1. Introduction

2. Materials and Methods

3. Results

3.1. Thermogravimetric Analysis

To identify decomposition of samples with increased temperature, heating rate was set to 10 K per minute. Thermogravimetric investigation shows that there are couple of decomposition processes happening in all investigated samples, which is expected as the dross is a complicated sample with many possible components. Most likely those points arise from hydroxide, carbonate and possibly from metallic Al melting. SOW has 5 points, RM also has 5 points whereas HDC has only 3 points. The initial mass loss (starting from point 1) is related to evaporation of moisture and water from the samples.

Figure 1. (a) SOW sample have 5 distinct changing points, (b) RM has 5 distinct point (c) on the other hand HDC only 3 distinct points.

Figure 1. (a) SOW sample have 5 distinct changing points, (b) RM has 5 distinct point (c) on the other hand HDC only 3 distinct points.

3.2. Elemental and Structural Analysis

3.2.1. SOW Sample

SEM and EA for sample SOW have been carried out in (Device)

The results are as follows;

As we can see in the Figure 2 the SOW sample consists of majority aluminium and alumina, but with substantial amount of impurities, mostly fluoride and sodium that comes from the material processing reactants in addition to Ca and Fe as regular additives to aluminium alloys. XRD analysis shows that crystalline structures are present from corundum to calcium fluoride, in addition to other components. It has amounts of nitride which changes the thermal properties of the material substantially and makes it hard to physically process. Even in our processing experience preparation of samples was very complicated as separation and grinding of samples took a lot of tools and time. Thus, especially this step material is very complicated to recycle in a conventional way.

Figure 2. SEM, EDS and XRD analysis of sample SOW, (a) EDS analysis of Al composition in one point (b) EDS measurement of Al composition in 2. point (c) Overall composition of elements in point 1 (d) overall EDS measured composition in point 2 (e) XRD diffractogram of sample (f) average value of atomic concentration percentage as measured by EDS.

Figure 2. SEM, EDS and XRD analysis of sample SOW, (a) EDS analysis of Al composition in one point (b) EDS measurement of Al composition in 2. point (c) Overall composition of elements in point 1 (d) overall EDS measured composition in point 2 (e) XRD diffractogram of sample (f) average value of atomic concentration percentage as measured by EDS.

3.2.1. RM Sample

SEM and EDS analysis of RM sample

Sample consists of Al, O, K, Mg, Ca, Na, F, V, Fe and Ti. The impurities are not so evenly distributed and the morphology differs a little bit as well depending on the measuring point. Although the impurities are a small part of the total concentration, some parts show a higher amount of fluorine. The Al content changes from 30% to 45 % and oxygen changes from 43 to 59%, other trace elements also change substantially. It is known that composition of samples will have substantial effect on the hydrogen production. On the other hand, crystalline structure shows lower number of components, such as Al nitride, oxide, and vanadium oxide. In comparison to SOW sample, the amount of crystalline components is lower, and it is mostly composed of oxides.

Figure 3. SEM, EDS and XRD analysis of sample RM. (a) SEM of the sample (b) ESD analysis with all element composition at once (c) Al composition of the sample (d) second overall view of all present element distribution (e) XRD of the sample (f) summary of average atomic concentration of the sample.

Figure 3. SEM, EDS and XRD analysis of sample RM. (a) SEM of the sample (b) ESD analysis with all element composition at once (c) Al composition of the sample (d) second overall view of all present element distribution (e) XRD of the sample (f) summary of average atomic concentration of the sample.

3.2.2. HDC Sample

The sample consists of following elements: Al, Mg, F, Si, Ca and O are evenly distributed across the sample area.

Figure 4. SEM, EDS and XRD analysis of sample HDC (a) ESD mapping of the sample (b) SEM image of the sample (c) ESD analysis with Al element composition (d) XRD of the sample (e) summary of average atomic concentration of the sample.

Figure 4. SEM, EDS and XRD analysis of sample HDC (a) ESD mapping of the sample (b) SEM image of the sample (c) ESD analysis with Al element composition (d) XRD of the sample (e) summary of average atomic concentration of the sample.

From EDS and SEM we see that HDC sample is somewhat homogeneous without visible crystalline structures, similarly that is visible in XRD results, as the crystalline forms Al metal, also Si, CaF₂ and MgO visible from found strong peaks in the diffractogram pattern.

