Understanding Sulfate Stability on Mars: A Thermo-Raman Spectroscopy Study

1. Introduction

Evidence from martian missions indicates that the planet once had surface water, which likely formed saline brine pockets as Mars lost its atmosphere and became cold and dry (Mancinelli, 2005; Mancinelli et al., 2004). These brine pockets could have served as refuges for life before drying into evaporite deposits, potentially harboring halophilic organisms (Barbieri and Stivaletta, 2011).

It is known that evaporite deposits can trap and preserve biological records over geological timescales, which makes them excellent candidates for finding evidence of past life (Barbieri and Stivaletta, 2011; Fischer et al., 2016). The preservation of microbial fossils and organic matter in these deposits on Earth suggests that similar processes could have occurred on Mars (Barbieri and Stivaletta, 2011; Bosak et al., 2021; Huidobro et al., 2023b). Therefore, to understand the potential habitability of Mars, terrestrial evaporite environments are being analyzed as martian analogs (Barbieri, 2013; Mancinelli et al., 2004).

Both remote sensing instruments and in situ measurements from landed spacecraft have detected evaporite minerals on Mars (Bannari et al., 2023; Tosca and McLennan, 2006). This fact has led to the search for evidence of ancient life in association with evaporite minerals on the red planet. An evaporite identified several times in the history of Mars landing exploration is gypsum [CaSO₄.2H₂O] (Harrisson, 2019). This compound was first identified from orbit in 2005 through the OMEGA hyperspectral imager onboard the Mars Express mission (Fishbaugh et al., 2007).

Apart from gypsum, the Chemical Camera (ChemCam) and the Chemistry and Mineralogy (CheMin) instruments, onboard the Curiosity rover, were able to demonstrate that some fracture fills of Yellowknife Bay sedimentary deposits (Mars) consisted of other Ca-sulfates (Nachon et al., 2014). Depending on the hydrogen and oxygen content of the compounds, the laser-induced breakdown spectroscopy (LIBS) analyses can differentiate between the hydration levels of Ca-sulfates. This is because the H-alpha emission line of these compounds usually appears at 656.5 nm and that of oxygen at 777.1 nm (Rai et al., 2019). In the case of the Yellowknife Bay fracture fills, even though the spectral line of these elements was detected and their respective intensities were very low, the presence of bassanite [CaSO₄. ½H₂O] was detected (Nachon et al., 2014; Rapin et al., 2016). Likewise, LIBS analyses detected the most common Ca-sulfate lines though did not identify any emission that could be associated with hydrogen, which thus confirmed the presence of the anhydrous calcium sulfate, anhydrite [CaSO₄] (Huidobro et al., 2023a; Nachon et al., 2014). Hence, the three hydration states of calcium sulfate coexist on Mars.

Apart from these common compounds, the Alpha PArticle X-Ray Spectrometer (APXS), ChemCam, and CheMin instruments onboard the Curiosity rover detected calcium sulfate together with other cations (Yen et al., 2017). After analysis of the APXS and LIBS elemental results, the molar ratio of sulfate/calcium was found to be greater than one, which suggested the presence of calcium with other metal sulfates, or mixed-cation sulfates (Arvidson et al., 2014; Grotzinger et al., 2014; McLennan et al., 2005; Nachon et al., 2014; Rapin et al., 2016; Squyres et al., 2012; Yen et al., 2017). If mixed-cation sulfates could form under evaporitic (hydrothermal and non-hydrothermal) conditions, they could be expected to be found in Jezero Crater and in Oxia Planum, sites where liquid water was present. Previous works (Huidobro et al., 2023a) suggested the possible presence of syngenite [K₂Ca(SO₄)₂.H₂O] and görgeyite [K₂Ca₅(SO₄)₆.H₂O] on Mars, since both mineral phases contain higher SO₃% content compared with CaO%, and K-bearing compounds were identified both on Mars (Meslin et al., 2013; Sautter et al., 2015) and martian meteorites (García-Florentino et al., 2021).

Syngenite and görgeyite are often found in mineral formations on Earth. Syngenite has been observed by visible and near-infrared spectroscopy in the Death Valley salt pan (Crowley and Clark, 1992) together with gypsum in efflorescent salts on K-bearing sandstones (Alexandrowicz and Marszałek, 2019), in metamorphic evaporite deposits (Borchet and Muir, 1964), and in volcanic caves due to interactions with guano (Benedetto et al., 1998). Görgeyite has also been found in terrestrial settings, including a Triassic evaporite sequence in Western Greece, salt deposits in Germany, and lower middle Triassic polyhalite rocks in Nongle in Sichuan Province in China (García-Florentino et al., 2021).

