Volume 13, Issue 5 p. 1317-1325
Full Paper
Open Access

Ruthenium-Catalyzed E-Selective Partial Hydrogenation of Alkynes under Transfer-Hydrogenation Conditions using Paraformaldehyde as Hydrogen Source

Marcus N. A. Fetzer

Marcus N. A. Fetzer

Department of Chemistry, University of Cologne, Greinstr. 6, D-50939 Köln, Germany

These authors contributed equally to this work.

Search for more papers by this author
Dr. Ghazal Tavakoli

Dr. Ghazal Tavakoli

Department of Chemistry, University of Cologne, Greinstr. 6, D-50939 Köln, Germany

These authors contributed equally to this work.

Search for more papers by this author
Prof. Axel Klein

Prof. Axel Klein

Department of Chemistry, University of Cologne, Greinstr. 6, D-50939 Köln, Germany

Search for more papers by this author
Priv.-Doz. Dr. Martin H. G. Prechtl

Corresponding Author

Priv.-Doz. Dr. Martin H. G. Prechtl

Department of Chemistry, University of Cologne, Greinstr. 6, D-50939 Köln, Germany

Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais 1, 1049-001 Lisboa, Portugal

Search for more papers by this author
First published: 04 February 2021
Citations: 21

Dedicated to Prof. Dr. Luis A. Oro on the occasion of his 75th anniversary in 2020.

Graphical Abstract

Protocol E: E-alkenes were synthesized with up to 100 % E/Z selectivity via ruthenium-catalyzed semi-hydrogenation of alkynes. Paraformaldehyde was used for the first time as the hydrogenation source in presence of water for partial transfer-hydrogenation of alkynes to E-alkenes. A simple setup and inexpensive catalyst makes this protocol to be a feasible and promising stereo complementary procedure to the well-known Z-selective Lindlar reduction in late-stage syntheses.

Abstract

E-alkenes were synthesized with up to 100 % E/Z selectivity via ruthenium-catalyzed partial hydrogenation of different aliphatic and aromatic alkynes under transfer-hydrogenation conditions. Paraformaldehyde as a safe, cheap and easily available solid hydrogen carrier was used for the first time as hydrogen source in the presence of water for transfer-hydrogenation of alkynes. Optimization reactions showed the best results for the commercially available binuclear [Ru(p-cymene)Cl2]2 complex as pre-catalyst in combination with 2,2-bis(diphenylphosphino)-1,1-binaphthyl (BINAP) as ligand (1 : 1 ratio per Ru monomer to ligand). Mechanistic investigations showed that the origin of E-selectivity in this reaction is the fast Z to E isomerization of the formed alkenes. Mild reaction conditions plus the use of cheap, easily available and safe materials as well as simple setup and inexpensive catalyst turn this protocol into a feasible and promising stereo complementary procedure to the well-known Z-selective Lindlar reduction in late-stage syntheses. This procedure can also be used for the production of deuterated alkenes simply using d2-paraformaldehyde and D2O mixtures.

1 Introduction

Semi-hydrogenation of alkynes is a very attractive and challenging reaction and a wide spectrum of homogeneous and heterogeneous catalytic approaches has been developed in this context.1-8 This transformation is especially interesting from the point of view that alkynes are readily accessible from the well-known and reliable Sonogashira reaction.9-13 Alkenes are most versatile building blocks which can be converted easily into many other functional groups.14-16 Olefins can be found also in technically or medically important natural products, bulk chemicals, fragrances, pharmaceuticals, agrochemicals, and polymers14-16 In all these cases, the control over the stereochemistry is crucial.17 Reduction of alkynes selectively to Z-alkenes can be carried out using the well-known Lindlar catalyst,18 and even base metal catalysts.19, 20 Various transition metal catalysts5, 10, 13, 15, 36-51 have been developed so far which in combination with different transfer hydrogenation sources21-31 or dihydrogen gas32-35 can be used for the stereoselective semi-hydrogenation (reduction) of alkynes to olefins. In this regard, selective formation of E-alkenes is more challenging and less well-studied.36-40 Traditionally, E-alkenes obtain from the Birch reduction using dissolved metals which precludes the use of base-labile and sensitive substrates.41-45 To date, there is a handful of widely applicable direct procedures for the selective reduction of alkynes to E-alkenes by means of catalytic semi- (transfer)hydrogenation.30, 46-49 Scheme 1a shows some of the successful E-selective alkene synthesis procedures using H2 gas as the hydrogenation source.39, 40, 50-53 Application of H2 gas in these examples, however, requires a more expensive setup and increases the safety issues. Transfer semi-hydrogenation could be therefore, a good solution for this problem. In 2016, Lindhardt etal. reported Ru-catalyzed E-selective alkyne transfer semi-hydrogenation at low temperatures and low H2 pressure (Scheme 1b) on a large substrate scope of alkynes.17 The reactions were carried out in a special two-chamber reactor (COwareTM) or in a continuous flow reactor. Different substrates containing various functional groups were reduced under these conditions with very high yields and chemo- and stereo-selectivity. However, simple and easily scalable protocols, avoiding the use of sophisticated equipment, for sustainable E-selective transfer hydrogenation of alkynes complementing the Lindlar-type Z-selective reactions are so far not at hand.

