrial/technical application for the semi‐hydrogenation of propyne to propene has been achieved with N‐doped carbon nanotubes on Ti₃C₂ sheets, embedded with cobalt nanoparticles. ^[74] Propyne could selectively be hydrogenated to propene with a selectivity of ∼96 % while achieving practically full conversion.

9. Nickel[ ¹¹ , ²⁷ , ⁷⁵ , ⁷⁶ ]

The earliest report for a Z‐selective semi‐hydrogenation using nickel was the usage of “P‐2 Nickel” by Herbert Brown in 1963, ^[77] only 11 years after Lindlar's original publication. ^[7b] A colloidal black catalyst was obtained after reduction of Ni(OAc)₂ with sodium borohydride in ethanol. Addition of ethylenediamine improved the selectivity of the reaction. A handful of alkynes were hydrogenated with near‐exclusive formation of the (Z)‐alkenes in high yields under mild conditions (1 bar H₂, room temperature). The “P‐2 Ni” method of (semi‐)hydrogenation is relatively well‐established and has found widespread applications in synthetic chemistry. These include, among others, the total synthesis of the macrolide aplyolide A, ^[78] the gypsy moth pheromone disparlure, ^[79] the hydroxylated polyunsaturated carboxylic acid 13R‐HDPA as a biochemical precursor to resolvins ^[80] as well as several other examples (Scheme 27). ^[81] The P‐2 nickel protocol is undoubtedly popular in complex molecule synthesis, not least because of its long history, operational simplicity and high accessibility due to the cheap, commercially available reagents being employed. This method generally delivers exceptional Z‐selectivities in lightly functionalized molecules, however the yields and functional group tolerance are mediocre. This necessitates careful planning of the synthetic route to include the corresponding step as early as possible. For instance, whereas alcohols, esters, acetals, benzylic ethers, and carboxylic acids are tolerated, aryl halides undergo facile hydrodehalogenation. ^[81d] Occasionally, mixtures of unreacted starting material, (Z)‐alkene and alkane are observed after consumption of 1 equiv. H₂. ^[81j] In other cases, the chemoselectivity might be higher, as shown impressively by the semi‐hydrogenation of hepta‐1,4‐diyn‐1‐yl‐trimethylsilane as an intermediate step in the synthesis of 13R‐HDPA (Scheme 27, bottom). ^[80] The catalyst is able to discriminate between the alkyl‐ and the trimethylsilyl‐substituted alkyne to selectively reduce the alkyl‐substituted triple bond.

Scheme 27.

Application of P‐2 Ni in the total syntheses of (−)‐aplyolide A, ^[78] (+)‐disparlure ^[79] and 13R‐HDPA ((R)‐13‐hydroxy‐7Z,10Z,13R,14E,16Z,19Z‐decosapentaenoic acid). ^[80] en=ethylene‐1,2‐diamine.

A similar transfer hydrogenation system was reported in 2017 by Wen and co‐workers. ^[82] In this user‐friendly protocol, the authors reduced NiCl₂ in methanol using NaBH₄ to form nickel nanoparticles, which were characterized via TEM (average particle diameter: 6.2±0.3 nm) The hydrogen atoms incorporated into the alkyne during the hydrogenation step originate from both the reductant NaBH₄ (as hydride) as well as methanol (as proton), which was proven by deuteration experiments. Low amounts of nickel (0.4 mol%) and 0.5–8.0 equiv. of NaBH₄ were sufficient for a quantitative conversion to the (Z)‐alkenes within minutes at room temperature (Scheme 28).

Scheme 28.

A simple nickel‐catalyzed transfer semi‐hydrogenation protocol for terminal and internal alkynes. ^[82] Conversions and product selectivities (Z/E/alkane) are given below each substrate. [a] 12 % 4‐octanol formed.

Very recently, Hale and co‐workers reported a diastereoselective semi‐hydrogenation using molecular hydrogen at ambient pressure and a nickel complex featuring a PSiP pincer ligand. ^[83] In most cases, (E)‐alkenes were obtained from the respective alkynes. For alkynes bearing electron‐deficient aryl substituents, the selectivity switched towards the (Z)‐isomer (Scheme 29). The authors propose a (Z)‐selective semi‐hydrogenation cycle in combination with a separate E/Z isomerization cycle via a hydrometallation/β‐hydride elimination pathway.

Scheme 29.

Stereoselective Ni‐catalyzed semi‐hydrogenation of internal alkynes with H₂. ^[83] Conditions: 1–2.5 mol% [Ni], 1 bar H₂, benzene‐d₆ (0.1 M), r.t., 4 h. ¹H‐NMR yields and E/Z ratios are given below each substrate.

The application of nickel nanoparticles as hydrogenation “catalysts” without hydrogen gas was previously shown by Alonso et al. ^[84] The Ni‐nanoparticles were prepared via reduction of NiCl₂ with 4 equiv. lithium in the presence of a catalytic amount of 4,4’‐di‐tert‐butylbiphenyl as electron carrier in THF. Two equivalents of ethanol or isopropanol were added as the hydrogen source for the semi‐hydrogenation step. This method requires the use of stoichiometric or superstoichiometric amounts of Ni and Li, which poses a major drawback.

