Butadiene from acetylene-ethylene cross-metathesis

Article abstract of DOI:10.1039/c5cc00853k

Acetylene to butadiene direct synthesis, via enyne cross-metathesis, is demonstrated with commercial ruthenium carbene catalysts. Using excess of ethylene, yields greater than 50% are obtained. High activity is observed in the first minute of the reaction (TOF > 800 h^-1 based on butadiene). Catalyst reusability and poisoning are discussed. This journal is

Full text of DOI:10.1039/c5cc00853k

ChemComm
View Article Online
View Journal
COMMUNICATION
Butadiene from acetylene–ethylene
cross-metathesis†
Cite this:DOI: 10.1039/c5cc00853k
Ioan-Teodor Trotus-, Tobias Zimmermann, Nicolas Duyckaerts, Jan Geboers and
Ferdi Schuth*
Received 29th January 2015,
Accepted 18th March 2015
¨
DOI: 10.1039/c5cc00853k
www.rsc.org/chemcomm
Acetylene to butadiene direct synthesis, via enyne cross-metathesis, is were applied also on an industrial scale. However, these were multi-
demonstrated with commercial ruthenium carbene catalysts. Using step syntheses, for example, acetylene - butynediol - butandiol -
excess of ethylene, yields greater than 50% are obtained. High activity butadiene.4 A one-step synthesis route from acetylene to butadiene,
is observed in the first minute of the reaction (TOF 4 800 hÀ1 based on on the other hand, has not been reported yet. Coupling reactions of
butadiene). Catalyst reusability and poisoning are discussed.
acetylene to introduce a diene functionality into a larger product are,
however, known.5,6
In recent years oil prices have fluctuated very strongly. Alternative
Enyne metathesis is an emerging organic synthesis technique
routes to oil-derived commodities thus could become increasingly which unites an alkyne moiety with an alkene moiety to form a
important in order to ensure that society’s needs are still met conjugated diene.7–10 This reaction became widely used with the
throughout a transition period from an oil-based chemical industry advent of ruthenium carbene type olefin metathesis catalysts;11–17
to a situation with a higher feedstock diversity, including coal, gas, however, enyne metathesis is much less understood than olefin
and renewables. Acetylene, which was at the heart of the chemical metathesis. Both the intramolecular (ring closing enyne
industry until roughly 50 years ago, could once again serve as a metathesis or RCEYM)18 and the intermolecular versions
platform molecule for the synthesis of various bulk chemicals, (enyne cross metathesis or EYCM)19 of this reaction are known,
considering that its main synthesis routes start from coal or natural but to the best of our knowledge no example of EYCM using
gas, which are available in larger reserves than oil. This was already unsubstituted acetylene has yet been reported. As the cross-
discussed in comprehensive reviews on this topic.1,2 The broad metathesis of ethylene and acetylene should produce butadiene
potential of this molecule is obvious, since it was used as a in a one-step reaction, we have undertaken an investigation into
feedstock for various processes. After acetylene lost its central the possibilities of effecting this transformation. Furthermore,
position in favor of oil-derived feedstocks, such as propylene and performing EYCM with ethylene and acetylene, the simplest
ethylene, significant advances have been made in catalysis and the molecules which can participate in this reaction, could offer
understanding of catalytic processes. It thus seems worthwhile to additional mechanistic insight into EYCM.
investigate once more the possibilities that acetylene presents in
the light of these developments.