Details are in the caption following the image

Previously reported catalytic reactions for E-selective semi-hydrogenation of alkynes compared to this work.17, 40, 50-52

Recently, the number of reports on the use of C1 molecules such as formic acid,21, 22, 24, 26, 28, 30, 37, 38, 54-57 formaldehyde,58-61 and methanol62 as hydrogen sources in transfer-hydrogenation reactions are increasing. Although it has been considered frequently as an inexpensive and safer substitute for pressurized hydrogen gas58, 63 in transfer hydrogenations and reductions, formic acid suffers from some drawbacks. High temperatures (up to 150 °C) often seem to be necessary in most of these reactions..37, 38, 54-57 And, its acidic nature causes some restrictions for acid sensitive substrates and thereby lowers functional group tolerance.37, 38, 54-57 To adjust the pH of the reaction media, addition of bases, such as amines is frequently inevitable. To overcome the problem of acidity of formic acid in such reactions, application of methanediol produced from formaldehyde in aqueous solutions could be suitable since it is less acidic and therefore, applicable for acid sensitive substrates and reactions under base-free conditions.58-60, 62-64

In continuation of our work to generate dihydrogen from aqueous formaldehyde,65, 66 or insitu generated aqueous formaldehyde from methanol62 and paraformaldehyde (pFA)66 for hydrogenations or as energy carrier, we have also used these hydrogen sources in combination with [Ru(p-cymene)Cl2] as pre-catalyst for transfer hydrogenation of alkenes (with methanol),62 for formaldehyde dismutation (with pFA in water),67 for reductive methylation (with pFA in water)61 and most recently for reductive deamination of nitriles to alcohols (with pFA in water).58 In terms of practical aspects, it turned out that the easy-to-handle non-volatile paraformaldehyde (oligomeric solid powder) is very convenient in comparison to the volatile C1-molecule competitors, methanol, aqueous formaldehyde and formic acid, while the hydrogen content of pFA is equal to formaldehyde referring to the monomer. Thus, pFA has a high potential to be used as liquid organic hydrogen carrier. Upon addition of water, pFA converts to methanediol with a hydrogen content of 8.4 % wt. which under the appropriate reaction conditions can be degraded to H2 and CO2 through an exothermic pathway.60, 63 On the other hand, pFA is a very cheap, stable and readily available solid which can be handled easily in industrial applications and in the lab and we are aiming to test its activity in other challenging hydrogenations or reductions.

Herein, we report, on a highly efficient E-selective, environmentally benign, base-free, operationally simple, mild and safe semi-hydrogenation of alkynes using a commercially available Ru catalyst and pFA as the hydrogen source (Scheme 1c). The presence of the commercially available 2,2-bis(diphenylphosphino)-1,1-binaphthyl (BINAP) proved to be necessary for reaching high conversions and selectivities. The reactions did not require high temperatures, long reaction times, inert conditions, degassed/dried solvents, or special set-ups. The use of safe, stable, cheap and solid pFA as hydrogen source is another promising feature of this procedure which makes it also feasible for industrial applications.

