IE58026B1 - Method of preparing active magnesium-hydride-magnesium-hydrogen-storer systems - Google Patents

Description

COMPLETE SPECIFICATION "METHOD OF PREPARING ACTIVE MAGNESIUM-HYDRIDE-MAGNESIUM-HYDROGEN-STORER SYSTEMS" ·-;·>/ /CAT'.C-.S Mo_____/UJ f.CIf·'iCATiQM Fli-EO 3ι-S/% STUDIENGESELLSCHAFT KOHLE mbh, a German Company organised under the laws of the Federal Republic of Germany, of Kaiser-Wilhelm-Platz 1, 4330 Mulheim/Ruhr, Federal Republic of Germany. -1The present invention ,concerns a method of preparing magnesium-hydride — magnesium—systems (MgH^-Mg systems) that can be employed as reversible hydrogen-storage systems.

The MgH2~Mg system is the most appropriate of all known metal-hydride and metal systems that can be used as reversible hydrogen-storage systems because it has the highest percentage by weight (7.65 % by weight) of reversibly bound hydrogen and hence the highest energy density (2332 Wh/kg; Reilly & Sandrock, Spektrum der Wissenschaft, April 1980, 53) per unit of storer.

Although this property and the relatively low price of magnesium make the MgH^-Mg seem the optimum hydrogen storer system for transportation, for hydrogen-powered vehicles that is, its unsatisfactory kinetics have prevented it from being used up to the present time. It is known for instance that’pure magnesium can be hydrogenated only under drastic conditions, and then only very slowly and incompletely. .The dehydrogenation rate of the resulting hydride is also unacceptable for a hydrogen storer (Genossar & Rudman, Z. _f. Phy s. Chem. , Heue Foige 115, 215 [1979] , and the literature cited therein).

Intensive efforts have been devoted in recent years to · . improve the hydrogenability of magnesium by doping or alloying it with such individual foreign metals as aluminum (Douglass, Metall. Trans. 6η, 2179 [1975]) indium (Mintz, ? Gavra, & Hadari, J. Tnoro. Nucl·. Chcrr. 40 , 765 [1S73]), or iron (Welter £ Rudman, Scripta Meta 1lurgica 16 , 285 [1932]) , with various foreign metals (German Offenlegungsschriften 2 346 672 and 2 846 673) , or with intermetallic compounds like Mg^li or Mg2Cu (Wiswall, Too Appl. Phys. 29, 201 [1978] and Genossar & Rudman, op. cit.) and LaMi^ (Tanguy et al., Mater. Res. Bull. 11, 1441 [1976)). 1.0 Although these attempts did improve the kinetics, certain essential disadvantages have not yet been eliminated from the resulting systems. The preliminary hydrogenation of magnesium doped with a foreign metal or intermetallic compound still demands drastic reaction conditions, and the system kinetics will be satisfactory and the reversible hydrogen content high only after several cycles of hydrogenation and dehydrogenation. Considerable percentages of foreign metal or of expensive intermetallic compound are also necessary to improve kinetic properties. Furthermore, the storage capacity of such systems are generally far below what would theoretically be expected for MgH^· A considerable advance with respect to the kinetics of : MgH^-Mg systems is a method, specified in European Patent 0 003 564, for the homogeneously catalytic hydrogenation of magnesium in which the magnesium is converted with hydrogen in the presence of a catalyst consisting of a halide of a metal of Subgroups IV-VIII of the periodic table and of a magnesium-organic compound or magnesium hydride, in the presence if necessary of a polycyclic aromatic or tertiary amine, and in the presence if necessary cf a magnesium halide MgX^ in which X = Cl, Br, or I.

The main advantage of this method, aside from the mild reaction conditions accompanying the preliminary hydrogenation of the magnesium, is the superior kinetics of the resulting system with respect to the subsequent cycles of dehydrogenation and hydrogenation. The magnesium can accordingly be charged with hydrogen during the subsequent dehydrogenation and hydrogenation cycles either without pressure or under only slightly increased pressure and at lower temperatures than with known systems of this type (with the Mg-Mg^-Ni system for instance). The storage capacity of an MgH2~Mg system obtained by homogeneous catalysis is also in the vicinity of the theoretical level.

The indicated advantages can be realized with as little as 0.05 % of transition metal by weight of magnesium, and beyond 20% little extra advantage is gained; preferably it ranges form 0.5 to 3%.

Now, a new; process that effectively improves the kinetics of Mgi^-Mg hydrogen-storer systems has, surprisingly, been discovered.

The process consists of doping a finely divided form of the magnesium hydride or metallic magnesium by exposing it to a solution of an appropriate transition-metal complex or of an appropriate transition-metal organic compound. An extremely fine distribution of the particular transition metal precipitates over the surface of the particles of magnesium hydride or magnesium and assumes the function of catalyst in the dehydrogenation and hydrogenation cycles.

Additionally it has surprisingly been discovered that, for the purpose of producing active magnesium-nydridemagnesium-hydrogene-storer systems, finely divided magnesium can be doped by using metallic transition metal, whereby doping can be achieved by mechanically mixing the metal powders.

