US 2623812 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
Filed April 30, -1948 E M D EBORALL ETAL PRODUCTION 0F HYDROGEN Dec. 30, 1952 UNITED STATES PATENT OFFIC PRODUCTION OF HYDROGEN Esther Myriam Doris Eborall, London, William Albert Baker, Harrow Weald, and Edwin Andrew Guthrie Lddiard, Farnham Royal, England; said Eborall and said Baker assignors to said Liddiard Application April 30, 1948, Serial No. 24,336 In Great Britain July 8, 1943 7 Claims.
This invention relates to the production of hydrogen gas, and the present application is a continuation-in-part of application No. 548,570 led August 8, 1944, now abandoned.
At the present time hydrogen gas is used in relatively large quantities in a large number of scattered sites, many of which are remote and relatively inaccessible, and it is usually necessary to transport the gas to the site in heavy expensive cylinders, with the result that heavy transport costs add to the cost of producing the hydrogen.
The object of the present invention is to obviate the above drawbacks and disadvantages by providing simple and effective means for generating the gas on the site where it is to be employed (.whether on land or at sea) although it will be understood that the process of our invention may be employed in any location where this procedure may seem desirable.
Our invention is based on the :observation that a surprisingly rapid evolution of hydrogen is obtained from a mixture of magnesium and iron, or iron oxide, when in contact with water, or sodium chloride solution, or sea-water. It is well known that hydrogen can be produced by the decomposition of water by contact with reactive metals and that the rate of decomposition can be increased by amalgamating or alloying with the reactive metal some other more noble metal. The mechanism of such as action is held to be that the surface of the less noble reactive metal when alone in contact with water tends to become polarised by the formation of a hydrogen lm. When, however, another more noble metal is present, whether as an impurity of alloying constituent, an electrolytic cell, sometimes of atomic dimensions, is believed to be produced, in which the more noble metal forms the cathode on which hydrogen may be discharged, while the less noble metal forms the anode, In general, the
nobility of a metal is judged by its electrode po-v tential. The electrode potential of the alkali metals is so high that they will reduce. aqueous solutions rapidly alone and without the necessity for the presence of another more noble metal. Magnesium (electrode potential +2.40 volts) and aluminium (electrode potential -f-l.7 volts) do not react with water at normal temperatures to any significant extent when they are in the pure state, but when they contain impurities, and particularly when in the presence of metals of low electrode potential, will react and evolve hydrogen. It is known that when aluminium is amalgamated with mercury (electrode potential 0.7986 v.) and brought into contact with water, reaction proceeds at an appreciable rate, and that the reaction is further accelerated by the presence in solution of a substance which will dissociate in water and render it more conducting, e. g. an alkali, acid, borate, phosphate or chloride. It is also well known that by `alloying magnesium with other more noble metals, e. g. nickel and copper, the rate of corrosion of magnesium with hydrogen evolution becomes more rapid. In all these cases, however, the reaction rates are insufficiently great to provide a useful and convenient method of generating hydrogen.
in seeking to find a convenient, rapid'and economical method of generating hydrogen from water, and particularly from sea-water, we made the surprising observation that mixtures of magnesium and iron evolved hydrogen at a more rapid rate than either magnesium or aluminium amalgamated with mercury. Furthermore, we made the surprising observation that magnesium in contact with heavily oxidised iron, or even in contact with pure ferric oxide, reacted at a more rapid rate than when in Contact with iron oxidised only by normal exposure to a dry atmosphere. It became clear in subsequent experim-ental work that the high efficiency of the magnesium/iron mixture was due to the presence on the iron of some oxide, since the reaction rate could be increased Aby oxidising the iron powder prior to mixing with the magnesium.
In previous teachings regarding the generation of hydrogen from magnesium or aluminium, it has been considered necessary to alloy or amalgamate the magnesium or aluminium with the more noble metal. Both aluminium and magnesium dissolve in mercury at room temperatures and the readily corrodible magnesium alloys which contain high proportions of nickel and copper and which, although not previously suggested as hydrogen generators, might reasonably be expected to be useful insuch an application, are produced by alloying nickel or copper with mag--l nesium in the molten state, in which there is mutual solubility. Iron has a very low solubility in molten magnesium at all temperatures within some hundreds of degrees of the melting point of magnesium, and iron would not therefore suggest itself as a suitable metal to be employed with magnesium for hydrogen generation, since it cannot readily be alloyed in substantial amounts with magnesium.
