Alpha-olefins – a question of balance balance

31 May 2004 00:02  [Source: ICB]

With PE comonomer growth outpacing the growth in higher fractions a market imbalance is due. This is encouraging refinement of full range processes. Producers face the continuing challenge of serving a broad marketplace, as Mark Morgan explains

Linear alpha-olefin (LAO) producers with full range plants are facing the continuing challenge of balancing production to provide products for very diverse markets. The process technology in commercial operation mostly uses ethylene oligomerisation and thus produces a distribution of even-numbered carbon numbers ranging from C4 (butene-1) to as high as C30+ (triacontene) in some cases. Although on-purpose processes can make certain discrete fractions, so-called full range processes still dominate supply.

Major producers serve a broad range of chemical industry segments from polyethylene comonomers (C4-C8), through synthetic lubricants (C10) and detergent intermediates (C12-C14), to oilfield chemicals, paper-sizing agents (C16-C18), lubricant additives (C20+), to wax rheological modifiers (C24+). In addition there are myriad fine and performance chemical intermediates.

Shell Chemicals’ alpha-olefin process also produces linear internal olefins (LIO) many of which are used in the production of detergent intermediates.

Each market segment has very different behaviours in terms of market size and growth, geography, fragmentation and need for technical service. The challenge of serving all is formidable.

Looking forward, the industry is evolving, with PE comonomer growth outpacing the growth in higher fractions, leading to substantial market imbalance. This has encouraged the development of full range processes with shorter carbon number distributions and new on-purpose processes.

There are five full-range processes in commercial operation. In the ChevronPhillips Chemical full-range process (originally developed by Gulf, which subsequently merged with Chevron) alpha-olefins are synthesised from ethylene using Ziegler chemistry. Since ChevronPhillips uses a dilute catalyst system, there are two basic steps to the olefin synthesis process: chain growth and displacement.

The process uses the single-stage combined growth and displacement technique. A characteristic of the single-step process is its broad carbon number distribution. A typical weight distribution will follow a reduced geometric series, increasing in molecular weight from C4.

The ChevronPhillips full-range process has limited flexibility but variations in process conditions, such as reactor temperature, can pivot the distribution around decene-1. Increasing temperature serves to increase the displacement reaction favouring lower molecular alpha-olefins.

The Shell alpha-olefins process is really two processes – alpha-olefins and metathesis for the production of internal olefins. The alpha-olefin or oligomerisation stage, involving combined growth and displacement reactions, behaves like ChevronPhillips, although a promoted nickel catalyst is used. Selected components from low and high parts of the LAO distribution are then processed by isomersation/disproportionation (metathesis).

The impact of the combined LAO/LIO processes is a peculiar looking distribution providing a range of olefins in the detergent range to serve downstream Linevol and Neodol detergent alcohol businesses.

BP employs a unique, modified Ziegler process using both a catalytic step and a stoichiometric stage. By recycling alpha-olefins to the chain-growth section, BP can peak production of olefins in the carbon range desired, though this introduces a degree of branching into the alpha-olefin product. It is believed that BP operates with some flexibility as to the relative position of the peak at octene-1. The process overall provides a high yield of comonomers and decene-1 for captive poly-alpha-olefin (PAO) production.

Both Idemitsu and Mitsubishi operate full range technologies producing long tail distributions. Idemitsu manufactures LAOs using a zirconium/aluminium complex with a similar distribution to ChevronPhillips.

In general, new full-range technologies under development focus on shortening the LAO distribution to focus on comonomer products. There are three technologies offered for licence: Axens Alphaselect, UOP Linear-1 and Sabic/Linde Alpha-Sablin.

Sabic is likely to commercialise Alpha-Sablin in the medium term. The process has substantial flexibility in distribution profile based on its two-component catalyst system. Axens, meanwhile, has tried to shorten the distribution. However, conceptually there are substantial similarities with commercial processes.

Regarding on-purpose production, butene-1 has been extracted from petrochemical and refinery C4 streams for many years. Two world-scale projects are nearing completion or in start-up, namely Degussa (Oxeno) at Antwerp, Belgium, and Petrokemya at Al Jubail, Saudi Arabia. The process is relatively straight forward and is a useful means of adding value to C4s as part of a wider C4 integration strategy.