One particular advantage of the method by using soluble transition metal complexes or organo-transition metal compounds is that even slight amounts of the precipitated transition metal provoke a powerful catalytic effect either immediately or subsequent to only a few cycles of dehydrogenation and hydrogenation and that this effect is maintained as the cycles continue. When on the other hand nickel is electrolytically deposited on magnesium (Eisenberg, Zagnoli, & Sheriden, Journ. Less Common Metals 74 323 (1980)) for example, its catalytic effect will decrease precipitously after only a few cycles. Another advantage is that even though only 3% or less., by weight of the particular transition metal in terirvs of the magnesium hydride or magnesium is qnough to obtain the desired catalytic effect, the H^-storage capacity of the resulting system will be relatively high.

The process by using metallic transition metal in finely divided form in accordance with the invention consists of mechanically mixing magnesium metal in a finely divided form and an appropriate transition metal in a finely divided form, e.g. by milling both materials in the absence or presence of an organic solvent, whereby magnesium is doped by said transition metal.

The elements of Subgroups IV-VII of the periodic table — titanium, vanadium, chromium, molybdenum, tungsten, iron, ruthenium, cobalt, rhodium, iridium, nickel, palladium, and platinum — are all appropriate transition metals.

Complexes or metal-organic compounds of nickel, palladium and iron are preferred transition-metal complexes or transition-metal organic compounds for the method in accordance with the invention. Especially preferred are bis-(1,5-cyclooctadiene)-nickel (COD2Ni), bis-(pallyl)-nickel, bis-(acetylacetonato)-nickel, bis(ethylato)-nickel, phosphane-nickel complexes, and tetracarbonylnickel. When the transition metals palladium and iron are employed as dopes, they are preferably added in n^-allyl, Λ^-cyclopentadienyl, olefin, phosphane, acetylacetonato, or carbonyl complexes. Complexes or metal-organic compounds of platinum, cobalt, rhodium, iridium, iron, ruthenium, chromium, molybdenum, tungsten, titanium, and vanadium can however also be employed to dope the magnesium or magnesium hydride .

Preferred transition metals for doping metallic magnesium by metallic transition metal in finely divided form are Ni, Pd and Fe, whereby nickel is especially preferred, but also the other transition metals as mentioned above can be employed for doping magnesium metal.

Magnesium hydride obtained by homogeneous catalysis as specified in the above-mentioned European Patent 0 003 564 (in which the magnesium is converted with hydrogen in the presence of a catalyst consisting of a halide of a metal of Subgroups IV-VIII of the periodic table and of a magnesium-organic compound or magnesium hydride, in the presence if necessary of a polycyclic aromatic cr tertiary amine, and in the presence if necessary of a magnesium halide MgXn in which X = Cl, Br, or I) is especially appropriate for doping in accordance with the present invention.

Commercially available magnesium hydride with kinetic properties made appropriate for dehydrogenation and hydrogenation by doping it with a transition metal compound like bis-COD-nickel (Ex. 9) for instance in accordance with the invention can however also be employed .

The material can be doped by using transition metal complexes or organo transition metal compounds in accordance with the invention in an aliphatic, cycloaliphatic, or aromatic hydrocarbon or in an aliphatic, cycloaliphatic, or aromatic ether like, for example, tetrahydrofuran (THF) in which the particular transition-metal complex or transition-metal organic compound is soluble cr partly soluble. As Example 1 will demonstrate, magnesium hydride obtained by homogeneous catalysis can be doped in situ with COD?Ni in THF in the presence of the homogeneous hydrogenation catalyst or subsequent to separation from the catalyst and THF in another solvent like toluol for example.

The chemical processes basic to the doping method by using transition metal complexes or organo transition metal compounds may vary according to dope, solvent, reaction conditions, and sometimes the presence of hydrogen.

There are four types of doping reaction: a) Doping by thermal decomposition of the dissolved transition-metal complex, occurring for example when the magnesium hydride is doped with COD^Ni in toluene at 100 - 110°C (Ex. 2) : MgH2 + COD2Ni 100-110° MgH2(Ni) (1) -2COD b) Doping by hydrogenation of the transition-metal complex: Hydrogen is introduced (Ex. 9) into the suspension or solution of toluene to accelerate the doping of MgH2 with COD2Ni in toluene at 20 °C. The nickel complex rapidly hydrogenizes autocatalytically into cyclooctane and nickel that precipitates in an extremely fine form on the particles of MgH2: °C MgH2 + COD2Ni + 4H2 y MgH2(Ni) (2) -2CSH16’ If commercially available metallic-magnesium powder is doped with.a transition metal, the amount of reversibly bound hydrogen will depend also on the particle size of the powder.

Thus, there will be 3.74% by weight of reversibly bound H2 when 50-mesh magnesium is employed (Ex. 4) , whereas, when fine-grained (325 mesh, Ex. 10) magnesium is employed, a reversible hydrogen uptake of 6.5% by weight, relatively close to theoretical, will be attained . c) Doping by immediate reaction of the magnesium hydride with the transition-metal organic compound: When MqHn is doped with bis-(^^-ally) -nickel or bis3 2 (-y^-ally)-palladium in toluene or TIIF for example, the transition metal will precipitate on the surface of magnesium even below room temperature, propene will form, with 70% or more of the propene calculated from the equations MgH2 + (^-C3H5)2 Ni ----> Mg(Ni) + 2CoHr J 0 (3) and MgH2 + (t^-C3H5)2 Pd ----> Mg(Pd)+ 2C3H6 (4) being detectable, and the magnesium hydride will be dehydrogenized.