The metathetic reaction pathways possible in an ethylene–
acetylene atmosphere with a ruthenium carbene catalyst are
Butadiene is mainly produced by steam cracking, which has depicted in Fig. 1. Assuming a typical enyne metathesis mecha-
been shifted to lighter feedstocks in recent years, thus leading to nism to be in operation,7 the key intermediates in this process
higher quantities of ethylene, but smaller quantities of the heavier are the ruthenium methylidene 1 and the ruthenium vinylcarbene 2.
products including butadiene. With butadiene being an important As acetylene polymerization with ruthenium carbene catalysts is
building block for synthetic polymers, pathways leading to this known to occur20 and polymerization of alkynes has been reported
molecule which are not dependent on steam cracking seem to be even for RCEYM,21 it seems that an excess of ethylene would be
desirable.3 In the past, particularly in Germany during the Second required in order to stop the reaction at the butadiene formation
World War, synthesis routes to butadiene starting from acetylene stage and thus suppress the formation of larger oligomers. It is
also worth mentioning that carrying out a RCEYM in an ethylene
atmosphere (Mori conditions) has been reported to produce higher
¨
yields for certain substrates compared to an inert gas atmosphere,22
although this is not always the case.21 The ethylene atmosphere is
generally required to suppress polymerization if sterically unde-
manding alkynes are used as substrates.
Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1,
¨
D-45470 Mulheim, Germany. E-mail: schueth@mpi-muelheim.mpg.de
† Electronic supplementary information (ESI) available: Detailed experimental
procedures, catalyst screening, solvent screening, influence of catalyst loading,
influence of using a butadiene containing feed. See DOI: 10.1039/c5cc00853k
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 2 Results of catalytic tests performed with 10 mg HG2 in 20 mL of
CH2Cl2 at a reaction time of 15 minutes using 680–850 mL (STP) mixture
of ethylene and acetylene, the total pressure of 11–14 bar. Conversion
based on acetylene.
Fig. 1 Metathetical reaction pathways in an acetylene–ethylene atmosphere.
The reaction setup used for the catalytic tests is described in
the ESI,† Fig. 1S. Caution: acetylene should be handled with
great care, since it can explosively decompose and is highly introduced at 80 1C. GC-FID analysis, of both the liquid phase and
explosive in mixtures with oxidizing compounds. After performing the gas phase, revealed that propylene, along with butenes, C5
a screening of some commercially available catalysts (ESI,† Fig. 2S) dienes and C6 polyunsaturated compounds are also formed in
and solvents (ESI,† Fig. 3S), we chose to study acetylene–ethylene these reactions, but they generally do not add up to more than 5%
EYCM in dichloromethane with the Hoveyda–Grubbs second of the butadiene formed, which does not lead to a closed carbon
generation catalyst (HG2). The influence of using different balance for acetylene. This suggests that a significant part of the
catalyst loadings was also tested in preliminary experiments acetylene is converted to higher polyenes, which cannot be properly
(Fig. 4S, ESI†). While not fully conclusive, the results of these quantified, because they are formed in low amounts each and have
experiments suggest that butadiene production scales with the low volatilities. Full analysis of the liquid phase is complicated by
amount of catalyst present, while undesired side reactions are the formation of insoluble polyacetylenes, and was not performed
zero order in catalyst concentration. Therefore, in all following for all experiments.
runs the highest catalyst loading of 10 mg (0.016 mmol), corre-
sponding to a catalyst concentration of 800 mM, was used in full conversion after 15 minutes, a time dependent study of the
order to maximize butadiene yield. reaction was performed. Fig. 3 presents a conversion vs. time
As many of the experiments presented in Fig. 2 did not reach
First the influence of the ethylene : acetylene ratio (E/A) and plot for EYCM at 80 1C with E/A = 32.5, showing also the yield,
of reaction temperature on the conversion of acetylene and the selectivity and amount of butadiene produced. Interestingly,
selectivity towards butadiene were studied. The results depicted more than 80% of the butadiene produced after 30 minutes is
in Fig. 2 show that the conversion as well as selectivity toward actually formed in the first minute of the reaction. This
the yield of butadiene, at identical reaction time, generally suggests that the catalyst has a reasonably high initial activity.
increase with E/A. While the increase in conversion can be If the reactor was scaled up to accommodate 1 L of solution and
explained by the presence of less acetylene in the gas mixture if the conditions of the first minute were sustained over one hour,
fed into the reactor, the increase in selectivity with E/A proves the butadiene productivity would be around 35 g per hour and
that excess ethylene is essential for this reaction.
