Hydrothermal Alteration of Etna Ash and Implications for Mars

: Volcanic activity represents one of the main factors controlling the geological evolution of Mars, whose morphology has remarkable counterparts on Earth. Studies on the Martian surface revealed the widespread occurrences of newly formed minerals originated by the hydrothermal alteration of volcanic rocks. In this work, we carried out a series of experiments to test the reactions occurring during the hydrothermal alteration of basaltic ash from Etna (Italy) as a possible similar reaction fully grown on the Martian rock. The volcanic ash used for the hydrothermal alteration experiments was collected during the eruption of Etna in 2001, and its composition shares similarities with Martian bedrocks. Ash was altered under hydrothermal conditions at initial pH 5 at two temperatures (150 and 200 ◦ C) and reaction times of 5, 10, and 31 days. After a number of runs, we attained analcime NaAlSi 2 O 6 · H 2 O. Our ﬁndings are in line with the hypothesis that zeolite on Mars probably originated from a low-temperature hydrothermal environment. The conclusions accord with the assumption that the analcime crystals recognized on Mars formed under the same conditions as those of our experimental setups. TEM EDS SAED: morphology (at high magniﬁcations), chemical composition, crystallinity of individual particles; DSC DTG: thermal stability, identiﬁcation.


Introduction
Previous studies on Martian rocks showed that the surface of Mars is mainly constituted of basaltic products and that processes controlling the modification of the surface during the geological past (Noachian period) are primarily water-related [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. The alteration of the mineralogical composition of Mars rocks had been inferred from orbital observations, terrestrial analogs, local surface analyses by rovers' science payloads, and the study of Martian meteorites [16][17][18][19][20][21]. The discovery of hydrated minerals showed that hydrous reactions took place over the story of the planet [1][2][3] and that the mineralogical evolution of Martian's soil materials is controlled by the evolution of the weathering conditions. Chevrier and Mathe [22] report that the evolution of the Martian surface passed through three steps corresponding to the dominant weathering process acting in that period: (i) a clay-type weathering process characterized by a thick H 2 O/CO 2 -rich atmosphere (Noachian); (ii) a sulfate-type acidic weathering process when water became scarcer (Hesperian); (iii) very slow weathering by strongly oxidizing agents (H 2 O 2 , O 2 ) in cold and dry conditions (today). The secondary mineral assemblage in the most recently altered terrains (late Noachian-Hesperian) is dominated by sulfate and iron oxides, by contrast to the Earth which is characterized by carbonate, clays, and quartz sediments. The difference reflects the nature of the chemical factor promoting weathering: CO 2 for the Earth and likely SO 2 for Mars [23]. In older terrain, by contrast, clay minerals as well zeolites and ferric oxides have been identified in units dating from the Noachian period, when it is supposed that the basaltic rocks of the surface underwent to different alteration phenomena [1,[3][4][5]. In particular, hydrothermalism played a key role in the formation of zeolite outcrops of Mars [6][7][8][9][10][11][12][13][14][15]. Indeed, there is evidence of the presence of phyllosilicate and zeolite deposits on the planet [16][17][18], indicating the widespread aqueous alteration of the basaltic surface, some of which took place under hydrothermal conditions. Zeolite on Mars was discovered by OMEGA in Noachian terrains around Mawrth Vallis and Nili Fossae [18,24,25], Hellas basin [11,12], Eridania basin [13,15], Utopia Rupes [14], as well as in scattered outcrops in the southern highlands [1,25], often in association with impact craters [4,25]. Zeolites are hydrated aluminosilicate minerals made from interlinked tetrahedra of silica (SiO 4 ) andalumina (AlO 4 ). This structure is negatively charged and attracts the positive cations and water molecules sited inside the heart of the framework structure [26][27][28][29]. In the present study, we focused on the occurrence of analcime (zeolite), suspected in Noachian terrains. The analcime (zeolite) was identified by Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) in craters to the north of the Hellas basin and to the west of Nile Fossae near the Antoniadi Basin [18,25]. Basaltic rocks have been identified in this locality by the analyses of the MGS-TES (Mars Global Surveyor Thermal Emission Spectrometer) [30][31][32][33][34]. Together with analcime, CRISM detected also a mineral assemblage composed of chlorite, hydrated silica, and Fe/Mg smectite. Considering that such a mineral assemblage is a product of hydrothermalism on Earth [35,36], it suggests that the hydrothermal circulation of fluids could represents their major formation process. Analcime is a sodium-rich zeolite (NaAlSi 2 O 6 ·H 2 O) formed in hydrothermal conditions, as investigated on Massif Central in France and Icelandic basalts [37][38][39]. Analcime formations also occur in highly alkaline lake waters [40], as a result of weathering of tephra [41], and, infrequently, as a primary magmatic product [39,42,43]. During the past decade, experimental synthesis, aimed at the formation of zeolite on Mars, has received remarkable attention. In particular, basaltic glass was altered under temperatures higher than 150-200 • C, in the presence of CO 2 but allowing a pH > 7, and it was demonstrated that the formation of specific zeolites depends on the composition of the glass as well as on that of the hydrothermal solution [44][45][46][47][48].
In the present study, we focused on the occurrence of zeolites, suspected in Noachian terrains. We carried out a series of laboratory experiments to test the hydrothermal conditions able to reproduce the formation of analcime from terrestrial starting material, in order to infer the possible mechanisms and conditions at the origin of the zeolite minerals. We used as starting material ash from the 2001 Etna eruption [49,50], which shares similarities with Martian rocks. In particular, Etna ash falls in a field of basalts similar to Mars basalts from Gusev Crater and to some samples of Martian meteorites and Spirit Rover [51][52][53] (see also Section 3.1). In this context, and in the light of the potential implications for Mars, this work intends to evaluate how the experimental conditions can affect the growth of analcime in terms of morphology, size, and thermal stability. In nature, the observation of hydrothermal alterations shows the importance of the following factors: temperature, pH of the solutions, chemical composition, nature (crystalline or amorphous) of the starting material, and condition of the system (closed or opened).
The Etna ash was altered under hydrothermal conditions at initial pH = 5 at two different temperatures (T: 150 and 200 • C) with run time (t) ranging between 5 and 31 days. These experimental conditions were adopted taking into account that: (i) relatively low-temperature hydrothermalism (<350 • C) is assumed to have played a key role in the formation of the zeolite outcrops of Mars [5]; (ii) the interaction of water and volcanic glass at low temperatures (usually <250 • C) readily yields zeolite minerals [54]; (iii) subcritical solvothermal synthesis (usually operating from 150 to 250 • C) has been well demonstrated by the successful synthesis of zeolites [6]; (iv) laboratory and theoretical studies demonstrate that acidic aqueous solutions played an important role in the dissolution of basaltic ashes on Mars [22,23,[55][56][57]. The water reactant is assumed to be moderately acid (pH = 5), since an acidic environment during the early Noachian Period is presumed [7,22]. It derived from the oxidation and dissolution of volcanically produced acid gas in pure water [57]. An additional aspect to be taken into account is that the atmosphere's composition in which weathering develops has an influence on the nature of the secondary phases. Indeed, the atmosphere can affect the pH of the hydrosphere. Previous studies have shown that the chemical composition of the Martian atmosphere, which is rich in CO 2 (95%), at adequate pressure influences the pH of solutions at equilibrium [58]. Thus, CO 2 /SO 2 -rich atmospheres were equilibrated with acid solutions.