Accordingly metallic transition metal added as finely divided powder is precipitated on the surface of the magnesium particles. d) Doping, by reducing the transition-metal complex with magnesium hydride: When magnesium hydride is doped with nickel(II) complexes likq bis-(acetylacetonato)- or bis-(ethylato)nickel, it can also be exploited to reduce the divalent nickel, to zero-valent nickel. In this method, however, a proportion of the magnesium hydride equivalent to the nickel (II) complex is lost as a storer because of the formation of the corresponding magnesium salt.

The technical advantages of the improved kinetic properties of these MgH2-Mg hydrogen-storer systems derive from the hydrogen being charged more rapidly, under lower pressures, and at lower temperatures and discharged more rapidly and under lower pressures than was previously possible. This reduces material and operating costs and makes the hydrogen storers more practical and reliable to handle.

The accompanying drawings refer together with relevant examples hereto: Figure 1 (Example 1) shows hycrogenation-dehydrogenaticn doped volume plotted against time for a/sample a ancL/comparison sample, per Example 1; Figure 2 (Example 1) shows a sec of curves of hydrogen release volume at different pressures plotted against temperature, per Example 2; Figure 3 (Example 4) shows a comparison of hydrogen release volume.at different times and temperatures for the first 3 cycles comparing a doped sample and an undoped sample; Figure 4 is similar to Figure 3 (Example 4) for cycles 2C-3O; and Figure. 5 is similar to Figure 3 (Example 4) for cycles 55-53.

The appropriateness of the MgH^ and Mg treated by the methods specified in the examples for dehydrogenation and hydrogenation were tested by submitting samples t'o a) one or more cycles of dehydrogenation and hydrogenation under normal pressure (normal-pressure test) and b) 30-40 or more cycles of dehydrogenation and hydrogenation under elevated pressure (high-pressure test) if necessary.

The normal-pressure tests were carried out with a thermovolumetric apparatus designed for that purpose. (3ogdancvic & Spliethoff, Chem.-Ιπσ.-Tech., 55 (2) 155 % 1983) .

To investigate the prooerties of the MgH_, and Me as H_ storers over long-term use, the samples were subjected to a series of dehydrogenation and hydrogenation cycles under slightly elevated (15 bars max) pressure. Λ completely automatic and electronically controlled apparatus developed especially for this purpose was employed.

The air-sensitive MgH^-Mg systems or metal complexes were tested in argon. The solvents were cleared of air and moisture and distilled in argon, was employed in all the tests, extracted from a cylinder for hydrogenation cycles.

Technical (99.9¾) hydrogen Fresh hydrogen was always the dehydrogenation and Example 1_ .0 kg (617 moles) of magnesium powder (particle size 0.3 mm, 50 mesh) were hydrogenated into magnesium hydride in 75 2. of THF in a stirrer vessel by the method specified in European Patent 0 003 564 with a titanium catalyst (mole ratio of Mg:anthracene:TiCl4 = 100:1:1) at 60-73’ C under an H2 pressure of 2 bars.

The H2 atmosphere of the vessel was replaced with an argon atmosphere for doping with nickel. 1.4 kg (5 moles) of solid COD2Ni were added to the in situ suspension of MgH2 in THF under the argon. The suspension was heated to 100° C for 4,hours while stirred. The magnesium hydride, doped with nickel, was filtered out, washed with THF and pentane, and (dried in a vacuum (0.2-0.4 mbars). The product was subjected to one cycle of dehydrogenation and hydrogenation (dehydrogenation at 230-370° C under 10-0.4 mbars and hydrogenation at 335-350° C under 5-10 bars) to clear it of organic residue. 14.0 kg of magnesium hydride were obtained in the form of a light gray pyrophoric powder composed of C 0.0, H 6.01, Mg 35.17, Ti 1.51, Cl 2.72, and N i 0.3 9 .

Normal-pressure test A 0.72-g sample of the product was dehydrogenated in a (1-bar) H2 atmosphere at 334° C and then hydrogenated at 230° C (both temperatures furnace temperatures) in the thermovolumetric apparatus. Curves b in Figure 1 represent the dehydrogenation and hydrogenation cycle.

Curves a in Figure 1 represent in comparison a dehydrogenation cycle for a (0.69—g) sample prepared by the same (titanium catalyst) method but not doped (dehydrogenation at 329° C under 1 bar and hydrogenation at 230° C under 1 bar). Comparison with the undoped MgH£ sample shows that doping with nickel considerably improves not only the dehydrogenation kinetics but also the hydrogenation of the resulting magnesium. The dehydrogenation and hydrogenation cycles can be repeated several times under normal pressure with the doped Mg!I_ sample and will indicate only minimum losses of i^-storage capacity.

*.» High-pressure test A 19.09-g sample of the nickel-doped magnesium hydride was subjected to a series of 42 dehydrogenation and hydrogenation cycles at various hydrogenation pressures and times (t^). The external dehydrogenation temperature (T^) was always 367° C and the external hydrogenation temperature (T. ) alwavs 267° C. Table I lists the results of the high-pressure test. Figure 2, top, illustrates typical curves of hydrogen release and uptake for cycles carried out under 10, 5, 3, and 2 bars of H2 pressure along with the associated sample-temperature curves. For comparison, a 34-cycle high-pressure test of a (17.34—g) sample of magnesium hydride prepared with a titanium catalyst but not doped was carried out under the Ίθ same conditions. Table Ia lists the results and Figure 2, bottom, illustrates curves for typical cycles.