From elemental analysis we see that depending on the processing stage of aluminium, dross has variations of compositions and structures, which in turn complicates recycling and reusing these materials. The simplified estimated energy consumption of 1t of dross is 454 kWh/t in oil and 72 kWh/t in electricity with yield of 0.45 t of metallic Al [14]. It is noteworthy that often the exact composition of used dross is not revealed, nonetheless it is quite complicated to compare results and composition as depending on the stage of sample collection, moreover various producers will most likely have variations of compositions that will depend on the individual processes, raw materials and treatments.

3.3. Hydrogen Production from Samples

The water aluminium reaction produces gases at a various rate, and as it turns out we have to see what gasses are being produced during the reaction with the electrolyte. Following are separation into each sample.

3.3.1. SOW Sample

SOW sample measuring the gas composition at the beginning of the reaction and at a later time, composition of the analyzed gases is visible in Table 1. We can see that 0.2% of methane was detected at the beginning of the reaction with, 97.4% hydrogen, 0.3% O₂, 1.2% N₂ and 0.1 % CO₂. At the end of the reaction there was still methane present 0.2%, water vapor 0.1% and 99.6% of hydrogen.

3.3.2. RM Sample

In the beginning of the experiment 98.3% of hydrogen was detected and 0.0% methane, in addition to N₂ 0.4%, O₂ 0.1%, and water 1.9%. Then the reaction was let to run till completion. From gas analysis one can see that methane is present in the first hour – however, it is impossible to find more than 0.09 wt% methane in the following reaction hours. Hydrogen composes the majority of the collected and analyzed gas as shown in the Table 2.

3.3.3. HDC Sample

H₂ production of HDC sample.

Pressure-temperature measurement shows steady production of gas and heat during sample - water reaction. Reaction starts with release of CO₂ and expected H₂ as seen in Table 3, then Sample was left in the reactor overnight to detect the maximum gas generation and allow full reaction with electrolyte. During the night Methane evolution was detected as well as nitrogen which decreases the gas ration, thus amount of H₂ decreases to 92.5%.

Methane, Ethane and CO₂ are the most present impurities from reaction, but CO₂ might be also introduced from the leaks of the flow reactor as N₂ is greatly introduced into the reactor. However, Methane is the strongest hazardous impurity of 0.4% of the gas. Gas was collected in 400 cc collection volume. This means that within 1st day only 1bar of H₂ was generated. The last MS spectra is after additional sample addition which might introduce air from vale opening – this also explains the increase in methane as much more “pristine” HDC surfaces are exposed to the electrolyte.

3.3.4. Comparison of Hydrogen Generation and Reaction Efficiency

O/N gas analysis was measured several times for each sample, average results are depicted in Table 4. The obtained results seem quite scattered due to inhomogeneous composition of the samples. Some particles are with high purity, on the other hand many parts of samples show high amount of oxygen and increase amount of nitrogen (as XRD patterns).

From the sample-water reaction, it was identified that AlN (as XRD detected) containing samples would result in the formation of ammonia. Thus, the application of PEM fuel cell is not viable for these types of materials as ammonia would degrade and poison the FC. Any contamination must be avoided if generated gas is to be used in a fuel cell. Ammonia (even at ppm level) is a real contaminant to the PEM hydrogen fuel cell [15]. 15 h exposure to 30 ppm NH₃ in the anode fuel caused a rapid drop in cell performance.

For initial experiments with 1M alkaline and elevated temperature (40 °C), the reaction was not as fast as expected. The slowest H₂ generation reaction was recorded from HDC sample, while RM showed the highest generation velocity/rate. However, we observed that some particles of SOW and RM samples did not react at all.

From the investigation of sample reaction efficiency, it seems that RM shows highest efficiency, even though the initial expectation is that this samples has roughly 40% Al, from composition we see that it is mostly aluminium oxide, XRD doesn’t show much of hydroxide. Other samples have impurities as Si or N, C which hinder the reaction. To increase the reactivity of samples and compare gas generation. After filing down the samples to comparable particle size the reaction efficiency is shown in the Figure 6a, where reaction efficiencies are depicted in Figure 6b and comparison of dross samples with Al waste from construction material in Figure 6c.

