Multi-Scale Minero-Chemical Analysis of Biomass Ashes: A Key to Evaluating Their Dangers vs. Beneﬁts

: A multi-methodic analysis was performed on ﬁve samples of ﬂy ashes coming from different biomasses. The aim of the study was to evaluate their possible re-use and their dangerousness to people and the environment. Optical granulometric analyses indicated that the average diameter of the studied ﬂy ashes was around 20 µ m, whereas only ~1 vol% had diameters lower that 2.5 µ m. The chemical composition, investigated with electron probe microanalysis, indicated that all the samples had a composition in which Ca was prevalent, followed by Si and Al. Large contents of K and P were observed in some samples, whereas the amount of potentially toxic elements was always below the Italian law thresholds. Polycyclic aromatic hydrocarbons were completely absent in all the samples coming from combustion plants, whereas they were present in the ﬂy ashes from the gasiﬁcation center. Quantitative mineralogical content, determined by Rietveld analysis of X-ray powder diffraction data, indicated that all the samples had high amorphous content, likely enriched in Ca, and several K and P minerals, such as sylvite and apatite. The results obtained from the chemo-mineralogical study performed make it possible to point out that biomass ﬂy ashes could be interesting materials (1) for amendments in clayey soils, as a substitution for lime, to stimulate pozzolanic reactions and improve their geotechnical properties, thus, on the one hand, avoiding the need to mine raw materials and, on the other hand, re-cycling waste; and (2) as agricultural fertilizers made by a new and ecological source of K and P. R.V,


Introduction
Fly ashes represent the amount of solid, inorganic residue left after the complete burning of biomasses. They are an integral part of plant materials and can have a wide range of elements. Due to the difference in raw materials as well as in the combustion plant type, the chemical physical characteristics of fly ashes can be quite different [1]. Moreover, the percentage of ashes produced during biomass burning varies according to the biomass type, ranging from a few units to about 10 dry weight percent [2,3].
Several papers [4][5][6] have shown that the principal chemical components of biomass ashes are silicon, aluminum, calcium, iron, magnesium, and sodium, together with valuable amounts of important plant nutrients, such as potassium and phosphorous.
The strong push towards the reuse of materials, together with the likely large increase of fly ashes from biomass over the last decades, suggest the need to take actions devoted to their re-use as second generation products [7]. Important examples include in X-ray powder diffraction analyses were performed on randomly oriented powder at room temperature using a Phillips PW-1830, equipped with a graphite monochromator on the outgoing rays, and using CuK α radiation (λ = 1.5406). The voltage and current intensity of the generator were 40 kV and 30 mA, the step size was 0.3 • , and the acquisition time for each step was 18 s. The diffraction patterns were collected in a Bragg-Brentano geometry (θ-2θ), in a range of 2θ between 5 • to 80 • . The Rietveld method [22], as implemented in GSAS-EXPGUI [23,24], was used for quantitative mineralogical analysis of the samples. A fraction of 10% by weight of crystalline silicon (Si) was added as external standard for the determination of the amorphous content. The background was fitted with a Chebyshev polynomial function and diffraction peaks were modeled with a pseudo-Voight function. Scale factor, lattice constants, and coefficients for the peak shapes were refined for each phase.

TGA and DTA
Thermogravimetric analyses (TGAs) were performed with a Netzsch STA 449F3. Approximately 50 to 100 mg of finely ground material was heated at a rate of 10 • C min −1 , under room atmosphere, from 20 • C up to 1000 • C. The Netzsch software was used to process the results at the Department of Chemistry of Perugia University, Italy.
Differential thermal analysis (DTA) is a thermal analysis technique in which the heat flow into or out of a sample is measured as a function of temperature or time while the sample is exposed to a controlled temperature program. It is a very powerful technique for the evaluation of material properties such as glass transition temperature, melting temperature, crystallization, specific heat capacity, cure process, purity, oxidation behavior, and thermal stability.

SEM Analyses
The surface morphology and the chemical composition were examined through field emission scanning electron microscopy (FE-SEM) using an LEO 1525 with a ZEISS AsB (angle-selective backscatter) detector (Oberkochen, Germany), available at the Department of Physics and Geology of the University of Perugia. The FE-SEM instrument was coupled with EDS microanalysis (BrukerQuantax EDS). Gold coating was applied for SEM observations.