The results of these two tests reveal that samples of both the doped and the undoped MgH2 prepared by homogeneous catalysis make appropriate reversible hydrogen storers under low H2 pressures (2-10 bars) . Both Ii2-storage capacity 2Q and kinetics; remain practically constant subject to measurement error for 42 and 34 cycles under these conditions.

Nevertheless, there were distinct differences in kinetics in favor of the doped sample (Fig. 2, top and bottom). The hydrogenation curves for low pressure (2 and 3 bars) reveal the greatest differences, with the hydrogenation times of the doped samples being at least 3 times shorter.

The slighter differences in hydrogenation under higher pressures result from heat transport becoming more and more significant as the rate-determining factor as hydrogenation becomes more rapid. co cn co O ro cn ro o cn Table 1. - Results of high-pressure test of Mg.sample doped with 0.091 nickel (Ex. . T » d 367 • C T « 267® C, t ’ 1.5. a Cycle Dehydrogation Hydrogenation Cycle Dehydrogenation llydrooena tion Ho. H_ Pressure Time t. H_ No. H, Pressure Time t. »», — 2 --1 n 2 (hours) it) [barsl (hours) ID (A) (bars| li) 1 13.7 10 2 14.8 22 14.9 3 2 14.5 2 14.8 II II 23 15.0 II II II 3 >1 · II II 1« 24 II II 14.4 4 14.7 · II 14.6 25b> · 14.5 5b) 14.6 ’ 14.5 26 ·' »1 · II ό .. II «1 27 14.9 II II 14.4 7 ·' 2 3 14.4 28 15.0 II II 14.5 8 14.7 «1 1* 29 II ii II II 9 .. II 30 15.1 II II 14.6 iob> 14.6 ·» II 31 15.0 II 1« 14.7 11 .. II II 32 15.1 II II · 12 14.5 5 1.5 II 33 15.0 II II 13 14.7 ·· 14.5 34 ·· 5 1.5 14.9 14 14.E 1· li 35 14.9 II II II 15b> 14.7 II ·· 36 II ·· 15.0 16 14.6 II 14.4 37 15.0 II 14.9 17 14.7 3 2 14.5 38 M Ii II ’* 18 II II II II 39 14.9 il II 19 14.8 II II 40 15.1 II <1 15.0 20 .. II 41 14.9 ·' II 21 1 ·· 42 aj 0 bars b) „. Figure " atmospheric pressure. c) Reversible H level after 42 cycles! 6.601 by weight.' Yield: 19.09 g. 2, top illustrates these cycles. co cn co O ro cn ro O cn cn able la. - Results of high-pressure test of an undoped sample of MgH^ (Ex. 1). - 367* C, - 267* C, t^ " hours.

Cyc 1 e No. Dehvdrogation 2 HI Pressure (bars)a) Hydroqenation Time t^ (hours) 2 in Cycle Ho. Dehydrogenation 2 ID Pressure (bars|a3 Hydrogenation Tine th (hours)H2 (1) 1 14.0 10 1 14 . 3 IB 13.4 3.5 4.5 13.8 2 13.6 ·' 1.5 14.4 - 1910 , 2.5 15.0 3 13.5 . -2 lix ... 20 3 5 14.0 .b) ** 13.6 II II 14.5 21 13.3 II 13.8 5 It II II 22 13.2 II II 13.6 6 5 14.0 23 II II 13.7 7 1( tt 14.1 24 II M M eb> *1 II li 25 13.2 ·· II 13..i 9 13.7 II II 14.0 26b) 13.0 II II 10 13.6 ' 1« 13.9 27 II II M 13.8 11 It II li II 2Θ 12.9 ·· 13.6 12 2 6 13.6 29 It «1 13.7 1 ?’ 13. 3 ii 11 30 II 10 2 _c) 1-1 tt it II 13.5 31 13.3 II il - 15 13.4 3.5 4.5 14.1 32 II . it - 16 13.5 II ii II 33 M IIM - 17 13.4 ·· 14.0 34 ** II _d) pressure. 1 0 bars " atmospheric ti Figure 2, bottom illustrates these cycles.

, Hot measured.

Reversible H level after 34 cycles: 6.93% by weight. Yield: 16.13 g Example 2 218.7 g (0.9 moles) of magnesium powder (particle size 0.3 mm, 50 mesh) were hydrogenated into magnesium hydride in 1.1 i of THF by the method specified in European Patent 0 003 564 with a chromium catalyst (mole ratio of Mg:anthracenexCrCl^ = 100:1:1) at 60 C under ? pressure of 20 bars. , The product was filtered out, washed with THF and pentane, and drxed in a (0.2-bar) vacuum at room temperature to constant weight. The yield was 252.2 g of MgK2 composed of C 4.4, H 7.0, Mg 84.2, Cr 0.75 and Cl 2.04%.

