Hydrogen production using Al-Sn alloys prepared by rapid solidification

Y.M.Abbas, A. Hassan Ibrahim, S. Mosaad, M.Orabi (1) Physics Department, Faculty of Science, Suez Canal University, Egypt ymabbas@gmail.com (1) Physics Department, Faculty of Science, Suez Canal University, Egypt Ahmedphysics.ah@gmail.com (1) Physics Department, Faculty of Science, Suez Canal University, Egypt saramosaad@windowslive.com (1) Physics Department, Faculty of Science, Suez Canal University, Egypt Ahmed_Abdel-Hamid@science.suez.edu.eg  Abstract


Introduction
The direct combustion of fossil fuels that produced air pollution is a dangerous environmental problem. Consequently, there is an urgent request for renewable and clean fuel alternatives for future energy source. Hydrogen, which displays high calorific value and clean combustion product, might be the first choice in the future, as it is a smart fuel for fuel cells where the electric energy is obtained directly by the electrochemical reactions of hydrogen and oxygen [1]. Furthermore, hydrogen has many wonderful advantages such as high energy density, non-polluted reaction products, and rich natural sources. Alternatively, hydrogen does not typically exist by itself in nature and must be generated from a number of compounds through a chemical reaction. Numerous methods can be used for hydrogen generation, such as the decomposition of fossil fuels, water splitting and a reaction of metals or alloys with water. Among the methods, especially, the technology based on the reaction of reactive metals with water to produce hydrogen has been explored broadly [2][3][4]. As an advanced application of a conventional technology, hydrogen production by corrosion of alloys and hydrolysis has been extensively studied [5]. In a metal reactions, Large amounts of hydrogen release occurs instantly after the metal is immersed, even in mild aqueous solutions. So, the reactions of metal allow a real-time hydrogen generation from aqueous mediums. A number of studies have surveyed the use of metals, including Zn, Al, and Mg, to produce hydrogen from aqueous solutions [6][7][8][9][10][11][12]. Amid the candidates, aluminum (Al) has been recognized as the most talented metal for hydrogen generation [13] because it is cheap and environmentally recyclable, produces hydrogen without carbon oxides, and provides a high hydrogen yield of up to 11.1 wt%. On the other hand, the inactive film can form easily on the Al surface during the hydrogen generation, which decreases the hydrogen production rate. Numerous methods such as the use of chemical additives in solutions and the addition of alloying elements in Al alloys have been considered to stop the creation of inactive film and accelerate the corrosion of the alloy surface. Especially, the hydrogen generation rate has been reported to be increased by the addition of elements such as Fe, Ni and Cu, owing to the precipitation of phases like Al2Cu, Al3Fe, and Al3Ni along the grain boundary of aluminum [14][15][16]. Moreover, Al-Si alloys were reported to have the I S S N 2 3 4 7 -3487 V o l u m e 1 3 N u m b e r 6 J o u r n a l o f A d v a n c e s i n P h y s i c s highest initial H2 generation in alkaline water. It has also been reported that hydrogen can be produced by the reaction of Si with an alkaline solution of NaOH [17]. However, another metal is needed to activate the reaction between Si and water to produce H2, due to the low activity of Si, and Al can be used for this persistence [18]. Tin (Sn) was chosen for preparing the Al alloy. Al-Sn alloy reactants were investigated by changing the alloy content and microstructure.

Characterization Techniques
The phase identification and crystal structure of the rapidly solidified Al-Sn alloys were analyzed by using X-ray powder diffractometer (X'Pert pro panalytical MRD) at the Ain Sham University, Egypt, and equipped with a CuKα target of (λ=1.54056Å) radiation and a scanning rate of 2 o (2θ)/min. The Rietveld refinement of the XRD peaks was analyzed through (FullProff Suit program 2.05). GFourier program (FullProff package) was used for the calculation and visualization of electron density within the unit cell. Average crystallite size was calculated by Scherrer's formula considering the position and broadening of the most intense diffraction peak in XRD spectra, beside using Williamson-Hall plot for calculation of the particle size and microstrain. The microstructure of the rapidly solidified Al-Sn alloys were observed by scanning electron microscopy (SEM /Hitachi S-4800) at the (Central Metallurgical Research Institute -Helwan-Egypt), utilizes an electron beam accelerated at 500 V to 30 kV, and features a maximum resolution of 1 nm. Maximum magnification is 800,000X. To observe SEM, the alloy sample surface was polished with the silicon carbide abrasive paper, followed by ultrasonic treatment in ethanol for 2 min to remove the oxide layer. TG/DTA 6300 system (Cairo university, Egypt) was used to characterize our samples, the samples exposed to temperatures up to 350 o C, with the temperature rise rate up to 20°C/min. The Electrical properties and Temperature dependence of resistivity had been performed for all prepared alloys using an RLC bridge (model: GRUNDIG programmable automatic RLC Meter) with a resistance measurement range (1 mΩ -99.9 MΩ). For H2 generation test a water displacement method was used to calculate the volume of the generated hydrogen for the Al-Sn melt spun alloys.