It is well known that temperature is a critical factor influencing the kinetics and stability of secondary minerals. A primary cause of temperature fluctuations on Mars and other celestial bodies is the impact of external projectiles on planetary surfaces. Such impacts subject the target rocks to extreme pressures, temperatures, and strain rates and they are able to induce irreversible chemical, mineralogical, and physical alterations in these materials. The temperature range produced by such impacts is influenced by various factors that include the mass of the projectile, its velocity, and its composition (Gucsik, 2008).

When a meteoroid passes through Earth’s atmosphere, it heats the atmospheric particles, causing the surrounding gases to glow brightly. Consequently, most meteoroids disintegrate before reaching the ground (Koschny et al., 2019). Those that reach the surface have also been exposed to high temperatures, which induce chemical, mineralogical, and physical changes.

Therefore, considering that temperature affects the stability of thermosensitive compounds (El-Shobaky et al., 1996; Kloprogge et al., 1992; Ostroff and Sanderson, 1959), it is possible to monitor these temperature-induced changes in the crystalline structure of evaporites. In this regard, Raman spectroscopy emerges as one of the most suitable techniques due to its high sensitivity in detecting small variations in the crystalline structure of compounds. It can distinguish between polymorphs, such as the three anhydrite Ca-sulfates (anhydrite I, α-CaSO₄, cubic; anhydrite II, β-CaSO₄, orthorhombic; anhydrite III, ɣ-CaSO₄, hexagonal) (Prieto-Taboada et al., 2014). Raman spectroscopy can also be used to study the hydration-dehydration processes of hydrated evaporites, as it can monitor the loss of water molecules in the Raman spectral region from 2900 to 4000 cm⁻¹.

In light of these considerations, this work focuses on studying the effects of high temperatures on the crystalline structures of certain sulfates (specifically gypsum, syngenite, and görgeyite) that have been either detected or are expected to be present on Mars as well as on Earth’s surface and in the subsurface. Recognizing the influence of high temperatures on the stability of these three sulfates is essential for a thorough understanding of their crystal changes and corresponding Raman band shifts. Beyond their significant role in advancing the geochemical characterization of meteorites and martian analogs, these novel findings support data analysis of current and upcoming missions equipped with Raman spectrometers, which include the SuperCam and SHERLOC instruments onboard the Perseverance rover, the Raman Laser Spectrometer (RLS) onboard the Rosalind Franklin rover, and the Raman spectrometer onboard the Martian Moons eXploration mission to Phobos. Notably, SHERLOC is equipped with a 248.6 nm laser for organic detection, while the other three Raman spectrometers, like the one used in this work, use 532 nm lasers for mineralogical analysis. The RLS is unique in that, unlike the others that perform remote surface analysis, it will analyze pre-ground samples extracted from depths of up to 2 m within the rover itself (Huidobro et al., 2022).

Ultimately, this research will deepen our understanding of sulfate stability in martian and terrestrial environments, enrich the science of planetary exploration, and provide a basis for interpreting spectral data from Mars with unprecedented clarity.

2. Materials and Methods

3. Results

Table 1 is a literature review that summarizes all of the Raman bands of the expected compounds, which include the main sulfates, such as gypsum, syngenite, and görgeyite, as well as secondary phases, such as bassanite, anhydrite I and II, anhydrite III, arcanite [K₂SO₄], and calcium-langbeinite [K₂Ca₂(SO₄)₃].

After evaluating the temperature versus Raman shift plots, the assigned thermosensitive Raman bands of gypsum were identified as 1008, 3405, and 3491 cm⁻¹. The thermossensitive Raman bands of syngenite were located at 1006 cm⁻¹ and that of görgeyite at 1004 cm⁻¹.

In the case of gypsum, Fig. 1 shows the evolution of the main Raman band (∼1008 cm⁻¹) as a function of the temperature. The ∼1008 cm⁻¹ Raman band shifts to lower wavenumbers when temperature increases, following the trend observed in Eq. 1.Band Position ( ∼ 1 00 8 cm − 1 ) = − 0.00 586 · T ( K ) + 1 00 9 . 33 R 2 = 0. 98(1)

OPEN IN VIEWER

FIG. 1.