Analysis of Ash and Mineral Phases
The starting material (ash) was first observed under an optical microscope (OM) (20x, 116 ZEISS, Thornwood, NY, USA) in order to characterize the components of the ash and their morphology. Successively, X-ray fluorescence (XRF) and X-ray powder diffraction (XRPD) analyses were made to determine the main mineral phases and the whole composition of major and minor elements of the ash. Transmission and scanning electron microscopy (TEM and SEM) coupled with EDS electron microprobes for chemical analysis, derivative thermogravimetry (DTG), and differential scanning calorimetry (DSC) analyses were conducted for a more detailed definition of the morphology, size, and chemical composition of the individual phases. XRPD patterns were obtained by a Bruker D8Advance (Bruker, Billerica, MA, USA) X-ray diffractometer with CuKa radiation, monochromated with a graphite sample, operating at 40 mA and 40 kV. Scans were collected in the range of 3-66 • (2θ), with a step-counting time of 3 s and a step interval of 0.02 • . DIFFRACplus EVA software (version 11.0., Bruker, Karlsruhe, Germany) was used to identify the mineral phases. An assessment of the mineralogical and amorphous phases present in the ash was attain by Rietveld refinements [59,60], tooled using TOPAS software V.4.2 (Bruker, Karlsruhe, Germany). XRF was used to determine the chemical composition of the ash, in terms of the major and minor elements (SiO 2 , Na 2 O, K 2 O, Fe 2 O 3 , MnO, Al 2 O 3 , MgO, CaO, P 2 O 5 , and TiO 2 ), by using a Bruker S8 Tiger WD X-ray fluorescence spectrometer (Bruker, Karlsruhe, Germany), with a rhodium tube (XRF beam of 34 mm and intensity 4 kW). The analysis was carried out using 6 g of specimen, on pressed powder pellets placed over boric acid (maximum working pressure 25 bar). Secondary electron SEM imaging was carried out using the Field-Emission Scanning Electron Microscope FEI QUANTA 200 (Thermo Fisher Scientific, 120 Waltham, MA, USA), equipped with an X-ray EDS suite comprising a Si/Li crystal detector, model EDAX-GENESIS 4000 (EDAX Inc., Mahwah, NJ, USA). Detector constants and work conditions were as follows: tilt angle 0 • and voltage 20 kV. Thermal analysis was carried out by Netzsch STA 449 C Jupiter (DSC/TGA: Netzsch STA 449 C Jupiter, Netzsch-Gerätebau GmbH, Selb, Germany). Heating was carried out in alumina crucibles at a heating rate of 10 • C/min under an air flow of 30 cm 3 ·min −1 . Samples were ground in an agate mortar, and about 10 mg was heated to a temperature in the range of 25-600 • C. Thermal analysis accuracy was checked by six repeated collections using a kaolinite reference samples, revealing good theoretical weight sensitivity of 0.10 µg, DSC detection limit <1 µW, and reproducibility (instrumental T precision of ± 1.2 • C). Transmission electron microscopy (TEM) analysis was performed on a JEOL JEM 1400 Plus (120 kV) (Jeol, Tokyo, Japan), equipped with a Jeol large-area silicon drift detector SDD-EDS (Jeol, Tokyo, Japan) and a double tilt holder to obtain structural data using selected-area electron diffraction (SAED). Powdered samples were deposited on a copper mesh grid coated with 200 A carbon film. Contribution of each analysis carried out on the starting material and the experimental run products: (i) XRPD: identification, crystallinity, semi-quantitative abundance; (ii) SEM/EDS: morphology, chemical composition, identification of individual particles; (iii) TEM/EDS/SAED: morphology (at high magnifications), chemical composition, crystallinity of individual particles; (iv) DSC/DTG: thermal stability, identification.