Hydrogen generation test
Rapidly solidified Al100-x-Snx alloy X= (0, 25, 55, 75 all in wt.%) was prepared using melt spinning technique at 800 º C. For the hydrogen generation tests, the prepared alloys were attached using epoxy resin and polished. The area of the specimen exposed to the environment was 2 cm 2 , The fixation of the sample was illustrated in fig.1. The hydrogen generation tests were performed in a 2 M NaOH solution at (100 to 110°C) for the Al-Sn melt spun alloys. This is because fast hydrogen generation from the hydrolysis of the Al alloy in bulk form in strong alkaline water. In a weak alkaline medium, the exclusion of passive films like Al2O3 and Al(OH)3 is not possible [19]. The temperature for the reaction was maintained by a digital thermocouple. The volume of producing hydrogen gas was measured using the water displacement method as cited elsewhere [20]. In this specific experiment, a 1000 ml measuring cylinder filled with water was employed in a plate partially filled with water and a silicon tube connected to the reaction flask was inserted inside the cylinder, as shown in Fig.2. As the H2 flow starts, the gas through the silicon tube comes in the cylinder. The H2 gas started to be generated at the top of the cylinder, and displaces water over its pressure.

Refinement Results and Discussion
In the Rietveld analysis of X-ray diffraction data of the four alloy systems (Pure-Al, Al25Sn75, Al75Sn25, and Al45Sn55), we employed the software (FULLPROF) using the Rietveld method. It is designed to refine simultaneously both the structural (lattice cell constants, atomic positions and occupancies) and microstructural parameters (crystallite size and lattice micro strain). The shape of the peaks in the experimental diffraction patterns was well described by an asymmetric pseudo-Voigt (PV) function. To simulate the theoretical x-ray diffraction patterns of the prepared alloy systems the following considerations for the different phases we made: 1-Identification of the phases by computer search-match to compare the actual experimental patterns with the International Centre for Diffraction Data (ICCD) database of known materials.
2-Index the diffraction pattern to determine the crystal system, unit cell dimensions and space group.
In each refinement, a total of more than twenty parameters were refined: zero shift, scale factor, background coefficients, three lattice constants, asymmetry factor, and parameter for the full width at half maximum.
The Rietveld plots of the refinements for pure-Al, Al25Sn75, Al75Sn25, and Al45Sn55 prepared alloys are given in fig.4 to Fig.7. In this figure the observed intensity data, y, are plotted in the upper field as points. The calculated patterns are shown in the same field as a solid line curve. The difference between calculated and observed patterns is shown in the lower field. The short vertical bar in the middle field indicates the positions of possible Bragg reflections.

Electron density Maps, and calculations
GFourier program was used for the calculation and visualization of electron density within the unit cell. The visualization is very useful in identifying the atomic positions of constituent elements within the unit cell for known or unknown crystals, i.e. denser the electron density contours indicate the position of a heavier element among the constituent elements in the unit cell. In the world of crystallography, You may wonder about this use of the Fourier transform. The scattering density ρ(x,y,z) is usually calculated according to the equation: Where,ρ(x,y,z) is the electron density at a point x,y,z in a unit cell of volume V, F(hkl) is the structure factor amplitude and αhkl is the phase angle of each Bragg reflection (h,k,l).
Two-dimensional maps are typically drawn with contours (and sometimes colour) to indicate different density levels, while three-dimensional maps employ a chicken-wire style mesh representing a single level. (The latter works very well in protein crystallography, where 99% of the atoms (excluding H) scatter near identically.) Some typical 2-dimensional maps of the plane z = 0 for Al and Sn phases are calculated using X-ray structure factors are shown below in Figures 8  and 9.  The zero level density contour is shown in black, while the colors red through to violet-brown region indicates increasing levels of electron density around Al and Sn phases.

Thermal analysis of Al-Sn Alloys
Differential Thermal Analysis (DTA) is performed for as quenched alloys in the temperature range from 300 to 1000 K at a heating rate of 10 K min -1 by a DTG-60H. Fig. (10.a) shows the DTA scan in the temperature range for rapidly solidified pure Al alloys. It shows that, no phase transformation has been observed before melting, which occurred at 931.2 K compared with 933.47 K for conventional Al. The enthalpy and entropy of fusion is estimated from the fc for pure Al rapidly solidified was found to be 7.07 kJ.kg   . This means that the enthalpy of RS pure Al is increased. Assuming that the formation of vacancies is the most important effect of rapid solidification in this case, the enthalpy increase due to the presence of vacancies is estimated to be ΔHv= 3.64 kJ.kg -1 . Similarly, we can calculate the increase in entropy due to the formation of vacancies: ΔSf CC -ΔSf RS = Ss RS -Ss cc , from which Ss RS -Ss cc = 11.47 -7.57 = 3.9 J.kg -1 .K -1 . The change in the free energy is given byΔGv = ΔHv − TΔSv, where T is the temperature just before melting (929.1 K). Substituting for ΔHv and ΔSv then we can obtain ΔGv = 7029.83 J.kg -1 . Fig. (10.b) shows the DTA curve for Al75Sn25, Al25Sn75 and Al45Sn55 alloys. The details of thermal analysis using DTA for all alloys are summarized in Table (1).

