Shell's Omega MEG process kicks off in South Korea

12 August 2008 00:00 [Source: ICB]

The first full-scale plant using Shell's catalytic MEG process has just come on stream. One ICIS reporter investigates what the technology offers

THE "HOLY Grail" for a process engineer could be the development of a technology that converts all the raw materials to the desired end product with the minimum theoretical energy consumption, no emissions and the lowest capital cost.

While this may be a fantastic but vain hope, in the highly competitive petrochemical industry, particularly with today's high feedstock and energy costs, just a few percentage points of improvement in process performance can push a technology into a leadership position.

In the case of monoethylene glycol (MEG) technology, the holy grail could be the complete conversion of ethylene oxide (EO) to MEG without any production of the higher glycols such as diethylene glycol (DEG). However, this near-impossible goal did not stop researchers at Japan's Mitsubishi Chemical developing a catalytic process that converts practically all the EO into MEG.

Shell Global Solutions, the oil major's global consulting arm, has integrated Mitsubishi's MEG process with its own EO technology to offer the OMEGA process. The first plant incorporating the integrated process is now operating in South Korea, with a second unit in Saudi Arabia due to start up in the next few months.

Piet van den Berg, licensing manager, Shell EO/EG (ethylene glycol) processes, explains: "The global MEG market is growing at 6-7%/year, driven by polyester fiber demand in Asia and demand for PET [polyethylene terephthalate] packaging resin. However, there has been a mismatch of demand, with the DEG market growing at less than 6%/year and so the trend has been to maximize the yield of MEG."

Van den Berg adds that there are two areas where MEG yields can be maximized: improving EO selectivity and MEG selectivity.

EO is produced by the direct oxidation of ethylene, with high purity oxygen over a catalyst containing silver at temperatures of about 230-270°C (440-518°F). A side reaction competing with the main reaction forms carbon dioxide (CO2) and water. This reaction is suppressed using an ethyl chloride moderator. The CO2 is recovered and removed from the process.

In the early 1960s, when today's conventional technology was first being commercialized, EO selectivity was around 65%, with the main by-product being CO2. With the latest catalysts, van den Berg notes that EO selectivity is now approaching 90%.

In the EO to EG step, excess water is used to increase the selectivity to MEG. In the conventional process, the EO-water mixture is heated to around 200°C and the reaction takes place in the aqueous phase under pressure. MEG is produced along with DEG, triethylene glycol (TEG) and other glycols.

The proportion of the higher glycols can be controlled using excess water to minimize the reaction between the EO and glycols, and the water:EO ratio is critical in determining the volumes of higher glycols produced. In Shell's conventional process, a MEG selectivity of 90% is achieved with a H2O:EO ratio (wt/wt) of 9:1. The water-glycol mixture from the reactor is fed to multiple evaporators where the water is recovered and recycled. The water-free glycol mixture is separated by distillation into the MEG and the higher glycols. This operation consumes a lot of energy and requires purification, storage and handling equipment for the by-products.

In the past 10 years, the main EG licensors, which include Shell, have carried out research in using ion exchange resins to enhance the MEG selectivity. According to van den Berg, these resins can achieve a selectivity of 95% and above, but there is still a need for excess water of a H2O:EO ratio (wt/wt) of 6:1.

TWO-STEP PROCESS

However, Mitsubishi Chemical took a different approach, which involved the use of a catalyst and water:EO ratio approaching stoichiometric levels. This conversion is carried out in two reaction steps in order to achieve a high selectivity to MEG.

The first step in the process is the reaction of EO with dissolved CO2 to produce ethylene carbonate (EC). The CO2 is obtained from the CO2 produced and recovered in the EO plant. However, at start-up, the CO2 is supplied from a liquid storage tank. The second step is the reaction of EC with water, present in slight excess, to form MEG. In this reaction, the CO2 consumed in the first reaction is released and recycled back to the EC reactor, resulting in no overall CO2 consumption.

The EC reaction is exothermic (24 kcal/gmole) and the EC hydrolysis reaction slightly endothermic (-2 kcal/gmole). Both take place in the liquid phase using a set of homogeneous catalysts such as phosphonium halide.

The mixture from the EC hydrolysis reactors is fed to a glycol dehydrator to remove the water. The glycol stream containing the catalyst is then sent to the separation section at the bottom of the MEG purification column.

The catalyst solution is recycled back to the EC reactor. A small bleed is taken from the catalyst recycle stream to limit the build-up of heavy components and fresh catalyst can be added. The MEG is flashed off and sent to the top section of the purification column, where the finished MEG is recovered. A stream of heavy glycols, mainly DEG, is separated and sent to storage for further processing.

From 1995, Mitsubishi spent six years of process research and development taking the technology from bench scale to a pilot plant and then to a semi-commercial unit. It first tested the process in a 1,000 tonne/year pilot plant and then built a 15,000 tonne/year demonstration plant to scale up the technology. The MEG produced was tested in a third-party PET plant to check it was fully compatible with MEG from a conventional process.

Shell Global Solutions acquired the Mitsubishi technology in 2002 and became its exclusive licensor. The two firms then carried out a joint design exercise to integrate the Mitsubishi MEG technology with the Shell EO process and to solve scale-up problems with the reactors and separators. Risk, health and safety assessments were also performed.

Van den Berg claims that the selectivity of EO to MEG for the Shell OMEGA process is 99.3-99.5% and it can produce up to 1.95 tonnes of MEG from 1 tonne of ethylene. This conversion compares to 1.53-1.70 tonnes produced by Shell's conventional process, depending on the EO catalyst used.

The OMEGA process is also claimed to have 10% lower capital costs than a conventional process of equal MEG capacity. Much of the savings are due to the elimination of the need to treat the by-products and waste water. Steam consumption is 20% lower, while 30% less waste water is produced.

Since 2004, Shell has sold five licences. The first commercial plant, with a capacity of 400,000 tonnes/year, using the OMEGA process was started up in Daesan, South Korea, for Lotte Daesan Petrochemical in May 2008. The project was implemented in only 29 months, notes Arthur Rots, Shell's EO/EG design group leader. From a kickoff meeting in January 2006, the basic design package was produced by July 2006. South Korea's Samsung Engineering was responsible for the engineering, procurement and construction phase, with mechanical completion in March 2008. Start-up was achieved on May 21, and the guarantee test runs completed on June 1.

The second OMEGA plant, with a MEG capacity of 600,000 tonnes/year, to be operated by PetroRabigh, a joint venture of Saudi oil company Saudi Aramco and Japan's Sumitomo Chemical, in Saudi Arabia, is due to be completed in late 2008. Shell will employ the process in its own 750,000 tonne/year plant in Singapore due for start-up in early 2010.