Hydrothermal Alteration of Ashes
For the present study, hydrothermal alteration experiments were performed using the solvothermal method. This method has been already widely employed for the synthesis of zeolites [61]. Ash fragments were crushed using a Bleuler Rotary Mill (Sepor, Los Angeles, CA, USA) at a speed of 900 revolutions per min (rpm) for 3 s. In all experiments, 0.70 g of powder and 2-4 mL of H 2 O (HCl was added to obtain pH = 5) were added into a Teflon-lined stainless-steel autoclave of 25 mL capacity. The hydrothermal treatments of the ashes were carried out in a polytetrafluoroethylene (PTFE) lined autoclave, at two different temperatures, 150 • C and 200 • C, with an autogenous pressure [62,63] and reaction times of 5, 10, and 31 days; after heating, the autoclave was cooled to room temperature naturally. Subcritical solvothermal synthesis has been well demonstrated by the successful synthesis of zeolites. It is worth mentioning that hydrothermal processes correspond to the alteration of the host rock by aqueous fluids (>100 • C), generally induced by volcanic activity. Subcritical solvothermal synthesis utilizes temperatures below the critical point of water (374 • C) [63,64].

Ashes Characterization
The volcanic ash used for the hydrothermal alteration experiments was collected during the eruption of Etna in 2001 [49,50]. In particular, ash samples were collected in between 18 July and 7 August, when explosive activity occurred in the form of fire fountaining, strombolian explosions, hydromagmatic explosions, and pulsing ash explosions [49]. It is worth remembering that the ashes were not altered by atmospheric agents or anything else, before being hydrothermally altered in the laboratory. Under the OM (Figure 1), the ash appeared to consist of single crystals, tachylyte, and sideromelane glass fragments. successful synthesis of zeolites. It is worth mentioning that hydrothermal processes correspond to the alteration of the host rock by aqueous fluids (>100 °C), generally induced by volcanic activity. Subcritical solvothermal synthesis utilizes temperatures below the critical point of water (374 °C) [63,64].

