Jan-2005

Increasing hydrogen production

The cost effectiveness of increasing refinery hydrogen capacity with compact convection reforming-based plants. Depending on available feedstocks, existing process configurations and operating conditions, medium-scale units are options

Ib Dybkjær, Haldor Topsøe

Article Summary

Environmental pressure has led to significant changes in the refinery industry. The overall composition of the refinery products has become lighter, since consumption of heavy fuel oil has decreased. Furthermore, the properties of transport fuels, in particular, are changing. Specifications for reformulated gasoline require less aromatics and olefins and impose constraints on light hydrocarbons and sulphur. New legislation for diesel requires deep desulphurisation to 10–50ppm sulphur. The requirement for removal of sulphur may be accompan-ied by a desire to remove aromatics.

In general, these trends result in an increase in the hydrogen to carbon (H:C) atomic ratio of the refinery products. At the same time, available oil resources on average have become heavier with higher content of sulphur and metals. This has created a large requirement for more hydrotreating (HDS, HDN, HDM) and hydrocracking.

Traditionally, the hydrogen consumption in refineries was covered to a large extent by hydrogen produced as a by-product from other refinery processes, mainly catalytic reforming, which is the conversion of paraffins into aromatics and hydrogen (not to be confused with catalytic steam reforming). As aromatics are not wanted in reformulated fuels, it means that less hydrogen is available from catalytic reforming.

In retrospect, there is a need for increased hydrogen production capacity in refineries. The increased capacity will, as mentioned previously, mainly be used in hydrotreating and hydrocracking. The increase is expected to be divided roughly equally between large plants serving hydrocrackers and similar projects, and medium-scale plants for hydrotreaters and other similar type units. The increasing demands can be satisfied by a number of actions such as:
- Optimise use of available hydrogen by careful analysis of the hydrogen network (hydrogen management)
- Revamp existing production facilities for increased capacity as required for peak shaving or to satisfy constant increased demand
- Install new production facilities as required
- Buy hydrogen “over-the-fence” from industrial gas suppliers.

[Madsen S W and Dybkjær I, Novel revamp solutions for increased hydrogen demands; ERTC London, November 2003]
If all requirements were to be covered by construction of new plants, construction of a total of five to ten new large-scale plants (50000–200000Nm3/h) and 20–40 new medium-scale plants (10000 –50000Nm3/h) could be required worldwide per year to supply additionally required hydrogen. This would include plants constructed by gas suppliers to support over-the-fence supply.

If the decision in a given situation is indeed to install new production capacity, several options are available. The choice will depend, among other factors, on required size of production and cost of available feedstocks. For medium to large-scale hydrogen demands, production on the basis of natural gas (or refinery off-gas) by steam reforming is the preferred solution. In areas affected by expensive hydrocarbon feedstocks, methanol may be a suitable alternative.

One possible scheme involves producing methanol in an area with very low priced natural gas and then transporting the methanol to the hydrogen plant location. A methanol based hydrogen plant is a simple unit and less costly than a natural gas and naphtha based plant.

Reforming technologies
The principal process for converting hydrocarbons into hydrogen is steam reforming, and involves the following reactions:
CH4 + H2O = CO + 3H2   
(-∆°298 = 206kJ/mol)          (1)
CO + H2O = CO2 + H2   
(-∆°298 = 41kJ/mol)          (2)
CnHm + nH2O = nCO + (m+2n ) H2   
                2
(-∆°298 =-1109kJ/mol for nC7H16)      (3)

Reaction (1) represents the steam reforming of methane. It is reversible and strongly endothermic, and maximum conversion is obtained at high temperature, high steam-to-methane ratio, and low pressure. The design of the steam reforming process is in part dictated by these constraints.
Feedstocks for steam reforming units may range from natural gas to kerosene, or even to diesel oil, and plant capacities are from less than 100Nm3/h hydrogen to more than 250000Nm3/h hydrogen equivalents in the form of methanol synthesis gas. For even higher capacities of up to 750000Nm3/h hydrogen equivalents (for fuel methanol production or Fischer-Tropsch (FT) synthesis), auto-thermal reforming is used alone or in combination with steam reforming.

Adiabatic pre-reforming is used for conversion of higher hydrocarbons in the feedstock. Furthermore, a pre-reformer removes any traces of sulphur present in the feed, leaking through the desulphurisation or from contaminated steam. It can be seen as a kind of pre-treatment, which preserves catalyst activity and allows for more stringent operating conditions in the downstream equipment, such as tubular reformers, convective reformers or autothermal reformers, depending on the process application.

Fired tubular reforming is the preferred option for large-scale hydrogen production. For example, the Topsøe tubular reformer comprises a radiant furnace based on the radiant wall concept, which ensures optimum use of high-alloy tube materials and long life of the reformer tubes, and a waste heat section for recovery of heat from the hot flue gas.