Granulometric Analysis
Particle sizes greater than 3 microns were determined by using an Accusizer TM Optical Particle Sizer Model 770 (Santa Barbara, CA, USA), whereas, for particle sizes lower than 3 microns, the dynamic light scattering technique was employed by using a NICOMP 380 ZLS zeta potential/particle sizer (PSS NICOMP; Santa Barbara, CA, USA). Approximately 10 mg of each sample was dispersed in 1 mL of bi-distilled water and sonicated for 10 s to improve the dispersion of the single particles.

EMPA and LA-ICP-MS Analyses
The samples were also analyzed in terms of bulk composition. In order to obtain homogeneous pellets, the ashes were heat-treated up to and beyond their fusion points. The heating was performed by putting a few grams of each specimen into a crucible made out of pure Pt and using a muffle capable of heating to 1600 • C. Once the samples were completely molten, they were quickly quenched at room temperature. The glass pellets obtained after the quenching were embedded with epoxy resin; the mounts were cut, ground, and highly polished down to a 1 µm grade diamond polishing paste. Minor and trace elements were analysed at the Department of Physics and Geology (Perugia University) by LA-ICP-MS using the analytical protocol reported in Ref. [25].
Before being analyzed, each polished mount had to be coated using graphite. EMPA-WDS analyses were carried out at the Electron Microprobe Laboratory of the Earth Science Department "Ardito Desio" (University of Milan, Milan, Italy), using the EMPA-WDS JEOL 8200 Super Probe with 5 WDS channels and one EDS channel under the following operating conditions: accelerating voltage, 15 kV; beam current, 5 nA; counting time, 30 s for all the elements and 10 s for the backgrounds. Calibration for chemical analysis was accomplished with a set of synthetic and natural standards, including: grossular (Al, Si, Ca), omphacite (Na), olivine (Mg), apatite (P), scapolite (Cl), pure Cr, rhodonite (Mn), fayalite (Fe), K-feldspar (K), and ilmenite (Ti).

PAHs (Polycyclic Aromatic Hydrocarbons)
To evaluate the possibly dangerous character of fly ashes and to determine their possible employments, polycyclic aromatic hydrocarbons measurements were performed.
The samples were analyzed with gas chromatography (Agilent mod. 6890) coupled with mass spectrometry (5973N, Agilent Technologies), using reference standards.

Granulometric Results
The results of the granulometric analysis are reported in Figure 1. The average sizes were quite similar for the five samples and equal to 23.8, 17.0, 21.0, 21.0, and 18 µm, from sample 1 to 5.
On the other hand, more relevant differences can be seen from the analysis of the granulometric curves reported in Figure 1, where the red lines highlight the positions of the 2.5 and 10 µm diameter particles. These values represent important limits as they identify the inhalation properties and, thus, the health hazard with regard to their inhalation capacities.
All samples contained very low percentages of particles with dimensions under 2.5 µm (less than 1 vol%), whereas the contents in particles with diameters less than 10 µm was quite different, varying from 5 vol% in sample 5 to 40 vol% in sample 4.

X-ray Diffraction Results
The quantitative mineralogical composition, as well as the amount of amorphous components, are reported in Table 2, whereas Figure 2 shows the X-ray diffraction profiles collected and calculated with Rietveld analyses. In particular, in the graphs, the red crosses represent the data collected and the green lines represent the calculated best-fitting profiles, obtained by adding in the refinements of the minerals reported in Table 2. The theoretical peak positions of each mineral are reported under the profile, i.e., each horizontal line with different bars represents the theoretical peak position for that mineral. The lowest bar line reports the theoretical peak position of metallic Si, added to each sample as an internal standard for the quantitative analysis. The pink profile in the lower part of each graph represents the difference profile generated by subtracting the measured profile from the calculated one [24]. benzo(g,h,i)perilene, were determined.

Granulometric Results
The results of the granulometric analysis are reported in Figure 1. The average sizes were quite similar for the five samples and equal to 23.8, 17.0, 21.0, 21.0, and 18 µm, from sample 1 to 5. On the other hand, more relevant differences can be seen from the analysis of the granulometric curves reported in Figure 1, where the red lines highlight the positions of the 2.5 and 10 µm diameter particles. These values represent important limits as they identify the inhalation properties and, thus, the health hazard with regard to their inhalation capacities.
All samples contained very low percentages of particles with dimensions under 2.5 µm (less than 1 vol%), whereas the contents in particles with diameters less than 10 µm was quite different, varying from 5 vol% in sample 5 to 40 vol% in sample 4.