As the samples are pre-treated with the file, the particles are more like spheric as cylindric, plates or films. It means, in the first degree of monodisperse approximation, the geometry of particles is described only by the single radius. With respect to volumetric bodies, at the chemical reaction the shrinking of the solid reactant occurs, therefore the radius is time-dependent, r(t). The base of the shrinking core model for a spherical particle is the expression that links the molar aluminium reaction rate dn_Al/dt on the surface with diminishing the size of a particular aluminium particle:

$$4\pi r^2\left(t\right)\frac{d{n}_{Al}}{dt}=-\frac{d}{dt}\left(\frac{4\pi }{3}\frac{{\rho }_{Al}}{M_{Al}}{r}^3\left(t\right)\right),$$

(8)

where ρ_Al ans M_Al are density and molar mass of the solid (aluminium). Eq. (8) is correct for the first and more intensive, so-called surface reaction rate step, being analyzed in the present work. This equation is widely solved out in literature [16], where it is shown that at a sufficient stirring:

$$\frac{r\left(t\right)}{r_0}={\left(1-f\right)}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.},$$

(9)

where r₀ is the initial radius of the solid particle, and f is normalized non-dimensional hydrogen yield (0 < f < 1), where f=1 means the maximal theoretical yield, calculated by ideal (or real) gas equation of state. At the experimental case of gaining the pressure p of yielded hydrogen in the constant volume V:

$$f=\frac{2}{3}\frac{V}{R{T}_a}(p-p_0)\frac{W_{Al}}{M_{Al}}.$$

(10)

Here W_Al is reacted aluminium mass, p₀ and T_a are the reference pressure and temperature, respectively. The coefficient 2/3 comes from the stoichiometry in reaction between solid and liquid, eq. (6). Summarizing Hiraki et al. [16], the reaction rate constant at the surface reaction step, k_s, has to be derived from the equation:

$$\left\{1-{\left(1-f\right)}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}\right\}{r}_0=\frac{M_{Al}}{{\rho }_{Al}}{\bullet c}_{alk}\bullet t\bullet k_s$$

(11)

The coordinates in Figure 6b,c are chosen so that the slope of each curve at each point means the reaction rate constant k_s. Table 5 displays the best possible detected constant values of k_s before the surface reaction step is affected by the second – mass transfer step of the chemical reaction, [13]. We prefer to sign units on x, y axes as mm*, s* because of non-dimensional multipliers.

Figure 6. (a) Hydrogen reaction efficiency for non-uniform samples, (b) Reaction efficiency for uniform size samples.

Figure 6. (a) Hydrogen reaction efficiency for non-uniform samples, (b) Reaction efficiency for uniform size samples.

In the Table 5 we are comparing industrial white dross H₂ generated efficiency compared to pure Al reaction with water. We have reported efficiency close to 100% of aluminium waste from window frame scrap[17]. Here for calculations, we have used Al density as 2700 g⸳cm^-3, initial particle radius is 0.06 mm, Al molar mass 27 g⸳mol^-1 and 1M NaOH electrolyte. As we can see the investigated samples as HDC barely reaches over 12%, which is consistent with the reaction rate and sample composition.