Additional nickel‐catalyzed transfer semi‐hydrogenation systems have been reported that utilize silane/water or methanol, ^[85] amines, ^[86] phosphinic acid, ^[87] or zinc/formic acid ^[88] as hydrogen sources. [Ni(η²‐C,C‐alkyne)(dippe)] complexes have been employed by Barrios‐Francisco and García for the semi‐hydrogenation of internal alkynes in 2009. ^[85] The complexes were formed using the dimeric [NiH(dippe)]₂ as precursor. Reduction of the alkyne in a stoichiometric nickel‐mediated reaction yielded the (Z)‐alkene in the presence of water and the (E)‐alkene in the presence of methanol as hydrogen sources. The phosphine ligand is oxidized irreversibly during the course of this reaction, thus preventing a catalytic application. This drawback could be circumvented with the addition of triethylsilane as an oxygen scavenger and hydrogen source. No silane was required when methanol was used as the hydrogen source due to the potential oxidation of methanol to formaldehyde. High reaction temperatures (180 °C) and long reaction times (48 h) were required for the reductions (shorter reaction times were possible for the reactions with methanol). Only limited mechanistic details were provided, particularly the origin of the E‐ or Z‐selectivity with different hydrogen sources remained unclear.

The same group later reported a similar system based on the [NiH(dippe)]₂ complex using amines as the hydrogen source. ^[86] High temperatures (180 °C) and long reaction times (72 h) were necessary for the desired outcome of the reaction; the nature of the solvent (coordinating or non‐coordinating) as well as the amine also played a crucial role in the conversion and selectivity. The (E)‐alkene was the main product in most cases. An initial formation of the (Z)‐alkene and subsequent isomerization under the reaction conditions was proposed as the main pathway (Scheme 30).

Scheme 30.

Proposed mechanism for the initial formation of the (Z)‐alkene with a [NiH(dippe)]₂ pre‐catalyst. The [Ni(dippe)(η²‐DPA)] (DPA=diphenylacetylene) was isolated and characterized by X‐ray crystallography. ^[86]

In 2014, Chen et al. reported the use of phosphinic acid as the hydrogen source, paired with [NiCl₂(dppp)] (dppp=1,3‐bis‐(diphenylphosphine)propane) in acetic acid for an (E)‐selective semi‐hydrogenation of internal alkynes. ^[87] It was hypothesized that Ni(0) is operating as the catalytically active species (Scheme 31, top). After initial formation of the (Z)‐alkene in the transfer hydrogenation step, the alkene undergoes a subsequent isomerization to its (E)‐alkene, as demonstrated with isomerization experiments (Scheme 31, middle). Near‐exclusive formation of the (E)‐alkenes was observed for most substrates in high yields after 4–48 h of reaction time (Scheme 31, bottom).

Scheme 31.

An E‐selective nickel‐catalyzed transfer semi‐hydrogenation of internal alkynes using phosphinic acid as the hydrogen source. ^[87] Reaction times, isolated yields and E/Z ratios are given below each substrate.

A transfer hydrogenation protocol that enables access to both the (E)‐ and (Z)‐isomer by variation of the nickel catalyst has been developed by Richmond and Moran in 2015. ^[88] Zinc in combination with formic acid was employed as the hydrogen source, hydrogenating internal alkynes at 120 °C in dioxane. To achieve a Z‐selective semi‐hydrogenation, NiBr₂ was used as the precursor, whereas [NiCl₂(dme)] (dme=1,2‐dimethoxyethane) combined with triphos (bis(diphenylphosphinoethyl)phenyl‐phosphine) gave predominantly (E)‐alkenes (Scheme 32). These starkly different reaction outcomes are rationalized by an inherently Z‐selective semi‐hydrogenation by means of either nickel catalyst, whereas the presence of triphos enables an additional Z‐to‐E‐isomerization cycle. Mechanistic insight was gained by isomerization experiments: (Z)‐Stilbene underwent complete isomerization to (E)‐stilbene in the presence of [NiCl₂(dme)] and triphos within 1 h at 120 °C, while no isomerization was observed when only NiBr₂ was present.

Scheme 32.

A nickel‐catalyzed transfer semi‐hydrogenation yielding (Z)‐ or (E)‐alkenes. ^[88] Reaction conditions: 5 equiv. Zn, 5 equiv. HCO₂H, dioxane, 120 °C, 16 h. Isolated yields and Z/E ratios are given below each substrate.

A similar reaction featuring excess zinc in H₂O as the hydrogen source was reported six years later by Li et al. ^[89] In the presence of 1,10‐phenanthroline and 5 mol% Ni(OTf)₂, aryl‐ or alkyl‐substituted alkynes were converted to primarily (Z)‐alkenes after 22–36 h at 80 °C in THF. The use of NiI₂ instead altered the selectivity towards (E)‐alkenes at slightly lower temperatures and longer reaction times (60 °C, 40–72 h, Scheme 33). As (Z)‐stilbene was not isomerized to (E)‐stilbene under the latter conditions, the authors ruled out a stepwise hydrogenation‐isomerization mechanism and hypothesized that the intermediate vinyl‐Ni complex, obtained after cis‐selective hydrometallation of the alkyne, undergoes direct isomerization without formation of intermediate (Z)‐alkene. This isomerization is hypothesized to be hindered in the presence of coordinating triflate due to the formation of intramolecular hydrogen bonds (Scheme 34).