X-ray Diffraction Results
The quantitative mineralogical composition, as well as the amount of amorphous components, are reported in Table 2, whereas Figure 2 shows the X-ray diffraction profiles collected and calculated with Rietveld analyses. In particular, in the graphs, the red crosses represent the data collected and the green lines represent the calculated best-fitting profiles, obtained by adding in the refinements of the minerals reported in Table 2. The theoretical peak positions of each mineral are reported under the profile, i.e., each horizontal line with different bars represents the theoretical peak position for that mineral. The lowest bar line reports the theoretical peak position of metallic Si, added to each sample as an internal standard for the quantitative analysis. The pink profile in the lower part of each graph represents the difference profile generated by subtracting the measured profile from the calculated one [24].

Thermal Analysis Results
The thermal analysis of the five investigated samples showed relevant differences for the weight losses. The results are given in Figure 3.
The first steps in the TGAs, at approximately 100 • C, demonstrated weight losses between 1 and 18 wt%, which were observed in all the samples excluding sample 5. These corresponded to endothermic peaks in the DTA ( Figure 4) and can be associated with the loss of hygroscopic water, absorbed after the combustion of oxides [26]. In sample 1, at approximately 400 • C, there was a weight loss of about 50 wt%, corresponding to an exothermic peak in the DTA spectra and associated with the combustion of residual organic matter.
The step at approximately 600 • C could have been related either to the well-known carbonate decomposition reaction (CaCO 3 to CaO + CO 2 ) or to the decomposition of calcium-phosphates (Ca 3 (PO 4 ) 2 = CaO + P 2 O 5 or K 3 PO 4 = 3K 2 O + P 2 O 5 ) where the phosphorous pentoxide is lost as gas.
On the whole, the total mass losses between 20 and 1200 C varied from 13.38 in sample 2 to 70.84 in sample 1.

Thermal Analysis Results
The thermal analysis of the five investigated samples showed relevant di the weight losses. The results are given in Figure 3.
The first steps in the TGAs, at approximately 100 °C, demonstrated w between 1 and 18 wt%, which were observed in all the samples excluding sam corresponded to endothermic peaks in the DTA ( Figure 4) and can be associa loss of hygroscopic water, absorbed after the combustion of oxides [25]. In approximately 400 °C, there was a weight loss of about 50 wt%, correspondin thermic peak in the DTA spectra and associated with the combustion of resi matter.
The step at approximately 600 °C could have been related either to the carbonate decomposition reaction (CaCO3 to CaO + CO2) or to the decompo cium-phosphates (Ca3(PO4)2 = CaO + P2O5 or K3PO4 = 3K2O + P2O5) where the p pentoxide is lost as gas.
On the whole, the total mass losses between 20 and 1200 C varied from 1 ple 2 to 70.84 in sample 1.

Electron Microscopy Results
The results of the FE-SEM analysis coupled with EDS semi-quantitative chemical analysis are shown in Figures 5-8, with interesting textural features shown together with some of the accessory minerals also found by XRPD analysis. Apatite crystals appeared as roundly shaped at sub-micrometric to deci-micrometric sizes (apatite in samples 2 and 5, Figures 5 and 7). Moreover, elongated K2O and KCl crystals are shown in samples 4 and 5 (Figures 6 and 7).
Interesting de-vitrification microstructures were also observed in sample 3 ( Figure  8). In particular, windmill structures and neoblasts surrounded by glass are shown.

Electron Microscopy Results
The results of the FE-SEM analysis coupled with EDS semi-quantitative chemical analysis are shown in Figures 5-8, with interesting textural features shown together with some of the accessory minerals also found by XRPD analysis. Apatite crystals appeared as roundly shaped at sub-micrometric to deci-micrometric sizes (apatite in samples 2 and 5, Figures 5 and 7). Moreover, elongated K 2 O and KCl crystals are shown in samples 4 and 5 (Figures 6 and 7). Interesting de-vitrification microstructures were also observed in sample 3 ( Figure 8). In particular, windmill structures and neoblasts surrounded by glass are shown.

Chemical Composition Results
The chemical analysis of the major elements in the studied fly ashes ( These data fit well with the FE-SEM-EDS microanalysis, where several apatite microcrystals were found in sample 2 ( Figure 5) and an elongated crystal of K 2 O in sample 4 ( Figure 6).
Potentially toxic elements (PTEs), namely As, Cd, Cr, Cu, F, Pb, Hg, Mo, Ni, Se, V, and Zn, are naturally present in the environment [27] and some of them, in low percentages, can be essential for the correct metabolism of biologic activity. However, when their concentrations increase, they can become toxic. Their accumulation can be caused by anthropogenic activities, such as mining and industrial and agricultural activities [28], and is dependent on their mobility. For most of the PTEs, the mobility increases with the decrease in pH, very rapidly at a pH lesser than 7, whereas at higher pH conditions it slows down. Beyond the pH, other soil parameters, including the organic matter (OM), redox considerations, the chemistry of soil particles, and the composition of clay, affect the distribution, concentration, and mobility of PTEs. They are considered as hazardous contaminants due to their ability to intoxicate and become aggregated and persistent in the environmental media. For these reasons, several legislative decrees define the concentration limits for PTEs in urban and industrial areas.