Four samples (a-d) weighing 21-23 g of this magnesium hydride were each suspended in 300 mz of toluene. The toluene suspension was treated with various amounts of solid COD2Ni (Table ;II). The batches were then stirred for 15 % hours at 100° C. The doped MgH2 samples were filtered out, washed with toluene and pentane, and dried in a (0.2-mbar) vacuum. Λ sample of undoped MgH2 was thermally treated in toluene at 100° C in the same way but without the addition of COD2Ni.

GJ Ul Go O co ui co o Table 2. Doping MgH2 with COD^.'i in toluene at 100° Sample MgH2 (gl COD2tIi (9) Doped MgH lg) Ni %a) calc. a 22.4 15.0 20.2 12.5 b 21.7 3.3 21.6 3.1 c 21.2 0. 74 20.8 ... - ' 0/7 <1 23.0 0.08 22.3 0.07 Composition of doped samples (*] Ni C H 12.8 6.0 6.1 3.3 5.6 6.5 0.8 4.9 6.9 0.07 4.8 7.0 Mg Cl Cr 73.5 0.46 b) -83.5 0.42 b) 85.4 0.39 b) 87.2 0.31 b) a) b) At 100*. doping. Cot determined.

I co I Uotmal-oressurc test Samples a-d, doped with nickel, were subjected to the dehydrogenation and hydrogenation cycle described in Example 1. The results of the normal-pressure test show that considerable improvement in hydrogenation kinetics can be achieved even at a dose of 0.8% nickel (Sample c). Doping with more nickel (Sample a or b) leads to only inconsiderable further improvement in the hydrogenation kinetics .

Norma1-pressure rest at various temperatures 0.79 g of Sample b (with 3.3% nickel) and 0.65 g of Sample e, which had not been doped with nickel, were dehydrogenated under normal pressure at 330° C. The resulting samples of active magnesium were hydrogenated at the temperature range of 100-283° C and under 1 bar of H2 pressure. As the results of this test indicates, both samples can be;hydrogenated at an impressive rate at as low as 150° C. Hydrogenation rates increase with temperature, reaching a maximum at 250-255° C for both samples. As temperature continues to rise the hydrogenation rates naturally drop because the systems are approaching equilibrium (235° C, 1 bar). Comparison of both samples indicates that the doped sample exhibits considerably better hydrogenation kinetics throughout the tested temperature range.

Ignition tests were conducted with a nickel-doped sample of active magnesium under various I! pressures. Sample b (3.3% nickel) was dehydrogenated at 330° C and under bar. Portions of 11.0 g of the resulting active · magnesium were hydrogenated in a 300-mi autoclave at various H2 pressures. The autoclave was heated up at a constant rate of 2° C/min and the temperature inside each sample measured. At pressures of 10 and 15 bars the ignition temperature of the sample was about 150° C. During the next 30 minutes the sample heated up briefly to 380-390° C as the result of hydrogenation heat, and after another 10 minutes the hydrogenation process was practically complete. At lower H2 pressures (0.7 and 2 bars) the ignition temperature was naturally higher (190-200° C) . A maximum temperature of 270-290° C was attained during hydrogenation. Maximum attainable hydrogenation temperature can, because it can not exceed the equilibrium temperature correspond'inc to the H2 pressure, I accordingly be controlled ahead of rime by setting a given H2 pressure.

* Example 3_ 72.3 g (3.0 moles) of powdered magnesium (particle size, 0.3 mm, 50 mesh) were hydrogenated into magnesium hydride in 0.35 I of THF by the method specified in European Patent 0 003 564 with a titanium catalyst (mole ratio of Mg:anthrace.ne:TiCl4 = 200 :1 : 1) at 60° C under a pressure of 80 bars. The product was filtered out, washed with THF and pentane, and dried in a high vacuum at room temperature to constant weight. The yield was 252.2 g of a product composed of C 3.09, Η 7.40, Mg 84.2, Mg 86.36, Ti 0.59, and Cl 1.42% (Sample f).

An 11.35-g sample of the resulting magnesium hydride was suspended in 100 mi of toluene. The suspension was treated with, 0.90 g of COD^Ni and stirred for 5 hours at 100° C. The doped magnesium hydride was filtered out, washed with toluene and pentane, and dried in a high vacuum at room temperature. The yield was 11.18 g of a product composed of C 2.90, H 6.79, Mg 86.10, Ni 1.54, Ti 0.59, and Cl 1.01% (Sample g) .

Normal-pressure test The results of the normal-pressure test show that the doped sample (Sampled g) had considerably better H^-storer properties with' respect not only to kinetics but also to reversibility t;han the undoped sample (Sample f) . Subsequent to 2-3 cycles of dehydrogenation and hydrogenation the amount of H2 released at 330° C had been completely recaptured at 230° C.

Example £ 300.0 g (12.3 moles) 0.3 mm, 50 mesh) were j of toluene. The of powdered magnesium (particle size heated in a vacuum and suspended in suspension was treated with 27.7 g (0.1 mclc) cf COD^Ni. Hydrogen was introduced at room temperature and under 1 bar into the yellow toluene solution while stirred and with the magnesium powder suspended in it until it lost color (about 2 hours) . The doped magnesium powder was filtered out, washed with pentane, and dried in a vacuum. The yield was 304.2 g of powdered magnesium doped with 2% nickel.

