Composition for Generating Hydrogen

Abstract

The invention provides particulate compositions, which generate hydrogen when contacted with water, the compositions comprising particles of: aluminium; one or more metal oxides; and one or more chloride salts of alkali metals or alkaline earth metals.

The invention also provides methods of preparing such compositions and methods of generating hydrogen by contacting the compositions with water.

Claims

1-25. (canceled)

26. A particulate composition, which generates hydrogen when contacted with water, the composition comprising particles of: 60 to 70% by weight of aluminium particles; 10 to 15% by weight of a group II metal oxide; 10 to 15% by weight of copper (II) oxide; 3.5 to 4.5% by weight of NaCl; 2.5 to 3.5% by weight of KCl; and 2.5 to 3.5% by weight of CaCl.sub.2.

27. The composition according to claim 26, wherein the group II metal oxide is CaO, BaO, MgO or a mixture thereof.

28. The composition according to claim 27, wherein the group II metal oxide is CaO.

29. A particulate composition according to claim 26, wherein a proportion of aluminium oxide has been removed from a surface of the aluminium particles.

30. A method of making a particulate composition according to claim 26, the method comprising milling a combination of aluminium particles, one or more metal oxides and a mixture of NaCl, KCl and CaCl.sub.2 chloride salts in a ratio by weight of 3.5-4.5 : 2.5-3.5 : 2.5-3.5 respectively.

31. A method according to claim 30, wherein the milling is conducted using a planetary ball mill.

32. A method according to claim 31, wherein the milling is conducted using 5 or more balls having a diameter of greater than 5 mm.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0107] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0108] FIG. 1 is a graph showing the effect of varying the metal oxide present in a milled composition containing aluminium particles, metal oxide and NaCl on the volume of hydrogen generated.

[0109] FIG. 2 is a graph showing the effect on hydrogen yield when using a combination of CaO and CuO as metal oxides in a composition containing aluminium particles, metal oxide and NaCl milled using a first milling programme.

[0110] FIG. 3 is a graph showing the effect on hydrogen yield when using a combination of CaO and CuO as metal oxides in a composition containing aluminium particles, metal oxide and NaCl milled using a second milling programme.

[0111] FIG. 4 is a graph showing the effect on hydrogen yield when varying the proportions of CaO and CuO in a composition containing aluminium particles, CaO, CuO and a combination of KCI, NaCl and CaCl.sub.2.

[0112] FIG. 5 is a graph showing the effect on hydrogen yield when varying the nature of the salt in a milled composition containing aluminium particles, CaO, CuO and the salt.

[0113] FIG. 6 is a graph showing the effect on hydrogen yield when using a combination of NaCl, KCl and CaCl.sub.2, compared to CaCl.sub.2 alone, in a milled composition containing aluminium particles, CaO, CuO and the salt(s).

[0114] FIG. 7 is a graph showing the effect on hydrogen yield when using a combination of NaCl, KCI and CaCl.sub.2, compared to no salts, in a milled composition containing aluminium particles, CaO and CuO.

[0115] FIG. 8 is a graph showing the effect of using various milled and non-milled combinations of aluminium particles, metal oxide(s) and salt(s) on hydrogen yield.

[0116] FIGS. 9 and 10 are graphs showing the effect of the milling conditions of the compositions of the invention on hydrogen yield.

[0117] FIG. 11 is a graph showing the effect of aluminium particle size on hydrogen yield.

[0118] FIG. 12 is a graph showing a comparison of hydrogen yield when recycled and ‘pure’ aluminium are used in the compositions of the invention.

[0119] FIG. 13 is a graph showing the volume of hydrogen generated by a composition of the invention when contacted with aqueous solutions of ethanol at various concentrations.

[0120] FIG. 14 is a graph showing the volume of hydrogen generated by a composition of the invention when contacted with aqueous solutions of ethylene glycol at various concentrations.

[0121] FIG. 15 is a graph showing the volume of hydrogen generated by a composition of the invention when contacted with aqueous solutions of urea at various concentrations.

EXPERIMENTAL SECTION

Methods

Particle Synthesis

[0122] In the Examples below, the following method was used to prepare the aluminium-containing compositions of the invention.