2 Results and Discussion

Inspired by our recent publication on reductive deamination of nitrile to alcohols,58 semi transfer hydrogenation of diphenylacetylene as the model substrate using aqueous pFA as hydrogen source in presence of the [Ru(p-cymene)Cl2]2 complex 1 as catalyst was tested initially (Table 1). In first experiments, adding no ligand (Entries 1 to 7) conversions up to about 30 % were obtained based on product formation determined by 1H-NMR through variation of the amount of pFA, temperature, and reaction time, although no stereo-selectivity was obtained under these conditions. This is remarkable in view of the report by Lindhardt etal. that semi transfer hydrogenation of diphenylacetylene with this catalyst using in-situ produced hydrogen gas or formic acid as the transfer hydrogenation source is not possible (Scheme 1).17 Reducing the reaction times did not enhance selectivity (Table 1, Entries 2–4). Increasing the temperature to 70 °C resulted in higher conversions but partial over-reduction occurred (Table 1, Entry 5). Decreasing the concentration of pFA was not beneficial (Table 1, Entries 6–7 and Table S3, SI).

Table 1. Optimization of the reaction parameters for the partial transfer hydrogenation of diphenylacetylene (DPA).

image

Entry

pFA (eq)

T [°C]

time [h]

ligand [mol %]

conv. [%]

E : Z:alkane

1

8

60

18

29

2

8

60

3

trace

3

8

60

6

10

4

8

60

12

18

5

8

70

18

36

6

7

70

18

21

7

5

70

18

<5

8

8

70

18

PhCN (10)

30

9

8

70

18

phen[b] (5)

10

8

70

18

NH2-PhCN[c] (10)

11

11

8

70

18

Br-PhCN[d] (10)

32

12

8

70

18

Ph-Py[e] (5)

37

13

8

80

18

Ph-Py[e] (5)

46

14[a]

8

80

18

PPh3 (4)

9

trace:0:>99

15[a]

8

80

18

(RO)3PO[f] (4)

68

28 : 27 : 45

16[a]

8

80

18

BINAP[g] (8)

100

71:0 : 29

17[a]

8

80

18

BINAP (4)

100

46:0 : 54

18[a]

8

80

12

BINAP (4)

100

54:0 : 46

19[a]

8

80

5

BINAP (4)

100

81:0 : 19

20[a]

8

80

2

BINAP (4)

100

>99 : 0 : 0

  • Reaction conditions: diphenylacetylene (1 mmol), 1 (1 mol%), pFA, ligand, toluene (2 mL), H2O (2 mL), T, time. [a] 2 mol% of 1 was used, [b] 1,10-Phenantroline, [c] 2-Aminobenzonitrile, [d] 4-Bromobenzonitrile, [e] 2-Phenylpyridine, [f] Tris(2-ethylhexyl) phosphate, [g] 2,2‘-bis(diphenylphosphino)-1,1‘-binaphthyl.

To investigate effects of the reaction medium on the reactions, different organic solvents were used in combination with water which resulted in the conclusion that a water toluene mixture, as well as 1,2-dichloroethane (DCE) are the best reaction media (Table S1, SI). Application of pure water resulted in only 57 % conversion while E/Z selectivity was 32 : 25 (Table S1). Toluene was selected for further experiments, while the toxic DCE was discarded. The ratio of toluene and water was varied in the next step and it was found that the best results can be obtained using a 1 : 1 ratio.

In the next steps, other first- and second- row transition metal catalysts were used instead of 1 in this reaction but in all cases, no better results were observed compared to the model reaction (Table S2) and therefore, optimization was continued using 1.

Different types of ligands were employed in the next step to increase the selectivity. Nitrogen-based ligands in most cases improved the conversions but had no significant effect on the selectivity (Table 1, Entries 8–13). To our delight, introduction of phosphorous-based ligands to the reaction medium had a significant effect on both the selectivity and/or conversion of the reaction (Entries 14–16). The best ligand in this series was BINAP with quantitative conversion and >99 % E-selectivity using 2 mol% of complex 1 and 8 mol% of BINAP (2,2‘-bis(diphenylphosphino)-1,1‘-binaphthyl). Decreasing the concentration of BINAP to 4 mol% resulted in higher reaction rates but also formation of diphenylethane as side-product (Table 1, Entry 17). In order to prevent over-reduction, the reaction time was reduced in the next step. The best results were obtained after 2 hours while no Z-stilbene or 1,2-diphenylethane was detected. Further decreasing the reaction time resulted in both lower conversion and lower selectivity.