High-pressure test A 15.0-g sample of the nickel-doped magnesium powder was subject to a series of 66 dehydrogenation and hydrogenation cycles varying in pressure, temperature (T^), and time (t^) . The autoclave dehydrogenation temperature (T^) was always 366° C and the dehydrogenation time (t^) hour. Table 3 shows the results of the test. -22Table 3 - Results of· high-pressure test of powdered magnesium sample doped with 2% nickel.

Td = 366° C, t, = 1 hour, a Cycle Hydrogenation Dehydrog. No.Th PressurebhH2H2 . l°C] [bars]a) [hours] mbl ' mb) 1 338 15 9 9.6 8.3 2-4 II n 2 5.4 4.5 5-6 264 ft 1 - It 7-41 338 It n 6.4 5.7 42-54 264 II II 6.1 5.8 55-60 It 5 2 6.5 6.6 61-62 II 15 1 5.6 5.7 63 1· II - 5.8 64-65 338 ft 6 8.1 7.6 66 ’·. II II 2 6.3C) * bars = atmospheric pressure b) 20° C under 1 bar .6 g of MgH2 3.74% by weight of H2 weighed out (subsequent to normal-pressure test).

Figure 3, top, illustrates the course of hydrogenation during the first three cycles and Figure 4, top, during dehydrogenation and hydrogenation cycles 20-23 at a hydrogenation temperature of 333° C and an pressure of 15 bars. The two figures also show the associated _ 23 the same sample-temperature curves. Figure 5, top, shows curves for the same sample during cycles 55-58 at a hydrogenation temperature of 264° C and a pressure of 5 bars. For comparison a high-pressure test lasting 31 cycles was carried out under similar conditions with a 15.1-g sample of undoped magnesium powder (manufactured by Ventron, 50,mesh). Table 3a and Figures 3, 4,. and 5, bottom, show the results.

Comparison of the curves in Figures 3, 4, and 5, top and bottom, indicates drastic differences in the kinetic behavior of the two samples in favor of the nickel-doped sample. These differences are extremely clear in both the initial (Fig. 3) and subsequent (Fig. 4) dehydrogenation and hydrogenation cycles. Particularly remarkable differences in hydrogenation kinetics in favor of the doped sample appear under lower pressures (5 bars, Fig. 5), with the doped sample., exhibiting especially superior kinetics in spite of the lower external temperature.

Table. 3a - Results of powdered-magnesium samp high-pressure le (Example 4). T. = 366° C, d test t, = 2 a of undoped hours.Th = 338° C, Cycle Hydrogenation Dehydrog. No . PressurefchH2H2 t [bars]a) [hours] [tibl . lt)b’ 1 15 9 - _0.2 2 11 7 6.2 5.8 3 II 2 4.8 4.9 4 n ll 5.5 5.3 5 It 1* 5.9 5.9 6 n fl 6.2 6.4 7 II ll 6.4 6.5 8 σ IT 6.6 6.7 9 5 10 8.0 8.1 10 w n 8.4 8.2 11-31 ; 15 2 7.4-3 .1 7.3-8.1 a) 0 bars = atmospheric pressure b’ 20° C under 1 bar Example 5. 1.79 g of nickel tetracarbonvl in 5 mt of toluene were added to 19.2 g of (undoped) magnesium hydride prepared as in Example 2 in 200 mt of toluene, which was then heated for 8 hours to 100° C (reflux coil: acetone and dry ice). The suspension was filtered and the doped magnesium hydride 2b washed-with toluene and pentane and dried in a vacuum (0.2 mbars). The yield was 20.0 g of a product composed of C 6.03, H 6.48, Mg 80.60, Ni 2.60, and Cl 1.11%.

The resulting MgH^ sample exhibited improved kinetic properties in the normal-pressure test in comparison with the undoped sample.

Examole 6 24.0 g (1.0 moles) of powdered magnesium (particle size: 0.3 mm, 50 mesh) were hydrogenated into magnesium hydride in 150 m 2, of THF by the method specified in European Patent 0 003 564 with a chromium catalyst (mole ratio of Mg:anthracene :CrCl^ = 100:1:1) at 20-24° C under a pressure of 80-100 bars. The product was filtered out, washed with TIIF1 and pentane, and dried in a high vacuum at room temperature to constant weight. The yield was 27.0 g of MgH2 composed of C 6.66, H 6.64, Mg 32.64, Cr 1.21 and Cl 2.95% (Sample h). 1.19 g of Sample h were placed in a '25-m£ flask connected to a cold trap and provided with a dropping funnel. The apparatus was evacuated. A solution cf 0.42 g (3.30 mmoles) of bis-(η 3-ally 1)-nickel in 4.0 m of toluene was added from, the funnel to the sample while it was being stirred at room temperature. The resulting gases were condensed in the cold trap (liquid N2) . After allowing the reaction to occur for 2.5 hours, the cold trap was defrosted, yielding 60 m?. of gas (20° C and 1 bar) consisting of 93.5% propone and 1.5% prepane (mole %, mass spectrography). After another 60 hours of reaction at room temperature the formation of another 43 -ral (20° C and 1 bar) consisting of 95.9% propene and 4.0% propane was demonstrated. On the whole, 70% of the allyl groups were determined to be propene, 11.4% propane, and 0.4% hexadiene-1,5 in the reaction of M9H2 with bis-(n3-allyl)-nickel. The solvent (toluene) was evaporated out in a vacuum and the residue dried in a high vacuum at room temperature. The product was composed of .