Linear regressions for the band position of the gypsum main Raman band from 313 to 673 K. (A) Shows that the gypsum dehydration to anhydrite occurred at 453 K. (B) Shows how the main Raman band of anhydrite III shifted toward lower wavenumber when increasing temperature. (C) Shows how the main Raman band of gypsum shifted toward lower wavenumbers when increasing temperature up to 453 K. At this point, the transformation to anhydrite started to occur (not shown). Moreover, the linear regression parameters were calculated to estimate the quality of the results. M is the slope, σ _m the uncertainty of the slope, b the intercept of the line, σ _b the uncertainty of the intercept, R ² the coefficient of determination (0–1), S _x/y is the standard error of the regression, and n the number of points used in the calculation.

However, this trend occurred up to the measured step-temperature value of 433 K. At 453 K (T _i), gypsum began to lose its hydration water molecules to become anhydrite III, with its main Raman band appearing at 1025 cm⁻¹ (Prieto-Taboada et al., 2014). Following this methodology, we determined that the transformation of gypsum into anhydrite occurred between 453 and 473 K. Above this temperature, the main Raman band of anhydrite III also underwent a shift toward lower wavenumber as the temperature increased, following Eq. 2, as seen in Fig. 1.Band Position ( ∼ 1 0 25 cm − 1 ) = − 0.0 1448 · T ( K ) + 1 0 32 . 55 R 2 = 0. 98(2)

It should be emphasized that during the dehydration process, there is an intermediate phase between gypsum and anhydrite III, known as bassanite. This phase was not observed in these experiments (Chio et al., 2004; Prasad et al., 2001) because the temperature ramps used were steep and the plateau times were short, simulating the shock event of a meteoroid impact. According to the work of Schmidt et al. (Schmid et al., 2020), slow ramps and very long plateaus favor dehydration. For example, as shown by Schmid in Fig. 5 (Schmid et al., 2020), at 383 K with a 4 h plateau, gypsum and bassanite coexist, but with a 20 h plateau, anhydrite III also coexists with gypsum and bassanite. At 10 degrees higher (393 K) with a 4 h plateau, gypsum and bassanite still coexist, but with a 20 h plateau only anhydrite III is present in the sample. Therefore, it is advisable to use these equations when the dehydration process is believed to be similar, as it has been found that the results can vary depending on the methodology used.

By monitoring the hydration –OH bands and following this methodology, the T _i of gypsum at 453 K was also proven. Supplementary Fig. S1 in Supplementary Material shows how the first and second –OH bands disappeared at this temperature due to anhydrite formation. On the one hand, the trend observed for the first –OH gypsum band (3401 cm⁻¹) consisted of a shift toward lower wavenumber when temperature increased. Nevertheless, the R ² remained below 0.80, so it could be said that there is not a good correlation between this band shift and temperature. On the other hand, the trend observed for the second –OH gypsum band (3494 cm⁻¹) consisted of the shift toward higher wavenumbers when temperature increased, following Eq. 3. This band shift was proved to be highly related with temperature according to the obtained R ² of 0.99.Band Position ( ∼ 3494 cm − 1 ) = 0.0 42 0 · T ( K ) + 3481 . 1 0 R 2 = 0. 99(3)

As occurred in the low-temperature tests (Huidobro et al., 2023a), the only temperature-sensitive syngenite band was the one that appeared at ∼1005 cm⁻¹. According to Eq. 4, band shifts toward lower wavenumbers were observed with increasing temperature.Band position ( ∼ 1 00 5 cm − 1 ) = − 0.0 1656 · T ( K ) + 1 0 1 0. 11 R 2 = 0. 99(4)

However, as can be seen in Fig. 2, Eq. 4 worked only up to its T _i at 633 K, since at this temperature the syngenite was totally transformed into anhydrite I/II with its main Raman band at 1017 cm⁻¹. This hypothesis was reinforced via analysis of the syngenite –OH band (∼3309 cm⁻¹), which, although it did not show any trend with temperature, also disappeared at 633 K.

OPEN IN VIEWER

FIG. 2.

Linear regressions for the ∼1006 cm⁻¹ syngenite band and for the ∼1004 cm⁻¹ görgeyite band from 313 to 673 K. (A) Shows how syngenite transformed into anhydrite, calcium-langbeinite, and arcanite at 633 K. (B) Shows the linear regression parameters for the syngenite main band shift (∼1006 cm⁻¹). (C) Shows that at 593 K görgeyite degraded; and (D) shows the linear regression parameters for the main görgeyite Raman band (∼1004 cm⁻¹). M is the slope, σ _m the uncertainty of the slope, b the intercept of the line, σ _b the uncertainty of the intercept, R ² the coefficient of determination (0–1), S _x/y is the standard error of the regression, and n the number of points used in the calculation.