Following these observations, we tried to test the catalytic activity and selectivity of the other Ru-based catalysts with structures similar to complex 1. The binuclear complexes 2–8 (close derivatives of 1) (Figure 1) were synthesized following our literature procedures61, 65, 67 and were employed in the reaction (Table 2). As we expected, most of these catalysts showed high activity in the model reaction. Similar results were obtained using complexes 24 compared with 1. As they were not significantly better, we focused on the use of 1 as a commercially available catalyst and Entry 20 (Table 1) was selected finally as the set optimized parameters resulting in full conversion and high selectivity after 2 hours reaction time, while a shorter reaction time of 1.5 hours gave lower conversion (91 %).

Details are in the caption following the image

The binuclear Ru complexes 18 used as pre-catalysts in this study.

Table 2. Comparison of the catalytic activity and selectivity of pre-catalysts 18 in the partial transfer hydrogenation of diphenylacetylene.

image

Entry

pre-catalyst

Conv. [%]

E : Z:alkane

1

1

100

100 : 0 : 0

2

2

100

100 : 0 : 0

3

3

100

99:0 : 1

4

4

100

94 : 6:0

5

5

84

66 : 21 : 13

6

6

58

59 : 41:0

7

7

10

8 : 2:0

8

8

33

17 : 16:0

  • Reaction conditions: diphenylacetylene (1 mmol), pFA (8 mmol, 240 mg), pre-catalyst (2 mol%), BINAP (4 mol%), H2O (2 mL), toluene (2 mL) at 80 °C for 2 hours.

To investigate the scope of our reaction, different kinds of aliphatic and aromatic alkynes were employed under the obtained optimized conditions (Figure 2). While other Ru-catalyzed systems reduced aliphatic or aryl-alkyl alkynes only in low yields (<25 %) and unselectively,68 semi transfer hydrogenation of dialkyl substituted alkynes as well as unsymmetrical aryl-alkyl substituted and sole aromatic ones was done successfully and with good to excellent E-selectivity using our catalytic system. The reaction times were also noticeably short.

Details are in the caption following the image

Partial transfer hydrogenation of aliphatic and aromatic alkynes under optimized reaction conditions. a Reaction conditions: alkyne (1 mmol), pFA (8 mmol, 240 mg), [Ru(p-cymene)Cl2]2 (1) (2 mol%), BINAP (4 mol%), H2O (2 mL), toluene (2 mL) at 80 °C for the appropriate time for full conversion if not otherwise stated (isolated yields given in brackets or *NMR yields). The ratios in second row in each case refer to E: Z: alkane ratios, respectively. E-selectivity in % at quantitative NMR yields (>99 %) if not otherwise stated; *NMR yields for: 10 h 95 %, 10k 96 %, 10 m 69 %, 10n 80 %, 10o 89 %, 10p 16 %, 10q 73 %, 10r 39 %. The isolated yield of 10e was lower owing to the lower boiling point of the product.

Functional group tolerance of the defined protocol is also appreciable. The presence of both electron-donating and electron-withdrawing groups on the aromatic ring can be tolerated very well although the reaction rates are higher using substrates bearing electron-donating groups. While proto-dehalogenation of halogen-containing alkynes is a known side-reaction in semi-hydrogenation using metal-based catalysts incl. Lindlar-type catalysts,19, 69 no sign of proto-dechlorination of substrate 9b was observed under our conditions offering additional flexibilities. Furthermore, reducible functional groups such as ester and nitro groups were tolerated successfully. However, for the nitro compound 9q yielding 10q around 10 % of the overreduction side-product was also detected in the mixture. It is noteworthy that the other catalytic systems used for the semi-hydrogenation of alkynes suffer from low chemoselectivity in regard to nitro containing substrates.50

The reaction of heterocyclic compound 9o to 10o resulted in good E : Z selectivity albeit it required longer reaction time. Low stereo-selectivity was observed, however, for the conversion of other heterocyclic compounds. For example, coordination of the pyridine group to the Ru center might deactivate the catalyst by the formation of Ru−N bonds. This assumption was confirmed via ESI MS analysis of the crude reaction mixture (Figure S67 SI). As the triple bond in this molecule is in para position of the pyridine ring, interaction of the triple bond with the active sites of the catalyst is probably hampered resulting in poor stereo-selectivity.