C 9.09, H 6.11, Mg 69.94, Ni 12.90, Cr 0.53, and Cl 1.20%.

The resulting nickel-doped MgH2 sample exhibited improved kinetic properties in comparison with the undoped sample (Sample h) in the normal-pressure test.

Example 7. 1.33 g of the MgH2 Sample h (Ex. 6) were doped as in Example 6 with 0.39 g (2.1 mmoles) of bis-( n3-allyl)-palladium at room temperature. 72.6% of the allyl groups in this reaction were determined to be propone, 1.2% propane, and 2.4% hexadiene-1,5. The pa 1ladium-dcped magnesium hydride was composed of C 6.34, Η 5.90, tlg 72.88 , ?d 13.07, Cr 0.47, and Cl 1. 13¾.

The palladium-doped MgH^ sample exhibited improved kinetic properties in comparison with the undoped sample (Sample h) in the normal-pressure test.

Example 8. 2.59 g of MgH2 Sample h (Ex. 6) were suspended in 20 ml of toluene and treated at -78° C with a -78° C solution of 0.22 g (1.25 mmoles) of tris- (n '-allyl) - iron in 20 ml of toluene. The temperature of the mixture was allowed while stirred to rise frcm -78° C to -6° C within 8 hours. It was then stirred at room temperature fcr 24 hours. The iron-doped MgH2 was filtered out, washed with toluene and pentane, and dried in a high vacuum. The iron-doped magnesium hydride was composed of C 9.45, H, 7.13, Mg 78.19, Fe 1.93, Cr 0.45, and Cl 2.561.

The iron-doped MgH9 sample exhibited improved kinetic properties in comparison with the undcped sample (Sample h) in the normal-pressure test. _ 28 _ Examole 9 1.86 g (6.3 mmoles) of CCD2Ni were added to 19.45 g of magnesium hydride (mf. by Alpha Products) in 100 mi of toluene. Hydrogen (1 bar) was introduced into the solution or suspension while the latter was stirred at room temperature. After 630 mj, of H2 (20° C and 1 bar) , (649 ml calculated for the hydrogenation of the CC5D2Ni) were taken up within 1.6 hours, hydrogen takeup was significantly slower. During the next 2.5 hours 230 mi of H2 (20° C and 1 bar) were taken up (hydrogenation of toluene), after z which hydrogenation was interrupted. The suspension was filtered out and the doped magnesium hydride washed with toluene and pentane and dried in a high vacuum. The yield was 19.90 g of a product composed of C 0.39, H 4.95, Me 84.02, and Ni 1.89%.

The doped MgIK. sample exhibited considerably improved kinetic properties in comparison with the undoped sample (from Alpha Products) in the high-pressure and normal-pressure tests.

Example 10 50.0 g (2.06 moles) of powdered magnesium (Alpha Products, 325 mesh) were suspended in 150 m of toluene. The suspension was treated with 4.77 g (17.4 mmoles) of CGD2Ni. Hydrogen (1 bar) was introduced at room temperature from an automatically recording gas burette (Chem. Ine. Techn. ,, loc. cit.) into the yellow toluene solution while it was being stirred and with the magnesium powder suspended in it until there was a sharp kink in the hydrogen-uptake curve subsequent to the uptake of 1.70 ι of H2 (20° C and 1 bar, 102% of th.). Hydrogenation took 1 hour. The doped magnesium powder was filtered out, washed with pentane., and dried in a vacuum. The yield was 50.5 g of nickel-doped magnesium powder.

High-pressure test A sample of 14.7 g of the resulting magnesium powder was subjected to a series of 35 hydrogenation and dehydrogenation cycles at different hydrogenation pressures and times (t^) . Dehydrogenation autoclave temperature (T^) was 362° C and hydrogenation autoclave temperature (T^) 264° C. The dehydrogenation time (t^) was a constant 1 hour. Except for the first hydrogenation, which was carried out at 337° C and lasted about 4 hours, % effective hydrogenation times (the time until the active 'j magnesium practically stepped taking up H9) were 1.2 hours at 1 bar, 0.3 hours at 2 bars, 0.3 hours at 3 bars, 0.6 hours at 5 bars, and 0.4 hours at 15 bars (excess H2 pressure), with the temperature inside . the sample always increasing to the equilibrium temperature corresponding to the given H^ pressure. The effective dehydrogenation times were 0.6 hours, with the temperature inside the sample falling below 300° C. The reversible hydrogen content varied in accordance with hydrogenation pressure and time during the cycles of hydrogenation and dehydrogenation between ^1.3 i, corresponding to 6.0¾ by weight, and 12.6 corresponding to 6.6% by weight of M9H2 ^20° C arid 1 bar) . A high-pressure test lasting 22 cycles was also conducted under the same conditions with a sample of 15.0 g of undoped powdered magnesium;(Alfa Products, 325 mesh) .

Comparison of the two tests demonstrated drastic differences in the kinetic behavior of the two samples in favor cf the nickel-doped sample. The first hydrogenation of the doped sample was at least 3 times faster and subsequent ✓ hydrogenations, at 5 and 3 bars, at least 10 and 15 times faster respectively than the hydrogenations of the undoped sample. The reversible hydrogen content of the undoped sample ranged from 6.3 to 6.8% by weight of MgH2 in accordance with hydrogenation pressure and time.