Hydrogen generation using the corrosion of Al-Sn alloys in an alkaline solution
Rapidly solidified Al100-x-Snx alloy, X= (0, 25, 55, 75 all in wt.%) were immersed in 2 M NaOH solution at 95 °C, and the volume of generating hydrogen of the solution by immersing the alloys was recorded, as shown in Fig.11.
The volume of hydrogen generation varied from 210 ml for pure Al to 531 ml for Al45Sn55 at 3.5 min. For the Al45Sn55 alloy, the amount of generating hydrogen was the highest at 531 ml among all the measurements at 95 °C. The asquenched melt spun 55 wt % Sn alloy showed saturation in the hydrogen generation rate after 3.23 min. While the 25 wt% and 75 wt% Sn alloys exhibited a saturation of hydrogen generation at 320 and 235 ml, respectively. The higher hydrogen generation for the melt spun alloys might be owing to the good distribution of Sn along the Al grains at the surface and the resulting shorter path of electron between Sn and Al in galvanic corrosion. In our work, the most important effect of the addition of Sn was the shortening of delay time, so that the reaction process started instantly after the sample came in direct contact with alkaline solution. This is owing to the creation of galvanic cells between the Al grain and close Sn phase. The existence of the Tin phase in grain boundaries can accelerate the corrosion of Al grains by galvanic effect and the consequential hydrogen production.

Fig.11. The Volume of hydrogen generation from 2 M NaOH solution of as-quenched melt spun Al-Sn alloys.
Also, the reaction kinetics increased with increasing the solution temperature. The delay time declined from tens of minutes to a few minutes with increasing Sn to Al weight ratio. The lesser delay time or shorter reaction duration was observed more strikingly in the higher Sn containing alloy. A simple mechanism was recommended for galvanic corrosion and H2 production in prepared Al-Sn alloys, as shown in Fig.12. through the corrosion development in the chemical etching of Al-Sn alloys because Sn is a better catalyst for H2 progression than Al [22]. Rendering to earlier studies [23], the metallic Sn particles covering the Al particles can act as cathodic centers to generate H2 via reaction (2), while Al acts as an anodic center for oxidation through reaction (1) by creating a galvanic cell. Reaction (3) indicates the complete reaction: This galvanic corrosion can make the hydrogen evolution on the surface of the Al-Sn alloys become larger than that of pure Al.

Microstructure of Al-Sn Alloys
The microstructure of the rapidly solidified Al-Sn alloys were examined. Figures(13.a) to (13.d) shows the microstructural analysis of the rapidly solidified Al-Sn alloys prepared by melting spinning technique, the grain distributions and average grain size were calculated using an SPM data visualization and analysis tool (Gwyddion 2.32), see table (2) . In our work we mainly focused on nonuniform distribution of Sn particles through the spinning process in a molten Al-Sn alloy. The Al-Sn alloys exhibited a network structure of a β-Sn second phase surrounding the Al grains.The Al phase nucleated and grew from the existing Al matrix, and left the Sn phase along the boundaries of the Al grains. In the SEM images, the white part represents the Sn phase, and the gray black part is the Al phase, with an increasing Sn content, the brighter Sn phase in grain boundaries became denser. Evidently, the Sn phase enclosed the granular Al matrix. In addition, the Al grain size decreased with an increasing Sn content.

Conclusion
Rapidly solidified Al100-x-Snx alloy X= (0, 25, 55, 75 all in wt.%) was prepared using melt spinning technique at 800 º C. the structure and microstructural analysis were characterized usind X-ray diffraction and rietveld analysis, which confirms the presense of two pases of pure Al with a well defined face centered cubic structure (FCC) of space group (F m -3 m) and β-Sn with a well defined tetragonal structure of space group (I 41/a m d).
From the Differential thermal analysis it is found that all the produce alloys are stable. Melting temperature reduced from 933.47 K for Al pure to 500.95 K in Al45Sn55.
Rapidly solidified Al100-x-Snx alloy, X= (0, 25, 55, 75 all in wt.%) were immersed in 2 M NaOH solution at 95 °C. The best volume of hydrogen generation which is (531 ml) is for Al55Sn45 alloy due to the good distribution of Sn along the Al grains at the surface and the resulting shorter path of electron between Sn and Al in galvanic corrosion.