[0123] Prior to milling, all powders were dried in a vacuum furnace (Townson and Mercer Ltd) for 24 hrs to remove any excess moisture. After drying, the powders were kept in a desiccator inside an oxygen-free glove box (Saffron Scientific Alpha) which was purged with 99.99% pure argon gas to ensure a moisture and oxygen-free environment. Inside the glove box, an oxygen sensor (SYBRON Taylor) was placed to measure the level of oxygen within an accuracy of ± 0.01%.

[0124] Below is a list of the components used in the milling process for the aluminium-containing compositions of the invention. [0125] Aluminium recycled (99.1 wt%, sieved further with 40 .Math.m, 75 .Math.m and 105 .Math.m mesh, obtained from iHOD USA). [0126] Aluminium pure (99.5 wt%, Alfa Aesar, 200 mesh, Fisher Chemical). [0127] Calcium oxide (99.0 wt% CaO, 65 .Math.m, Fisher Chemical). [0128] Copper oxide (99.0 wt% CuO, nanoparticles, ACROS Organics). [0129] Barium oxide (90.0 wt% BaO, nanoparticles, ACROS Organics). [0130] Potassium chloride (99.5 wt% KCI, 65 .Math.m, Fisher Chemical). [0131] Calcium chloride (80 wt% CaCl.sub.2, 280 .Math.m, VWR Chemical). [0132] Sodium chloride (98.0 wt% NaCl, 150 .Math.m, Fisher Chemical).

[0133] Unless stated otherwise, recycled aluminium powder as received from iHOD USA was used. This aluminium powder contained a blend of different particle sizes and therefore the recycled aluminium was sieved to provide 3 different particle size ranges to establish the effect of different particle sizes on hydrogen yield. For this purpose, sieves BS410/1986 (Endecott Test Sieve shaker E.F.L Mark II with Endecott’s Ltd) with sizes ranging from 3 .Math.m to 300 .Math.m were employed. The sieves were placed in descending size order on top of each other and on the top-most sieve (300 .Math.m mesh size) aluminium powder was dispensed. The sieving process was carried out for 48 hrs.

[0134] After separation had taken place, sieves corresponding to particle diameters of 40 .Math.m, 70 .Math.m and 100 .Math.m were selected. The particles in the 40 .Math.m sieve had a diameter between 40 .Math.m and 50 .Math.m, the particles in the 70 .Math.m sieve had a diameter between 70 .Math.m and 80 .Math.m and the particles in the 100 .Math.m sieve had a diameter of between 100 .Math.m and 110 .Math.m.

[0135] Powder preparation for milling was performed under anaerobic condition inside a glove box before being transferred to a planetary ball mill device for milling. All percentage weights of the components of the composition are given as a weight percentage with reference to the total weight of the composition.

[0136] For the ball milling, a ball-to-powder ratio of 10:1 by weight was used. Eight milling balls (spherical stainless-steel balls 7 mm diameter) and 3 g of aluminium powder along with chosen additives were placed into a 50 ml stainless steel milling jar while inside the glove box. The sealed assembly from the glove box was then transferred to a planetary ball mill device (Retsch PM-100). The total weight of the milling jar was adjusted with a counter balance on the milling machine station to avoid imbalance and rattling during high-speed milling.

[0137] Different milling programmes were set up in which the direction of rotation of the mill and the milling speeds were altered. Details of the milling programmes used are provided in Table 1 below:

TABLE-US-00001 Milling Programmes Milling Programme Total milling time Milling period Speed of milling Break between milling periods Directions of milling 1a 1 hr and 38 min 1 min 258 rpm 30 sec Anticlockwise/ Clockwise 1b 1 hr and 38 min 1 min 518 rpm 30 sec Anticlockwise/ Clockwise 1c 2 hr and 38 min 1 min 518 rpm 30 sec Anticlockwise/ Clockwise 1d 2 hr and 24 min 1 min 518 rpm 30 sec Anticlockwise/ Clockwise 2a 1 hr and 38 min 1 min 258 rpm 5 sec Anticlockwise/ Clockwise 2b 1 hr and 38 min 1 min 518 rpm 5 sec Anticlockwise/ Clockwise

[0138] Programmes 1a to 1d differed in milling speed and total milling time only and consisted of 1 min milling, a 30 sec break followed by a further 1 minute of milling with rotation in the opposite direction and another 30 sec break. This was repeated until a total milling time 1 hr and 38 min (for programmes 1a and 1b) and 2 hr and 24 min (for programmes 1c and 1d) was reached.