liter. While one might argue that a 32.5 to 1 ethylene–acetylene
It is worth noting that with careful choice of reaction mixture (around 3% acetylene) seems impractical for large scale
conditions yields for butadiene higher than 50% were obtained in application, it should be considered that ethylene produced
these experiments. Turnover numbers (TON) based on the amount from steam crackers can contain up to 2% acetylene, which
of butadiene produced are only between 3 and 24. Increasing needs to be removed because it acts as a poison for ethylene
temperature seems to have a positive effect on butadiene polymerization catalysts. This is typically done via selective
productivity, which could be due to a higher fraction of the hydrogenation over palladium based catalysts.26 It might be
catalyst overcoming the initiation barrier at higher temperatures, as profitable, if instead of hydrogenating the acetylene to ethylene,
recent reports suggest that for this type of catalyst incomplete one could convert it to butadiene. While this would require a
activation can occur.23–25 In addition, although care was taken to further separation step to recover the newly produced buta-
introduce the same amount of gas in all experiments, this becomes diene from the ethylene, depending on the prices of the
quite difficult at higher temperatures, due to the increase in vapor different products this could possibly be an attractive option
pressure of the solvent. This leads to the amount of gas mixture considering that a C4 separation unit is anyway present in the
introduced at 40 1C being around 1.25 times higher than that downstream processing of a steam cracker.26 However, in order
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
Fig. 4 Reusability of the catalyst at 40 1C with 10 mg HG2 in 20 mL of
CH2Cl2 using 800–850 mL (STP) E/A = 32.5 mix at a total pressure of 11 bar.
Fig. 3 Conversion time plot at 80 1C with 10 mg HG2 in 20 mL of CH2Cl2
using 680–730 mL (STP) E/A = 32.5 mix at a total pressure of 14 bar.
exposed to each of the three compounds alone for 2 minutes at
40 1C, in quantities comparable to those which are present
during a typical catalytic test, then the solvent was purged with
for this process to be useful, the catalyst needs to be either very nitrogen for 10 minutes to remove the dissolved gas, the solvent
cheap, or very stable in that it can easily be applied to a continuous volume was adjusted to 20 mL by adding fresh dichloromethane
process. In addition, basically full conversion of the acetylene is and the pretreated catalyst was tested in EYCM with ethylene
required because the unconverted acetylene would still act as a and acetylene. As blank, a nitrogen pretreatment followed by
poison in the downstream olefin polymerization reaction.
nitrogen flushing and the addition of fresh solvent was also
In order to assess the reusability of the catalyst an experiment performed to exclude the possibility of other influences. As can
was designed where the reaction was performed at 40 1C for be observed from Fig. 5, the nitrogen pretreatment provides the
2 minutes in 20 mL of dichloromethane with 0.016 mmol of HG2 same result as the catalytic reaction within the experimental
at 11 bar with a mixture of E/A = 32.5. After 2 minutes the gas error without any pretreatment.
phase was removed and the solvent was flushed with nitrogen
However, when the catalyst is pretreated with ethylene,
for 10 minutes to remove the dissolved ethylene, acetylene and acetylene or butadiene the amount of butadiene produced
butadiene which were determined by GC analysis. Then the drops to around 60% of that which is obtained in the absence
solvent volume was adjusted to 20 mL by adding fresh dichloro- of pretreatment. The differences in conversion and selectivity
methane and the reaction was performed once again for 2 minutes, are quite close to the experimental error (plus–minus eight
then the process was repeated for a second reuse. The results are percent points, see ESI†), however, the loss of butadiene
shown in Fig. 4. The amount of butadiene produced in the second productivity is significant, and it shows that in fact all three
run is around 26% of that produced in the first run, and the key compounds in the reaction have a negative effect on
amount of butadiene produced in the third run is around 32% of butadiene production. For butadiene, this is corroborated by
that produced in the second run.
the fact that the addition of butadiene to the feed during
The loss in butadiene productivity and acetylene conversion reaction results in lower butadiene productivity (Fig. 5S, ESI†).