Ashes Characterization
The volcanic ash used for the hydrothermal alteration experiments was collected during the eruption of Etna in 2001 [49,50]. In particular, ash samples were collected in between 18 July and 7 August, when explosive activity occurred in the form of fire fountaining, strombolian explosions, hydromagmatic explosions, and pulsing ash explosions [49]. It is worth remembering that the ashes were not altered by atmospheric agents or anything else, before being hydrothermally altered in the laboratory. Under the OM (Figure 1), the ash appeared to consist of single crystals, tachylyte, and sideromelane glass fragments.   successful synthesis of zeolites. It is worth mentioning that hydrothermal processes correspond to the alteration of the host rock by aqueous fluids (>100 °C), generally induced by volcanic activity. Subcritical solvothermal synthesis utilizes temperatures below the critical point of water (374 °C) [63,64].

Ashes Characterization
The volcanic ash used for the hydrothermal alteration experiments was collected during the eruption of Etna in 2001 [49,50]. In particular, ash samples were collected in between 18 July and 7 August, when explosive activity occurred in the form of fire fountaining, strombolian explosions, hydromagmatic explosions, and pulsing ash explosions [49]. It is worth remembering that the ashes were not altered by atmospheric agents or anything else, before being hydrothermally altered in the laboratory. Under the OM (Figure 1), the ash appeared to consist of single crystals, tachylyte, and sideromelane glass fragments.   XRPD pointed out that the volcanic ash consists of plagioclase (44%), augite (39%), amorphous glass (10%), and olivine (7%), in a decreasing order of abundance ( Figure 3). XRF analyses revealed that the ash is basaltic, with 46.7 wt.% of SiO 2 and 4.66 wt.% of Na 2 O + K 2 O (Table 1).
Minerals 2020, 10, x FOR PEER REVIEW 5 of 15 XRPD pointed out that the volcanic ash consists of plagioclase (44%), augite (39%), amorphous glass (10%), and olivine (7%), in a decreasing order of abundance ( Figure 3). XRF analyses revealed that the ash is basaltic, with 46.7 wt.% of SiO2 and 4.66 wt.% of Na2O + K2O (Table 1).  The chemical analysis of Etna ash is plotted on the binary TAS Alkali vs. Silica (Total Alkali vs. Silica) diagram; [65], which is a typical presentation for the classification of Earth igneous rocks [66]. In the same diagram (Figure 4), we plotted the chemistries of several Martian rocks acquired from orbital data and in situ [51,52]. In particular, we displayed the composition of the "Irvine" specimen, a mildly alkaline basalt discovered on the flanks of Columbia Hills (Gusev crater, Mars) obtained by the Alpha Particle X-ray Spectrometer (APXS; [53]) as well as an average of the chemical igneous composition of some samples of the Gale Crater Mars [67] obtained by the ChemCam Laser-Induced Breakdown Spectrometer (LIBS) in the MSL Curiosity rover. Moreover, the composition of the Martian meteorite Northwest Africa (NWA 6234) analyzed using the CAMECA SX-100 electron probe micro-analyzer, consistent with the composition of previously studied Martian meteorite basalts [68], was also reported in the TAS diagram ( Figure 4). Etna ash (green square) fell in the field of basalts close to the Mars basalts from Gusev Crater (red triangle) and to some samples of Martian meteorites (grey full circles) and Spirit Rover (black full circles) ( Figure 4). In this regard, basalt was chosen to depict the igneous rock type thought to cover some part of the Martian bedrock as ascertained from in situ XRF measurements of soils at the Viking landing sites [69] and from analysis of remotely sensed thermal emission data from Mars [30,70,71].  The chemical analysis of Etna ash is plotted on the binary TAS Alkali vs. Silica (Total Alkali vs. Silica) diagram; [65], which is a typical presentation for the classification of Earth igneous rocks [66]. In the same diagram (Figure 4), we plotted the chemistries of several Martian rocks acquired from orbital data and in situ [51,52]. In particular, we displayed the composition of the "Irvine" specimen, a mildly alkaline basalt discovered on the flanks of Columbia Hills (Gusev crater, Mars) obtained by the Alpha Particle X-ray Spectrometer (APXS; [53]) as well as an average of the chemical igneous composition of some samples of the Gale Crater Mars [67] obtained by the ChemCam Laser-Induced Breakdown Spectrometer (LIBS) in the MSL Curiosity rover. Moreover, the composition of the Martian meteorite Northwest Africa (NWA 6234) analyzed using the CAMECA SX-100 electron probe micro-analyzer, consistent with the composition of previously studied Martian meteorite basalts [68], was also reported in the TAS diagram ( Figure 4). Etna ash (green square) fell in the field of basalts close to the Mars basalts from Gusev Crater (red triangle) and to some samples of Martian meteorites (grey full circles) and Spirit Rover (black full circles) (Figure 4). In this regard, basalt was chosen to depict the igneous rock type thought to cover some part of the Martian bedrock as ascertained from in situ XRF measurements of soils at the Viking landing sites [69] and from analysis of remotely sensed thermal emission data from Mars [30,70,71].