With the recent start-up of the Q-Chem I project hexene-1 as well as butene-1 can be made on purpose from ethylene. Q-Chem I uses the ChevronPhillips ethylene trimerisation process, but there are already several Axens Alphabutol ethylene dimerisation units in commercial operation.

At Secunda, South Africa, Sasol operates a world-scale coal gasification and synthetic fuels complex using its iron-based Fischer Tropsch process. The complex produces synthetic gasoline and FT waxes with a diverse product portfolio including solvents, creosols and high grade waxes.

The synthetic gasoline stream contains a rich resource of alpha-olefins. Sasol, the largest chemical producer in South Africa, began recovering pentene-1 and hexene-1 from its coal-to-synthetic fuels complex at Secunda in the early 1990s adding octene-1 in 1998. During 2003 Sasol began extracting C11/C12 fractions to serve captive detergent alcohols production. Another octene-1 line is under construction with plans for speciality alcohols in due course.

Whilst there is still considerable research activity into ethylene trimerisation, olefin metathesis has become the focus of attention at ABB Lummus. It has developed a process to convert butylenes contained in petrochemical C4 streams into hexene-1 with ethylene and propylene co-products. The process, designated OCT, involves a complex number of steps, especially if crude C4s are the starting point.

Crude C4s can be hydrogenated selectively to provide a butylenes-rich C4 stream. Isobutylene can be removed via two-stage MTBE or a de-isobuteniser column. The process scheme proposed by Nexant Chem Systems includes Butenex (from Krupp Uhde) and super-base catalysed olefin isomerisation. The effect of this is a considerable increase in capital investment, but also an increase in process flexibility to take C4s from various sources. A different configuration of process units could be used for a dedicated cracker and processes such as Butenex may not then be needed.

A process has been developed but never commercialised for the conversion of butadiene into octene-1 via a complex series of chemical reactions beginning with butadiene telomerisation with acetic acid. However, it could form part of a C4s to comonomer LAO complex together with butene-1 and now hexene-1 production via OCT. The conceptual crude C4s to comonomer complex provides 150 000 tonne/year of octene-1, 84 000 tonne/year of hexene-1 and 100 000 tonne/year of butene-1.

However, the economics look promising on an octene-1 basis and a total alpha-olefins basis with revenues suggesting a 10% return is possible assuming no margin is taken on butadiene extraction. Further optimisation could vastly improve economic performance. Such a complex would be suitable for regions with worldscale naphtha crackers, particularly western Europe and east Asia, possibly China. In reality this complex is likely to be a very site-specific LAO process solution.

Process economics

Full-range LAO plants make a broad range of products, usually with a peak in the comonomer range. In calculating the cost of total LAO production it is clear that the cost is dominated by ethylene. On this basis there is little to choose between full-range processes, especially if you compare them on a standard location such as the US Gulf Coast or mainland western Europe.

An alternative approach is to look at revenues on an annualised basis. Taking a closer look at long and peaked LAO distributions from both a carbon number and revenue distribution perspective, it is evident that in the long distribution case there is a double peak in the revenue distribution.

LAO pricing tends to be formula based up to C12 and from then on pricing is set on a value in use basis. C16 and C18 are high value fractions and in demand, hence the second revenue peak. In the peaked distribution, with its emphasis on comonomers, octene-1 revenues are very pronounced in the revenue distribution. Therein lies the balancing challenge, as producers would like to make more high revenue fractions but they cannot do this without making and trying to place the lower value fractions. The compromise is an operating rate that maximises total revenues over all fractions.

It is possible to calculate production costs (ethylene, utilities, manpower, etc) and superimpose the revenues from comonomer fractions and higher fractions. The impact of the technology is clearly seen in the revenue balance between comonomer fractions and higher fractions.

If the costs of producing hexene-1 via OCT and polyethylene comonomers (octene-1, hexene-1 and butene-1) are compared with full-range LAO and Sasol processes on a per tonne basis, it shows that new technologies like OCT have the potential to be very competitive in terms of LAO supply.

The alpha-olefins industry is at a crossroads, given the disparity between polyethylene comonomer growth and that of higher fractions. Full-range alpha-olefin producers face the continued challenge of serving a broad marketplace with a technology that has not really evolved to take into account turbulent changes in global markets.