could be due to the accumulation of polyunsaturated compounds of Another interesting aspect is that during pretreatment with
low volatility in the solvent which would lead to an increase in the acetylene or ethylene no butadiene was observed, thus proving
number of possible nonproductive reaction pathways.27 Additionally, that the butadiene obtained in the reaction indeed comes from
ruthenium methylidenes 1 are known to be quite unstable,28,29 and a reaction of acetylene and ethylene, not from an unknown
the ruthenacyclobutane intermediate in the nonproductive reaction side-reaction involving either of the two reactants alone. Also
of ethylene with the ruthenium methylidene was shown to decom- noteworthy is that during the ethylene pretreatment an amount
pose, producing propylene,30 which we also observed in trace of propylene equal to around 30% of the catalyst amount was
amounts (vide infra). Decomposition of the catalyst could thus be formed. Considering that after ethylene pretreatment the
an extra factor responsible for the observed decrease of activity when amount of butadiene formed is around 60% of the amount
reusing the catalyst solution.
formed in the absence of pretreatment, this fits together with
The detrimental effect of ethylene on catalyst stability has the findings of van Rensburg30 that propylene is formed upon
been previously reported for olefin metathesis.31–33 However, ethylene treatment, leading to a complex which is inactive for
direct comparison is not possible since the experimental setup metathesis. Propylene formation was not observed upon acetylene
as well as the conditions and the actual reaction being studied are or butadiene pretreatment.
significantly different. It was thus investigated to which extent
We have shown for the first time EYCM reaction using
ethylene, acetylene and butadiene contribute to the observed loss acetylene and the first directed one-step synthesis of butadiene
in productivity. For these experiments, the catalyst solution was first from acetylene. For this reaction it is essential to use a high E/A
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
4 J. W. Copenhaver and M. H. Bigelow, Acetylene and Carbon Monoxide
Chemistry, Reinhold Publishing Corporation, New York, 1949.
5 J. R. Kong and M. J. Krische, J. Am. Chem. Soc., 2006, 128,
16040–16041.
6 V. M. Williams, J. R. Kong, B. J. Ko, Y. Mantri, J. S. Brodbelt, M.-H.
Baik and M. J. Krische, J. Am. Chem. Soc., 2009, 131, 16054–16062.
7 S. T. Diver, Coord. Chem. Rev., 2007, 251, 671–701.
8 S. T. Diver and A. J. Giessert, Chem. Rev., 2004, 104, 1317–1382.
9 C. Fischmeister and C. Bruneau, Beilstein J. Org. Chem., 2011, 7,
156–166.
10 S. T. Diver and J. R. Griffiths, Olefin Metathesis, John Wiley & Sons,
Inc., 2014, pp. 153–185.
11 A. Fu¨rstner, L. Ackermann, B. Gabor, R. Goddard, C. W. Lehmann,
R. Mynott, F. Stelzer and O. R. Thiel, Chem. – Eur. J., 2001, 7,
3236–3253.
12 E. L. Dias, S. T. Nguyen and R. H. Grubbs, J. Am. Chem. Soc., 1997,
119, 3887–3897.
13 M. S. Sanford, J. A. Love and R. H. Grubbs, J. Am. Chem. Soc., 2001,
123, 6543–6554.
14 M. Scholl, T. M. Trnka, J. P. Morgan and R. H. Grubbs, Tetrahedron
Lett., 1999, 40, 2247–2250.
Fig. 5 Influence of catalyst pretreatment at 40 1C with 10 mg HG2 in 20 mL
15 M. Ulman and R. H. Grubbs, J. Org. Chem., 1999, 64, 7202–7207.
16 S. B. Garber, J. S. Kingsbury, B. L. Gray and A. H. Hoveyda, J. Am.
Chem. Soc., 2000, 122, 8168–8179.
of CH2Cl2 using 800–850 mL (STP) E/A = 32.5 mix at a total pressure of 11 bar.