Chemical Composition Results
The chemical analysis of the major elements in the studied fly ashes (Table 3) indicated that the main component for all of them was CaO, followed by SiO2 and then Al2O3 or, in one case, by MgO. The component P2O5 was dominant for sample 2, reaching approximately 9 wt%, whereas for sample 4 the concentration of K2O reached 5 wt%. Chemical analyses of PTEs are listed in Table 4 for samples 2, 3, 4, and 5. It was impossible to homogenize sample 1 as it burned during annealing even before melting. In Figure 9, the values of some selected PTEs (V, Cr, Co, Ni, Zn, Pb, Cu, As, Hg, and Se), together with Cl and S, are reported, as measured in the four analyzed samples of fly ashes. The ratios of the measured concentrations of PTEs and the limit concentrations of these same metals for urban and industrial sites, following the Italian legislation (legislative decree 52/2006 Tables 1 and 5, part V), are shown in Figures 10 and 11, respectively.
All samples had PTE concentrations below the legislative limit for industrial sites and only sample 2 had Cu and Zn, and sample 4 Cu, with values above the legislative limits.   The total amount of the PAHs naftalene, acenaftene, acenaftilene, fluorene, fenatrene, antracene, fluoratene, pirene, benzoantracene crysene, benzo(b,k,j) fluorantene, benzo(e)pirene, benzo(a)pirene, indenopirene, dibenzo(a,h)antracene, and benzo(g,h,i)perilene (Table 5) were below the detection limits for all the investigated samples, with the exception of sample 1, which came from a gasification plant. The ratios of the measured concentrations of PTEs and the limit concentrations of these same metals for urban and industrial sites, following the Italian legislation (legislative decree 52/2006 Tables 1 and 5, part V), are shown in Figures 10 and 11, respectively. The ratios of the measured concentrations of PTEs and the limit concentrations of these same metals for urban and industrial sites, following the Italian legislation (legislative decree 52/2006 Tables 1 and 5, part V), are shown in Figures 10 and 11, respectively.
All samples had PTE concentrations below the legislative limit for industrial sites and only sample 2 had Cu and Zn, and sample 4 Cu, with values above the legislative limits.   . Ratios between the measured concentrations of V, Cr, Co, Ni, Zn, Pb, Cu, As, Hg, and Se in the four samples ( Figure 9) and the values for the Italian limits of those metals for urban sites. Figure 11. Ratios between the measured concentrations of V, Cr, Co, Ni, Zn, Pb, Cu, As, Hg, and Se in the four samples ( Figure 9) and the values for the Italian limits of those metals for industrial sites.
The total amount of the PAHs naftalene, acenaftene, acenaftilene, fluorene, fenatrene, antracene, fluoratene, pirene, benzoantracene crysene, benzo(b,k,j) fluorantene, benzo(e)pirene, benzo(a)pirene, indenopirene, dibenzo(a,h)antracene, and benzo(g,h,i)perilene (Table 5) were below the detection limits for all the investigated samples, with the exception of sample 1, which came from a gasification plant. . Ratios between the measured concentrations of V, Cr, Co, Ni, Zn, Pb, Cu, As, Hg, and Se in the four samples ( Figure 9) and the values for the Italian limits of those metals for industrial sites.
All samples had PTE concentrations below the legislative limit for industrial sites and only sample 2 had Cu and Zn, and sample 4 Cu, with values above the legislative limits.

Discussion
The comparison of the multi-methodological analyses performed here on five samples coming from fly ashes of different vegetal biomasses allows us to outline the following considerations: 1.
The PTE concentrations were evaluated for the potential human and ecological risks, since their accumulation increases the toxic hazard [29,30]. In all of the samples of fly ashes from biomasses, the PTE concentrations analyzed were below the limits indicated in Italian legislation. Moreover, the PAH content was undetectable for all samples with the exception of sample 1, which was a unique sample coming from a gasification plant.

