Sep-2016

High efficiency zero export steam reforming

Development of a new hydrogen production technology based on steam methane reforming.

HOLGER SCHLICHTING, DIETER ULBER, SÉBASTIEN CADALEN, SÉBASTIEN DOUBLET and LAURENT PROST
Air Liquide

Article Summary

Air Liquide operates a large fleet of hydrogen production plants worldwide providing hydrogen, CO and syngas to customers in the refining and chemicals industry. The Global Engineering and Construction branch offers proprietary hydrogen production technologies to the market. Steam methane reforming (SMR) is the technology most widely applied to produce hydrogen from natural gas and light hydrocarbons. The technology typically produces high pressure steam as a byproduct. This article describes the drivers for the development of a new hydrogen production technology based on SMR. SMR-X technology is dedicated to locations where steam is of low economic value. The technology as well as the development effort are described, the main results are shared, and the new technology is explained.

Steam reforming is extensively applied in industry to convert natural gas and hydrocarbon streams into pure hydrogen. Figure 1 shows a typical configuration for a hydrogen plant. The feedstock is converted catalytically in the presence of steam in tubular reactors at high temperature. The energy for the endothermic reforming reactions is provided by heat transfer from the firebox in which fuel is burned. The hot reformed gas and the flue gas released from the firebox are used for pre-heating various streams such as the feed, the combustion air and the fuel, and also to produce high pressure steam. This steam is utilised for the reforming process itself while the surplus is exported to other, nearby users. The amount of export steam can be adapted by process optimisation to the user’s needs over a wide range.

The SMR process can be considered as a heat exchanger network. The design of the heat exchanger network for optimised energy recovery of the hot streams is pivotal for an efficient plant layout. The energetic efficiency of the overall SMR process can be defined by the following equation:

                                                          [1]

where the nominator represents the energy flows of the product hydrogen and the export steam respectively and the denominator is the sum of the energy of natural gas feed and fuel streams. It can be shown that the overall theoretical efficiency of the SMR process increases with increasing steam export.¹

A pinch study enables analysis of the energy efficiency of a heat exchanger network. The composite curve displays the amount of heat that is transferred at each temperature from the hot streams to the cold streams. Figure 2 shows a typical composite curve for the SMR process, which exhibits a pinch point typically below 200°C.

Figure 3 depicts the theoretical efficiency of a standard SMR configuration, assuming an ideal heat exchange with 0K pinch temperature and a commercial case with 25K pinch temperature, respectively. The amount of export steam per volume of hydrogen product has two practical limits. Reducing steam export below 5 kg/kgH2 results in a surplus of tail gas from the hydrogen purification unit, which is used as fuel gas in the firebox. Export of excess fuel gas is normally neither an efficient nor an economic option. Maximum steam export is reached when the flue gas is released with ambient temperature. Higher steam export can only be reached by additional firing, for instance by duct firing.

Improvement of physical efficiency often has no economic advantage. Energy integration and optimisation of heat exchanger networks reduce the steam demand on process plants, often leading to low economic value of export steam coming from the hydrogen plant. Therefore plant layouts become more attractive when they provide highest efficiency for hydrogen production while minimising export steam. This can even mean reducing export steam to zero or making the hydrogen plant a net importer of steam.

Various options exist to reduce the export steam of a standard SMR plant. For instance, a pre-reformer with reheat can be introduced, operating conditions can be fine tuned, or the high temperature reformed gas can be used to preheat the reformer feed and to produce steam. It is more efficient to utilise the high temperature heat of the reformed gas for endothermic natural gas reforming reactions. This leads to the concept of a heat exchange reformer, which is the basis of SMR-X technology.

In this technology, the hot reformed gas flows in an inner tube arrangement counter-current to the feed flow through the catalyst bed, thereby providing a portion of reaction heat (see Figure 4). Approximately 20% of the energy required for the SMR reactions can be provided by this internal heat exchange. The lower temperature of the reformed gas leaving the reactor leads to significantly lower steam production in the process gas boiler. In addition, less energy has to be transferred from the firebox to the reformer tubes, resulting in significantly lower flue gas flow and consequently lower steam production in the flue gas boiler. Zero export steam SMR plants can be easily designed with reduced steam production in both boilers.

The challenge of such technology is to specify the material grade of the inner heat exchanger tubes since they are operating in process conditions prone to metal dusting (MD) corrosion risk. MD is a complex and catastrophic corrosion phenomenon affecting alloys exposed to reducing and highly carburising gases in the temperature range 400-800°C. This results in alloy disintegration to a fine dust of metal and carbon particles.2 MD corrosion is thermodynamically possible below the carbon formation temperature; this can be expressed by the chemical equilibrium temperature of the Boudouard reaction (see Equation 2) and the carbon monoxide reduction reaction (see Equation 3), respectively:

                  [2]
    
                  [3]

Since both of these reactions are exothermic, MD corrosion potential increases during gas cooling while material degradation occurs in a narrow temperature range with a maximum rate between 600°C and 700°C, depending on gas composition, process parameters and the alloy (see Figure 5). The lower temperature limit is kinetically controlled mainly by the activation energies of the reactions. The higher temperature limit is controlled either by the thermodynamic potential or, as proposed in [3] and [4], by the formation of a protective chromium oxide layer, which is favoured at high temperature.

Therefore the design of a reliable heat exchange reformer requires the selection of an appropriate material for the inner tubes. In addition, an accurate model is required to predict the gas and metal temperatures in order to assess MD corrosion potential for the whole range of the plant’s operating conditions. A typical engineering approach of adding design margins bears the risk that the equipment has to be operated in the range of the highest level of MD corrosion attack, leading to potentially short equipment lifetime and premature failures.

To address the technical challenges involved, Air Liquide decided to execute long term demonstration tests in its multipurpose, commercial scale SMR pilot plant. The results validate SMR-X technology’s performance in the long term and provide reliable data sets used for model validation.