Scheme 33.

Scope of the nickel‐catalyzed semi‐hydrogenation reported by Li and co‐workers in 2021, affording (Z)‐ or (E)‐alkenes. ^[89] Reaction times, ¹H‐NMR yields vs. 1,3‐benzodioxole and Z/E ratios are given below each substrate.

Scheme 34.

Proposed divergent mechanisms of the semi‐hydrogenation shown in Scheme 30 above depending on the anion. ^[89]

An alternative example for a nickel‐catalyzed system with switchable selectivity was reported in 2019 by Murugesan and co‐workers. ^[90] Nickel(II) nitrate hexahydrate was used as nickel source; exclusive use of this salt in acetonitrile at 30 bar H₂ and 120 °C gave a heterogeneous catalyst for Z‐selective semi‐hydrogenation of internal and terminal alkynes (selectivities up to 99/1), a small amount of alkanes being observed for the latter. In the presence of multidentate ligands such as triphos, mostly (E)‐alkenes were observed. A homogeneous nickel catalyst was hypothesized in this case, as the ligand is able to stabilize the intermediate nickel hydride species. A hydrogenation‐isomerization mechanism was suggested (Scheme 35).

Scheme 35.

Proposed mechanism based on DFT calculations for the Z‐selective semi‐hydrogenation of internal alkynes and the isomerization to the (E)‐alkene catalyzed by a homogeneous nickel‐triphos catalyst. ^[90] The total positive charge of each nickel(II) complex was omitted for clarity.

A further improvement in the field of E‐selective semi‐hydrogenation was contributed by Thiel et al., ^[91] where NiI₂ or Ni(OTf)₂ in combination with dppf (1,1′‐bis(diphenylphosphino)‐ferrocene) as ligand offered a generally applicable catalytic protocol for the synthesis of (E)‐alkenes. Under similar conditions to the triphos‐system (30 bar H₂, 100 °C, dioxane, 16 h), ^[90] a multitude of alkynes were selectively hydrogenated, most remarkably including alkyl‐substituted alkynes, which have not been previously successfully hydrogenated. A similar hydrogenation‐isomerization mechanism is postulated.

Akin to the iron‐nanoparticle‐catalyzed semi‐hydrogenation in ionic liquids (see chapter 7), ^[50] a very related system containing nickel nanoparticles was later reported by Konnerth and Prechtl. ^[92] The nickel nanoparticles were directly prepared via thermal decomposition of [Ni(cod)₂] (cod=1,5‐cyclooctadiene) in the nitrile‐containing ionic liquid. Compared to the iron system, milder reaction conditions (1–4 bar H₂, 30 °C) were effective for Z‐selective hydrogenations of internal and terminal alkynes.

An alternative pathway employing Ni nanoparticles is the preparation from [Ni(cod)₂] in the presence of the CO₂ adduct of the dimethylimidazolium cation (Me₂Im‐CO₂) under H₂ atmosphere, generating NHC‐stabilized Ni‐NPs by decarboxylation of Me₂Im‐CO₂. ^[93] The small and uniform nano‐particles could be immobilized on carbon nanotubes (CNT); the immobilized catalyst was subsequently utilized for the Z‐selective semi‐hydrogenation of internal alkynes. Low catalyst loadings (3 mol%) and mild reaction conditions (5 bar, 50 °C, 16 h) were sufficient. The Ni‐NPs@CNT were tested in recycling experiments, showing only a minor loss of activity after four runs.

Murugesan et al. reported the preparation of monodisperse graphene‐encapsulated nickel nanoparticles for the Z‐selective semi‐hydrogenation of internal and terminal alkynes in 2019. ^[94] The nanoparticles were prepared via pyrolysis of a Ni(OAc)₂/fructose/SiO₂‐mixture at 800 °C. The heterogeneous catalyst could be used in multiple reaction cycles with a broad substrate scope (Scheme 36). Remarkable chemoselectivities were observed, as challenging substrates containing aldehydes, anhydrides or protic groups were well tolerated.

Scheme 36.

Substrate scope of the Z‐selective semi‐hydrogenation of alkynes using a Ni‐fructose@SiO₂ catalyst. ^[94] Isolated yields and Z/E ratios are given below each substrate.

A novel approach to Z‐selective nickel‐catalyzed semi‐hydrogenation was presented in 2021 by Valiente and co‐workers. ^[95] The authors employed dilute sulfuric acid as the hydrogen source, while the reaction itself is carried out electrochemically using a nickel foam cathode at room temperature within 3–8 h. Generally good yields and selectivities were obtained for a variety of functionalized substrates (Scheme 37a). The reaction most likely occurs at the surface of the nickel cathode, which is initially protonated due to the acidic medium. The alkyne is cis‐selectively hydrometallated and the free (Z)‐alkene is released upon 2e‐reduction, regenerating the active Ni surface. Redox neutrality is achieved by concomitant water oxidation (Scheme 37b).

