Promotion Effects in Co-based Fischer-Tropsch Catalysis

BY FERNANDO MORALES AND BERT M. WECKHUYSEN Department of Inorganic Chemistry and Catalysis, Utrecht University, Debye Institute, Sorbonnelaan 16, Utrecht 3584 CA, The Netherlands

1 General Introduction

1.1 Fischer-Tropsch Synthesis. – Franz Fischer, head of the Max-Planck Institut fur Kohlenforschung in Miilheim (Germany) and Hans Tropsch, a co-worker of Fischer and professor of chemistry in Prague (Czech Republic), Miilheim (Germany) and Chicago (Illinois, USA), discovered in 1922 a catalytic reaction between CO and H2, which yields mixtures of higher alkanes and alkenes. This invention made it possible for Germany to produce fuels from its coal reserves and by 1938 9 Fischer–Tropsch (F–T) plants were in operation making use of, e.g, cobalt-based F–T catalysts. The expansion of these plants stopped around 1940, but existing plants continued to operate during World War II. It is worthwhile to notice that in 1944, Japan was operating 3 F–T plants based on coal reserves. Whilst being a major scientific as well as a technical success, the F–T process could not compete economically with the refining process of crude oil, becoming important starting from the 1950s. All this coincided with major discoveries of oil fields in the Middle East and consequently the price of crude oil dropped. Although a new F–T plant was built in Brownsville (Texas, USA) in 1950, the sharp increase in the price of methane caused the plant to shut down. Thus, due to bad economics F–T technology became of little importance for the industrial world after World War II and no new F–T plants were constructed. An exception was South-Africa, which started making fuels and chemicals from gasified coal based on the F–T process a half century ago due to embargoes initiated by the country's apartheid policies. Till today, South Africa's Sasol (South African Coal, Oil and Gas Corporation, Ltd.), building its first commercial F–T plant in 1955, is known as a major player in this field.

It is remarkable to notice that there is today a renewed interest in F–T technology mainly due to:

(i) The rising costs of crude oil. For some time now, the oil prices are well above $50 per barrel.

(ii) The drive to supply environmentally friendly automotive fuels, more in particularly, the production of synthetic sulphur-free diesel, especially interesting for the European car fleet.

(iii) The commercialisation of otherwise unmarketable natural gas at remote locations. CO2 emission regulations will certainly lead in the future to a ban on natural gas flaring near crude oil production wells.

This all has led to the recent decisions on major investments by big petrochemical companies, such as Shell and ExxonMobil, to built large scale F–T plants in Qatar. This will result in an important shift from crude oil to natural gas as feedstock for the production of fuels and chemicals in the decades to come. Industry projections estimate that by 2020 5% of the production of chemicals could be based on F–T technology with methane instead of crude oil refining operations. All this is especially promising in view of the long-term reserves of coal, which are estimated to be more than 20 times that of crude oil and coal is still used as the carbon source at the largest and economically successful F–T complex, namely the plants Sasol One to Three near Sasolburg in South Africa. A picture of a Sasol Fischer–Tropsch plant is shown in Figure 1.

The stoichiometry of the F–T process can be derived from the following two reactions, the polymerization reaction to produce hydrocarbon chains (1), and the water-gas shift reaction (2):

CO + 2 H2 -> -(-CH2-)- + H2O (1)

CO + H2O <-> H2 + CO2 (2)

The overall stoichiometry in case reaction (2) is completely driven to the right is:

2CO + H2 -> -(-CH2-)- + CO2 (3)

With ΔH227 = -204.8 kJ, while the maximum attainable yield is 208.5 g of alkenes CnH2n per Nm3 of a mixture of 2 CO and H2 for complete conversion., The CO/H2 is usually called synthesis gas, or in short syngas. The production of syngas, either by partial oxidation or steam reforming, can account for over 60% of the total cost of the F–T complex since the gasification process is highly endothermic and therefore a high-energy input is required. It should also be clear that the carbon source used, being it either coal or natural gas, is available at low cost, while the gasification of methane is much more efficient than that of coal since coal simply has a much lower hydrogen content. The syngas produced is then fed into a F–T reactor, which converts it into a paraffin wax that is subsequently hydrocracked to make a variety of chemicals, at present mostly diesel, but also some naphtha, lubricants and gases. A scheme of the F–T reaction process, including syngas production and hydrocracking of the wax, is given in Figure 2.

The F–T reaction involves the following main steps at the catalyst surface:

(i) The adsorption and maybe dissociation of CO;

(ii) The adsorption and dissociation of H2;

(iii) Surface reactions leading to alkyl chains, which may terminate by the addition or elimination of hydrogen, giving rise to either paraffin or olefin formation.

(iv) Desorption of the final hydrocarbon products, which can be considered as the primary products of the F–T process.

(vi) Secondary reactions taking place on the primary hydrocarbon products formed due to, e.g., olefin readsorption followed by hydrogenation or chain growth reinitiation.

Various detailed mechanisms have been proposed and this matter still remains a controversial issue in the literature.

Some of the scientific questions that arise are:

(i) Does the adsorbed CO molecule first dissociate into chemisorbed carbon and oxygen atoms? The chemisorbed carbon formed can then be hydrogenated to surface methyl and methylene groups in subsequent steps. Chain growth occurs by stepwise addition of Ci monomers to a surface alkyl group.

(ii) Is the adsorbed CO molecule hydrogenated to a CHO or HCOH species, which inserts in the growing hydrocarbon chain?

(iii) Is CO directly inserted in the growing chain and then subsequently hydrogenated?

It should be clear that a discussion on the F–T mechanism is beyond the scope of this paper and we refer the interested reader to several review papers on this topic. In this respect, it is noteworthy to mention the...