The hydrogenation of the nitrile-containing substrate 9p was found problematic as only 16 % of the corresponding E-alkene was formed alongside with the phenylmethanol derivative 10 p’ (Scheme 2). This is in line with our previously report on reductive deamination of nitriles to alcohols.58 The product ratio in Scheme 2 probably means that the nitrile reduction occurs faster than the reduction of triple bond, in line with the coordination of the CN moiety to the Ru center preventing the approach of the triple bond to the Ru and subsequent hydrogenation. In addition, we could synthesize cinnamyl alcohol, nitro cinnamic acid, carbamates and thio carbamates (Figure 2).

Details are in the caption following the image

Reduction of nitrile containing alkyne 9p under optimized conditions.

In order to get some mechanistic insight, the model substrate, diphenylacetylene, was reacted under the optimized conditions with H2O replaced by D2O and pFA replaced by d2-pFA. After 3 hours, the deuterated product 10u was obtained with >99 % E-selectivity and full conversion (Scheme 3a). In another experiment, normal pFA was used in combination with D2O as the hydrogen sources for this reaction (Scheme 3b). 71 % E-selectivity was achieved in the production of deuterated product under these conditions. A combination of d2-pFA and H2O, however, showed the possibility of H/D exchange during the catalytic cycle as it has been observed also in our previous study (Scheme 3c).58, 66 In this case, non-deuterated E-stilbene was formed as the main product. The use of D2O alongside with pFA gave superior yields for the formation of deuterated alkenes over the combination of H2O and d2-pFA.

Details are in the caption following the image

Isotope labelling experiments. Semi transfer hydrogenation of diphenylacetylene under standard conditions using a) D2O and d2-pFA; b) D2O and normal pFA; and c) d2-pFA and H2O. The degree of deuteration was determined by comparison of 1H NMR and 2H NMR as well as GC-MS analyses. In parenthesis, the average isolated yields of two experiments have been reported.

Unfortunately, these experiments did not reveal the origin of the E-selectivity in our reaction. One possibility could be the initial formation of kinetically favorable Z-product followed by its subsequent fast isomerization to the thermodynamically stable E-alkene. To verify this assumption, the formation of all possible products and the consumption of the substrate diphenylacetylene with time in the model reaction were examined by running several batch reactions under exactly the same conditions but different reaction times. This should allow to determine intermediates in the initial reaction period. We found that indeed Z-stilbene forms initially and isomerizes over time to the desired E-stilbene (Figure 3).

Details are in the caption following the image

Conversion of diphenylacetylene to E- and Z-alkene over time. The molecules were unequivocally detected using 300 MHz 1H NMR spectroscopy.

To figure out the factors affecting this isomerization, the isomerization of the commercially available Z-stilbene was examined under varying conditions (Table 3). Under the standard optimized conditions obtained in this work, quantitative conversion of Z-stilbene to E-stilbene (96 %) and 1,2-diphenyl ethane (4 %) was achieved in 2 hours. In the absence of pFA, the isomerization did not occur. Thus, pFA is either involved in the structure of the catalyst which is responsible for the isomerization or the reaction goes through the usual metal-hydride pathway via intramolecular hydrogen transfer of the Ru-coordinated olefin.62 The third reaction was run without [Ru(p-cymene)Cl2]2 and as expected, no isomerization occurred after 2 hours. In the absence of the BINAP ligand, isomerization was very slow and after 2 hours only 6 % of the E-stilbene was formed and more than 81 % of the starting material was detected. Thus, the BINAP ligand is vital for the transfer semi-hydrogenation of alkynes to alkenes and for the isomerization step. On this basis, one can say that BINAP is part of the structure of the catalyst species involved in the rate determining step of the isomerization where fast Z to E isomerization is enforced by the high bulkiness around the Ru center. The last reaction was performed in toluene as solvent without water. After 2 hours, only 40 % of the Z-stilbene was converted which means that the presence of water does not seem to be crucial for the isomerization, but the reaction proceeds markedly slower in its absence. It is noteworthy that in this case, no alkane was detected which means that water is one of the main sources of hydrogen production under these conditions and no hydrogenation can occur in its absence which is in agreement with our previous studies for hydrogen production from pFA.66