Scheme 37.

a) Substrate scope of the electrochemical Z‐selective semi‐hydrogenation of internal alkynes using a Ni foam cathode. ^[95] CC=carbon cloth. Reaction times, ¹H‐NMR yields vs. 2,3,5,6‐tetrachloronitrobenzene and Z/E ratios are given below each substrate. b) Proposed mechanism for the electrochemical Z‐selective semi‐hydrogenation of internal alkynes using a Ni foam cathode. ^[95]

A similar electrochemical approach was recently developed by Lee et al., ^[96] who utilized [Ni(bpy)₃]²⁺ as pre‐catalyst. Based on cyclic voltammetry, UV and IR spectro‐electrochemistry, a mechanism was proposed (Scheme 38): [Ni(II)(bpy)₃]²⁺ is reduced to [Ni(0)(bpy)₂] which readily coordinates an alkyne to give a nickelacyclopropene complex. Alternating protonation and single electron reduction is believed to operate. The isolated nickelacyclopropene was shown to release the (E)‐alkene upon protonation. The absence of alkene hydrogenation indicates rapid dissociation of the alkene. Moderate yields of alkenes were obtained with moderate Z/E ratios. Only phenylpropargyl alcohol afforded high yield, stereoselectivity, and Faraday efficiency.

Scheme 38.

Electrochemical nickel‐catalyzed semi‐hydrogenation of alkynes by sequential protonation‐reduction events. ^[96]

A heterogenous catalyst containing zeolite‐immobilized nickel was applied in the hydrogenation of acetylene to ethylene by Chai and co‐workers. ^[97] The removal of trace acetylene impurities in gas feed stock of ethylene is a vital step prior to its polymerization to polyethylene (cf. section 7). Therefore, highly selective and active heterogenous catalysts which are robust and recyclable are desirable – both criteria are fulfilled in this case. Isolated Ni(II) centers confined in the zeolite structure were postulated to act as the catalytically active sites. The local electrostatic fields within the zeolite structure in combination with the nickel were hypothesized to induce heterolytic dihydrogen cleavage.

The pyrolysis of Ni(OAc)₂ ⋅ 4 H₂O in the presence of melamine yielded Ni₃N nanorods on an N‐doped carbon solid support, as was reported by Shi and co‐workers in 2020. ^[98] Terminal and internal alkynes were successfully semi‐hydrogenated in a Z‐selective fashion (Table 4). Sensitive functional groups such as carbonyls, esters or alcohols were well tolerated under the reaction conditions. As a downside, rather high catalyst loadings (7.8 mol%) and elevated H₂ pressures (20 bar) were required.

Table 4.

Solid‐supported Ni₃N nanorods as catalyst for the Z‐selective semi‐hydrogenation of terminal and internal alkynes. ^[98]

Entry	Alkyne	Time [h]	Conversion (alkene) [%]	Z/E
1		1	>98 (87)	–
2		5	99 (98)	98/2
3[a]		15	>99 (96)	95/5
4		3	>99 (89)	95/5
5[b]		11	>99 (99)	>99/1
6		2	>99 (96)	>99/1

Conditions: [a] 120 °C, 30 bar H₂; [b] 120 °C, 40 bar H₂.

In 2020, nickel‐containing bimetallic heterogenous catalysts were reported to be effective for semi‐hydrogenation of alkynes.[ ⁹⁹ , ¹⁰⁰ ] Ramirez and Lu utilized nickel combined with rare earth metals (Ga, La, Y, Lu) for E‐selective hydrogenation. ^[96] Comparatively mild reaction conditions were possible (Scheme 39). Only a limited substrate scope with few functional groups was reported, however, a detailed kinetic and mechanistic investigation was presented by the authors (Scheme 40). Depending on the metal the nickel is combined with (electronic and steric effects, Lewis acidity), different reaction pathways have been identified. The rare earth metal functions as a promoter that alternates electronic characteristics of the nickel center and thus influences the substrate binding properties of the catalyst while additionally providing a free coordination site for the phosphine ligand. The latter is a crucial feature, as the nickel center requires a free coordination site in order to coordinate the alkyne, which is enabled by the transfer of a phosphine ligand from the Ni to the rare earth metal.

Scheme 39.

Ni−M‐bimetallic catalysts for E‐selective semi‐hydrogenation of internal alkynes. ^{[99] 1}H‐NMR yields vs. 1,3,5‐trimethoxybenzene and E/Z ratios for M=Y are given below each substrate. [a] 60 h; [b] 3.5 mol% catalyst.

Scheme 40.

Mechanistic proposals for E‐selective semi‐hydrogenation of alkynes with bimetallic nickel‐M catalysts (M=Ga, Y, Lu): Hydrogenation (left) and E/Z‐isomerization (right). ^[99] Ph, iPr groups on N and P atoms are omitted for clarity. Top: Ni−Ga catalyst; H₂ coordinates to Ni, followed by alkyne coordination; phosphine ligand then dissociates. Bottom: Ni−M catalysts (M=Y, Lu); alkyne is coordinated first, followed by oxidative addition of H₂ and phosphine coordination to M.