In addition to the formation of anhydrite, at 633 K two possible anhydrous sulfates could be formed: the potassium sulfate arcanite [K₂SO₄] (Cheng et al., 2013) and the mixed calcium and potassium sulfate langbeinite [K₂Ca₂(SO₄)₃] (Morey et al., 1964), whose main Raman bands appear at 985 and 994 cm⁻¹, respectively (Cheng et al., 2019; Madariaga et al., 2020; Pekov et al., 2022). Although the Raman bands observed above 633 K are consistent with those of these two anhydrous sulfates, the lack of secondary Raman bands makes this characterization incomplete and inconclusive.

Figure 3 shows the Raman spectrum of pure syngenite at 313 K; the spectrum of heated syngenite at 633 K, the temperature at which anhydrite I/II was detected; and the spectrum of heated syngenite at 653 K, where anhydrite I/II and the potential anhydrous potassium sulfates were detected.

OPEN IN VIEWER

FIG. 3.

Raman spectra of pure syngenite at 313 K (A); heated syngenite at 633 K (B), where anhydrite I/II starts to form (979 and 999 cm⁻¹ Raman bands correspond to heated syngenite and 1017 cm⁻¹ to anhydrite); and heated syngenite at 653 K (C), where anhydrite I/II (1018 cm⁻¹), and the potential anhydrous sulfates [arcanite (987 cm⁻¹) and calcium-langbeinite (978 cm⁻¹)] appear. S = syngenite; A = anhydrite.

With regard to the behavior of görgeyite upon heating, only the main Raman band (∼1007 cm⁻¹) underwent changes with temperature (see Fig. 2). Once again, this linear regression, as in the previous cases, was based on the band shift toward lower wavenumbers when temperature increased. However, this trend, which can be seen in Eq. 5, was followed only up to 593 K. At this temperature, the main band adopted unusual positions that could not be assigned to any specific phase.Band position ( ∼ 1 00 7 cm − 1 ) = − 0.0 2113 · T ( K ) + 1 0 11 . 38 R 2 = 0. 98(5)

As explained above, reversibility tests were conducted to confirm the potential rehydration of the compounds upon returning to RT, particularly when the sample stabilized under terrestrial conditions.

Figure 4 illustrates how, after heating the gypsum to its T _i (453 K), dehydration to anhydrite III occurred. This fact is evidenced by the band shift from 1007.6 cm⁻¹ at RT (purple point) to 1022.5 cm⁻¹ at 453 K (green point).

OPEN IN VIEWER

FIG. 4.

Plot of the main Raman band position of calcium sulfate versus time after heating up to gypsum’s T _i. The purple point represents the gypsum band at RT; the green point shows the transition from gypsum to anhydrite III during the heating at 453 K; and the orange points depict the positions of the Raman band after heating gypsum to its T _i and during different periods of cooling. G = gypsum; A III = anhydrite III; B = bassanite; RT = room temperature.

The orange points of Fig. 4 correspond to the wavenumbers of the main band after heating to 463 K and after waiting during different periods at RT. The first orange point represents the Raman measurement taken after the initial waiting period of 15 min, by which time RT had already been reached. After this time, calcium sulfate begins to rehydrate under ambient terrestrial environmental conditions, transforming into the hemihydrate calcium sulfate bassanite, whose main Raman spectral band appeared in this study at 1012 cm⁻¹. Considering that the RT band position of bassanite is 1015 cm⁻¹, it should be noted that the bands that appeared at 1012 cm⁻¹ correspond to a metastable bassanite affected by temperature.

After periods of 1 and 3 weeks, the stabilized bassanite (1015 cm⁻¹) coexisted with gypsum (1008 cm⁻¹). This correlation was lost after 4 weeks, since the complete rehydration to gypsum occurred as evidence by the detection of a single band at 1008 cm⁻¹.