Table 3. Z to E isomerization of Z stilbene.

image

Isomerization test conditions

Z-stilbene [%]

E-stilbene [%]

DPE*[%]

1+BINAP+pFA+H2O

0

96

4

1+BINAP+H2O (without pFA)

100

0

0

BINAP+pFA+H2O (without 1)

100

0

0

1+pFA+H2O (without BINAP)

81

6

13

1+BINAP+pFA (without H2O)

60

40

0

  • Reaction conditions: diphenylacetylene (1 mmol), pFA (8 mmol, 240 mg), catalyst 1 (2 mol%), BINAP (4 mol%), H2O (2 mL), toluene (2 mL) at 80 °C for 2 hours. *DPE=diphenylethane

In a further experiment, the model reaction was repeated in a COware® reactor to see if the reaction goes through a hydrogenation or a transfer hydrogenation pathway. Complex 1, pFA, H2O and toluene were added to one chamber where hydrogen is supposed to be produced; while the second chamber contained diphenylacetylene, precatalyst 1, BINAP, H2O and toluene. In this setup, H2 produced in the first chamber can be transferred to the second one to hydrogenate the alkyne substrate. The results of this reaction show that after 3 hours, only 49 % of the starting material has been hydrogenated with a Z : E ratio of 1 : 1 which implied no selectivity under these conditions. Conducting this reaction for a longer time showed that the reaction in COware does not reach the full conversion even after 24 hours. The results showed that around 60 % conversion is obtained after 24 hours while the E: Z selectivity is 10 : 6, respectively, at this point. Therefore, it can be concluded that although semi-hydrogenation of alkyne is possible via hydrogenation using insitu produced H2 gas in a COware, the obtained stereoselectivity is pretty much lower than the one observed under the optimized conditions reported in this work and the reaction needs a lot more times to be completed Therefore, the presence of pFA and BINAP at the same time in the reaction mixture is crucial for the semi-hydrogenation as well as for the Z to E isomerization in agreement with the product distribution in Table 3. One possible argument in this context is that the coordination of BINAP, as a strong coordinating ligand, to the Ru center by replacing the p-cymene ligands affects the electron density of the complex such that the hydrogenation cycle can be facilitated. It can also increase the steric hindrance around the metal center resulting in the facile Z to E isomerization.

With the aim of revealing the nature of the catalytic species involved in the reaction mechanism and to extend the utility of this protocol in late-stage and natural product synthesis, extend ESI-MS analyses were conducted in the next step. The ESI-MS experiments were done using aliquots of the sample reaction mixture after 30 and 60 minutes. The presence of the hydride species A and B (Figure S68 and S69, SI) in both reaction mixtures according to the ESI-MS data shows that the reaction follows the same catalytic pathway as reported in our previous works for hydrogen generation from aqueous pFA solution.58, 61, 66 The existence of species C and D can also be confirmed from the signal at 1057.064 and 1067.094, respectively (Figure S70 and S71, SI). These species can be formed via ligand exchange between p-cymene ligands and BINAP respectively diphenylacetylene. It should be noted that no catalytic species could be found in the ESI-MS data which can be assigned to any monomeric species containing any combination of the catalyst and the active reaction ingredients. Therefore, we assumed that the formiato-bridge dimeric species are the active catalytic species in this reaction.

Based on the results obtained from our ESI-MS analyses as well as the mechanistic experiments and our previous studies in this context, a tentative pathway is proposed for the transfer semi hydrogenation of alkynes (Figure 4). The mechanism contains two different catalytic cycles. In the first catalytic cycle, reforming pFA in water results in the formation of the formiato-bridge complexes E and F which are responsible for the production of hydrogen, carbon dioxide and formic acid over the course of the reaction. In the presence of BINAP, alkyne, the insitu formed formic acid and water, some of the molecules of the catalyst precursor 1, respectively E, can also form G, the active catalytic species for the second catalytic cycle where alkyne is hydrogenated in the presence of the insitu formed hydrogen via the intermediate species H to produce Z-alkene. Fast isomerization of Z-alkene to its corresponding E-alkene in the next step delivers the desired product while regenerating the active catalytic species G.