We found that a convenient method of preparing a substance to generate hydrogen rapidly from water, and particularly from salt water, was to take magnesium in the form of powder and mix it with iron powder in a finely divided form, carrying with it a superficial layer of oxide, which may, however, form by exposure to the atmosphere, or with iron oxide, or with a mixture of iron powder and iron oxide, and incorporate with the mixture about 5% of a water-soluble inorganic chloride, such as ammonium chloride, and then to compress the mixt-ure into blocks or pellets of suitable size. We also found it -convenlent to yadd a small quantity of graphite to the mixture to act as a lubricant, and a little kerosene to facilitate binding.
As evidence of the remarkably rapid rate of hydrogen generation from the magnesium iron, or magnesium iron oxide mixtures, as compared with other means of generating hydrogen from metallic couples, we quote the following experimental results:
corporate into the mixture about 5% of a soluble chloride, such as ammonium chloride, a small proportion of graphite to act :as a lubricant, and a little kerosene to facilitate binding. We then compress the mixture into `blocks or pellets of suitable size. The rate of reaction of the mixture when exposed to Water will vary with the proportions of magnesium and iron, or iron oxide, and, while the iron content may vary from 1% to 40%, it is generally unnecessary to exceed an iron content of 30%. Satisfactory results are obtained with about iron. The mixture when pressed into pelletsshould be stored in a'irand moisture-tight containers, since it will otherwise deteriorate on exposure to a moist atmosphere and generate hydrogen which may be a source of danger. For this reason, it is an advantage if the constituents of the mixture including the soluble chloride, are not markedly hygroscopic, since the presence of hygroscopic substances in Gos evolved in ccs/gm. ot mixture in- .\Iaterial Tested Medium 5 min. 10 min. 30 min.
l. Pellets containing 70% ol magnesium and iron powder Pellets containing 76% magnesium mixturel and 24% cast iron powder.
. Pellets containing 99% magnesium mixture1 and 1% ferrie oxide.
3. Pellets containing 95% magnesium mixture l and 5% ferrie oxide.
. Pellets containing 90% magnesium mixture l and 10% ferrie oxide.
. 98% aluminium powder, 1% mercurio oxide and 1% sodium hydroxide.
85% magnesium, 15% nickel alloyY 70% magnesium, 30% copper alloy...
70% magnesium, 30% nickel alloy.
70% magnesium, 30% mercury alloy 2.
Loss than 1l] o Less than 3 in 24 hours 1 The magnesium mixture referred to in tests 2 to 5 consisted of 92% magnesium powder, 6.6% ammonium chloride and 1.4% graphite.
2 To produce the magnesium mercury alloy, magnesium metal po The water used Was initially at room temperature (20 C.) but it will loe apparent that it will be heated by the heat generated by the reaction. External heat was not applied.
Comparison between test 1 and tests 6, 7, 8 and 9 show the greater reaction rate of magnesium/iron 4as compared with magnesium/nickel, magnesium/copper, aluminium/mercury, despite the fact that in all cases the difference in electrode potential between the metals concerned is greater than with magnesium/iron, while in the ease of magnesium/mercury alloy, the rate of generation was negligible. 2 with tests 3, 4 and 5 show that reaction rate increases with increase of ferric oxide and that the mixture incorporating only 1% ferrie oxide reacts more rapidly than one contain-ing 24% of cast iron powder.
`Other experimental work showed that the reaction rate obtained with iron powder was increased by oxidising the iron by heating in air, and that the reaction rate with hydrogen-cleaned sponge iron, which nevertheless would have revoxidised to some extent, was less than with ordinary cast iron powder.
In carrying our invention into effect in one convenient manner, we take magnesium or magnesium alloy, in the form of powder or finely divided swarf, or turnings, and mix this with iron oxide, or iron, in the form of lil-ings or granulated powder, or reduced or mechanically powdered iron of such la neness to pass the 200 mesh of the British Standards sieve. We may also in- Comparisons of test l wder was added to mercury in the proportions of These were ground together at room temperature place, and the mixture was subsequently heated until all the merthe mixture will increase the liability to deteriorate on storing in moist atmospheres.