17 J. S. Kingsbury, J. P. A. Harrity, P. J. Bonitatebus and A. H. Hoveyda,
J. Am. Chem. Soc., 1999, 121, 791–799.
18 T. J. Katz and T. M. Sivavec, J. Am. Chem. Soc., 1985, 107, 737–738.
19 R. Stragies, M. Schuster and S. Blechert, Angew. Chem., Int. Ed. Engl.,
1997, 36, 2518–2520.
20 D. E. Schuehler, J. E. Williams and M. B. Sponsler, Macromolecules,
2004, 37, 6255–6257.
21 A. G. D. Grotevendt, J. A. M. Lummiss, M. L. Mastronardi and
D. E. Fogg, J. Am. Chem. Soc., 2011, 133, 15918–15921.
22 M. Mori, N. Sakakibara and A. Kinoshita, J. Org. Chem., 1998, 63,
6082–6083.
ratio in order to suppress unwanted acetylene polymerization;
significant quantities of butadiene are only observed if E/A is
greater than 20. However, catalyst stability and/or selectivity
pose some problems for this synthesis, since pretreatments
with any of the three components of the reaction, i.e. acetylene,
ethylene and butadiene, have been shown to decrease butadiene
production.
This study only provides proof of concept that direct butadiene
formation is possible following this pathway. For a practical
application, however, a catalyst needs to be developed which has
higher stability towards ethylene, lower acetylene polymerization
activity, and a preference for reacting only with the smallest
molecules (ethylene and acetylene), in order to avoid loss of
selectivity due to further reactions of butadiene and loss of activity
due to nonproductive metathesis events.
23 T. Vorfalt, K. J. Wannowius, V. Thiel and H. Plenio, Chem. – Eur. J.,
2010, 16, 12312–12315.
˜
´
24 F. Nu´nez-Zarur, X. Solans-Monfort, R. Pleixats, L. Rodrıguez-
Santiago and M. Sodupe, Chem. – Eur. J., 2013, 19, 14553–14565.
25 D. J. Nelson, S. Manzini, C. A. Urbina-Blanco and S. P. Nolan, Chem.
Commun., 2014, 50, 10355–10375.
26 H. Zimmermann and R. Walzl, Ullmann’s Encyclopedia of Industrial
Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2000.
27 I. C. Stewart, B. K. Keitz, K. M. Kuhn, R. M. Thomas and
R. H. Grubbs, J. Am. Chem. Soc., 2010, 132, 8534–8535.
28 S. H. Hong, M. W. Day and R. H. Grubbs, J. Am. Chem. Soc., 2004,
126, 7414–7415.
29 S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day and
R. H. Grubbs, J. Am. Chem. Soc., 2007, 129, 7961–7968.
30 W. Janse van Rensburg, P. J. Steynberg, W. H. Meyer, M. M. Kirk and
G. S. Forman, J. Am. Chem. Soc., 2004, 126, 14332–14333.
31 Z. Lysenko, B. R. Maughon, T. Mokhtar-Zadeh and M. L. Tulchinsky,
J. Organomet. Chem., 2006, 691, 5197–5203.
´
T. Z. is grateful for a Kekule scholarship from the Fonds der
Chemischen Industrie. J. G. is grateful for a research fellowship
from the Alexander von Humboldt Foundation. Part of this
research was funded by EVONIK Industries which is gratefully
acknowledged.
Notes and references
¨
32 J. Scholz, S. Loekman, N. Szesni, W. Hieringer, A. Gorling,
1 I.-T. Trotus-, T. Zimmermann and F. Schu¨h, Chem. Rev., 2014, 114,
1761–1782.
2 H. Schobert, Chem. Rev., 2014, 114, 1743–1760.
3 R. H. Nielsen, ON-Purpose Butadiene Production, IHS Chemical, 2012.
M. Haumann and P. Wasserscheid, Adv. Synth. Catal., 2011, 353,
2701–2707.
33 J. M. Bates, J. A. M. Lummiss, G. A. Bailey and D. E. Fogg, ACS Catal.,
2014, 4, 2387–2394.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015