Details are in the caption following the image

Proposed pathway for the transfer semi hydrogenation of alkynes.

3 Conclusion

Highly selective partial transfer-hydrogenation of both aliphatic and aromatic substituted alkynes was achieved using the commercially available [Ru(p-cymene)Cl2]2 (1) complex as pre-catalyst alongside with the ligand 2,2-bis(diphenylphosphino)-1,1’-binaphthyl (BINAP) (1 : 1 ratio per Ru atom and BINAP) in a 1 : 1 mixture of water and toluene. Paraformaldehyde (pFA), as a safe, stable, easy to handle, cheap and commercially available reagent has been used for the first time as the hydrogen source in this reaction. This protocol is complementary to the traditional Lindlar reduction resulting in the production of Z-alkenes with very high chemo and stereo-selectivity. Mechanistic investigations showed that E-selectivity in our reactions is a result of Z to E isomerization of the formed alkenes occurring only in the presence of the catalyst 1 and pFA while the presence of BINAP is not crucial but markedly speeds up the isomerization. Through our procedure, E-alkenes can be synthesized without the use of inert conditions, dry solvents and special setups with up to 100 % selectivity. This work also highlights the importance of ligand-metal interactions in catalytic procedures such that, the reaction cannot be done selectively in the absence of BINAP and any of the other tested ligands (N and P donors). The procedure can be also used for the production of deuterated alkenes using D2O instead of H2O. Due to the use of pFA as a neutral reagent as well as base-free conditions, the functional group tolerance of the presented method is high and many different substrates can be exposed to these conditions. All in all, we believe that this method holds great promises for future application in late-stage synthesis.

Experimental Section

Materials

Paraformaldehyde (pFA) was purchased from Alfa Aesar. d2-pFA and [RuCl2(p-cymene)]2 were purchased from Sigma Aldrich. All chemicals were used without further purification. Deionized water was used for all experiments. The hydrogenation reactions were carried out without precautions against moisture or oxygen unless otherwise stated.

Instrumentation

NMR spectra were recorded with Bruker Avance II 300 (1H NMR 300 MHz, 13C NMR 75 MHz) using TMS as reference. 2H NMR spectra were recorded with Bruker Avance III 500 using TMS as reference. Hexamethyldisilane was employed as the internal standard for calculating the NMR conversions and yields. Data were reported as chemical shifts multiplicity singlet (s), doublet (d), triplet (t), multiplet (m). High resolution ESI-MS(+) was performed on a Thermo Scientific LTQ Orbitrap XL. GC-MS measurements were performed on Agilent Hewlett Packard 6890 Series Plus chromatograph. A HP 5973 Series was used as mass detector and hydrogen employed as the carrier gas.

Analytical methods

Electrospray ionization mass spectrometry (ESI-MS). High resolution ESI-MS experiments were performed (resolution 30.000) using THERMO Scientific LTQ Orbitrap XL mass spectrometer with a FTMS analyzer. All measurements were done on positive ion mode. Therefore, some species could appear as protonated (m/z [M+H]+), diprotonated (m/z [M+2H]2+), sodium adduct (m/z [M+Na]+) or as a combination (m/z [M+H+Na]2+). Some of the ESI-MS results reveal that, during the reaction, free coordination sites could be occupied by different ligands present in the reaction mixture.

Gas chromatography- mass spectroscopy (GC-MS). GC-MS experiments were performed using an Agilent HP6890 system coupled with a mass detector (MSD) 5937N. Dihydrogen is used as carrier gas with a flow rate of 14 mL/min and a pressure of 0.3 bar. An Agilent 19091S-4335 HP-5 MS (30 m ×0.25 mm) was used as capillary tube.