The second example of a bimetallic nickel catalyst includes polyvinylpyrrolidone‐stabilized nickel‐iron nanoparticles that were prepared by reduction of NiCl₂ ⋅ 6 H₂O and FeSO₄ ⋅ 7 H₂O in aqueous solution using NaBH₄. ^[100] Different Ni‐to‐Fe ratios have been prepared and investigated: While the pure iron catalyst was inactive under the aqueous reaction conditions, the pure nickel and the bimetallic Ni₁Fe₁ and Ni₃Fe₁ catalysts gave primarily alkanes. With a Fe/Ni ratio of 3 : 1, the Ni₁Fe₃ catalyst showed highest semi‐hydrogenation activity, which could be further improved towards higher Z‐selectivities with further optimization and screening of additives. The addition of n‐butylamine and the iron content in the nanoparticles are essential for regulating the free adsorption sites for alkenes to prevent overreduction and improving the chemoselectivity.

10. Copper[ ¹¹ , ²⁷ , ¹⁰¹ , ¹⁰² , ¹⁰³ , ¹⁰⁴ ]

Interestingly, most examples of powerful copper‐catalyzed semi‐hydrogenations of alkynes have emerged in the last decade. One of the pioneering reports for a direct semi‐hydrogenation using dihydrogen was published by Semba et al. in 2015. ^[105] Simple copper(I) salts, such as CuCl in combination with phosphine ligands (PPh₃), were used in the presence of alkoxides (LiO ^t Bu) and a proton source ( ⁱ PrOH). A straightforward mechanism involving a heterolytic cleavage of the hydrogen followed by cis‐hydrocupration of the alkyne and final protonolysis of the vinyl copper species has been suggested (Scheme 41). Another related method utilized tethered alkoxidocopper(I) complexes formed in situ as catalysts for a highly Z‐selective semi‐hydrogenation of internal alkynes without over‐reduction. ^[106] NHC ligands bearing an alkoxide group were used as ligands to avoid the use of protic additives and additional alkoxides. An intramolecular heterolytic hydrogen bond cleavage was proposed to occur under the reaction conditions, forming a copper hydride species and a free alcohol group (Scheme 42), analogous to the mechanism depicted in Scheme 41. Further improvements were made with other NHC ligands by the same group. Use of the IPr ligand gave an air‐stable NHC‐copper(I) hydroxide complex for Z‐selective semi‐hydrogenation that circumvents the need for preactivation. ^[107] However, elevated hydrogen pressures (up to 100 bar) are necessary (Scheme 43).

Scheme 41.

Proposed mechanism of the Cu‐catalyzed Z‐selective semi‐hydrogenation of internal alkynes involving a copper(I) hydride species and an alcohol as the sources of hydrogen. ^[105]

Scheme 42.

A heterolytic hydrogen bond cleavage with a tethered copper(I) alkoxide complex. ^[106]

Scheme 43.

IPrCuOH‐catalyzed semi hydrogenation of internal alkynes. ^[107]

An alternative approach was presented in 2016 in a report by Wakamatsu et al. ^[108] The authors employed [Cu(O ^t Bu)(SIMes)] as the catalyst for a Z‐selective semi‐hydrogenation, generated in situ from [CuCl(SIMes)] and NaO ^t Bu in an octane/dioxane mixture (Table 5). This system functioned under atmospheric hydrogen pressure, albeit at elevated temperatures (100 °C). Esters were well tolerated, but no desired product was obtained in the presence of an aromatic bromide or nitrile.

Table 5.

[Cu(O ^t Bu)(SIMes)]‐catalyzed Z‐selective semi‐hydrogenation of internal alkynes. ^[108]

Entry	Alkyne	Alkene [%][a]	Z/E
1		68	>99/1
2		60	97/3
3[b]		66	93/7
4		80	92/8
5		88	90/10

[a] Isolated yields; [b] CuCl used instead of [CuCl(SIMes)].

Similar to the bimetallic Fe‐catalyst reported by Karunananda and Mankad (see chapter 7), a copper analogue of the same catalyst for the E‐selective semi‐hydrogenation of alkynes was reported. ^[56]

Silica‐supported copper nanoparticles were described extensively as heterogenous catalysts by the Copéret group. ^[109] The immobilized copper nanoparticles were prepared by the treatment of SiO₂ with the mesitylcopper(I) pentamer in toluene and subsequent hydrogenation at 300 °C. ^[109a] The catalyst was assessed for a variety of different alkynes, yielding only low selectivities for some substrates. This drawback was averted by the use of tricyclohexylphosphine (PCy₃) as the ligand: With this modification, both high yields and Z‐selectivities were achieved. Detailed mechanistic experiments were reported in order to evaluate the influence of the ligand.[ ^109b , ^109c ] Parahydrogen‐induced polarization experiments (PHIP) ^[109b] showed the feasibility of pairwise hydrogen addition on the surface with a significant influence of the PCy₃ ligand (0.2–0.6 % pairwise hydrogen addition to 1‐butyne without PCy₃ and ≥2.7 % with PCy₃). This result and the increased selectivity in the presence of the ligand were explained by a selective binding of the phosphine ligand to catalytically active, but more unselective sites. The coordination of PCy₃ on the surface also impedes the migration and scrambling of coordinated hydrogen. With further adsorption and kinetic experiments, ^[109c] it was shown that both alkynes and phosphine ligands, possess a higher adsorption coefficient than alkenes, thus preventing an over‐reduction to alkanes. This also explains the improved selectivity with phosphine ligands, but also other ligands, such as carbenes. With the previous results in hand, the same group investigated silica‐supported copper nanoparticles additionally modified with NHC ligands for the Z‐selective semi‐hydrogenation of alkynes. ^[110] Again, high selectivities and full conversion of alkynes was achieved.