[0139] Programmes 2a and 2b were used to test the importance (if any) of the intermediate break time where the break time was set to 5 sec instead of 30 sec as it was for Milling Programme 1a and 1b.

Measuring Yield of Hydrogen

[0140] In the Examples below, the following method was used to measure the amount of hydrogen liberated upon reaction of the aluminium-containing compositions of the invention with water (or other selected liquids).

[0141] A Pyrex® glass tube (60 ml, inner diameter: 21 mm) was used as the reaction vessel. A rubber stopper with 2 holes acted as a sealant for the connections. One of the holes in the stopper provided the exit channel for the hydrogen that was liberated in the reaction whereas the other hole was used to insert a thermocouple (k-type) connected to a digital data logger (Picotech, Model: 2204) in order to monitor the temperature.

[0142] Before the start of the reaction, the vessel was thoroughly purged with pressurised argon gas in order to keep the concentration of oxygen in the vessel as low as possible. 0.3 g of an aluminium-containing composition (prepared using the method described above) was added to the reactor followed by 9 ml water (or other liquid as specified in the Examples below) at 25° C. which was added using a syringe. The reactor vessel was wrapped with an insulating polystyrene sheet. The mixing of water and the composition was accomplished by agitation using a small capsule-shaped stirrer bar (5 mm, 1 g) and a magnetic stirring plate (IKA-RH-Basic 2) used to set the agitation speed at 300 rpm. The size and the weight of the stirrer allowed free movement of particles inside the reactor.

[0143] The hydrogen gas generated was passed through a series of stainless steel pipes (internal diameter: 7 mm) with three elbow compression joints and one push-fit joint to avoid any gas leakage.

[0144] Two methods were employed to measure the rate of hydrogen generation and the total amount of hydrogen generated; one being inverted column method and the other involving the use of a gas mass flow meter. The gas mass flow meter had ± 0.01 ml accuracy in the flow range of 0-10 ml/min. The gas flow meter was pre-calibrated for hydrogen gas.

[0145] To ensure that dry gas entered the gas flow meter, a reinforced plastic tube joint (5 cm × 3 cm) containing a desiccant (silica gel) was attached to a gas mass flow meter (Aalborg GFM-17). The hydrogen produced was recorded via a data logger connected to a PC using the relevant Pico Logger software with sample intervals of 1 sec. The connections to the data logger enabled both the hydrogen flow rate and temperature to be read and recorded simultaneously. In order to analyse the quality of the gas produced, a gas-tight syringe was used to collect the gas and introduce the gas into a gas analyser (gas chromatogram, GC).

[0146] The% hydrogen yield values as reported below were calculated based on the theoretical maximum amounts of hydrogen that could be liberated from a 0.3 g composition containing 65% by weight of aluminium (i.e. 0.195 g of aluminium) - unless stated otherwise. This amount corresponds to 264.8 mL of hydrogen gas at 20° C. and 1 atm pressure (101,325 Pa).

EXAMPLES

Example 1

Comparison of Hydrogen Yields With Different Metal Oxides

[0147] Compositions were prepared comprising aluminum particles (diameter: 70 .Math.m to 80 .Math.m, obtained as described above), sodium chloride (NaCl) and various metal oxides in the proportions shown in Table 2. The selected metal oxides for this study were barium oxide (BaO), calcium oxide (CaO) and copper oxide (CuO). The powders were milled using Milling Programme 1b, as described in the Methods Section above, using a mill speed of 518 rpm and a total milling time of 1.1 hr.

[0148] The yield of hydrogen after 1000 seconds is shown in FIG. 1 and Table 2 below. The% hydrogen yield shown in Table 2 is relative to the maximum theoretical yield of hydrogen for the aluminum contained in the composition.

TABLE-US-00002 Powder composition with different metal oxide additives Powder composition (wt%) % Hydrogen Yield after 1000 sec Al 65%, BaO 25%, NaCl 10% 4.5% Al 65%, CaO 25%, NaCl 10% 3.8% Al 65%, CuO 25%, NaCl 10% 1.4%

[0149] In FIG. 1, it can be seen when using BaO, hydrogen gas was produced instantly and a total of 12 ml of hydrogen was generated in 1000 sec (corresponding to 4.5% hydrogen yield). For CaO and CuO, the yields of hydrogen were much lower and the generation of hydrogen was minimal after 600 and 400 sec respectively.