We nd that when these pellets are exposed to the action of salt water, the rate of the reaction may in some cases be such that gas equivalent to approximately times the volume of the original pellet, or block, is evolved per minute. We have also found that the incorporation in the mixture of a small proportion, for example, 5% of a soluble chloride, as indicated above. increases the reaction rate Somewhat, and markedly increases the reactivity of the compressed mixture with relatively pure water. The total volume of gas which can be produced from this mixture approximates very closely to the theoretical volume, assuming complete reaction of the magnesium present in the mixture with the water, that is to say, a mixture containing 1 lb. of magnesium would produce approximately 400 litres or 14 cubic feet of hydrogen.
With the normal method of transporting hydrogen at the present time in cylinders, it is usually found that a cylinder capable of holding 150 cubic feet of hydrogen weighs approximately lb., and it will be seen that by means of our invention 120 lb. of 30% magnesium powder-20% iron powder mixture will produce approximately 1350 cubic feet of hydrogen so that' the advantage of our invention over the normal method as regards transport is approximately 9:1. The mixture can be transported to the site where it is desired to generate hydrogen, and by allowing the mixture to come in contact with water,
salt solution, or sea-water, in a simple generator, hydrogen can be Very quickly and conveniently produced.
A suitable generator may be constructed on the lines of an acetylene generator or, as shown in the drawing, we may employ a cylinder or bell a open at the bottom and having an outlet i: at the top controlled by a tap c, there being a grid d, grating or constriction at a convenient point in the length of the cylinder upon Which the pellets or blocks e of mixture may be placed. When the bell with the pellets therein is immersed in water f, salt solution, or sea-water, hydrogen gas is immediately generated and may be led to the point at which it is to be used.
When no further gas is required, the outlet is closed, whereupon the pressure within the tube rises and thus depresses the water below the level of the blocks or pellets, so that generation ci hydrogen ceases. So soon, however, as the outlet tap is again opened, the water rises within the tube and gas generation is resumed.
l. In the manufacture of hydrogen, the process which comprises mixing iinely-divided metallic magnesium with from 1 to 40 per cent of a nelydivided iron material selected from a group consisting of metallic iron superficially oxidized and iron oxide, and contacting the resulting mixture with liquid water, whereby hydrogen is evolved in quantity corresponding substantially to the theoretical amount which would be produced by reaction of the finely-divided magnesium with water.
2. The process of claim 1 wherein the water contains a water-soluble inorganic chloride dissolved therein.
3. The process of claim 1 wherein the mixture of nely-divided magnesium and iron material is pelleted prior to contacting it with water.
4. 'I'he process of claim 3 wherein the pellets contain a water-soluble inorganic chloride in addition to the magnesium and iron material.
5. A pellet for generating hydrogen upon admixture with an excess of water, which comprises a compressed mixture of finely-divided metallic magnesium and from 1 to 40 per cent of a finelydivided iron material, selected from a class consisting of metallic iron supercially oxidized and iron oxide, the pellet also containing a small amount of a water-soluble inorganic chloride to increase the reaction rate when the pellet is contacted with water.
6. The pellet of claim 5 wherein the inorganic chloride constitutes about 5 per cent by weight of the pellet.
7. In the manufacture of hydrogen, the process which comprises mixing nely-divided metallic magnesium with from 1 to 40 per cent of a nelydivided iron material selected from a group consisting of metallic iron supercially oxidized and iron oxide, and contacting the resulting mixture with an excess of water at substantially atmospheric temperatures, whereby hydrogen is evolved in quantity corresponding substantially to the theoretical amount which would be produced by reaction of the finely-divided magnesium with water.
ESTHER MYRIAM DORIS EBORALL. WILLIAM ALBERT BAKER. EDWIN ANDREW' GUTHRIE LIDDIARD.
REFERENCES CITED The following references are cf record in the le of this patent:
Theoretical and Inorganic Chemistry, vol. 1, pages 278-281; vol. 4, pages 251, 267; andvol. 13, pages 312, 702, Longmans, Green & Co., N. Y., publishers.
Becks The Technology of Magnesium and its Alloys, page 313; 1940 Ed., F. A. I-Iughes & Co., Ltd., London.
Liddells Handbook of Non-ferrous Metallurgy, 2nd Ed. 1945, page 55, McGraw-Hill Book Co., Ltd., N. Y.