XRPD
XRPD results showed that analcime was obtained during four runs, and a variable quantity of unreacted ash was also detected in all runs, including plagioclase NaAlSi3O8-Ca(Al2Si2)O8, pyroxene (Ca,Na)(Mg,Fe,Al)(Si,Al)2O6, and olivine (Mg,Fe)2SiO4 ( Figure 5). Ash did not react at 150 °C either after 5 or 10 days of hydrothermal alteration. Analcime NaAlSi2O6·H2O started forming at 200 °C after 5 days of alterations (run Et3), when very tiny amounts of analcime crystals were obtained.

XRPD
XRPD results showed that analcime was obtained during four runs, and a variable quantity of unreacted ash was also detected in all runs, including plagioclase NaAlSi3O8-Ca(Al2Si2)O8, pyroxene (Ca,Na)(Mg,Fe,Al)(Si,Al)2O6, and olivine (Mg,Fe)2SiO4 ( Figure 5). Ash did not react at 150 °C either after 5 or 10 days of hydrothermal alteration. Analcime NaAlSi2O6·H2O started forming at 200 °C after 5 days of alterations (run Et3), when very tiny amounts of analcime crystals were obtained.

XRPD
XRPD results showed that analcime was obtained during four runs, and a variable quantity of unreacted ash was also detected in all runs, including plagioclase NaAlSi 3 O 8 -Ca(Al 2 Si 2 )O 8 , pyroxene (Ca,Na)(Mg,Fe,Al)(Si,Al) 2 O 6 , and olivine (Mg,Fe) 2 SiO 4 ( Figure 5). Ash did not react at 150 • C either after 5 or 10 days of hydrothermal alteration. Analcime NaAlSi 2 O 6 ·H 2 O started forming at 200 • C after 5 days of alterations (run Et3), when very tiny amounts of analcime crystals were obtained.

XRPD
XRPD results showed that analcime was obtained during four runs, and a variable quantity of unreacted ash was also detected in all runs, including plagioclase NaAlSi3O8-Ca(Al2Si2)O8, pyroxene (Ca,Na)(Mg,Fe,Al)(Si,Al)2O6, and olivine (Mg,Fe)2SiO4 ( Figure 5). Ash did not react at 150 °C either after 5 or 10 days of hydrothermal alteration. Analcime NaAlSi2O6·H2O started forming at 200 °C after 5 days of alterations (run Et3), when very tiny amounts of analcime crystals were obtained.  At 200 • C, the increase of the reaction time from 5 to 10 days (run Et4) led to an increase in analcime production (Table 2). Under the same experimental conditions (T, water-to-rock ratio (W/R)), an increase in reaction time from 10 (run Et4) to 31 days (run Et5) enhanced the reactions towards a higher amount of analcime (Table 2). This result was inferred by the intensity (400) of analcime reflection at 8.060 (2θ) ( Table 2); indeed, the reflection obtained from analcime synthesized in a shorter reaction (runs Et3, Et4) was lower in intensity, while a higher intensity was observed when analcime was formed in a longer reaction (run Et5). Table 2. Experimental setup and list of the products obtained for each run in order of decreasing abundance, as detected by XRPD, TEM/EDS, and SEM/EDS. Full width at half-maximum (FWHM). P = plagioclase; Py = pyroxene; O = olivine; A = analcime. Et6 ash was altered with the addition of water from 2 mL to 4 mL. Water with initial pH = 5; water-to-rock ratio (W/R). In run Et5, an increase in water from 2 mL to 4 mL at 200 • C (run Et6) improved the analcime yield. The results also showed that the sharpness of the main peak (400), which depicts the degree of crystallinity of analcime, increased with increasing reaction time and W/R. In particular, rising the reaction time from 5 to 31 days, under the same temperature of 200 • C (Et3-Et5), caused a decrease of the FWHM value from 0.154 • to 0.147 • ( Table 2). A similar trend occurred when water was increased from 2 to 4 mL (Table 2), at 200 • C. Anyway, a complete reaction was not achieved, since unreacted starting materials (plagioclase, pyroxene, and olivine) were present in all runs. After the hydrothermal reaction, all the diagnostic reflections of the plagioclase, pyroxene and olivine were preserved ( Figure 5).

