Iron and Chromium Catalysts in the Thermal Decomposition of Sulfuric Acid in the SI Hydrogen Cycle

[NNadir]

From my personal perspective, hydrogen is useless as a commercial retail fuel, and talk of hydrogen cars, (or HYPErcars) is just garbage thinking, and is only credible for those people credulous enough to read books like Winning the Oil Endgame, which bristles with platitudes, hand waving, wishful thinking and other useless delusional consumer car CULTure tripe. (It would seem that the Gulf of Mexico didn't "win the oil endgame," since the only ending going on here is the end of some ecosystems, and possibly species.) That said hydrogen is very useful as an industrial captive fuel, and could be useful for phasing out the dangerous fossil fuel oil, the destructive potential of which is being obviated yet again, not that anyone seems to recognize the critical importance of phasing it out. The most useful tool for phasing out dangerous fossil oil is the miracle fuel DME, which is actively being commercialized in Asia. It is readily accessible whenever one has hydrogen via the hydrogenation of carbon dioxide.

The spontaneous thermal decomposition of water to give hydrogen and oxygen is only thermodynamically favorable at temperatures in excess of 4000C, still impracticable from a materials science standpoint.

Even were it possible to achieve these temperatures industrially, one would still face the spontaneous recombination of the resultant hydrogen and oxygen as the mixture cooled.

To avoid these limitations many thermochemical cycles that operate at much lower temperatures to produce hydrogen have been extensively investigated and refined. These include the UT-3 (CaBr2) cycle, Ferrite Cycles, Tin Oxide cycles, Manganese Cycles, Cerium based cycles, and Germanium cycles.

I have a cute cycle of this type of my own, but won't discuss it here.

Many of these cycles all have merits and demerits, the main demerit for many of them being the necessity for solids handing. Continuous systems are ideal from an economic standpoint, and continuous systems generally rely on fluid phases, either gas, liquid or supercritical fluid, which is neither gas nor liquid but has properties of both.

However the most famous cycle, by far, is the "SI" cycle, or sulfur iodine cycle which consists of the following series of reactions:

1) H₂SO₄ -> SO₃ + H₂O

2) SO₃ -> 1/2 O₂ + SO₂

3) I₂ (l) + SO₂ (g) + 2H₂O (l) → 2HI (l) + H₂SO₄.

4) 2HI -> H2 + I2.

The net reaction is the decomposition of water to give hydrogen and oxygen, this in such a way as the two gases are physically separated.

Reaction 3 is known as the Bunsen reaction. It is exothermic. All of the other reactions are endothermic and require the input of various amounts of heat.

Reactions 1 and 2 are often combined (and they are expected to be accomplished in a single reactor and thus this is justified) into a single reaction which is written like this:

(5) H₂SO₄ -> 1/2 O₂ + SO₂ + H₂O

Reaction 5, an exothermic reaction, involves fairly high temperatures, the highest in the SI cycle, typically around 800C - 900C.

Although this reaction temperature is easily obtainable in industrial systems, it would be ideal to reduce the temperature to a lower level to assure system life, particularly because sulfuric acid is a corrosive compound, although not so corrosive that it isn't widely distributed for use in the car CULTure.

A recent paper in the scientific literature discusses this approach through the use of catalysts to lower the temperature at which the reaction takes place.

Here is the abstract of the paper: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V3F-4PK8FXR-1&_user=10&_coverDate=01%2F31%2F2008&_rdoc=46&_fmt=high&_orig=browse&_srch=doc-info(%23toc%235729%232008%23999669998%23678911%23FLA%23display%23Volume)&_cdi=5729&_sort=d&_docanchor=&_ct=70&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=e50ae40825e8a2cfb8e241069df1b238">International Journal of Hydrogen Energy 33 (2008) 319 – 326

The title of the article, which is written by Indian chemists, is "Catalytic decomposition of sulfuric acid on mixed Cr/Fe oxide samples and its application in sulfur–iodine cycle for hydrogen production."

Here are some choice excerpts:

Given the prospect of a lack of fossil resources and limitations on the release of greenhouse gases, only water and biomass are the two candidate raw materials for hydrogen production. The two processes that have the greatest likelihood of successful massive hydrogen production using water as the raw material are electrolysis and thermochemical cycles. The thermochemical cycles are processes where water is decomposed into hydrogen and oxygen via chemical reactions using intermediate elements which are recycled. The sum of all the reactions is equivalent to the dissociation of the water molecule. Because they only use heat without having it to convert it to electricity, these cycles have the potential of a better efficiency than electrolysis and hence have the potential to significantly reduce the cost for hydrogen production from water.
So far numerous cycles <1–4> have been investigated (almost 200), among them sulfur–iodine (S–I) cycle, originally proposed by General Atomics <5>, is being looked at as a promising process for hydrogen production due to its higher efficiency(∼56%).

