(143a) 1-Octene from Butadiene – New Technology Based on Heterogeneous Catalysts

1-octene has been used as one of the basic comonomers in the production of polyethylene. Its current availability is limited. At this time, there is no proven commercial process that allows making 1-octene only with high conversion and high selectivity. The main source of this comonomer is a product mixture from oligomerization of ethylene resulting from Linear Alpha Olefin (LAO) process. The LAO processes generate a broad range of olefininc hydrocarbons. The main drawback of this process is in the production of a mixture of products with a limited possibility of influencing its composition. Recently, Sasol claimed the development of selective ethylene tetramerization technology (1-3), and Dow Chemical started up installation converting butadiene into 1-octene. There is only scarce information about these developments.

There were several attempts made by academia and industrial companies to develop a process based on known chemistry - telomerization of butadiene with telomers such as: water, alcohols, or organic acids. The 1-octene synthesis from butadiene is a three-step process. First, and the most difficult step is the dimerization of butadiene connected with the addition of water (telomerization) to make 2,7-octadienol (reaction 1), or with the addition of methanol to make 1-methoxy-octadiene-2,7 (reaction 2).

1. 2 CH₂=CH-CH=CH₂ + H₂O → CH₂=CH-CH₂-CH₂-CH₂-CH=CH-CH₂-OH

2. 2 CH₂=CH-CH=CH₂ + CH₃OH → CH₂=CH-CH₂-CH₂-CH₂-CH=CH-CH₂-OCH₃

During this step, it is critical to have high selectivity catalyst to avoid the formation of higher oligomers of butadiene and isomeric diene derivatives. Catalyst, nature of electron donor, and reaction medium are critical parameters for this stage. Several patents have been granted to Dow Chemical (4), Kuraray (5), Oxeno (6,7), and Shell (8).

In the second step (reaction 3), the unsaturated ether or alcohol is hydrogenated to 1-substituted octane. This reaction uses classical hydrogenation catalysts (e.g. nickel, palladium, or platinum on alumina) working effectively under mild reaction conditions.

3. CH₂=CH-CH₂-CH₂-CH₂-CH=CH-CH₂-O-CH₃ → CH₃-(CH₂)₆-CH₂-O-CH₃

The last stage is cracking of 1-methoxy-octane (1-MOAN) to 1-octene and methanol (reaction 4). A very selective catalyst is required to avoid the migration of double bond along the carbon chain.

4. CH₃-(CH₂)₆-CH₂-O-CH₃ → CH₃-(CH₂)₅-CH=CH₂ + CH₃OH

According to the patent literature several catalysts based on gamma-alumina were claimed. Alumina modified with alkali metal ions, looked like an especially good catalytic system for cracking (5-7) ethers or alcohols.

Telomerization

Our approach to the development of a new heterogeneous process for the manufacturing of 1-octene from butadiene was focused on the use of solid catalysts to avoid costly procedures associated with the recovery of homogeneous catalysts, and handling large volumes of organic solvents. For the telomerization step we developed palladium containing catalysts active and selective in telomerization of butadiene with methanol. During the preparation and activation of the catalysts, it is essential to maintain palladium atoms in an ionic form. Therefore catalysts need to be activated in an oxidizing atmosphere at elevated temperature (450-850^oC). Reduction of palladium ions by hydrogen or by high temperature treatment caused total loss of catalytic activity in telomerization reaction.

These catalysts have been showing very good stability and selectivity toward 1-methoxy-octadiene-2,7 (1-MODE), exceeding 95% in the mixture with 3-methoxy-octadiene-1,7 (3-MODE) (11). These results were achieved in a fixed bed reactor system under continuous flow of reactants. Residence time of 1 hour was significantly shorter than that reported for batch reactor systems of 6-18 hours (9,10). At 85^oC under the pressure of 1.35 to 2.0 Mpa, we achieved between 60 to 85% conversion of butadiene, with the overall 1-MODE yields above 60%. The main by-products of the telomerization step were oligomers of butadiene (octatriene and vinyl-cyclohexene). Developed catalysts for telomerization demonstrated good stability, not changing activity over several hours of operations. In addition, these catalysts could be on-line regenerated without loosing activity and selectivity.

Hydrogenation

Hydrogenation of separated from telomerization step products 1-MODE occurs readily over either Pd or Ni catalysts. Seven commercially available catalysts were tested and compared under similar conditions. Five Pd catalysts compared under isothermal conditions gave a range of activity and selectivity towards the desired product, 1-methoxyoctane (1-MOAN). Two Ni catalysts were compared under adiabatic conditions.

In our study, Ni catalysts showed higher activity compared to Pd catalysts under similar conditions. This was clearly shown as a correlation between conversion and flow rate/residence time. Nickel catalysts maintained minimum requirements for activity and selectivity at much higher space velocities compared to Pd catalysts. Nickel containing catalysts demonstrated also longer lifetime.

Cracking of 1-methoxy-octane

For the third step, the cracking of the 1-MOAN to 1-octene is more problematic as the produced 1-octene tends to isomerize to internal octenes over solid oxide catalyst. Suppressing the isomerization of 1-octene to the minimum has been a focal point of our efforts. Several commercially available alumina catalysts were tested. The best results were achieved in the presence of a large pore alumina having an average pore diameter larger than 150 Angstroms (12). Using this catalyst, we achieved in the temperature range of 280-330^oC, conversions above 65%, and very high selectivities of 1-octene formation exceeding 96%. The large pore alumina maintained its activity and selectivity for a long period of time. Long-term runs showed only small decline in catalyst performance over a period of 800 hours. This high selectivity allows using the 1-octene fraction directly as a comonomer for ethylene polymerization. With the increase of cracking temperature there was a slow decline in the selectivity of 1-octene. The selectivity could be improved by selective blockage of acid sites by the addition of small amounts of ammonia (13).

The development of technology on lab scale was not fully successful in recovering methanol according to reaction (4) above. Cracking products contained di-methyl ether as the main component.

Bibliography

1. PCT Pat. Appl. WO 2004/056478, WO 2004/056479, and WO 2004/056480; July 8, 2004; Sasol Technology Ltd.

2. M.J. Overett, K. Blann, A. Bollmann, J.T. Dixon, F. Hess, E. Killian, H. Maumela, D.H. Morgan, A. Neveling, S. Otto; Chem. Commun., 2005, 622-624.

3. M.J. Overett, K. Blann, A. Bollmann, J.T. Dixon, D. Haasbroek, E. Killian, H. Maumela, D.S. McGuinness, D.H. Morgan; J.Am.Chem.Soc.; 127, 10723-30 (2005).

4. US Pat. 5,254,782; October 19, 1993; Dow Chemical Co.

5. US Pat. Appl. 20060111594; May 25, 2006; Kuraray Co., Ltd.

6. US Pat. Appl. 2004/0059170; March 25, 2004; Oxeno Olefinchemie Gmbh

7. PCT Pat. Appl. WO 03/031379; April 17, 2003; Oxeno Olefinchemie Gmbh

8. Can. Pat. 2,010,325; January 25, 2000; Shell Canada

9. Benvenutti F., Carlini C., Raspolli Galetti A.M., Sbrana G., Marchionna M., Patrini R.; J.Mol.Catal., 137 (1999), 49

10. Lee B.I., Lee K.H., Lee J.S.; J.Mol.Catal., 156 (2000), 283

11. US Pat. Appl. 20070112238, May 17, 2007; NOVA Chemicals S.A.

12. US Pat. Appl. 20070203381, August 30, 2007; NOVA Chemicals S.A.

13. US Pat. Appl. 20070100186, May 3, 2007; NOVA Chemicals S.A.