To estimate the approximate ratio between bassanite and gypsum after 1 and 3 weeks, a deconvolution of the thermosensitive Raman band was conducted on the average spectrum (obtained from the average of 20 spectra) using a 50% Lorentzian-Gaussian curve. The spectrum obtained after 1 week indicated the sample material analyzed (based on the respective areas measured under the spectral curves) consisted of45.91 ± 0.12% gypsum and 54.09 ± 0.12% bassanite. The spectrum obtained after 3 weeks showed a slight increase in gypsum content to 52.21 ± 0.12%, while the bassanite content decreased to 47.79 ± 0.12%. This trend aligns with the fourth-week observation, where all salt content had fully transformed to gypsum, which indicated complete rehydration.

Figure 5 shows the deconvolution process conducted after 3 weeks, along with the ratio content of bassanite/gypsum, calculated from the areas measured under the curve.

OPEN IN VIEWER

FIG. 5.

Raman spectrum after 3 weeks of cooling (in red), where the band resulting from the combination of the main bands of bassanite and gypsum appear. A band deconvolution, based on 50% Lorentzian and Gaussian, revealed the combination of a gypsum band at 1008.9 cm⁻¹ with an area of 225,760, and a bassanite band at 1015.1 cm⁻¹ with an area of 206,645. Given that the total area of the broad band is 432,405, it can be estimated that the mass proportion of gypsum is 52.21 ± 0.12 % and that of bassanite is 47.79 ± 0.12 %.

These spectral interpretations were in agreement with our analysis of the position of the gypsum –OH hydration bands. In this regard, Table 2 shows the hydration bands at six different waiting times. Hydration Raman bands that appear at ∼3401 and ∼3494 cm⁻¹ belong to gypsum, so, according to Fig. 4, they appear before heating and after 1, 3, and 4 weeks of waiting. With regard to the bands that were detected at ∼3553 and ∼3614 cm⁻¹, they only occurred when bassanite was present (15 min, 30 min, 1 h, 1 week, and 3 weeks).

Regarding syngenite, and as mentioned above, after exceeding T _i (633 K), syngenite degraded into anhydrite I or II. However, after subjecting another portion of syngenite sample to only 20 degrees above its T _i to perform the reversibility tests, not all the syngenite transformed to anhydrite; this resulted in a mixture of anhydrite I/II and syngenite (see Table 3). It should be noted that the main Raman bands of syngenite (∼981 and ∼1005 cm⁻¹, Table 1) after heating were observed to be shifted toward lower wavenumber, which agrees with our observations discussed in the previous section: the Raman wavenumbers decrease as temperature rises.

However, at RT and after waiting 10 and 15 min post-heating, no narrow Raman bands were observed; instead, the bands appeared broad. This observation suggests no phase change occurred, and that the change in the spectral line characteristics was due to a loss of crystallinity.

After 1 h, the resulting compound recovered its crystallinity and showed a single band at 994 cm⁻¹, which could correspond to the mixed anhydrous calcium and potassium sulfate: calcium-langbeinite. The absence of other ʋ₁ bands of calcium-langbeinite, such as those at 1011 or 1021 cm⁻¹, prevents a conclusive identification and characterization of this phase.

After 2 h, the 994 cm⁻¹ Raman band was still observed with another small band at 985 cm⁻¹, which may be attributed to the anhydrous potassium sulfate: arcanite. However, as in the previous assignment, the absence of secondary bands and the lack of a good sharpness of the main band make a definitive diagnosis impossible. This observation, together with the lack of hydration evidenced by Raman bands, is consistent with a lack of rehydration.

After 1 week of waiting, the Raman bands of syngenite were clearly identified, along with the two additional bands at 985 and 994 cm⁻¹. As previously mentioned, these bands could potentially correspond to arcanite and calcium-langbeinite; however, this attribution remains uncertain due to the lack of secondary bands. According to Table 3, rehydration started after 1 week and was completed after 3 weeks with the appearance of the syngenite hydration bands. The results obtained for 4 weeks were the same as those for 3 weeks; they revealed total rehydration.

Regarding görgeyite, after it was heated 20 degrees over its T _i (593 K), the position of the temperature-sensitive Raman bands (∼1007 and ∼1013 cm⁻¹) decreased. However, at RT and after 15 min time, these bands returned to their regular position, which demonstrated complete reversibility. In addition, as can be seen in Table 4, after 15 min at RT another band at 994 cm⁻¹ appeared. As stated before, it could belong to calcium-langbeinite, though the lack of secondary bands makes its identification impossible. The coexistence of both compounds remained over time up to 4 weeks. At this point, the entire compound turned entirely into the hydrated form as görgeyite.