From a technology standpoint newer processes such as Alpha-Sablin show promise in being able to focus on lower carbon numbers. However, margins do not currently justify re-investment except in unique locations where competitively priced ethylene is available. Clearly the Middle East is one such location where competitively priced ethane-derived ethylene exists.

There will continue to be investments in butene-1 extraction as part of developing C4 operations and in the right location on-purpose hexene-1 via OCT may also prove profitable. Looking forward, combining polyethylene comonomer production using a combined C4 approach may also be possible in the right location. In developing economies like China, comonomer demand is mainly butene-1 with only a little hexene-1. As major PE producers seek to invest in China there will be an increasing requirement for comonomers and the ability to integrate octene-1, hexene-1 and butene-1 into a C4 operation linked to a worldscale liquids cracker, may well prove viable.

Where to expect market development

By the end of 2003 the global consumption for linear alpha-olefins (LAO) (C4C20+) amounted to around 3.5m tonne. The industry has been hit in recent years by the global slowdown in chemicals consumption across virtually all segments of the alpha-olefins market.

Total consumption for butene-1, hexene-1 and octene-1 accounts for around 55% of total alpha-olefin consumption, while PE comonomers account for the remaining 45% (excluding polybutene-1).

As a whole alpha-olefin markets are still very much centred on the developed economies of the US, western Europe and to a lesser extent Japan. These markets consume the full range of alpha-olefins for higher performance PE, poly-alpha-olefin lubricants and advanced detergent formulations. That is not to say that developing economies do not show potential. The growth in PE production in the Middle East and east Asia is driving consumption for butene-1, mainly for commodity grades. There are developments in more specialised polyolefin products, but these are still relatively slow. In selected higher fractions like tetradecene-1 there is a growing east Asian market in alpha-olefin sulphonates, etc.

Overall, the consumption of alpha-olefins is growing at around 4.8%/year over the long term with 2003 as the base year. This appears high because 2001 and 2002 were difficult years with a severe slowdown in 2001 and only a very modest recovery in 2002 and 2003.

The divergence in consumption growth between PE and higher fractions is clear with implications for future supply patterns and technology development.

The average operating rate of the industry is currently in the 70-75% range, which is very low for a commodity chemicals business. Even with the closure of the Spolana facility in the Czech Republic operating rates will only improve slightly this year and next. In 2003, BP closed its plant at Pasadena, Texas, US, twice, once to decouple the closed alcohols unit and the other more extended period because of market conditions. BP has also recently announced its intentions to sell its alpha-olefins business.

New projects are planned by Sabic and ChevronPhillips/Qatar Petroleum (Q-Chem II) in the latter part of the decade. These investments too could impact future operating rates and in the meantime some further capacity rationalisation of full range plants is probably needed.

Additional diagrams and process flow charts are available from Mark Morgan, senior consultant and technology director EAMS, Nexant Chem Systems. E-mail mmorgan@nexant.com

COMMERCIAL FULL RANGE TECHNOLOGY

ChevronPhillips	Major complex at Cedar Bayou, Texas, US
	Licensed to Spolana Neratovice, Czech Republic (now closed)
	New Q-Chem II project announced for Qatar, post 2005
BP (formerly Amoco/	Oldest unit at Pasadena, Texas, US
Albemarle/Ethyl process)	European facility at Feluy
	Newest facility at Joffre, Canada
	Licensed to Nizhnekamsk, Russia
Shell	Integrated LAO/LIO and detergent alcohols at Geismar, Louisiana, US
	Integrated LAO/LIO and detergent alcohols at Stanlow/Carrington, UK
	Idemitsu at Ichihara, Japan
	Mitsubishi at Kurashiki, Japan
SOURCE: NEXANT CHEM SYSTEMS

< previous article(ICIS Podcast: Chemical News Central 2 November 2009)

For the latest chemical news, data and analysis that directly impacts your business sign up for a free trial to ICIS news - the breaking online news service for the global chemical industry.

Get the facts and analysis behind the headlines from our market leading weekly magazine: sign up to a free trial to ICIS Chemical Business.

Printer Friendly