Example 2

Use of a Combination of Metal Oxides

[0150] Compositions were prepared comprising aluminum particles (diameter: 70 .Math.m to 80 .Math.m, obtained as described above), sodium chloride (NaCl) and various metal oxides in the proportions shown in Table 3. The selected metal oxides for this the study were calcium oxide (CaO), copper oxide (CuO) and equal proportions of CaO and CuO (but with the total weight of metal oxides being kept to 25% of the total composition). For this study, all the powders were milled using Milling Programme 1b or 1d, as described in the Methods section above.

TABLE-US-00003 Powder composition with different metal oxide additives. Powder composition (wt%) Milling Programme % Hydrogen Yield after 1000 sec Al 65%, CaO 25%, NaCl 10% 1b 3.7% Al 65%, CuO 25%, NaCl 10% 1b 1.5% Al 65%, CaO 12.5%, CuO 12.5 %, NaCl 10% 1b 4.2% Al 65%, CaO 25%, NaCl 10% 1d 2.2% Al 65%, CuO 25%, NaCl 10% 1d 1.8% Al 65%, CaO 12.5%, CuO 12.5 %, NaCl 10% 1d 5.3%

[0151] Table 3 and FIG. 2 show the hydrogen yields of the three compositions prepared by Milling Programmes 1b and 1d. For the composition containing the combined metal oxide milled using Milling Programme 1b, a total of 11 ml hydrogen was produced after 1000 sec which is comparable to the previous use of the BaO additive, (see Example 1).

[0152] It can be seen that when a combination of the two metal oxides (CaO and CuO) were used, there was more of an immediate, albeit slower rise in the generation of hydrogen whereas there was a delay in production of hydrogen when CaO or CuO were used separately in the mixture. CuO was also observed to produce a lower volume of hydrogen over 1000 seconds.

[0153] Table 3 and FIG. 3 show the hydrogen yields of the three compositions prepared by Milling Programme 1d. Milling Programme 1d differed from Milling Programme 1b in that the total milling time was increased from 1.1 hr to 2.4 hr. In FIG. 3, it can be seen that the composition containing the combined metal oxides produced 13 ml hydrogen after 1000 sec while the compositions containing only CaO or CuO produced only 6 ml and 5 ml, respectively. In addition, it was noted that the high reaction rate seen previously for the CaO sample when it was milled for 1.1 hrs had also been affected, with it resulting in an inferior hydrogen yield after 1000 sec.

Example 3

Varying the Metal Oxide Ratios

[0154] To further explore the increased hydrogen yield when using combined metal oxide additives, different ratios of the two metal oxides were tested. Compositions were prepared comprising aluminum particles (diameter: 70 .Math.m to 80 .Math.m,obtained as described above), a mixture of sodium chloride (NaCl), potassium chloride (KCI) and calcium chloride (CaCl.sub.2) and various metal oxides in the proportions shown in Table 4. For this study, all the powders were milled using Milling Programme 2a.

TABLE-US-00004 Powder compositions with different ratios of CuO and CaO Powder composition (wt%) % Hydrogen Yield after 10,000 sec Al 65%, CaO 12.5%, CuO 12.5%, NaCl 4%, KCl 3%, CaCl.sub.2 3% 85 Al 65%, CaO 8.75%, CuO 16.25%, NaCl 4%, KCl 3%, CaCl.sub.2 3% 53

[0155] The volume of hydrogen produced by samples with a CuO: CaO ratio corresponding to 65 wt %: 35 wt% (sample 65-35) was compared to 50 wt% CuO and 50 wt% CaO (sample 50-50). The hydrogen flow rate and the volume of hydrogen generated by each composition can be seen in FIG. 4. Sample 50-50 displayed a higher flow rate than sample 65-35. This was approximately twice as high, e.g. at 1000 sec (rate of 0.04 ml/s for sample 50-50 versus 0.02 ml/s for sample 65-35.)