In addition to copper catalysts that utilize gaseous H₂, numerous examples for transfer semi‐hydrogenations have been reported. One early report from 2014 focused on the conversion of terminal aryl‐ and alkyl‐alkynes with phosphinic acid in the presence of copper(II) citrate and catalytic amounts of urotropine, which inhibits the homocoupling of the terminal alkynes (Scheme 44). ^[111] Functional groups such as ethers, halides, esters, carboxylic acids, and internal alkenes and alkynes were tolerated under the reaction conditions. Mechanistic proposals involve the initial reduction of the copper(II) species to a copper(I) hydride. After coordination of the alkyne, a copper vinyl complex is formed after cis‐hydrometallation, which is then protonated and reduced by phosphinic acid. A potential side reaction involves the (reversible) formation of a copper alkynyl species, which was proven by deuteration experiments – all H atoms in the terminal alkene were nearly fully deuterated, which necessitates the prior deuteration of the terminal alkyne under the reaction conditions. A wide variety of copper‐catalyzed transfer semi‐hydrogenation systems employ the silane derivative polymethylhydrosiloxane (PMHS) as the hydride source and an alcohol (mostly ^t BuOH) as the proton source. ^[112] Examples with their respective copper catalyst, ligand, and reaction conditions are given in Table 6 with the appropriate literature reference. A special case of silane‐mediated semi‐hydrogenations has been reported by Duan et al., who developed a protocol for the conversion of alkynyl amides to Z‐configurated α,β‐unsaturated amides. ^[113] Specifically, an excess of Me(EtO)₂SiH is used in combination with 10 mol% of CuCl₂ and dppf in the presence of excess LiOMe in THF. At room temperature, a multitude of (hetero)aryl‐substituted alkynyl amides is converted to the (Z)‐alkene within 8 h in good to excellent yields. Unfortunately, the authors did not report the amount of (E)‐α,β‐unsaturated amide formed. Similar to such silane‐based protocols, a Z‐selective system based on an alcohol and boron was reported by Bao et al. in 2019. ^[114] In situ formed IMes‐Cu‐t‐butoxide was combined with bis(pinacolato)diboron and ethanol to form a copper(I) hydride intermediate, which then further reacts with the alkyne at room temperature within 5 h in a similar fashion to the mechanisms described earlier in this review. Similarly, an alternative B₂pin₂‐mediated approach was presented in a recent report by Huang and co‐workers. ^[115] The Cu catalyst is formed in situ by the action of 4,4’‐bipyridine on Cu(OAc)₂ 2H₂O. Contrary to Bao et al., ^[114] only a sub‐stoichiometric amount of KO ^t Bu (0.3 equiv.) is required and methanol is used as the hydrogen source as opposed to ethanol, but higher temperature (60 °C) and longer reactions times (18 h) were required.

Scheme 44.

Mechanistic proposal for the copper‐catalyzed transfer semi‐hydrogenation of terminal alkynes using phosphinic acid H₃PO₂. ^[111]

Table 6.

Cu‐catalyzed transfer semi‐hydrogenations of alkynes using combinations of silane/alcohol as formal hydrogen sources. ^[112]

Entry	Cu/L catalyst	Conditions	Alkyne Yield alkenes [%] (Z/E)
1 [112a]		2 mol % [Cu] C6H14/THF 1 : 1 r.t.–65 °C, 20 h
2[ 112b , 112c ]		2 mol% [Cu] PhMe 45 °C, 8 h
3 [112d]		5 mol% [Cu] PhMe r.t. −50 °C, 3–30 h
4 [112e]		5 mol % [Cu] C6D6 r.t. −65 °C, 2 h

[a] PMHS=polymethylhydrosiloxane.

With the air‐stable [CuOH(IPr)] catalyst developed by the Teichert group, ammonia‐borane has been used as hydrogen source for a Z‐selective transfer‐hydrogenation. ^[116] No additional bases or alcohols were required. Additionally, this protocol was successfully applied in the hydrogenation of α,β‐unsaturated esters. With the ambition to entirely avoid the production of stoichiometric amounts of Si‐ or B‐containing waste, a transfer semi‐hydrogenation protocol with an alcohol as the sole hydride and proton source was reported in 2019 by Teichert and co‐workers. ^[117] [CuCl(SIMes)] with stoichiometric amounts of NaO ^t Bu in ⁱ PrOH enabled Z‐selective semi‐hydrogenations of internal alkynes at 140 °C and prolonged reaction times (24 h).