Example 11 29.4 g magnesium powder (maximum particle size 50 um, 270 mesh, origin: Hoechst) and 0.6 g nickel powder (Inco Co.) were milled in 200 ml toluene during 16 hours under argon in a glassball mill. The metal powder was filtered, washed with pentane and dried in vacuum (0.2 mbar). 27.3 g of a non-pyrophorous magnesium powder doped with about 2% by weight nickel were isolated in this manner.

For the application as hydrogen storage material 16.2 g of this powder were hydrogenated during 21 hours at 266°C and 5 bar hydrogene pressure and thereafter dehydrogenated (8.2 1»2; 20°C' 1 bar) at 344°c- The hydrogenation and dehydrogenation steps were repeated in which case only 3 hours were required for the hydrogenation.

During the following 32 hydrogenation-dehydrogenation cycles (4 cycles at 15 bar, 10 cycles at 3 bar, 7 cycles at 1 bar, 6 cycles at 5 bar, and 5 cycles at 10 bar) hydrogenation and dehydrogenation were carried out at 260 and 349°C, respectively, the corresponding periods of time being 2 hours and 1.5 hours. The kinetics during these processes substantially remained unchanged. During the last dehydrogenation 7% hydrogen by weight of the magnesium hydride were released.

It will be understood that the specification and examples are illustrative but not limitative of the present invention and that other embodiments within the spirit and scope of the invention will suggest themselves to those skilled in the art. 1. A method of preparing active magnesium hydride/ magnesium/hydrogen storer systems which can reversibly take up H2 by grinding of metallic magnesium in finely divided form with a transition metal from sub-group IV to VIII of the periodic table, wherein the transition metal is used in a proportion of up to 3% by weight, grinding being carried out in the presence of a non-polar organic solvent.

Claims (19)

1.CLAIMS

2. A method according to Claim 1, wherein nickel is used as a 10 transition metal.

3. A method according to Claim 1 or Claim 2, wherein the solvent is toluene.

4. is

5.A method carried out in according to any of Claims 1 to 3, wherein grinding a glass ball mill. 15 5. A method of preparing active magnesium hydride/ magnesium/hydrogen storer systems, substantially as described herein by way of example.

6. An active magnesium hydride/magnesium/hydrogen storer system prepared by the method of any of the preceding claims. Dated this 3rd day of May, 1990. BY:- TOMKINS & CO., Applicants' Agents, (Signed) 5. , Dartmouth Road, DUBLIN 6 - 33 tris-( η 1 -allyl)-iron, ferrocene, and an iron carbonyl.

7. A method as in Claims 1 through 4, characterized in that comolexes or metal-organic compounds of the transition 5 metals platinum, cobalt, rhodium, iridium, ruthenium, chromium, molybdenum, tungsten, titanium and/or vanadium are employed as dopes.

8. A method according to Claim 1, wherein the material contacted is magnesium hydride obtained by reacting magnesium with hydrogen in the presence of a catalyst 15 consisting of a halide of a metal of Subgroups IV-VIII of the periodic table or a magnesium-organic compound or magnesium hydride and optionally, in the presence of a polycyclic aromatic compound. 20 . .

9. x\ method according to Claim 8, wherein during the reacting of magnesium with hydrogen in the presence of a catalyst there is also present a polycyclic aromatic amine, a tertiary amine, MgCl_, MgBr ? and/or Mgl_. 25 ; z - z

10. A method according to Claim 1, wherein the material contacted is metallic magnesium in the form of particles 3q ranging in size from about 0.3-0.045 mm.

11. A method according to Claim 1. wherein the solvent of the solution comprises an aliphatic, cycloaliphatic or or aromatic hydrocarbon or an aliphatic, cycloliphatic or aromatic ether.

12. A method according to Claim 1, wherein the contact with the magnesium hydride or mctaLlic magnesium is effected at a temperature of about -50° C to +-150° C.

13. A method according to Claim 1, wherein the contact with the magnesium hydride or metallic magnesium is maintained for a time sufficient to deposit the transition metal in from about 0.05 to 20% by weight of the magnesium.

14. A method according to Claim 1, wherein the contact with the magnesium hydride or metallic magnesium is maintained for a time sufficient to deposit the transition metal in from about 1 to 47. by weight of the magnesium.

15. A method of preparing an active magnesium hydridemagnesium-hydrogen-storer system which can reversibly take up comprising contacting finely divided metallic magnesium with a transition metal of Subgroups IV to VIII of the periodic table in a finely divided form by milling both metals in the presence or absence of an unpolar organic solvent, whereby magnesium is doped by said transition metal.

16. A method according to claim 15, comprising milling magneslumpowder together with nickel powder in a glassball mill in the presence of an aliphatic, cycloaliphatic or aromatic hydrocarbon, preferrably toluene.

17. A method of preparing an active magnesium hydridemagnesium hydrogen storer system, substantially as described herein by way of example.

18. An active magnesium-hydride or magnesium hydrogenstorer which can reversibly take up H 2 produced by the process of any of the preceding claims. -35

19. In the reversible hydrogenation and dehydrogenation of magnesium which cycles between magnesium and magnesium hydride, the improvement which comprises employing magnesium or magnesium-hydride which has been produced by the process of any of claims 1 - 17.