[0156] The difference can also be seen in the generated hydrogen volume on the right-hand y-axis of FIG. 4, where the sample 50-50 produced 220 ml (85% H.sub.2 yield) after 10,000 sec compared with 140 ml (53% H.sub.2 yield) for sample 65-35 after the same period of time.

Example 4

Use of a Combination of Chloride Salts

[0157] Each of NaCl, KCI and CaCl.sub.2 were milled together with aluminium powder and CaO and CuO in equal proportions as listed in Table 5. This mixture of CaO and CuO is referred to below as MO. The powers were milled using Milling Programme 1a, as described in the Methods Section above.

TABLE-US-00005 Composition of additives in the sample Powder composition (wt%) % Hydrogen Yield after 1000 sec Al 65%, CaO 12.5%, CuO 12.5%, NaCl 10% 5.7% Al 65%, CaO 12.5%, CuO 12.5%, KCl 10% 5.3% Al 65%, CaO 12.5%, CuO 12.5%, CaCl.sub.2 10% 8.3% Al 65%, CaO 12.5%, CuO 12.5%, NaCl 4%, KCI 3%, CaCl.sub.2 3% 12.8%

[0158] As FIG. 5 and Table 5 show, it is clear that by using 10 wt% CaCl.sub.2, hydrogen gas is generated both more immediately and in a greater amount compared to NaCl and KCI within the first 1000 sec of reaction. At 1000 sec the CaCl.sub.2 sample had generated 22 ml of hydrogen compared with 15 ml for NaCl and 14 ml for the KCI sample.

[0159] The three salts were mixed together to determine the effect of using a combination of chloride salts. The mixture (hereinafter referred to as “PO”) contained three salts; CaCl.sub.2, NaCl and KCl in a ratio of 3:4:3 respectively. Furthermore, to investigate if there was synergistic effect, salt additive PO was tested against CaCl.sub.2. Milling Programme 1a was used to mill both compositions.

[0160] It can be seen from Table 5 and FIG. 6 that the hydrogen yield is increased when using a mixture of the three chloride salts compared with CaCl.sub.2 only. After only 600 sec, the composition containing PO had generated 22 ml of hydrogen gas compared to 13 ml for the composition containing CaCl.sub.2 only. This can be compared to 9 ml from NaCl or KCI from demonstrating its superiority over them.

[0161] To further explore the effect of salt additives, two powders were prepared. One contained all the additives, i.e. (Al+MO+PO) and other which had no salt additive, i.e. powder (Al+MO). These are called “No PO” and “With PO” in the results, respectively.

[0162] Here it was necessary to adjust the weight% accordingly. The absence of salt in the sample No PO was adjusted by increasing the portion of metal oxides to keep the Al:MO ratio 65:35. Powders were milled using Milling Programme 1a at 258 rpm and reacted with deionised water at 25° C. for 10000 sec.

TABLE-US-00006 Effect of removing salts from the composition Powder composition (wt%) % Hydrogen Yield after 4000 sec Al 65%, CaO 12.5%, CuO 12.5%, NaCl 4%, KCI 3%, CaCl.sub.2 3% 50% Al 65 wt%, CaO 17.5 wt%, and CuO 17.5 wt% 19%

[0163] From Table 6 and FIG. 7, it can be seen that milled “No PO” powders only produced 48 ml of H.sub.2 in 4000 sec and after that stopped producing any further hydrogen. On the other hand, the PO-containing powder displayed an increased hydrogen yield. In the first 4000 sec, the “With PO” sample generated 130 ml (50% yield) while only 48 ml of hydrogen (19% yield) was generated for the “No PO” sample.

[0164] Another important observation is that for “No PO” sample the reaction rate is slow for the first 1700 sec and then increases rapidly until 3000 sec reaction time where the reaction then appears to come to a halt.

Example 5

Combined Effect of Metal Oxides and Chloride Salts

[0165] To investigate the importance of milling and the additives to the volume of hydrogen produced, it was decided to prepare three samples via milling and a further sample without milling.