In 2021, Moran and co‐workers described the Z‐selective semi‐hydrogenation of internal alkynes catalyzed by copper nanoparticles, using ethylene glycol as the hydrogen source in the presence of an excess of KOH (Scheme 45). ^[118] The Cu‐NPs were previously prepared by the reduction of CuSO₄ with NaBH₄ in a basic water/ethylene glycol mixture. A variety of aryl‐ and alkyl‐substituted alkynes bearing functional groups such as amines, ketones or halides were hydrogenated to the respective (Z)‐alkene with good to excellent yields and selectivities. Alkanes were not observed. Elevated temperatures (150 °C) are necessary, however, the authors achieved reduction of reaction time from 2 h with conventional heating to 30 min under combined microwave and ultrasound irradiation.

Scheme 45.

Scope of the Z‐selective semi‐hydrogenation of internal alkynes using Cu nanoparticles and ethylene glycol as hydrogen source. Isolated yields and Z/E ratios for conventional heating below each substrate. Yields and selectivities are similar with combined microwave and ultrasound radiation. ^[118]

Park et al. presented an alternative Z‐selective copper nanoparticle approach, using 10 mol% copper(II) sulfate and 3 equiv. ammonia‐borane in ethanol to generate the CuNPs under the reaction conditions. ^[119] This user‐friendly protocol delivers excellent yields and Z‐selectivities after 3 h at 40 °C, while tolerating critical functional groups such as amines, alcohols or heteroarenes. Both internal and terminal as well as aryl‐ and alkyl‐substituted alkynes are successfully semi‐hydrogenated. Usually, yields and Z‐selectivies >90 % were achieved, albeit with trace amounts of alkane occasionally being detected. At the same time, Kusy and Grela presented a similar approach, using CuCl₂ 2H₂O instead to generate the CuNPs in situ using ammonia‐borane in a water/THF mixture. ^[120] This approach gave similarly satisfactory results as the system by Park et al. Though with slightly higher temperatures (60 °C) and reaction times of up to 24 h, the authors achieved a lower catalyst loading (2 mol%) with a lower excess of H₃N‐BH₃ (2 equiv.) and presented a larger, more diverse substrate scope, additionally including halides, esters and acetals.

An innovative photochemical approach using a Cu@TiO₂ catalyst has been reported by Kominami et al., which enabled a Z‐selective transfer‐hydrogenation of internal alkynes at room temperature under UV irradiation. ^{[121] i} PrOH served as the hydrogen source.

11. Zinc[ ¹²² , ¹²³ ]

Besides examples where Zn merely acts as a solid support for other metal catalysts (e. g. iron nanoparticles immobilized on ZnO), ^[50] there is only a single report with zinc being directly involved as co‐catalyst. ^[124] The reduction of [PdCl₂(PPh₃)₂] with zinc metal in the presence of ZnI₂ afforded a catalytic system capable of an E‐selective semi‐hydrogenation of internal alkynes. Potential mechanistic ideas involve an initial Z‐selective hydrogenation step (heterogeneous catalyst) and a subsequent Z/E‐isomerization (likely a homogeneous process). It was postulated that ZnI₂ facilitates the reduction of in situ formed [PdI₂(PPh₃)₂] to heterogeneous Pd species and that ZnI₂ increases the E‐selectivity. Additionally, ZnI₂ seems to reduce the amount of alkene over‐hydrogenation and therefore improves the chemoselectivity of the overall process.

12. Summary and conclusions

For the early transition metals (scandium, titanium, vanadium, and chromium) as well as zinc only few examples for selective alkyne semi‐hydrogenations have thus far been reported. Manganese, iron, cobalt, nickel, and copper, on the other hand. were intensely investigated in the past decade. Given the multitude of astonishing breakthroughs and the sustainability of 3d metals, a decline in research interest is highly unlikely.

Various heterogenous systems based on nanoparticle catalysis as well as homogenous systems were explored in this review. Heterogenous catalysis is dominated by Z‐selectivity, while homogenous catalysis may lead to both Z‐ and E‐isomers. In the latter case, the (E)‐alkene is typically formed via a cis‐hydrogenation/isomerization sequence featuring a (Z)‐alkene intermediate. Several reports of homogenous catalysts bearing multidentate pincer ligands were presented. Their high popularity stems from their ability to effectively prevent the often highly reduced metals from agglomerating while still offering open coordination sites for the alkyne and hydrogen.

With Brown's discovery of P‐2 Nickel in 1963, ^[77] it quickly rose to fame to become the most common non‐noble‐metal approach to alkyne semi‐hydrogenation in synthetic applications, which is still in use today.[ ⁷⁸ , ⁷⁹ , ⁸⁰ , ⁸¹ ] Since then, substantial progress has been made in the field of nickel‐catalyzed semi‐hydrogenations in terms of yields, stereoselectivities, chemoselectivities and sustainability. Noteworthy examples of the publications explored in this review, without any claim to completeness, include the Ni‐fructose@SiO₂ catalyst by Beller et al. that enabled very high yields, stereoselectivities, and chemoselectivities (Scheme 36). ^[94] At 10 bar H₂ pressure and 100 °C in a high‐pressure reactor, the conditions tolerated halides, anhydrides, and aldehyde functions and continuously gave >99 % Z‐selectivities. However, preparation of the catalyst requires high‐temperature pyrolysis under conditions that may be out of reach for many laboratories. Another catalyst type that exhibited high levels of activity and selectivity is the solid‐supported Ni₃N nanorods by Shi and co‐workers (Table 4). ^[98] Chemoselective reactions proceeded in the presence of aryl bromides, nitriles, esters, alcohols and ketones. However, elevated pressures of H₂ (20 bar) were required. The nanorods need be prepared by pyrolysis of nickel(II) nitrate under precise conditions, which required an additional working step and proper equipment.