The results showed that the temperature-sensitive Raman bands systematically shifted toward lower wavenumbers as temperature increased. This is because when temperature increases, the bonds of the molecules are expanded; consequently, the strength in the bond decreases, which causes a decrease in energy (Huidobro et al., 2023a).

In the case of gypsum, it was observed that the increase in temperature caused shifts in the main Raman band toward lower wavenumbers up to 453 K. At this temperature, the hydration water molecules of gypsum evaporated, leaving the anhydrous calcium sulfate anhydrite.

As stated in the introductory section, anhydrite can be present as three different polymorphous species with different structures (anhydrite I, II, and/or III), and Raman spectroscopy can distinguish between these compounds. Although anhydrite III was identified through its main Raman band at 1024 cm⁻¹, it is the least stable anhydrite polymorph (Prieto-Taboada et al., 2014), which was demonstrated through reversibility tests. These tests demonstrated the rapid rehydration of anhydrous calcium sulfate into bassanite (hemihydrate calcium sulfate) upon heating. During the waiting period, it was noticed that atmospheric rehydration persisted until reaching equilibrium between bassanite and gypsum. This outcome was expected, as the coexistence of these three minerals (gypsum, bassanite, and anhydrite) has been observed in nature (Vaniman et al., 2018).

Similar to gypsum, the main band of syngenite exhibited a displacement toward lower wavenumbers as the temperature rose. The resulting linear regression between Raman shift and temperature showed excellent quality, boasting an R ² value of 0.99. Nevertheless, such linear regression can only be followed up to its T _i of 633 K, the temperature at which syngenite begins to undergo a transformation into stable anhydrite (I or II). After heating up to its T _i, it was found that the resulting compounds were not perfectly crystallized, which resulted in the absence of well-defined Raman signals until 1 h of waiting. These experiments show that complete rehydration of syngenite under terrestrial conditions occurred after 3 weeks of waiting. In the meantime, other anhydrous mineral phases may occur, but their definite characterization was impossible due to the lack of secondary Raman bands in the experimental spectra.

As in previous cases, the position of the main Raman band of görgeyite decreased as the temperature increased. The linear regression obtained was of excellent quality (R ² = 0.98) and was monitored up to a T _i of 593 K. Reversibility tests demonstrated that after just 15 min of waiting, all the bands of görgeyite, including the hydration bands, were observed together with another band that could correspond to an intermediate mineral phase. The coexistence of both compounds persisted for up to 3 weeks, with complete rehydration occurring after 4 weeks.

Understanding the reversibility of anhydrous to hydrated sulfates under regular terrestrial conditions is crucial as an initial approximation of potential processes on Mars. This knowledge enables the design of experiments to explore possible hydration mechanisms that could occur under martian conditions. While martian surface conditions differ significantly from those on Earth, insights into rehydration under terrestrial conditions help establish baseline reactions that may still take place on Mars.

5. Conclusions

Secondary minerals, formed from the alteration of primary minerals, can undergo further alteration to form new secondary minerals. Temperature, pressure, and strain rates are crucial in these processes, which often result from meteoroid impact shock events on both Earth and Mars.

Taking this into account, studying the relationship between secondary minerals and temperature is crucial for understanding the mineralogical alterations caused by shock events. In this sense, this work focuses on the Raman shift and mineralogical transformation of secondary minerals such as gypsum, syngenite, and görgeyite, found on Earth and expected on Mars. To do so, temperature ramps that simulated shock events from 313 to 675 K were programmed to monitor the Raman transition of these minerals. This is why it should be noted that these results are related to the temperature ramps, the plateau times, the analytical monitoring technique, and the nature of the samples, among other factors, which differ from other studies.

By plotting the wavenumber of the temperature-sensitive Raman bands versus temperature, high-quality linear regressions (R ² >0.98) were achieved. These regressions will transform the way Raman spectra are interpreted in both Mars and Earth environments by incorporating temperature as a determining factor, a consideration previously overlooked for syngenite and görgeyite.

Therefore, the regressions presented in this work offer valuable insight into the thermal behavior of Raman spectra and open new ways for temperature identification and band prediction in various technological and scientific applications. This knowledge will improve our understanding of material properties and facilitate more accurate and efficient Raman analysis as a function of temperature by estimating the Raman band or the temperature at which the measurements were conducted.

In addition, the reversibility tests will shed light on a previously unresolved issue in both space and terrestrial fields. This work demonstrated that complete rehydration of heated gypsum, syngenite, and görgeyite (above their own T _i) was achieved after 1 month at RT, under a relative humidity of 80% or higher.