TABLE-US-00007 Comparison of Compositions Name Composition Milling Programme Hydrogen Yield after 10,000 sec Al + MO Al 65%, CaO 17.5%, CuO 17.5% Milling Programme 1a 94% Al + PO Al 65%, NaCl 14%, KCI 10.5%, CaCl.sub.2 10.5% Milling Programme 1a 0.02% Al + MO + PO Al 65%, CaO 12.5%, CuO 12.5%, NaCl 4%, KCI 3%, CaCl.sub.2 3% Milling Programme 1a 85% Al + MO + PO Al 65%, CaO 12.5%, CuO 12.5%, NaCl 4%, KCI 3%, CaCl.sub.2 3% No milling 54%

[0166] From FIG. 8, it can be seen that whilst the total yield of hydrogen after 10,000 seconds was slightly greater for the composition containing aluminium and MO (Al+MO) only compared to the composition with both additives (Al+PO+MO), for the composition containing both additives, the rate production of hydrogen was fairly constant for the first 6,000 seconds, after which point the rate of production steadily decreased. By contrast, for the composition containing aluminium and MO only, for the first 2,000 seconds the amount of hydrogen generated was low. This was followed by a sharp rise where large quantities of hydrogen were generated is a short period of time between 2,000 and 5,000 seconds. Therefore, whilst the overall hydrogen yield was slightly higher for the Al+MO composition than for the Al+PO+MO composition, the Al+PO+MO composition has the advantage that the rate of hydrogen generation is much more constant. It is therefore envisaged that this composition would be more useful in an apparatus where a steady rate of hydrogen generation is required over a period of 2 to 3 hours.

[0167] Without milling, the same composition produced only 700 ml hydrogen per gram of aluminium after 10000 sec, corresponding to an approximate hydrogen yield of 54%.

[0168] For the same reaction time, sample (Al+MO+PO) had already produced a volume of hydrogen of 400 ml/g Al. Furthermore, when 0.3 g of (Al+PO+MO) was allowed to react with 9 ml water for 12000 sec, it produced a total of 235 ml which correspond to a hydrogen yield of 90% per amount of metal reacted.

[0169] The hydrogen yields when either the metal oxides or the PO salt mix were omitted were significantly reduced.

Example 6

Effect of Milling Conditions on Hydrogen Yield

[0170] The effect of varying the milling conditions on the hydrogen yield of the milled compositions was studied. The compositions contained aluminum powder (40 .Math.m to 50 .Math.m, obtained as described above) 65%, calcium oxide 12.5%, copper (II) oxide 12.5%, NaCl 4%, KCI 3% and CaCl.sub.2 3%.

[0171] As can be seen in FIG. 9, when the powder prepared at 258 rpm was reacted with deionised water, hydrogen generation occurred progressively across the whole 1000 sec and was still ongoing at 10000 sec regardless of milling durations. The volume of hydrogen for compositions milled at 258 rpm for total milling times of 1.1, 1.77 and 2.4 hrs were 220 ml, 170 ml and 230 ml respectively. This corresponded to respective hydrogen yields of 85%, 65 % and 88%.

[0172] Results for compositions milled at 518 rpm showed no progressive hydrogen generation and after 1000 sec only ~13 ml of hydrogen was generated (corresponding to a yield of 4.9%). After 10000 sec no further hydrogen appeared to be produced.

[0173] In addition, three milling programmes - 1a, 1b and 2a (described in the Methods Section above) - were compared for their effects on hydrogen production. Similar to previous study all the compositions of the additives (Al 65 wt%, MO 25 wt%, Salt 10 wt%) including particle aluminium particles size, i.e. 40 .Math.m were kept constant.

[0174] As can be seen in FIG. 10, there is a striking difference in hydrogen production between three different Milling Programmes. Milling Programme 2a produces far less hydrogen (total of 80 ml, 30% yield) after 10000 sec compared with Milling Programme 1a (220 ml, 85% yield). However, Milling Programme 1b produced the lowest volume with only 13 ml of hydrogen (5% yield).

Example 7

Effect of Aluminium Particles

[0175] The effects of using recycled aluminium rather than non-recycled aluminium and the aluminium particle size used in the compositions of the invention were also investigated.

[0176] Recycled aluminium (provided by iHOD USA LLC) with particle size 3-200 .Math.m was sieved to obtain representative batches of particles having diameters of 40 .Math.m, 75 .Math.m and 105 .Math.m sizes prior milling. The different sized batches were then mixed with the additives (CaO 12.5%, CuO 12.5%, PO 10%) and milled using Milling Programme 1a.