In contrast to its neighbor in the periodic table, Cu catalysts have only recently been discovered as semi‐hydrogenation catalysts with good levels of activity and stereocontrol under mild conditions. Examples include the use of copper nanoparticles generated by reduction of CuSO₄ using NaBH₄ by Moran and co‐workers, ^[119] bearing close resemblance to P‐2 Ni. The complete absence of dihydrogen or sophisticated hydrogen sources makes this protocol, employing available ethylene glycol as the H₂ source, easily accessible and sustainable. Excellent functional group tolerance towards ketones, aryl chlorides, free amines and N‐heterocycles significantly expand the possible substrate palette.

The use of a hydrido manganese complex by the Kirchner group ^[39] (Scheme 13, Table 2) is a rare, remarkable example of an E‐selective semi‐hydrogenation. High yields, selectivities and functional group tolerance paired with low‐cost reagents and relatively mild conditions makes for an operationally simple protocol. However, inconveniently, the pre‐catalyst must be synthesized in multiple steps. Another striking example stems from the cobalt realm: The system developed by Li and co‐workers ^[70] (Scheme 25) features switchable stereoselectivity directed merely by the presence or absence of the phosphine ligand. The mild conditions tolerate various vulnerable functions such as halides, nitriles and esters, as well as free alcohols and amines. Moreover, water is used as the sole hydrogen source.

The direct use of water under either reductive or electrolytic conditions as opposed to using pre‐produced (green) hydrogen is a strong candidate for future‐proof protocols with further improvement in sustainability and efficiency. In this regard, the recent resurgence of electrochemical synthesis has enabled innovative protocols of stereoselective alkyne hydrogenations.[ ⁹⁵ , ⁹⁶ ] The ability to fine‐tune the elemental steps of single‐electron transfer and protonation by independent variations of electrode materials, electrolytes, conditions, metal‐ligand complexes, and Brønsted acids gives rise to an enormous optimization potential. We firmly believe that electrochemical hydrogenation protocols will receive substantial research interest in the near future.

Overall, a range of problems and drawbacks of 3d‐metal‐catalyzed semi‐hydrogenation pose challenges which must be overcome before mainstream use is conceivable. Several (reduced) 3d metal complexes are highly susceptible to water and oxygen, requiring inert conditions with rigorous exclusion of both. Moreover, their significant oxophilicity and general vulnerability to coordinating atoms lead to functional group tolerances and chemoselectivities that are often poor to mediocre. Particularly functions that are easily reduced, such as halides, ketones, aldehydes, esters, nitriles, or acidic substrates with alcohols or carboxylic acids frequently lead to poor yields.

However, enormous progress has been made in the past years, with significantly milder reaction conditions having been achieved. From the frequently reported harsh conditions (>50 bar H₂, >100 °C, long reaction times), high catalyst loadings, laborious catalyst preparation and low yields and selectivities in early publications, remarkable improvements in all areas have been accomplished. A plethora of new reports now demonstrate ambient conditions with atmospheric hydrogen pressure or transfer hydrogenation reagents, permitting a broader application with a simpler reaction set‐up. This allows for several user‐friendly protocols using commercially available precursors under accessible conditions which might threaten to dethrone the Lindlar catalyst as the method of choice for alkyne semi‐hydrogenation in the foreseeable future.

Conflict of interest

The authors declare no conflict of interest.

Biographical Information

Bernhard J. Gregori hails from the picturesque banks of the Bavarian Danube. He studied Chemistry in Regensburg and graduated with a M.Sc. thesis on Fe‐catalyzed hydrogenations. He continued his work on catalytic reductions at the University of Hamburg and was minted Dr. rer. nat. in 2021. Bernhard recently joined Alex Breder's team at the University of Regensburg.

Biographical Information

Mattis‐Ole W. S. Schmotz is a native Hamburg kid. As an enthusiastic student at the University of Hamburg, he won the Best B.Sc. Student Prize and was awarded a fellowship of the fast track Ph.D. program of the Hamburg Research Academy. Mattis is engaged with research on photochemistry and catalysis.

Biographical Information

Axel Jacobi von Wangelin was born in Berlin and saw the graffiti‐free side of the wall collapsing. He later enjoyed vibrant college lives in Erlangen, Utah, Cardiff, Rostock, Stanford, Cologne, and Regensburg. Axel is currently Professor of Molecular Chemistry at the University of Hamburg. His group's research spans all aspects of catalysis with special emphasis on earth‐abundant metals and light‐driven reactions.

Dedicated to Matthias Beller on the occasion of his 60^th birthday

B. J. Gregori, M.-O. W. S. Schmotz, A. Jacobi von Wangelin, ChemCatChem 2022, 14, e202200886.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.