Syntheses and reactions

The complexes 28 were synthesized and characterized according to our previous reports.61, 65

Syntheses of alkynes via Sonogashira reaction. The alkyne starting materials were synthesized based on a literature procedure70 with slight modifications as shown below (detailed information in the Supporting Information):

General Procedure 1 (GP1)

[Pd(PPh3)2Cl2] (72.0 mg, 0.10 mmol, 2 mol%) was added to an oven dried Schlenk tube under an Ar atmosphere followed by CuI (26 mg, 0.135 mmol, 3 mol %), 20 mL of dry and degassed THF, 5 mL of dry NEt3 and 4.4 mmol of the halogen containing compound. Then, 4 mmol of the corresponding terminal alkyne were added. The resulting reaction mixture was further degassed and stirred at room temperature for 48 h under Ar. The progress of the reaction was controlled using TLC. Upon completion of the reaction, 15 mL of distilled water was added to the mixture and the reaction was extracted with ethyl acetate (4×8 mL). The combined organic layers were dried over Na2SO4, filtered and evaporated under high vacuum. The crude material was purified using silica gel column chromatography (silica gel from ACROS 60 Å (0.035–0.070 mm)) with appropriate eluents (refer SI) to give the expected product.

General Procedure 2 (GP2)

[Pd(PPh3)2Cl2] (72.0 mg, 0.10 mmol, 2 mol%) was added to an oven-dried microwave vial under an Ar atmosphere followed by CuI (26 mg, 0.135 mmol, 3 mol %), 10 mL of dry and degassed THF, 3 mL of dry NEt3 and 4.4 mmol of the corresponding aryl/alkyl halide. The mixture was stirred for 10 min under Ar. To this mixture, 4 mmol of the terminal alkyne was added. The resulting reaction mixture was heated to 60 °C for 3 h in an Anton Paar Monowave 300 microwave. Then, 10 mL of distilled water was added and the reaction mixture was extracted with EtOAc. The combined organic layers were dried over Na2SO4, filtered and evaporated under reduced pressure. The crude material was purified using silica gel column chromatography to give the expected product.

General Procedure for the semi-hydrogenation of alkynes

A 20 mL crimp-top headspace-vial equipped with a magnetic stir bar was loaded with pFA (240 mg, 8.00 mmol), [Ru(p-cymene)Cl2]2 complex (12.2 mg, 0.02 mmol, 2 mol%), BINAP (24.9 mg, 0.04 mmol, 4 mol%) and 1 eq. (1 mmol) of an appropriate alkyne. A mixture of 2 mL water and 2 mL toluene was used as solvent. The reaction mixture was stirred at 80 °C in a heating block. The progress of reaction was controlled either by TLC or NMR of the crude. After the appropriate time, hexamethyldisilane as internal standard was added and the mixture was stirred for 5 min at room temperature prior to analysis. The mixture was diluted with ethyl acetate before addition of water. The aqueous layer was then extracted three times with EtOAc (3*10 mL). The organic layers were combined, dried over Na2SO4 and filtered. If needed, the crude product was purified over silica gel with cyclohexane, cyclohexane/dichloromethane or cyclohexane/ethyl acetate as eluents.

General procedure for determination of E: Z ratios

The reported E: Z ratios are based on the GC-MS analyses as well as NMR spectra. In each case, after completion of the reaction, hexamethyldisilane as internal standard was added to the crude and the mixture was stirred for 5–10 min. A small portion of the crude was then passed through a short silica column and washed with ethyl acetate. This sample was then analyzed with GC-MS after evaporation of solvent and dilution with absolute ethanol, methanol or ethyl acetate. Comparing the integration of the GC-MS signals and their retention times with the ones from the pure products, the E: Z ratios were obtained. At the same time, NMR spectra of the crude were recorded. As the chemical shift of the olefinic protons as well as their coupling constants in E and Z products are different, the integration of the signals of these protons was used also as the reference for determination of E: Z ratios. It should be noticed that the results of GC-MS and NMR spectra were close which confirms the accuracy of the reported values with two independent methods.

Acknowledgements

We gratefully acknowledge financial support provided by the MIWF-NRW (NRW-returnee program), the Heisenberg-Program (DFG), the COST Action “C−H Activation in Organic Synthesis (CHAOS)”, the Ernst-Haage-Prize of the Max-Planck-Institute for Chemical Energy Conversion. Moreover, we acknowledge the Alexander-von-Humboldt Foundation (G.T.). In addition, we acknowledge L. E. Heim, S. Vallazza and A. Thesing for previously performed syntheses of Ru complexes. Open access funding enabled and organized by Projekt DEAL.

    Conflict of interest

    The authors declare no conflict of interest.