SEM/TEM/EDS
In low-magnification images, the ash appeared to consist of different phases, both crystalline and amorphous, as verified by SAED, in agreement with the results of XRPD. As regards morphology, volcanic amorphous ash fragments showed low contrast in TEM images and a ring-shaped SAED pattern (Figure 6a). The amorphous ash particles were detected between the edges of the crystals (Figure 6a). The chemical composition of the glass is consistent with the elements being present in the stoichiometric composition of analcime (Si, Al, Na), in addition to the presence of low amounts of Fe and Ca. The most evident and abundant crystals corresponded to plagioclase, pyroxene, and olivine, characterized by well-defined edges and crystal size typically greater than 500 nm (Figure 6a). At 150 • C, the altered ash looked very similar to the untreated one, with apparently well-preserved morphology and size of the ash particles. However, at higher magnification, TEM investigations revealed that at 150 • C (run Et1, Et2), tiny variations on ash morphology occurred after both 5 and 10 days of alteration, and as a matter of fact, the ash contour started to be less defined and seemed to peel off in many layers (Figure 6b), although under SEM, no edge modifications could be detected in the runs Et1 and Et2. By increasing the time of hydrothermal synthesis, the alteration tended to become more evident under TEM. Ash contour began to completely fade after 5 days of alteration (run Et3), and the thickness of single fragments increasingly reduced, with the simultaneous appearance of spherical analcime crystals (Figure 6c). Under the same synthesis conditions (T, W/R), an increase in reaction time from 5 (run Et4) to 10 days (run Et4) improved the crystals size (Figure 6d). At 200 °C and reaction time of 31 days (run Et5), analcime crystallized into well-defined crystals (Figure 6e). TEM observations of Et3, Et4, and Et5 products testified a more intense alteration of the ash particles, whose dimensions and width tended to be progressively reduced. For the last run products (run Et6), the alteration engaged not only the amorphous ash contour but also the inner part of amorphous ash fragments. Indeed, a further increase in W/R (run Et6), led to an increase in analcime crystallinity, and newly formed euhedral analcime crystals were visible both under TEM and SEM (Figures 6f and 7). EDS/TEM qualitative chemical analysis showed that the composition of several analcime crystals from all runs consisted of Na, Al, O, and Si (Figure 6f). This result is in accordance with the idealized chemical formula of analcime: NaAlSi2O6·H2O, although very a low amount of Ca was observed consisting of impurities derived from the starting materials (Figure 6f), as often also observed in natural samples [72]. The growth of analcime minerals on the top of the ash surface was further verified by SEM/EDS investigations (Figure 7a-c). Analcime assumed a cubic morphology [73,74], with size ranging from a few nm (Figure 7d,e) to some micrometer in width (Figure 7a-c). However, the SEM images showed that the analcime crystals were not equal in size and that unreacted phases (plagioclase, pyroxene, and olivine) were still present. The size of single crystals sometimes reached a maximum length of about 1.5 µm. As regards EDS/SEM chemical analysis, analcime was present in small crystals, often less than 1 µm in size, finely associated/covered with the starting materials (Figure 7). This prevented their pure EDS chemical characterization, leading to mixed analyses (analcime and unreacted phases). Another aspect to consider is that EDS micro-analysis is semiquantitative, and the synthesized crystals should be polished in order to obtain an accurate evaluation; however, as shown in Figure 7, the crystals obtained in this study were too small to be polished. Ash contour began to completely fade after 5 days of alteration (run Et3), and the thickness of single fragments increasingly reduced, with the simultaneous appearance of spherical analcime crystals (Figure 6c). Under the same synthesis conditions (T, W/R), an increase in reaction time from 5 (run Et4) to 10 days (run Et4) improved the crystals size (Figure 6d). At 200 • C and reaction time of 31 days (run Et5), analcime crystallized into well-defined crystals (Figure 6e). TEM observations of Et3, Et4, and Et5 products testified a more intense alteration of the ash particles, whose dimensions and width tended to be progressively reduced. For the last run products (run Et6), the alteration engaged not only the amorphous ash contour but also the inner part of amorphous ash fragments. Indeed, a further increase in W/R (run Et6), led to an increase in analcime crystallinity, and newly formed euhedral analcime crystals were visible both under TEM and SEM (Figures 6f and 7). EDS/TEM qualitative chemical analysis showed that the composition of several analcime crystals from all runs consisted of Na, Al, O, and Si (Figure 6f). This result is in accordance with the idealized chemical formula of analcime: NaAlSi 2 O 6 ·H 2 O, although very a low amount of Ca was observed consisting of impurities derived from the starting materials (Figure 6f), as often also observed in natural samples [72]. The growth of analcime minerals on the top of the ash surface was further verified by SEM/EDS investigations (Figure 7a-c). Analcime assumed a cubic morphology [73,74], with size ranging from a few nm (Figure 7d,e) to some micrometer in width (Figure 7a-c). However, the SEM images showed that the analcime crystals were not equal in size and that unreacted phases (plagioclase, pyroxene, and olivine) were still present. The size of single crystals sometimes reached a maximum length of about 1.5 µm. As regards EDS/SEM chemical analysis, analcime was present in small crystals, often less than 1 µm in size, finely associated/covered with the starting materials ( Figure 7). This prevented their pure EDS chemical characterization, leading to mixed analyses (analcime and unreacted phases). Another aspect to consider is that EDS micro-analysis is semi-quantitative, and the synthesized crystals should be polished in order to obtain an accurate evaluation; however, as shown in Figure 7, the crystals obtained in this study were too small to be polished.