4. Discussion

5. Conclusions

References

1. Supporting S, Resource M U.S. Geological Survey Mineral Resources Program— 2007,. Program.
2. Capuzzi S, Timelli G Preparation and melting of scrap in aluminum recycling: A review 2018,. Metals (Basel). 8:249.
3. Blomberg J, Söderholm P The economics of secondary aluminium supply: An econometric analysis based on European data 2009,. Resour Conserv Recycl Volume 53, pp. 455–463. [CrossRef]
4. John A. S. Green Aluminum Recycling and Processing for Energy Conservation and Sustainability 2007,.
5. Padamata SK, Yasinskiy A, Polyakov P A Review of Secondary Aluminum Production and Its Byproducts 2021,. JOM 73:2603–2614.
6. Bell S, Davis B, Javaid A, Essadiqi E Final Report on Refining Technologies of Aluminum Enhanced Recycling, Action Plan 2000 on Climate Change, Minerals and Metals Program-The Government of Canada Action Plan 2000 on Climate Change Minerals and Metals Program, managed by the Minerals and Meta 2003.
7. Smith YR, Nagel JR, Rajamani RK Eddy current separation for recovery of non-ferrous metallic particles: A comprehensive review 2019,. Miner. Eng. 133:149–159.
8. Schlesinger ME Aluminum Recycling 2006,. CRC Press.
9. Mesina MB, De Jong TPR, Dalmijn WL Improvements in separation of non-ferrous scrap metals using an electromagnetic sensor 2003,. Phys Sep Sci Eng Volume 12, pp. 87–101. [CrossRef]
10. Coates G, Rahimifard S Modelling of post-fragmentation waste stream processing within UK shredder facilities 2009,. Waste Manag Volume 29, pp. 44–53. [CrossRef]
11. Venkoba Rao B, Kapur PC, Konnur R Modeling the size-density partition surface of dense-medium separators 2003,. Int J Miner Process Volume 72, pp. 443–453. [CrossRef]
12. Hiraki T, Takeuchi M, Hisa M, Akiyama T Hydrogen production from waste aluminum at different temperatures, with LCA 2005,. Mater Trans Volume 46, pp. 1052–1057. 1057. [CrossRef]
13. Mezulis A, Richter C, Lesnicenoks P, et al. Studies on Water–Aluminum Scrap Reaction Kinetics in Two Steps and the Efficiency of Green Hydrogen Production 2023,. Energies Volume 16, pp. 5554. [CrossRef]
14. Ingason HT, Sigfusson TI Processing of Aluminum Dross: The Birth of a Closed Industrial Process 2014,. Jom Volume 66, pp. 2235–2242. [CrossRef]
15. Cheng X, Shi Z, Glass N, et al. A review of PEM hydrogen fuel cell contamination: Impacts, mechanisms, and mitigation 2007,. J. Power Sources 165:739–756.
16. Hiraki T, Takeuchi M, Hisa M, Akiyama T Hydrogen production from waste aluminum at different temperatures, with LCA 2005,. Mater Trans Volume 46, pp. 1052–1057. [CrossRef]
17. Urbonavicius M, Varnagiris S, Mezulis A, et al. Hydrogen from industrial aluminium scraps: Hydrolysis under various conditions, modelling of pH behaviour and analysis of reaction by-product 2024,. Int J Hydrogen Energy Volume 50, pp. 431–446. [CrossRef]
18. Hong JP, Wang J, Chen HY, et al. Process of aluminum dross recycling and life cycle assessment for Al-Si alloys and brown fused alumina 2010,. Trans Nonferrous Met Soc China (English Ed Volume 20, pp. 2155–2161. [CrossRef]
19. David E, Kopac J Hydrolysis of aluminum dross material to achieve zero hazardous waste 2012,. J Hazard Mater Volume 209–210, pp. 501–509. [CrossRef]
20. Olivares-Ramirez JM, de Jesus AM, Jimenez-Sandoval O, Pless RC Hydrogen Generation by Treatment of Aluminium Metal with Aqueous Solutions: Procedures and Uses 2012,. In: Hydrogen Energy - Challenges and Perspectives. IntechOpen.
21. Zhao Z, Chen X, Hao M Hydrogen generation by splitting water with Al-Ca alloy 2011,. Energy Volume 36, pp. 2782–2787. [CrossRef]
22. Küp Aylikci N, Mert SO, Aylikci V, et al. Microhydrogen production with water splitting from daily used waste aluminum 2021,. Int J Hydrogen Energy Volume 46, pp. 28912–28924. [CrossRef]
23. Varnagiris S, Urbonavicius M Hydrogen generation kinetics via hydrolysis of Mg2Ni and Mg2NiH4 powders 2021,. Int J Hydrogen Energy Volume 46, pp. 36323–36335. [CrossRef]
24. Alviani VN, Hirano N, Watanabe N, et al. Local initiative hydrogen production by utilization of aluminum waste materials and natural acidic hot-spring water 2021,. Appl Energy Volume 293, pp. 116909. [CrossRef]
25. Elsarrag E, Elhoweris A, Alhorr Y The production of hydrogen as an alternative energy carrier from aluminium waste 2017,. Energy Sustain Soc Volume 7, pp. 1–14. [CrossRef]
26. Raabe D, Ponge D, Uggowitzer P, et al. Making sustainable aluminum by recycling scrap: The science of “dirty” alloys 2022,. Prog Mater Sci Volume 128, pp. 100947. [CrossRef]
27. Dangtungee R, Vatcharakajon P, Techawinyutham L Aluminium dross neutralization and its application as plant fertilizer 2021,. In: Materials Today: Proceedings. Elsevier, pp 2420–2426.