[0177] In FIG. 11, the plotted results show the effect that the particles size has on the production of hydrogen. It can be seen that for the compositions made from recycled aluminium, particle size does have an effect on the yield of hydrogen. The smallest starting Al particle size, 40 .Math.m, showed the highest hydrogen generation followed by 75 .Math.m, whereas 105 .Math.m was considerably slower and produced the least amount of hydrogen of them all.

[0178] At 10000 sec reaction time, the 40 .Math.m batch had produced 220 ml, the 75 .Math.m batches produced slightly less of 172 ml and the largest sized recycle aluminium particle batch of 105 .Math.m only produced 90 ml hydrogen corresponding to a hydrogen yields of 85%, 66% and 35 %, respectively.

[0179] To continue the study, a 40 .Math.m recycled Al batch was compared to 10 .Math.m-diameter aluminium particles (obtained from Fisher Chemicals, 99.9% purity) named “Fisher Al”. For comparison, powder compositions were kept same as for above experiments, i.e. (Al 65 wt%, CaO 12.5 wt%, CuO 12.5 wt% and PO 10 wt%) and both powders were prepared using Milling Programme 1a (258 rpm). The volume of hydrogen produced from each sample can be seen in FIG. 12.

[0180] A distinctive reaction lag time of up to 2000 sec was observed in the case of Fisher Al particles, but a much shorter lag was witnessed for the Recycled Al 40 .Math.m sample. The flow rate of hydrogen generated from the Fisher Al particles continued to rise until the 2800 sec mark, after which a levelling off was observed. The amount of hydrogen generated by the Fisher Al corresponded to 85% hydrogen yield compared to 220 ml by “Recycled Al 40 .Math.m” corresponding to 92% hydrogen yield, both after 10000 sec reaction time.

Example 8

Reaction of Compositions With Other Liquids

[0181] The reaction of the compositions of the invention with aqueous solutions of ethanol, ethylene glycol and urea were investigated to determine the suitability of the compositions to generate hydrogen in environments where clean water is not readily available. The compositions were prepared according to the methods described above using aluminium particles (65 wt%) with a diameter of 70 - 80 .Math.m. The compositions also contained 12.5 wt% CaO, 12.5 wt% CaO and 10 wt% PO salt mix and were prepared using Milling Programme 1a (258 rpm).

[0182] In FIG. 13, the results of the hydrogen formation reactions with different concentration of ethanol solutions are displayed. It can be seen that regardless of the concentration, ethanol solutions were able to produce hydrogen gas. With the highest concentration of 0.68 M, 25 ml of hydrogen gas was liberated in a 1000 sec reaction, corresponding to a hydrogen yield of 9.4%.

[0183] In FIG. 14, the results of the hydrogen formation reactions with different concentrations of ethylene glycol and an industrial commercially available antifreeze (Q8 antifreeze, ethylene glycol content of >90% according to the product specification) are displayed. It can be seen that increasing the concentration of ethylene glycol up to 0.77 M appears to improve the hydrogen formation. In contrast, the commercial antifreeze produced the least amount of hydrogen.

[0184] Urea solutions were prepared according to D.F. Putnam, Composition and concentrative properties of human urine. NASA Report (1971). Urea (CH.sub.4N.sub.2O, Mr = 60.05 g/mol) powder was mixed with deionised water to make solutions of concentrations 0.101 M (0.66 g), 0.15 M (0.9 g), and 0.05 M (0.35 g). As the concentration was increased, the salts weight percentage, i.e. NaCl and KCI wt% were also increased and were dissolved into the urea solution to represent the actual levels of human urine. Once the mixing had finished the beaker in which the solutions were kept were tightly sealed and stored in an inert atmosphere at 17° C. to avoid any oxidation and ammonia formation.

[0185] As shown in FIG. 15, the highest concentration of urea, i.e. 0.15 M liberated 43 ml of hydrogen in a 1000 sec reaction. This corresponds to a hydrogen yield of 16%. The use 0.15 M solution of urea generated a larger volume of hydrogen compared to when deionised water was used by approximately 10 ml (3.8%).

[0186] The embodiments described above and illustrated in the accompanying figures and tables are merely illustrative of the invention and are not intended to have any limiting effect. It will readily be apparent that numerous modifications and alterations may be made to the specific embodiments shown without departing from the principles underlying the invention. All such modifications and alterations are intended to be embraced by this application.