2.
Gasification is a process that converts either biomass [31] or fossil fuel-based carbonaceous materials into gases. The process consists of the reaction of the feedstock material, at high temperatures (typically > 700 • C) without combustion, via control of the amount of oxygen and/or steam present in the reaction. The resulting gas mixture is called syngas (from synthesis gas) or producer gas and its largest fraction is composed of nitrogen (N 2 ), carbon monoxide (CO), hydrogen (H 2 ), and carbon dioxide (CO 2 ). Due to the flammability properties of the H 2 and CO, syngas is itself a fuel. The thermal analysis performed in this work showed a strong weight loss, more than 50 wt%, at approximately 400 • C for sample 1. Indeed, during the process some material remained un-combusted, likely due to the large amount of PAH content in that sample.

3.
The granulometric analysis indicated that the PM2.5 fraction, namely the amount of particles with average dimensions below 2.5 µm, which can represent a danger to human health due to their potential of being inhaled, was significantly lower than 1%; the amount of particles with diameters less than 10 µm was quite different, varying from 5% in sample 5 to 40% in sample 4.

4.
The high phosphorous content, as an apatite mineral, as well potassium content, as a sylvite mineral, suggests that these materials may represent good amendments for agricultural lands once the bioavailability of these elements is proved and verified. The electron microscopy analysis showed that the crystals usually have very small dimensions, from the nanometer scale to a few microns. Moreover, the Rietveld analysis indicated that the amorphous content in all samples was very high, ranging from 47.6 wt% in sample 1 up to 84.7 wt% in sample 5. Both the above-mentioned characteristics are good indicators of a large reaction surface, one the most important parameters for bioavailability. 5.
The high Ca content revealed by the chemical analysis did not have a counterpart in the mineralogical composition of the samples, as relevant amounts of Ca minerals were not observed in the XRPD data processing. Thus, the high Ca content is supposedly stored in the amorphous phase, which was quite abundant in the studied fly ashes. 6.
The high Ca content measured in the samples suggests that these materials can be used as amendments in soils, usually clayey soils, with low geotechnical properties. In these soils, the traditional addition of CaO, coming from the decarbonation reaction of carbonate minerals, improves the mechanical properties as pozzolanic reactions induced by the highly alkaline environment promote the formation of new binding compounds such as calcium silicate hydrate minerals (C-A-H; C-S-H) [32][33][34]. In a high pH environment, natural pozzolanas, rich in silicon and amorphous phase, promote pozzolanic reactions as they increase the availability of silicon and alumina [35]. The large amount of amorphous content, evidenced by the Rietveld analysis, and the chemical compositions of the analyzed ash samples suggest that at least some of them could be successfully used in combination with or as substitutions for traditional binders in soil stabilization. Such use would require the environment to reach the alkalinity needed to promote pozzolanic reactions, with or without the addition of alkaline activators, which are normally used for soil treatment by means of fly ashes [35].

Conclusions
The addition of Ca from fly ashes in soil with poor geotechnical properties may represent a good example of end of waste. In fact, in this way, two waste materials, namely fly ashes and clayey soils, can be reconverted into a good second-generation product, thus avoiding the removal and transport of the soils and the opening of new caves for the extraction of carbonate, as well as reducing the CO 2 emissions from transports and de-carbonatation reactions.
The present study indicates that the dimensions and the chemical compositions of the studied fly ashes, namely the low concentrations of PTEs and PAHs (with the exception of the sample from the gasification plant), as well as the high CaO content in the amorphous fraction, represent favorable factors that can enhance interest in the applicability areas of these products. In fact, biomass fly ashes could be used as amendments in clayey soils to substitute for lime and stimulate pozzolanic reactions and improve their geotechnical properties. Further experiments to check the effect of the addition of fly ashes with high CaO content in clayey soils are in progress.
Moreover, the application of P fertilizer in intensive crop and animal production systems has increased pressure on finite global reserves of phosphate rocks [36], and their availability worldwide has recently drawn important criticisms due to both the high depletion of natural reserves and their geo-localization in limited areas.
Biomass fly ashes demonstrate high levels of phosphorous and potassium contents, which are present in nano/micro-crystals (as shown by electron microscopy) that biomasses have assimilated during their life cycle. The possibility that biomass fly ashes could be used as agricultural fertilizes opens an interesting perspective on a new and ecological source of K and P.
The above-mentioned scenarios fit well with the principles of the circular economy, in which waste materials, such as biomass fly ashes, could be recycled and used as secondgeneration products, thus avoiding the need to mine natural raw materials, such as carbonated and phosphate rocks.