DSC/DTG Characterization
Analcime obtained in run Et6 was characterized by DSC/DTG to evaluate its thermal stability. Et6 was chosen being the run producing the highest amount of analcime.
The DSC curve of analcime exhibited one main endothermic peak at about 399 °C (Figure 8), linked to analcime breakdown, in agreement with literature data [75,76]. The endothermic peak can be related to the compensating cation-water binding energies. Analcime dehydrates between 350 °C and 450 °C, as shown in the DTG curve at about 399 °C. The DTG peak occurred as a consequence of the transfer of H2O molecules through the six-member rings during dehydration [28,75]. The DSC peaks below 110 °C are related to the liberation of H2O adsorbed on the samples' surface. Other phases (unreacted materials, i.e., pyroxene, plagioclase, and olivine) did not contribute to the thermal reaction, due to their high thermal stability [77].

DSC/DTG Characterization
Analcime obtained in run Et6 was characterized by DSC/DTG to evaluate its thermal stability. Et6 was chosen being the run producing the highest amount of analcime.
The DSC curve of analcime exhibited one main endothermic peak at about 399 • C (Figure 8), linked to analcime breakdown, in agreement with literature data [75,76]. The endothermic peak can be related to the compensating cation-water binding energies. Analcime dehydrates between 350 • C and 450 • C, as shown in the DTG curve at about 399 • C. The DTG peak occurred as a consequence of the transfer of H 2 O molecules through the six-member rings during dehydration [28,75]. The DSC peaks below 110 • C are related to the liberation of H 2 O adsorbed on the samples' surface. Other phases (unreacted materials, i.e., pyroxene, plagioclase, and olivine) did not contribute to the thermal reaction, due to their high thermal stability [77].

DSC/DTG Characterization
Analcime obtained in run Et6 was characterized by DSC/DTG to evaluate its thermal stability. Et6 was chosen being the run producing the highest amount of analcime.
The DSC curve of analcime exhibited one main endothermic peak at about 399 °C (Figure 8), linked to analcime breakdown, in agreement with literature data [75,76]. The endothermic peak can be related to the compensating cation-water binding energies. Analcime dehydrates between 350 °C and 450 °C, as shown in the DTG curve at about 399 °C. The DTG peak occurred as a consequence of the transfer of H2O molecules through the six-member rings during dehydration [28,75]. The DSC peaks below 110 °C are related to the liberation of H2O adsorbed on the samples' surface. Other phases (unreacted materials, i.e., pyroxene, plagioclase, and olivine) did not contribute to the thermal reaction, due to their high thermal stability [77].