Further on we here a description of the limitation I described above:

Sulfuric acid can be decomposed with or without a catalyst, while the decomposition of SO₃ has been known to be a catalytic reaction. Apparently only a few and partial kinetic data for the global process can be found in the literature <6–8>. Some indications regarding the kinetics of sub-reaction (Eq. (4)) can be found in some theoretical and experimental studies of the reverse reaction <9–11>. For the sub-reaction (5) some incomplete and scanty data are available even in the presence of catalytic agents. In order to decrease the operating temperature, transition metals or their oxides seem to be interesting candidates as sulfuric anhydride decomposition catalysts <12>. The activity of metal oxides for the sulfuric acid decomposition was studied in the range of 600–950 ◦C in a fixed bed reactor and the order of activity was reported as follows: Pt ≈ Cr₂O₃ > Fe₂O₃ >CeO₂ >NiO>Al₂O₃ <13>. However, the ever rising demand and increasing cost make noble metals less preferred for the above applications. Recently, oxides samples of Fe₂O₃ and Pd–Ag alloy have been tested to evaluate their effect on dissociation efficiency by Barbarossa et al. <14>. Dokiya et al. <6> reported on the use of several metal oxides as catalysts for sulfuric acid decomposition but only at two temperatures, viz 1093 and 1133 K. Although simple oxide catalysts are active and can be a substitute for noble metal catalysts but their limitations of thermal instability, and sintering effects on repeated use, prevent them from the actual application...

...With an objective to develop certain Cr substituted Fe₂O₃-based compositions which may serve as structurally stable and catalytically active materials for the sulfuric acid decomposition reaction, present study was taken up in which the Fe_2(1-x)Cr_2xO₃ samples were synthesized and characterized by powder X-ray diffraction (XRD), surface area measurements and vibrational spectroscopy. The redox behavior and reproducibility toward repeated reduction and oxidation cycles have been monitored by recording multi temperature-programmed reduction/oxidation (TPR/TPO) cycles. The catalytic activity was evaluated for sulfuric acid decomposition in temperature range of 400–800 ◦C as a function of Cr-substitution. The SO₂- present in effluent gases was identified and monitored by IR and mass spectrometry. The phase stability of used samples was analyzed by XRD and IR.

Then there's a whole bunch of stuff about making catalysts, and a number of technical descriptions of physical measurements of its properties under various conditions, as well as putative chemical mechanisms for the operations of the catalysts.

The best catalyst was a non-stoichiometric catalysts whose formula can be expressed Fe_1.8Cr_0.2O₃. This catalyst is reported to have driven the reaction to approximately 90% completion at a temperature of 750C.

All Fe_2(1-x)Cr_2xO₃ amples facilitated the thermal decomposition of sulfuric acid in temperature range of 550–800^oC much lower as compared to uncatalysed reaction. The maximum conversion was obtained at 800 ◦C with the onset of formation of SO₂ at 600 ◦C. Solid solution of 10% Cr substitution in Fe₂O₃ lattice has not only enhanced reproducibility toward repeated reduction–oxidation cycles but also prevented the deactivation of the catalyst during use in sulfuric acid decomposition reaction. Fe_1.8Cr_0.2O₃ was found to be promising active and stable mixed oxide catalyst for the above reaction. The amount of SO₂ formed was found to increase with increase in reaction temperature as confirmed from IR and mass spectrometry. The presence of XRD and IR lines due to sulfate species on the used samples confirms their formation and decomposition as the rate determining step in the mechanism of SO₃ → SO₂ conversion on oxides.

Cool, or hot, or something.

[phantom power]

1. Is 800C --> 750C a big deal for an industrial chemist?

[NNadir]

2. Yes, for a variety of reasons.

The effects of temperature on chemical reactions is given by the Arrhenius equation, which is of the form k_r = A e^(-E_a/RT). (A is a constant particular to the reaction, although under some circumstances it isn't really constant, but is itself a function of T.) Here E_a is the activation energy which is a function of the path a particular reaction takes, R is the gas constant, and T is the absolute temperature and k_r is the rate of the reaction.

(An interesting side light about Arrhenius, the winner of the 1903 Nobel Prize in chemistry, is that he predicted climate change in the 19th century.)

Note that for the decomposition of sulfuric acid (or sulfur trioxide gas) E_a is different with a catalyst than it is in the absence of the catalyst. A catalyst effects a change in reaction mechanism, and thus activation energy.

That said, the mechanism of other reactions are unchanged, and one such reaction that one needs to consider is the corrosion reaction involved in the materials. These reactions may be slow - indeed if the system works they must be slow - but they are not zero. From the form of the Arrhenius equation we see that rate increases asymptotically to a maximum value, but the function is non-linear; it's an exponential term. This will shorten the time that the reaction vessels remain intact. It will also involve higher pressures. The hope is to design systems that will operate for decades, even considerable fractions of a century.

If one is processing these materials on a billion ton scale, which is clearly the required scale, the mere heat capacity of the process requires a greater energy expenditure. Even if the fuel is something reasonable - and that would limit the case to nuclear energy - the fuel is not free. As a practical matter, however, any SI (or other thermochemical process) will be something like a "combined cycle" process used in the dangerous natural gas industry. The step down from the high temperature reaction will involve cooling to generate steam and electricity. Even so...

Note that the reaction need not be driven to completion and effectively really can't be, because of the high affinity of sulfuric acid for water and water is a side product of the decomposition. At some point one needs to distill the water out of the acid recycling a small amount of waste. There may be a trade off between accepting a 75% yield as opposed to a 90% yield if one can go even lower, say to 600 C.

The SI process has been the subject of huge international attention. Over the last few years several hundreds of papers on the subject have been published, and I've read or skimmed a pretty fair fraction of them.

I don't think it's the best of all possible cycles, but it will be, I think, commercial, assuming that humanity lasts much longer as it chases the last drop of oil and gas in a stupid lemming like adventure.