Scale metal membranes

Palladium-based dense metallic membranes have been known to be completely selective for hydrogen permeation and are used in commercially available small-scale hydrogen purification units (e.g., Johnson Matthey, 2007 REB Research, 2007 Power + Energy, 2007 ATI Wah Chang, 2007). These hydrogen purification units typically use palladium-alloy... 

From 1943 to 1945, Graham s law of diffusion was exploited for the first time, to separate U235F6 from U238F6 as part of the Manhattan project. Finely microporous metal membranes were used. The separation plant, constructed in Knoxville, Tennessee, represented the first large-scale use of gas separation membranes and remained the world s largest membrane separation plant for the next 40 years. However, this application was unique and so secret that it had essentially no impact on the long-term development of gas separation. 

Recently, attempts have been made to reduce the cost of palladium metal membranes by preparing composite membranes. In these membranes a thin selective palladium layer is deposited onto a microporous ceramic, polymer or base metal layer [19-21], The palladium layer is applied by electrolysis coating, vacuum sputtering or chemical vapor deposition. This work is still at the bench scale. 

To conclude this section, it is necessary to state that Pd and Pd-based membranes are currently the membranes with the highest hydrogen permeability and selectivity. However, the cost, availability, their mechanical and thermal stabilities, poisoning, and carbon deposition problems have made the large-scale industrial application of these dense metal membranes difficult, even when prepared in a composite configuration [26,29,33-37],... 

Inorganic membrane development is still in progress [57] (see also Section 14.2.2). Microporous silica membranes have been developed at several universities and research institutes. Membrane selectivities of 15 and 20 for the separation of H2 from CO2 have been reported. Even higher selectivities for H2 arid CO, CH4 and N2 have been measured [20,57]. Most measurements reported in the literature have been performed on a laboratory scale. However, it has been shown that it is possible to upscale these microporous ceramic membranes to, at least, bench scale [31,57]. With other membranes such as noble (Pd) metal membranes and dense ceramic membranes very high and almost infinite selectivities for hydrogen are possible [58]. The permeation of these membranes is generally smaller than the permeation of microporous membranes. 

Fig. 9.7 Scale-up design for a planar, all metal membrane system from Union Carbide Patent 3,336,730,22 August 1967 [12], Precedence for a 10 miUion cubic feet (25 tons) per day hydrogen plant using metal membranes ...

Fig. 9.9 A tantalum metal heat exchanger shown as a concept for scale-up of closed-one-ended all-metal membrane tubes (NORAM) (Courtesy C. Brereton, J. Lockhart, and W. Wolfs, NORAM)...

This chapter will address the special considerations that apply to incorporating dense, hydrogen-permeable metal membranes into practical membrane modules for commercial and industrial use. It is organized to present a brief historical overview, a general review of hydrogen-permeable metal membranes, scale-up from laboratory test-and-evaluation membrane modules to commercial membrane modules, membrane module design and construction, and commercial applicability. 

As discussed elsewhere in this text, there are two types of dense, hydrogen-permeable metal membranes to consider from the perspective of module scale-up and design thin metal foils and permselective metal layers formed on a porous support. Another class of hydrogen-permeable inorganic membranes - dense proton-conducting ceramic membranes - are still under development and are addressed in Chapter 2. 

As with any engineering exercise, when the objective is to successfully scale-up and develop a product for commercial and industrial applications, the focus must be on achieving acceptable economics. Of course, economics involves the upfront capital expenditure, as well as any ongoing maintenance costs and fuel or other utiUty costs. One of the potential advantages of dense metal membranes for hydrogen purification is that the membranes are preferably operated at signifi-... 

For a well designed membrane module of reasonably large scale, the initial cost should be dominated by the cost of the palladium content of the thin metal membrane (assuming a palladium alloy comprises the permselective layer). The module itself will be made of steel, most likely a nickel alloy, with or without significant chromium addition. The cost of the steel and the assembly labor should not exceed the cost of the palladium alloy membrane. 

Scale matters. We have seen that scale may be used to facilitate reconstruction of structures with nano-components, but it has also shown that scale is important when simulation takes place. When calculated correctly, properly, or if you like, usefully, transport effective coefficients can be determined and even compared to experimental data. However, in some cases new approaches may need to be considered. Here, approaches like mesoscopic physics, or a model of multiple scattering with effective media approximation (EMA) for condensed matter, based on the approach of atomic cluster, may play important roles. Recently, a review (Debe, 2012) was discussed on the different approaches that scientists and fuel cell developers in general, are using in order to have better and cheaper catalysts. Many have made a great impact on CL structures. Some approaches included supporting material but others considered unsupported catalysts too. The aspect ratio of particles has been recognized as a relevant factor. Metallic membranes, meshes, and bulk materials have also been considered of which the structural features will impact on the final structure and functionality of fuel cell technology. Local structures and at different levels of scale are still subjects of interest in many scientific works (Soboleva et al, 2010). 

Cost reduction is a key issue to be addressed in promoting the widespread use of metal membranes for hydrogen separation. In fact, despite the proven applicability of Pd-based membranes in several hydrogen production processes, the high cost of the Pd limits the use of these membranes to niche applications, typically small scale applications where highly pure hydrogen is required. 

In order to maintain a definite contact area, soHd supports for the solvent membrane can be introduced (85). Those typically consist of hydrophobic polymeric films having pore sizes between 0.02 and 1 p.m. Figure 9c illustrates a hoUow fiber membrane where the feed solution flows around the fiber, the solvent—extractant phase is supported on the fiber wall, and the strip solution flows within the fiber. Supported membranes can also be used in conventional extraction where the supported phase is continuously fed and removed. This technique is known as dispersion-free solvent extraction (86,87). The level of research interest in membrane extraction is reflected by the fact that the 1990 International Solvent Extraction Conference (20) featured over 50 papers on this area, mainly as appHed to metals extraction. Pilot-scale studies of treatment of metal waste streams by Hquid membrane extraction have been reported (88). The developments in membrane technology have been reviewed (89). Despite the research interest and potential, membranes have yet to be appHed at an industrial production scale (90). 

---