Discussion and Conclusions
In this work, we provide a set of hydrothermal alteration experiments performed on basaltic ash presumably similar to Martian rocks, in order to mimic the physical condition for similar reactions on Mars. After a few runs, we obtained analcime. The results showed that the yield of analcime depended mainly on the temperature and the nature of the starting materials (i.e., amorphous and crystalline). In fact, as known, amorphous materials are more reactive than crystalline ones [77]. In this work, the amorphous material consisted of volcanic glasses of tachylite and sideromelane, which are considered products of relatively rapidly cooled magma, whose groundmass varies from a glassy and sub-aphyric texture to a micro-and/or crypto-crystalline one [49,78]. Ash pyroclasts, like tachylite and sideromelane, are subjected to physical and chemical alteration during syn-eruptive and post-eruptive processes. Our results revealed that analcime formed to the detriment of the amorphous material present in the ash, while silicate (plagioclase, pyroxene, and olivine) already occurring in the ash, did not take part in the hydrothermal reaction. As a matter of fact, during reactions at low T (150 • C), with reaction times corresponding to 5 and 10 days of alteration, the formation of analcime was inhibited, while at 200 • C analcime started to form.
As reported [79] for similar silicate syntheses, the increment of temperature induced higher SiO 2 reactivity, thus improving the reaction between the starting elements. Moreover, Berger et al. [80] showed that at a temperature of 475 • C, a supercritical CO 2 -H 2 O fluid led the recrystallization of a basalt glass in few days. Therefore, the solubility of silicate glass was confirmed to increase as the temperature rose from 150-200 • C to 475 • C, under hydrothermal conditions. These findings demonstrate how the chemical reactivity of usually insoluble starting materials can be hugely increased by increasing the temperature. The absence of a kinetic model for our experiment makes speculative the extrapolation of prediction of glass alteration over geological time [80,81]. However, it is not excluded that the low-temperature (150 • C) alteration of glass after hundreds of years could produce analcime as it occurs at 200 • C. The widespread occurrence of clay minerals discovered in Mars's Noachian terrains, indicating a long-duration interaction between water and rock [7], could support this hypothesis. The outcome confirmed that both fluid system (W/R > 1) and temperature (or reaction time) are essential for reaching a thorough amorphous glass alteration. The initial acidic pH likely promoted alkali leaching from the basaltic glass, and so the formation of secondary analcime, at least at high temperature. The final pH of the solution was not measured but is supposed to be alkaline, considering the low W/R ratio used in our experiments and as reported in previous literature on experimental studies on basalt glass alteration [81][82][83]. For example, Berger et al. [81] demonstrated that basalt glass alteration reached the steady state in solutions at pH 9.5, at which silica solubility increased, and the solution became enriched in both silica and alkali, promoting the precipitation of zeolite. In this context, we hypothesize that the leaching process preceding the glass network dissolution, promoted analcime formation.
Overall, in this work we point out that the highest yield of analcime occurred at 200 • C, with a reaction time of 31 days. Therefore, the experimental results are consistent with a low-temperature hydrothermal formation of analcime on Mars. These temperatures were very similar to those reported by Saetre et al. [44] for the synthesis of Ca-zeolite and lower amounts of analcime, wh used basaltic glass as a starting material.
The possible mechanisms by which analcime formed, such as circulations of fluids triggered by impact or by lava cooling, acid water/vapor, H 2 O/CO 2 -rich atmosphere, are currently absent on Mars. Today in fact, in dry atmosphere, the alteration is tenuous. As expected from its low dielectric constant, CO 2 does not interact with the silicate framework, and no secondary minerals originate. Indeed, in the present Martian environment, analcime may exist only as a metastable mineral. However, in the Noachian age, during which an H 2 O/CO 2 -rich hydrosphere/atmosphere was present, the formation of analcime could have been promoted by a low-temperature (<350 • C) hydrothermal fluid alteration of a basalt glass, primarily related to an impact crater setting or to lava cooling [5,7,14,18,25,45,[84][85][86][87]. A modeled Martian impact inferred that hydrothermal systems give rise to circulating fluids and water vapor [88,89] at temperature ranges that meet the conditions used in our experiments. We inferred also that a hydrothermal solution with acidic pH interacted with basaltic glass during the Noachian period, in agreement with literature data [7,22,90,91]. In particular, Wang et al. [86] indicated that a possible acidic environment (i.e., pH 4-6) in the past would have promoted the formation of kaolinite-type phyllosilicates, while Peretyazhko et al. [91] experimentally demonstrated that the weathering of a chemically representative Martian basaltic glass, with water at pH 4 under oxic atmosphere, led to the formation of Fe, Mg-smectites. In our experiments, the formation of Fe, Mg-phyllosilicates was inhibited in favor of that of zeolites formation. The nature of the secondary phase at low W/R, as in these experiments, is mainly constrained by temperature and pH, which in turn affect the Si/alkali ratio and product solubility, for zeolite at a temperature of 150-200 • C and final basic pH, for Fe, Mg-phyllosilicates at higher temperatures and acid pH [45]. In this regard, our research could shed new light on defining whether a part of the aqueous solution during the early Martian alteration was mainly basic or acidic.
In this context, analcime can be recorded as a mineralogical marker that constrained the coupled aqueous-geological history of Mars. On Earth, hydrothermal environments containing primitive bacteria are thought to correspond to the expected location of the origin of life [92][93][94]. Similarly, the evidence of hydrothermal phenomena on Martian rocks could be an indication of the development of a biochemical system as a precursor of life.